Maturation of Fibrinolytic Bacillopeptidase F Involves both Hetero-and Autocatalytic Processes

Dongheng Meng,a,c Meihong Dai,a Bi-Lin Xu,a Zhong-Shu Zhao,a Xiaoliang Liang,a Mingqiu Wang,a Xiao-Feng Tang,a,b Bing Tanga,b

State Key Laboratory of Virology, College of Life Sciences, Wuhan University, Wuhan, Chinaa; Hubei Provincial Cooperative Innovation Center of Industrial Fermentation,Wuhan, Chinab; College of Life Science and Technology, Xinjiang University, Urumuqi, Chinac

Bacillopeptidase F (Bpr) is a fibrinolytic serine protease produced by Bacillus subtilis. Its precursor is composed of a signal pep-tide, an N-terminal propeptide, a catalytic domain, and a long C-terminal extension (CTE). Several active forms of Bpr have beenpreviously reported, but little is known about the maturation of this enzyme. Here, a gene encoding a Bpr (BprL) was clonedfrom B. subtilis LZW and expressed in B. subtilis WB700, and three fibrinolytic mature forms with apparent molecular massesof 45, 75, and 85 kDa were identified in the culture supernatant. After treatment with urea, the 75-kDa mature form had the samemolecular mass as the 85-kDa mature form, from which we infer that they adopt different conformations. Mutational analysisrevealed that while the 85-kDa mature form is generated via heterocatalytic processing of a BprL proform by an unidentifiedprotease of B. subtilis, the production of the 75- and 45-kDa mature forms involves both hetero- and autocatalytic events. Fromin vitro analysis of BprL and its sequential C-terminal truncation variants, it appears that partial removal of the CTE is requiredfor the initiation of autoprocessing of the N-terminal propeptide, which is composed of a core domain (N*) and a 15-residuelinker peptide, thereby yielding the 45-kDa mature form. These data suggest that the differential processing of BprL, either het-erocatalytically or autocatalytically, leads to the formation of multiple mature forms with different molecular masses orconformations.

Bacillus species produce many extracellular proteases at the endof the log phase of growth (1, 2). One of the extracellular

proteases produced by B. subtilis, bacillopeptidase F (Bpr), is en-coded by bpr (3), bpf (4), or hspK (5), and its expression is depen-dent on the protein DegU-P (6). Bpr from B. subtilis 168 isrequired for the processing of the glutamic acid-specific endopep-tidase Mpr, encoded by mpr (7). However, inactivation of bpfshows that Bpr is not essential for either growth or sporulation ofB. subtilis 168 (4). Bpr is the main component of NKCP, which isa purified protein layer that is produced during B. subtilis nattofermentation and possesses fibrinolytic and antithrombotic activ-ities (8). Fibrinolytic activity has also been reported for Bprs fromB. amyloliquefaciens (9) and B. licheniformis (10, 11).

Bprs belong to the superfamily of subtilisin-like serine pro-teases (subtilases) (12). They are synthesized as large precursors(�1,430 amino acid residues) consisting of a signal peptide, anN-terminal propeptide, a catalytic domain that shares approxi-mately 30% sequence identity to subtilisins, and a long C-terminalextension (CTE) of approximately 900 amino acid residues (3–5,9). The signal peptide of a subtilase precursor mediates the en-zyme’s secretion across the cytoplasmic membrane and is cleavedby a signal peptidase to generate the proform. Subsequently, theN-terminal propeptide, which usually functions as an intramolec-ular chaperone to facilitate the correct folding of the mature do-main and as a potential inhibitor of the mature enzyme, is auto-processed from the proform to release the active mature enzyme(13). In the case of Bpr, several active forms with apparent molec-ular masses ranging from 33 kDa to 90 kDa have been identified(3–5, 8, 9, 14–16). The multiple active forms of Bpr may resultfrom the processing of the N-terminal propeptide and the trun-cation of the CTE (4, 5), but the molecular basis for Bpr matura-tion is unclear. In addition, the CTE of Bpr is not essential forenzyme activity or secretion (4), and its function is unknown.

Recently, we have isolated a fibrinolytic B. subtilis strain, LZW,

from soil. In this study, a gene encoding a bacillopeptidase F(bprL) was cloned from this strain and expressed in B. subtilisWB700 and Escherichia coli to investigate the in vivo and in vitroprocessing mechanisms of this enzyme. We conclude that thematuration of BprL involves both hetero- and autocatalytic pro-cesses and that either the N-terminal propeptide or the CTE isprocessed in a stepwise manner to yield several active matureforms with different molecular masses and conformations. In ad-dition, we found that the CTE plays an important role in main-taining BprL in an inactive state. Deletion analysis revealed thatpartial removal of the CTE is required for the initiation of auto-processing of the enzyme to generate the 45-kDa mature form.

MATERIALS AND METHODSStrains and growth conditions. B. subtilis LZW was isolated from soil atWuhan University, China, and has been deposited in the China Center forType Culture Collection (CCTCC) under accession number CCTCCAB2015270. B. subtilis WB700 (trpC2 nprE aprE epr bpr �mpr::ble �nprB::bsr �vpr::ery), a strain deficient in seven extracellular proteases (17), wasused as the host for expression. E. coli DH5� and E. coli BL21(DE3) wereused as the hosts for cloning and expression, respectively. Bacteria were

Received 19 August 2015 Accepted 19 October 2015

Accepted manuscript posted online 23 October 2015

Citation Meng D, Dai M, Xu B-L, Zhao Z-S, Liang X, Wang M, Tang X-F, Tang B.2016. Maturation of fibrinolytic bacillopeptidase F involves both hetero- andautocatalytic processes. Appl Environ Microbiol 82:318 –327.doi:10.1128/AEM.02673-15.

Editor: R. M. Kelly

Address correspondence to Bing Tang, tangb@whu.edu.cn.

Supplemental material for this article may be found at http://dx.doi.org/10.1128/AEM.02673-15.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

crossmark

318 aem.asm.org January 2016 Volume 82 Number 1Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

grown in 50 ml of Luria-Bertani (LB) medium in a 250-ml flask at 37°Cwith shaking (180 rpm). Where necessary, the medium was supplementedwith 100 �g/ml ampicillin or 30 �g/ml kanamycin.

Fibrin plate assay. Fibrin plates were prepared as described by Kim etal. (18) with modifications. Briefly, 2.5 ml of bovine fibrinogen (Sigma, St.Louis, MO, USA) solution (2.2 mg/ml in 10 mM sodium phosphate buffer[pH 7.8]; sterilized by filtration) was mixed with 10 ml of sterilized LBmedium containing 1.5% agar (all percentages are wt/vol unless otherwiseindicated) that was warmed in a 45°C water bath. After the addition of 0.5ml of bovine thrombin (Sigma) solution (20 NIH U/ml in 10 mM sodiumphosphate buffer [pH 7.8]; sterilized by filtration), the mixture waspoured into a petri dish and kept at room temperature for 1 h to formfibrin clots. Thereafter, bacterial strains were inoculated on the fibrinplates and incubated at 37°C for 24 h. Colonies with surrounding clearzones were classified as fibrinolytic strains.

DNA manipulation, plasmid construction, and mutagenesis. Thegenomic DNA of B. subtilis LZW and plasmids were prepared by standardprotocols (19) and used as templates for PCR. The oligonucleotide se-quences of the primers used in this study are listed in Table 1. The geneencoding the BprL precursor was amplified from B. subtilis LZW genomicDNA with LA-Taq DNA polymerase (TaKaRa, Otsu, Japan) by PCR usingprimers P-F and P-R (Table 1), which were designed based on the se-quence of the bpr gene of B. subtilis 168 (3, 4). The amplified DNA frag-ment was then ligated into the pMD18-T vector (TaKaRa) to constructthe plasmid pMD18-bprL (Table 2) for sequencing.

The E. coli-B. subtilis shuttle plasmid pBE2 (20) was used as the vectorfor the production of BprL in B. subtilis WB700. The DNA fragmentcontaining the coding region of the BprL precursor (Fig. 1) and its ownpromoter (see Fig. S1 in the supplemental material) was amplified fromthe plasmid pMD18-bprL by PCR with Pfu DNA polymerase (Fermentas,Burlington, Canada) using primers B-F and B-R1 (Table 1) and insertedinto the PstI-BamHI restriction site of pBE2 to construct the expressionplasmid pBE2-bprL for BprL (Table 2). The expression plasmid pBE2-bprL�C912 for the CTE deletion variant BprL�C912 (Table 2) was alsoconstructed as described above by cloning the PCR product obtained withprimers B-F and B-R2 (Table 1).

To produce recombinant proteins in E. coli BL21(DE3) and to facili-tate the DNA manipulation, plasmid pETx was constructed based on theplasmids pET26b and pET28a (Novagen). Briefly, the 85-bp DNA frag-ment, which contained a NheI restriction site between the NdeI and XhoIsites of pET28a, was amplified by PCR using primers U-F and U-R (Table1) and then ligated into the NdeI-XhoI restriction site of pET26b, therebyyielding the plasmid pETx (Table 2). The genes encoding the proforms of

BprL and its sequential C-terminal truncation variants (BprL�C511,BprL�C678, and BprL�C912), as well as the CTE fragment (BprLC912)(Fig. 1) were amplified from pMD18-bprL by PCR using primer pairslisted in Table 2. The amplified genes were then inserted into the NheI-XhoI restriction site of plasmid pETx to generate expression plasmids fortarget proteins (Table 2).

The QuikChange site-directed mutagenesis method (21) was em-ployed to construct expression plasmids for active-site variants of BprL(BprLS258A), BprL�C678 (BprL�C678-S258A and BprL�C678-S258C),and BprL�C912 (BprL�C912-S258A and BprL�C912-S258C) usingprimer pairs and templates listed in Table 2. The sequences of all recom-binant plasmids were confirmed by DNA sequencing.

Expression, purification, and in vitro processing. B. subtilis WB700cells harboring expression plasmids for BprL and its variants were grownat 37°C with constant shaking (180 rpm) in LB medium containing 30�g/ml of kanamycin. When the culture was in the late log phase (24 h), itwas centrifuged at 10,000 � g for 10 min at 4°C. The supernatant wascollected and used for sodium dodecyl sulfate-polyacrylamide gel electro-phoresis (SDS-PAGE), immunoblot analyses, and a fibrin overlay assay.

E. coli BL21(DE3) cells harboring expression plasmids for BprL and its

TABLE 1 Oligonucleotide primers used in this study

Primer Nucleotide sequencea

P-F 5=-ATCCCGACATTCCTAAGAAACCGTA-3=P-R 5=-TGCTTCATAAGTATAAAGCAAGCTC-3=B-F 5=-TTGCACTACTGCAGGTCTTTTCCCTATTGCTTCCTTCGTTTAG-3=B-R1 5=-ACTGTCCGGGATCCTTAGTGGTGGTGGTGGTGGTGATTTTCTGTGTTGATATTAAG-3=B-R2 5=-ACTGTCCGGGATCCTTAGTGGTGGTGGTGGTGGTGTTTCCCTAAACCATCTGTAACAGC-3=U-F 5=-CGCGGCAGCCATATGGCTAG-3=U-R 5=-GTGGTGGTGCTCGAGTGCGG-3=E-F1 5=-TTGCACTAGCTAGCAGCAGTAAAGTCACCTCACCTTC-3=E-F2 5=-TTGCACTAGCTAGCGCGGAAGGACAAGTTTCTGTAGT-3=E-R1 5=-ACTGTCCGCTCGAGTTTATCCTCTCCCGGAATTAAGCCG-3=E-R4 5=-ACTGTCCGCTCGAGCTCTCCTGTCGTTTTATTGATGACCG-3=E-R5 5=-ACTGTCCGCTCGAGTAAATCAATTTCTTCGTCCGTCC-3=E-R6 5=-ACTGTCCGCTCGAGTTTCCCTAAACCATCTGTAACAGCG-3=SA-F 5=-GGACGGCACAGCAATGGCAGGGCCGCATGTATC-3=SA-R 5=-CCCTGCCATTGCTGTGCCGTCCCATCCATCCTC-3=SC-F 5=-GGACGGCACATGCATGGCAGGGCCGCATGTATC-3=SC-R 5=-CCCTGCCATGCATGTGCCGTCCCATCCATCCTC-3=a Underlining indicates the restriction enzyme sites used in this study. The mutated nucleotides are in bold. Italic indicates the His6 tag-encoding DNA sequences.

TABLE 2 Primer pairs, templates, and restriction enzyme sites used inplasmid construction

PlasmidPrimerpair Template

Restrictionsite

pMD18-bprL P-F/P-R Genomic DNApBE2-bprL B-F/B-R1 pMD18-bprL PstI-BamHIpBE2-bprL�C912 B-F/B-R2 pMD18-bprL PstI-BamHIpBE2-bprLS258A SA-F/SA-R pBE2-bprLpETx U-F/U-R pET28a NdeI-XhoIpETx-bprL E-F1/E-R1 pMD18-bprL NheI-XhoIpETx-bprL�C511 E-F1/E-R4 pMD18-bprL NheI-XhoIpETx-bprL�C678 E-F1/E-R5 pMD18-bprL NheI-XhoIpETx-bprL�C912 E-F1/E-R6 pMD18-bprL NheI-XhoIpETx-bprLC912 E-F2/E-R1 pMD18-bprL NheI-XhoIpETx-bprLS258A SA-F/SA-R pETx-bprLpETx-bprL�C678-S258A SA-F/SA-R pETx-bprL�C678pETx-bprL�C678-S258C SC-F/SC-R pETx-bprL�C678pETx-bprL�C912-S258A SA-F/SA-R pETx-bprL�C912pETx-bprL�C912-S258C SC-F/SC-R pETx-bprL�C912

Maturation of Bacillopeptidase F

January 2016 Volume 82 Number 1 aem.asm.org 319Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

variants were grown at 37°C until the optical density at 600 nm reachedapproximately 0.6. Production of recombinant protein was induced bythe addition of 0.4 mM isopropyl-�-D-thiogalactopyranoside, and thecultivation was continued at 37°C for 4 h. The cells were harvested, sus-pended in 50 mM sodium phosphate buffer (pH 7.5) containing 0.5 mMNaCl (referred to here as buffer A), and disrupted by sonication on ice.After centrifugation at 13,000 � g for 10 min, the insoluble fraction wascollected and solubilized in buffer A containing 8 M urea (urea buffer).The solubilized fraction was centrifuged at 13,000 � g for 10 min, afterwhich the supernatant was collected and subjected to affinity chromatog-raphy on an Ni2�-charged chelating Sepharose Fast Flow resin (GEHealthcare) column equilibrated with urea buffer. The column waswashed with 40 mM imidazole in urea buffer. Bound proteins were elutedwith 100 mM imidazole in urea buffer. The elution fraction was thendialyzed overnight (4°C) against buffer A to remove the imidazole andurea and allow the denatured protein to refold. Thereafter, the sample wasincubated at 30°C to allow for enzyme processing. During the incubation,aliquots of the sample were subjected to SDS-PAGE analysis and azoca-seinolytic activity assay.

Protein concentrations were determined using the Bradford method(22). The amounts of target proteins in the samples containing degrada-tion products or contaminating proteins were further assessed by bandintensity measurement following SDS-PAGE, using bovine serum albu-min as a standard. The enzyme solution was concentrated with a MicronTM-3 centrifugal filter (Millipore, Bedford, MA, USA) as needed.

Enzyme activity assay. Azocaseinolytic activity of the enzyme wasdetermined at 37°C for 1 h in 400 �l of reaction mixture, which con-sisted of 200 �l of enzyme sample and 0.25% azocasein in buffer A. Thereaction was terminated by the addition of 400 �l of 40% trichloro-acetic acid (TCA) to the reaction mixture. After incubation at roomtemperature for 15 min, the mixture was centrifuged at 13,400 � g for10 min, and the absorbance of the supernatant was measured at 335nm (A335) in a 1-cm cell. One unit of activity was defined as theamount of enzyme required to increase the A335 by 0.01 unit/minunder the conditions described above.

SDS-PAGE and fibrin overlay assay. SDS-PAGE was carried outwith the glycine-Tris system (23). To prevent self-degradation of theprotease during sample preparation (boiling) or electrophoresis, thesamples for conventional SDS-PAGE analysis were precipitated with20% TCA. After incubation at room temperature for 15 min, the pre-cipitated proteins were recovered by centrifugation at 13,400 � g for10 min and resuspended in ice-cold acetone, and the proteins were

recovered by centrifugation at 13,400 � g for 10 min. The proteinswere then air dried, boiled in the loading buffer unless otherwise in-dicated, and subjected to SDS-PAGE. After electrophoresis, the gel wasstained with Coomassie brilliant blue G-250. Where necessary, target-stained protein bands were excised from the gel, and the gel slices wereincubated at 80°C for 20 min in urea buffer to extract proteins. Theextracted proteins were precipitated with 20% TCA and then subjectedto SDS-PAGE after boiling in the loading buffer containing 8 M urea toinvestigate the influence of urea treatment on electrophoretic mobilityof the protein in SDS-PAGE.

For activity staining of fibrinolytic enzyme, samples without TCAtreatment were mixed with 5� loading buffer and subjected to SDS-PAGE without boiling. Meanwhile, a fibrin-polyacrylamide gel was pre-pared as described previously (24). The SDS-polyacrylamide gel and thefibrin-polyacrylamide gel were subjected to a fibrin overlay assay using theprocedure described by Blumentals et al. (25), except that the fibrinolyticreaction was carried out at 37°C for 15 h in buffer A. To determine themolecular masses of the fibrinolytic enzymes, protein bands on a replicateSDS-polyacrylamide gel were visualized by staining with 0.25 M KCl asdescribed previously (26), and the target bands corresponding to the fi-brinolytic bands, as determined by fibrin overlay assay, were excised fromthe gel. The proteins were extracted from the gel slices with urea buffercontaining 10 mM phenylmethylsulfonyl fluoride (PMSF) (Sigma), pre-cipitated with 20% TCA, and subjected to immunoblot analysis after boil-ing in loading buffer containing 8 M urea.

Immunoblot analysis. BprL�C912 comprises the N-terminal pro-peptide and catalytic domain, and BprLC912 comprises the CTE fragmentof BprL. They were produced in E. coli BL21(DE3) and served as antigensfor the generation of antibodies in rabbits. Antisera against BprL�C912(anti-NM) and BprLC912 (anti-C) were prepared and used for immuno-blot analysis as described previously (27). In some cases, the His tagmonoclonal antibody (Novagen) was used for immunoblot analysis asdescribed previously (27).

N-terminal sequencing. After separation by SDS-PAGE, the proteinswere electroblotted onto a polyvinylidene difluoride membrane andstained with Coomassie brilliant blue R-250. The target protein bandswere excised and subjected to N-terminal amino acid sequence analysisusing a PPSQ-33A protein sequencer (Shimadzu, Kyoto, Japan).

Nucleotide sequence accession number. The sequence of the regioncoding for BprL, including the open reading frame (ORF) and its flankingsequences, has been deposited in GenBank under accession numberKT259045.

FIG 1 Schematic representation of the primary structure of the recombinant BprL precursor. The signal peptide (S), the N-terminal propeptide (N), the catalyticdomain (CD), the C-terminal extension (CTE), and the fused His tag (H) are indicated. The locations of the active-site residues, D33, H80, and S258, are shown.Double-headed arrows represent the regions of the recombinant variants of BprL analyzed in this study. A partial sequence of the core domain of the N-terminalpropeptide (N*), the amino acid composition of the linker peptide, and partial N- and C-terminal sequences of CD are shown at the top. The first four identifiedresidues of the processed products (85 and 45 kDa) of BprL produced in B. subtilis WB700 and those coming from in vitro processing of BprL �C912 (47 and 45kDa) are underlined. The vertical open arrow indicates the possible heterocatalytic cleavage site that leads to partial removal of the CTE and triggers subsequentautoprocessing of the enzyme into the 45-kDa mature form.

Meng et al.

320 aem.asm.org January 2016 Volume 82 Number 1Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

RESULTSCloning and sequencing of the gene encoding BprL from Bacil-lus subtilis LZW. Using the primers P-F and P-R (Table 1), a4,793-bp DNA fragment was amplified from the genome of B.subtilis LZW. DNA sequencing of this fragment revealed that theORF encoding the BprL with 1,433 amino acid residues (Fig. 1) ispreceded by a DegU-dependent promoter (6) and is followed by aputative transcription termination site (see Fig. S1 in the supple-mental material). A comparison with other known Bprs suggeststhat the BprL precursor is composed of a signal peptide of 30residues, an N-terminal propeptide of 164 residues, a catalyticdomain of 327 residues, and a long CTE of 912 residues (Fig. 1).The BprL precursor shares 98%, 97%, and 70% amino acid se-quence identities with Bpr precursors from B. subtilis natto (5), B.subtilis 168 (3), and B. amyloliquefaciens CH86-1 (9), respectively.The nonconserved residues are located mainly within the CTE(see Fig. S2 in the supplemental material). While the catalytic do-

mains of BprL and B. subtilis 168 Bpr have an identical amino acidsequence, it differs from that of B. subtilis natto Bpr by only threeamino acid substitutions (see Fig. S2 in the supplemental mate-rial).

Production and processing of BprL and its variants in B.subtilis. The culture supernatant of the control strain, B. subtilisWB700(pBE2), was compared with that of B. subtilisWB700(pBE2-bprL), producing BprL, by SDS-PAGE. At leastfour additional protein bands with apparent molecular masses of85, 75, 45, and 33 kDa were detected in the latter (Fig. 2A). Toinvestigate whether these additional protein bands were derivedfrom BprL, the culture supernatant of B. subtilis WB700(pBE2-bprL) was subjected to immunoblot analysis using anti-NM oranti-C antibodies. While the 45- and 33-kDa proteins were de-tected only with anti-NM and anti-C antibodies, respectively, the85- and 75-kDa proteins were detected by both antibodies (Fig.2A). This indicates that all four proteins are processed products of

FIG 2 SDS-PAGE and immunoblot analyses of the products of BprL and its variants produced by B. subtilis strains. (A) B. subtilis WB700 (Bs) harboring pBE2or the expression vector pBE2-bprL, pBE2-bprLS258A, or pBE2-bprL�C912, as well as B. subtilis LZW, was grown in LB medium at 37°C for 24 h. The culturesupernatants were subjected to conventional SDS-PAGE and anti-NM or anti-C immunoblot analysis. The closed arrowheads indicate the major processedproducts. The 150- and 110-kDa products of BprLS258A are indicated by open arrowheads. (B) The proteins in the culture supernatant of B. subtilis WB700(pBE2-bprL) producing BprL were boiled in loading buffer with (�U) or without (U) 8 M urea and subjected to SDS-PAGE and anti-NM immunoblotanalysis. The 75-kDa product band (lane U) was excised from the SDS-polyacrylamide gel, and the proteins were extracted with buffer A containing 8 M urea,precipitated with 20% TCA, boiled in loading buffer containing 8 M urea, and then subjected to SDS-PAGE analysis (�U=). (C) The culture supernatants of B.subtilis WB700 (pBE2-bprL�C912) producing BprL�C912 (lanes 1 and 2) and the mature form of BprL�C912 purified by Ni2� affinity chromatography (lane3) were subjected to SDS-PAGE (lanes 1 and 3) and anti-His immunoblot analysis (lane 2).

Maturation of Bacillopeptidase F

January 2016 Volume 82 Number 1 aem.asm.org 321Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

BprL; among them, the 45-kDa product lacks the CTE and the33-kDa product is a fragment of the CTE. The SDS-PAGE profileof the proteins did not change markedly after incubation of theculture supernatant of B. subtilis WB700(pBE2-bprL) at 40°C for9 h (data not shown), suggesting that the four proteins are stableprocessed products of BprL. Interestingly, pretreatment of thesample with urea-containing loading buffer slowed migration ofthe 75-kDa product on SDS-PAGE, so that its migration rate wasconsistent with that of the 85-kDa product (Fig. 2B, lane �U).Moreover, after extraction from the gel with urea solution, the75-kDa product also displayed an apparent molecular mass of 85kDa (Fig. 2B, lane �U=). We interpret these results to mean thatthe 75-kDa product adopts a conformation different from that ofthe 85-kDa product. Also, it is relatively resistant to acid (TCA)treatment but sensitive to urea treatment.

The first four residues of the 45-kDa product were identified asATDG by N-terminal sequencing, indicating that the N-terminalpropeptide has been processed at the peptide bond Lys1-Ala1

(Fig. 1). N-terminal sequencing of the 85-kDa product revealedtwo polypeptides, with the first four residues identified as those ofthe 45-kDa product (ATDG) and the residues (SSSP) upstream ofthe peptide bond Lys1-Ala1, respectively, implying that the gen-eration of the 85-kDa product involves multiple processing of theN-terminal propeptide (Fig. 1).

Besides the 85-, 75-, 45-, and 33-kDa products, many minorprotein fragments were detected by anti-NM and/or anti-C anti-bodies in the culture supernatant of B. subtilis WB700(pBE2-bprL) but were absent in that of B. subtilis WB700(pBE2) (Fig. 2A),indicating that they were processing intermediates or degradationproducts of BprL. Many more of these minor fragments were de-tected by anti-C antibodies than by anti-NM antibodies, whichsuggested that the CTE of BprL is more sensitive to proteolysisthan its catalytic domain. The culture supernatant of strain LZWhad an immunoblot pattern of the processed products from BprLthat was similar to that for B. subtilis WB700(pBE2-bprL) (Fig.2A), suggesting that BprL undergoes multiple proteolytic process-ing in vivo.

It is known that replacing the catalytic residue Ser with Alaabolishes the activity of subtilases but does not affect enzyme fold-ing (28, 29). To investigate whether the multiple proteolytic pro-cessing of BprL occurs auto- or heterocatalytically, the culturesupernatant of B. subtilis WB700(pBE2-bprLS258A) producingthe active-site variant BprLS258A was subjected to SDS-PAGEand immunoblot analyses. Three protein bands with apparentmolecular masses of approximately 150, 110, and 85 kDa weredetected (Fig. 2A). The 150- and 110-kDa products may representthe secreted proform with a calculated molecular mass of 151.5kDa and a processed intermediate of BprLS258A. Since the 150-and 110-kDa products were not detected in the culture superna-tant of B. subtilis WB700(pBE2-bprL) (Fig. 2A), it is apparent thatthe active site of BprL mediates the conversion of the 150- and110-kDa proteins to other products. Nevertheless, the 85-kDaproduct and some of the minor protein fragments were found inculture supernatants of both B. subtilis WB700(pBE2-bprLS258A)and B. subtilis WB700(pBE2-bprL) (Fig. 2A), indicating that theirproduction is independent of the active site of BprL but relies onproteolytic activity elaborated by the host. Moreover, N-terminalsequencing of the 85-kDa product of BprLS258A revealed twopolypeptides with the first four amino acid residues identified asATDG and SSSP, which are identical to those of the 85-kDa prod-

uct of BprL. These results demonstrate that the in vivo processingof BprL involves both auto- and heterocatalytic events.

When the BprL�C912 precursor was produced in B. subtilisWB700, the 45-kDa product was detected in the culture superna-tant by SDS-PAGE and immunoblotting using anti-NM antibod-ies (Fig. 2A), confirming that the CTE is not required for thesecretion of BprL. In addition, the 45-kDa product could be de-tected by anti-His tag antibody and purified by Ni2� affinity chro-matography (Fig. 2C), suggesting that no further C-terminalprocessing occurred during the conversion of BprL�C912 to the45-kDa product. Therefore, the 45-kDa product contains only the327-residue catalytic domain with a C-terminal His tag and rep-resents the smallest mature form of BprL, similar to the activecomponent of NKCP from B. subtilis natto (8). The predictedmolecular mass of the smallest mature form of BprL is 34.4 kDa,and it is predicted to have a pI of 3.8. The predicted molecularmass is lower than its apparent molecular mass of 45 kDa as de-termined by SDS-PAGE analysis; this is a common phenomenonwith acidic proteins (30).

Detection of fibrinolytic activity of BprL. The fibrinolytic ac-tivity of BprL was evidenced by the finding that a clear zone wasformed around the colony of B. subtilis WB700(pBE2-bprL)grown on the fibrin plate (Fig. 3A). In contrast, the clear zone didnot appear around the colonies of B. subtilis WB700, B. subtilisWB700(pBE2), or B. subtilis WB700(pBE2-bprLS258A) (Fig. 3A).In addition, the formation of a clear zone around the colony of B.subtilis WB700(pBE2-bprL�C912) suggests that the CTE is notrequired for fibrinolytic activity of BprL (Fig. 3A). A fibrin overlayassay revealed fibrinolytic bands in the culture supernatants(without TCA treatment) of B. subtilis WB700(pBE2-bprL), B.subtilis WB700(pBE2-bprL�C912), and B. subtilis WB700(pBE2)(Fig. 3B). There were four bands associated with B. subtilis WB700(pBE2-bprL), two with B. subtilis WB700(pBE2-bprL�C912), andone with B. subtilis WB700(pBE2) (Fig. 3B). The fibrinolytic bandc was present in the culture supernatants of all three strains (Fig.3B), indicating that it is an unknown fibrinolytic enzyme of thehost. The fibrinolytic protein bands a, b, and d were extractedfrom the gel, precipitated with TCA, and then subjected to immu-noblot analysis using anti-NM antibodies after pretreatment withurea-containing loading buffer (Fig. 3C). We found that the fi-brinolytic band d represented the 45-kDa product of BprL (Fig.3C, lane d). The fibrinolytic bands a and b, which were present inthe culture supernatant of B. subtilis WB700(pBE2-bprL) but ab-sent in that of B. subtilis WB700(pBE2-bprL�C912) (Fig. 3B),showed an apparent molecular mass of 85 kDa after treatmentwith urea (Fig. 3C, lanes a and b). Therefore, the fibrinolytic bandsa and b corresponded to the 85-kDa and 75-kDa products of BprL,respectively. These results demonstrate that the 85-, 75-, and 45-kDa products are three mature forms of BprL with fibrinolyticactivity.

In vitro processing of BprL and its variants. Our next objec-tive was to assess whether BprL can mature properly in vitro. Therole of the long CTE in enzyme maturation was also investigated.We constructed recombinant proforms of BprL and its sequentialC-terminal truncation variants (BprL�C511, BprL�C678, andBprL�C912) and active-site variant (BprLS258A), as well as theCTE fragment (BprLC912) (Fig. 1). The recombinant proteinswith a C-terminal His tag were produced in E. coli, purified underdenaturing conditions, dialyzed at 4°C overnight, and incubatedat 30°C to allow enzyme refolding and processing. BprL and

Meng et al.

322 aem.asm.org January 2016 Volume 82 Number 1Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

BprL�C511 were degraded after dialysis and incubation, as shownby SDS-PAGE analysis (Fig. 4A), and had very weak proteolyticactivity (data not shown). Under the same conditions, BprLS258A(band e) was not degraded but was converted into a product (bandf) with slightly lower apparent molecular mass than band e (Fig.4A). No proteolytic activity was detected in the sample ofBprLS258A (data not shown), and band f could be converted intoband e after treatment with 8 M urea (Fig. 4A). Therefore, wesuggest that bands e and f represent the two forms of BprLS258Awith different conformations. Apparently, BprL and BprL�C511do not fold properly and adopt an unstable active conformationthat undergoes autodegradation. Moreover, immunoblot analysisshowed that the main processed products (85, 75, and 45 kDa) ofBprL detected in the culture supernatant of B. subtilis WB700(pBE2-bprL) were absent in the in vitro processing sample of BprL(Fig. 4D), demonstrating a major difference between in vivo and invitro processing of BprL. In contrast, BprL�C912 and BprL�C678were converted into the 45-kDa mature form, detected by an-ti-NM antibodies (Fig. 4B, C, and D), indicating that the twoproteins were properly folded and autoprocessed. These resultssuggest that partial removal of the CTE (more than 511 residuesfrom the C terminus) is required for the initiation of autoprocess-

FIG 3 Fibrinolytic activities of BprL and its variant. (A) B. subtilis WB700 (Bs),B. subtilis WB700 harboring pBE2 or the expression vector pBE2-bpr, pBE2-bprLS258A, or pBE2-bprL�C91, and B. subtilis LZW were grown on fibrinplates at 37°C for 24 h. (B) The culture supernatant, without TCA treatment orboiling, of B. subtilis WB700 harboring pBE2 or the expression vector pBE2-bprL or pBE2-bprL�C912 was subjected to SDS-PAGE (lanes s) followed by afibrin overlay assay (lanes f). The closed arrowheads indicate the BprL- andBprL�C912-derived products with fibrinolytic activity. (C) The proteinsbands that correspond to the fibrinolytic bands a, b, and d were extracted froma replicate SDS-polyacrylamide gel, precipitated with 20% TCA, boiled inloading buffer containing 8 M urea, and subjected to anti-NM immunoblotanalysis.

FIG 4 In vitro processing of BprL and its variants. (A to C) The denaturedsamples (lanes U) of BprL and its variants (0.5 �M) were dialyzed overnight at4°C against buffer A to allow the denatured proteins to refold (lanes R) andthen incubated at 30°C. At the time intervals indicated, samples were with-drawn and subjected to conventional SDS-PAGE analysis and azocaseinolyticactivity assay. The refolded sample of BprLS258A (lanes 2d) was supplementedwith 8 M urea and then incubated at 30°C for 24 h (lanes 2d=), followed bySDS-PAGE analysis. The two forms of BprLS258A (e and f) are indicated (A).Relative activity was calculated with the highest measured activity defined as100% (B and C). (D) The culture supernatant (24 h) of B. subtilis WB700harboring an expression vector for BprL (lanes 1) and the in vitro processingsamples (48 h) of BprL (lanes 2), BprL�C678 (lanes 3), and BprL�C912 (lanes4) were subjected to conventional SDS-PAGE and anti-NM or anti-C immu-noblot analysis. The closed arrowheads indicate the major processed productsof BprL. The positions on the gels of the proforms (P and P=), the intermediates(Ia, Ia=, and Ib=), the 45-kDa mature form (M), the processed core domain ofthe N-terminal propeptide (N*), and the degradation product of the CTE(C5=) are indicated.

Maturation of Bacillopeptidase F

January 2016 Volume 82 Number 1 aem.asm.org 323Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

ing of BprL into the active 45-kDa mature enzyme and demon-strate the importance of the CTE in maintaining BprL in an inac-tive state.

BprL�C912 was cleaved into a 47-kDa intermediate (Ia) andN* after dialysis, and Ia was further processed into the 45-kDamature enzyme (M) as incubation time increased (Fig. 4B). Thefirst four residues of the 47-kDa Ia and the 45-kDa M were deter-mined to be SSSP and ATDG (Fig. 1), respectively, demonstratingthat the 15 N-terminal residues of Ia were truncated to yield M.Therefore, the N-terminal propeptide of BprL is composed of acore domain (N*) and a 15-residue linker peptide (Fig. 1). Theautoprocessing of linker peptide-containing N-terminal propep-tides of subtilisins involves initial cis processing of N* and subse-quent trans processing of the linker peptide (28). Likewise, theremoval of the N-terminal propeptide of BprL proceeded in astepwise manner by cis processing of N* at Lys16-Ser15 andtrans processing of the linker peptide at Lys1-Ala1 (Fig. 1). Inaddition, the degradation of N* was accompanied by the release ofproteolytic activity (Fig. 4B), suggesting that the N-terminal pro-peptide of BprL is an inhibitor of the enzyme. Similar toBprL�C912, BprL�C678 was initially converted into an interme-diate (Ia=) by cleavage of N*, followed by processing of Ia= into Mvia another intermediate (Ib=) (Fig. 4C). BprL�C678 has 234more residues in its CTE than BprL�C912, and we suggest that theconversion of Ia= into M involved trans processing of both thelinker peptide and the partial CTE. The appearance of the inter-mediate Ib= and several smaller fragments that were less than 30kDa as well as the accumulation of degradation product C5= dur-ing the conversion (Fig. 4C) support this suggestion.

Unlike BprL�C912 and BprL�C678, which could be autopro-cessed into active 45-kDa mature enzyme, the active-site variantsBprL�C912-S258A and BprL�C678-S258A remained unpro-cessed after refolding and incubation alone (Fig. 5A and B, M),providing strong evidence that the active site of BprL mediatesautoprocessing of the N-terminal propeptide. In contrast, bothBprL�C912-S258A and BprL�C678-S258A were converted into a45-kDa mature form when incubated with supplemental 45-kDamature enzyme (Fig. 5A and B, �M), indicating that the N-ter-minal propeptide of the proform could be intermolecularlycleaved by the active mature enzyme. The size of the major cleavedN-terminal propeptide of BprL�C912-S258A or BprL�C678-S258A (Fig. 5A and B, �M, lane R) was similar to that of thecis-processed N* of BprL�C912 or BprL�C678 (Fig. 4B and C),suggesting that the initial cleavage of the N-terminal propeptideoccurs mainly within the linker peptide of the proform. In the caseof BprL�C678-S258A, which contains a partial CTE, both theN-terminal propeptide and the partial CTE were processed bysupplemental 45-kDa mature enzyme; the appearance of the in-termediate Ib= suggests that the partial CTE is trans processed in astepwise manner (Fig. 5B). These results demonstrate that thecatalytic domain in the variant of the BprL proform with a full orpartial deletion of the CTE is properly folded and adopts a prote-olysis-resistant conformation, allowing active protease to transprocess the variant into the 45-kDa mature form.

In subtilases, the replacement of the catalytic residue Ser withCys greatly reduces enzyme activity but does not eliminate itsautoprocessing activity, so that the maturation process is blockedafter the initial cis processing of the N-terminal propeptide (28,29). Similarly, the variant BprL�C912-S258C or BprL�C678-S258C itself was converted into the intermediate Ia or Ia= by cis

processing of N* but could not be further processed into the 45-kDa mature form (Fig. 5C and D, M). We found that the cis-processed N* of BprL�C912, BprL�C912-S258C, BprL�C678, orBprL�C678-S258C coeluted with Ia or Ia= from an Ni2�-chargedcolumn (data not shown), implying that the cleaved N* was stillbound to Ia or Ia= to form an autoprocessed complex (N*-Ia orN*-Ia=), as has been reported for other subtilisins (28, 29). In thepresence of the active 45-kDa mature enzyme, either N*-Ia orN*-Ia= was trans processed into the 45-kDa mature form (Fig. 5Cand D, �M), indicating that the cleaved N* and the partial CTE inthe autoprocessed complex are more susceptible to proteolysisthan the catalytic domain.

DISCUSSION

Based on these results, we propose the following mechanism forBprL maturation in B. subtilis. Maturation proceeds via both

FIG 5 trans processing of BprL variants. The refolded samples (5 �M) (lanesR) of BprL�C912-S258A (A), BprL�C678-S258A (B), BprL�C912-S258C(C), and BprL�C678-S258C (D) were incubated at 30°C in the absence (M)or presence (�M) of the 45-kDa mature enzyme (0.5 �M) (lane M). At thetime intervals indicated, samples were withdrawn and subjected to conven-tional SDS-PAGE analysis. The positions on the gels of the proforms (P andP=), the intermediates (Ia, Ia=, and Ib=), the 45-kDa mature form (M), theprocessed core domain of the N-terminal propeptide (N*), and the degrada-tion product of the CTE (C5=) are indicated.

Meng et al.

324 aem.asm.org January 2016 Volume 82 Number 1Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

hetero- and autocatalytic steps that yield several fibrinolytic ma-ture forms with apparent molecular masses of 45, 75, and 85 kDa.Partial removal of the CTE of the proform may trigger autopro-cessing of the N-terminal propeptide, which is composed of a coredomain (N*) and a 15-residue linker peptide. Complete removalof the N-terminal propeptide and the CTE leads to the generationof the 45-kDa mature form.

The 85-kDa mature form is generated via heterocatalytic pro-cessing of the proform by an unidentified protease of B. subtilis.Data from N-terminal sequencing of the 85-kDa mature form leadus to suggest that the N-terminal propeptide is processed at twocleavage sites within the linker peptide. Similarly, proteolyticcleavage of the N-terminal propeptide within the linker peptide ofBpr from B. subtilis 168 or B. subtilis natto leads to generation of an80-kDa or a 90-kDa mature form (4, 5). The serine protease in-hibitor PMSF does not prevent the production of the 80-kDa ma-ture form of Bpr in B. subtilis 168 (4), a finding that may indicatethat another protease(s) participates in the processing of the en-zyme. Heterocatalytic removal of the N-terminal propeptide viaproteolytic cleavage at the linker peptide has been described forother subtilases, notably the WF146 protease from thermophilicBacillus sp. strain WF146 (28) and the protease CDF from Ther-moactinomyces sp. strain CDF (27). In this context, the linker pep-tide of subtilases comprises a proteolytically sensitive region, al-lowing trans processing of the N-terminal propeptide by otherproteases. Moreover, the BprL with its N-terminal propeptide de-leted has a predicted molecular mass of 133.2 kDa, which is higherthan that of the 85-kDa mature form. This indicates that in addi-tion to the N-terminal propeptide, part of the CTE has been pro-cessed during the conversion of the BprL proform to the 85-kDamature form, although the exact C-terminal cleavage site remainsto be determined.

The production of the 75- and 45-kDa mature forms involvesboth hetero- and autocatalytic events. BprL alone is not autopro-cessed into the 75- and 45-kDa mature forms in vitro, and it maybe that an unknown protease of B. subtilis participates in the invivo conversion of BprL into the two mature forms. The in vitroanalyses of BprL and its C-terminal truncation variants reveal thatpartial removal of the CTE is required for the generation of the45-kDa mature form. Because the variant BprL�C678, but notBprL�C511, can be autoprocessed into the 45-kDa mature form,the in vivo production of the 45-kDa mature form in B. subtilis ismost likely triggered by heterocatalytic processing of the CTE; thecleavage site would then be located between Leu561 and Glu728(Fig. 1). After partial removal of the CTE, the 45-kDa mature formis generated autocatalytically by cis processing of the core domainof the N-terminal propeptide (N*), and subsequent trans process-ing of both the linker peptide and the remaining part of the CTE.In comparison with the 45-kDa mature form, the 75-kDa matureform contains part of the CTE. However, it is unlikely that the75-kDa mature form is an intermediate during the conversion ofBprL to the 45-kDa mature form, since both forms are stable mat-uration products of BprL. An alternative possibility is that the CTEof BprL contains several cleavage sites for an unidentified pro-tease(s) of B. subtilis and that differential proteolytic processing ofthe CTE leads to multiple intermediates with different conforma-tions. In some intermediates, after autoprocessing of the N-termi-nal propeptide, the remaining part of the CTE is susceptible toproteolysis and would be trans processed completely to generatethe 45-kDa mature form. In other intermediates, the remaining

part of the CTE is resistant to proteolysis such that their C terminieither would not be processed further or would only be truncatedpartially after autoprocessing of the N-terminal propeptide,thereby yielding the 75-kDa mature form. Notably, the 75-kDamature form, which displays an apparent molecular mass of 85kDa after urea denaturation, differs from the 85-kDa mature formin its resistance to acid treatment. This may be due to its adoptionof a conformation different from that of the 85-kDa mature form(heterocatalytic maturation product). In this context, differentialprocessing of BprL, either heterocatalytically or autocatalytically,leads to the formation of mature forms with different molecularmasses or conformations.

Multiple mature forms have been described for other CTE-containing subtilases, such as Vpr (31, 32) and Epr (33, 34) fromB. subtilis and pyrolysin from Pyrococcus furiosus (35, 36). Matu-ration of these enzymes also involves differential proteolytic pro-cessing of the CTE. The mature form of Vpr isolated from B.subtilis has a molecular mass of 28.5 kDa (31), but in vitro analysisshows that this enzyme can be autoprocessed into multiple matureforms with fibrinolytic activity and ranging from 40 to 68 kDa,although the 28.5-kDa form has not been detected (32). It may bethat another protease(s) of B. subtilis contributes to the in vivoproduction of the 28.5-kDa form of Vpr, similar to the case of the45-kDa mature form of BprL. Nevertheless, BprL differs from Vprin that BprL alone is not processed into stable mature forms;rather, it requires another protease(s) to partially cleave the CTEand thus to mature properly.

There are at least 27 proteases in the membrane, cell wall, andmedium of B. subtilis that can process or degrade secretory pro-teins (1, 2). Two major extracellular proteases (AprE and NprE)account for more than 95% of the total extracellular protease ac-tivity of B. subtilis (37), while five minor extracellular or cell wall-associated proteases (Epr, Vpr, Bpr, Mpr, and WprA) are respon-sible for most of the remaining activity (31, 33, 38, 39). Theprincipal function of extracellular proteases, including Bpr, isthought to be supplying nutrients for bacterial growth by degrad-ing extracellular proteins, although additional physiological func-tions have also been described for some of these proteases. Forinstance, Epr (40) and Vpr (41, 42) are involved in swarmingmotility of B. subtilis. AprE, WprA, and Vpr are reported to play arole in processing and activation of the antibiotic peptide subtilin,and a function for Bpr and Epr in this process cannot be ruled out(43). Notably, single-cell analyses revealed that only part of thevegetative subpopulation of B. subtilis cells highly and transientlyexpresses genes encoding Bpr and AprE (44). As these proteasesand their degradation products freely diffuse within the liquidgrowth medium, all cells within the entire population are expectedto benefit from their activities, suggesting that B. subtilis employscooperative or even altruistic behavior (44). Seven of the knownproteases, specifically AprE, Bpr, Epr, Mpr, NprB, NprE, and Vpr,are not required for BprL processing, as their genes are inactivatedin B. subtilis WB700 (17). However, we do not exclude the possi-bility that some or all of these proteases are involved in Bpr pro-cessing in wild-type B. subtilis and may accelerate Bpr maturation.Interestingly, Bpr is required for the processing of Mpr (7), whichis a glutamic acid-specific endopeptidase (45). Here, we foundthat Bpr itself requires an unidentified protease for maturation,and perhaps there exists a protease activation cascade that regu-lates extracellular proteolytic activity in B. subtilis. The long CTEof BprL is not essential for enzyme activity and secretion but plays

Maturation of Bacillopeptidase F

January 2016 Volume 82 Number 1 aem.asm.org 325Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

an important role in maintaining BprL in an inactive state andthus may contribute to the regulation of the protease activationcascade in B. subtilis. We noticed an unknown fibrinolytic enzymein the culture supernatant of B. subtilis WB700. The identificationof this enzyme and its possible relation to BprL maturation awaitsfuture experimental endeavor.

FUNDING INFORMATIONNational Natural Science Foundation of China provided funding to BingTang under grant numbers 31270099 and 31470185. National Infrastruc-ture of Natural Resources for the Science and Technology Program ofChina provided funding to Xiao-Feng Tang under grant number NIMR-2014-8.

REFERENCES1. Tjalsma H, Antelmann H, Jongbloed JD, Braun PG, Darmon E, Doren-

bos R, Dubois JY, Westers H, Zanen G, Quax WJ, Kuipers OP, Bron S,Hecker M, van Dijl JM. 2004. Proteomics of protein secretion by Bacillussubtilis: separating the “secrets” of the secretome. Microbiol Mol Biol Rev68:207–233. http://dx.doi.org/10.1128/MMBR.68.2.207-233.2004.

2. Tjalsma H, Bolhuis A, Jongbloed JDH, Bron S, van Dijl JM. 2000. Signalpeptide-dependent protein transport in Bacillus subtilis: a genome-basedsurvey of the secretome. Microbiol Mol Biol Rev 64:515–547. http://dx.doi.org/10.1128/MMBR.64.3.515-547.2000.

3. Sloma A, Rufo GA, Rudolph CF, Sullivan BJ, Theriault KA, Pero J.1990. Bacillopeptidase F of Bacillus subtilis: purification of the protein andcloning of the gene. J Bacteriol 172:1470 –1477. (Erratum, 172:5520.)

4. Wu XC, Nathoo S, Pang AS, Carne T, Wong SL. 1990. Cloning, geneticorganization, and characterization of a structural gene encoding bacillo-peptidase F from Bacillus subtilis. J Biol Chem 265:6845– 6850.

5. Yamagata Y, Abe R, Fujita Y, Ichisima E. 1995. Molecular cloning andnucleotide sequence of the 90k serine protease gene, hspK, from Bacillussubtilis (natto) no. 16. Curr Microbiol 31:340 –344. http://dx.doi.org/10.1007/BF00294696.

6. Tsukahara K, Ogura M. 2008. Characterization of DegU-dependent ex-pression of bpr in Bacillus subtilis. FEMS Microbiol Lett 280:8 –13. http://dx.doi.org/10.1111/j.1574-6968.2007.01019.x.

7. Park CH, Lee SJ, Lee SG, Lee WS, Byun SM. 2004. Hetero- and auto-processing of the extracellular metalloprotease (Mpr) in Bacillus subtilis. JBacteriol 186:6457– 6464. http://dx.doi.org/10.1128/JB.186.19.6457-6464.2004.

8. Omura K, Hitosugi M, Zhu X, Ikeda M, Maeda H, Tokudome S. 2005.A newly derived protein from Bacillus subtilis natto with both antithrom-botic and fibrinolytic effects. J Pharmacol Sci 99:247–251. http://dx.doi.org/10.1254/jphs.FP0050408.

9. Kwon GH, Park JY, Kim JS, Lim J, Park CS, Kwon DY, Kim JH. 2011.Cloning and expression of a bpr gene encoding bacillopeptidase F fromBacillus amyloliquefaciens CH86-1. J Microbiol Biotechnol 21:515–518.http://dx.doi.org/10.4014/jmb.1010.10061.

10. Hwang KJ, Choi KH, Kim MJ, Park CS, Cha J. 2007. Purification andcharacterization of a new fibrinolytic enzyme of Bacillus licheniformis KJ-31, isolated from Korean traditional Jeot-gal. J Microbiol Biotechnol9:1469 –1476.

11. Choi NS, Yoo KH, Hahm JH, Yoon KS, Chang KT, Hyun BH, MaengPJ, Kim SH. 2005. Purification and characterization of a new peptidase,bacillopeptidase DJ-2, having fibrinolytic activity: produced by Bacillus sp.DJ-2 from Doen-Jang. J Microbiol Biotechnol 15:72–79.

12. Siezen RJ, Leunissen JA. 1997. Subtilases: the superfamily of subtilisin-like serine proteases. Protein Sci 6:501–523.

13. Khan AR, James MNG. 1998. Molecular mechanisms for the conversionof zymogens to active proteolytic enzymes. Protein Sci 7:815– 836.

14. Kato T, Yamagata Y, Arai T, Ichishima E. 1992. Purification of a newextracellular 90-kDa serine proteinase with isoelectric point of 3.9 fromBacillus subtilis (natto) and elucidation of its distinct mode of action. Bio-sci Biotechnol Biochem 56:1166 –1168. http://dx.doi.org/10.1271/bbb.56.1166.

15. Roitsch CA, Hageman JH. 1983. Bacillopeptidase F: two forms of aglycoprotein serine protease from Bacillus subtilis 168. J Bacteriol 155:145–152.

16. Boyer HW, Carlton BC. 1968. Production of two proteolytic enzymes by

a transformable strain of Bacillus subtilis. Arch Biochem Biophys 128:442–455. http://dx.doi.org/10.1016/0003-9861(68)90050-7.

17. Ye R, Yang LP, Wong SL. 1996. Construction of protease-deficientBacillus subtilis strains for expression studies: inactivation of seven extra-cellular proteases and the intracellular LonA protease, p 160 –169. In Pro-ceedings of the International Symposium on Recent Advances in Bioin-dustry. Korean Society for Applied Microbiology, Seoul, South Korea.

18. Kim W, Choi K, Kim Y, Park H, Choi J, Lee Y, Oh H, Kwon I, Lee S.1996. Purification and characterization of a fibrinolytic enzyme producedfrom Bacillus sp. strain CK11-4 screened from Chungkook-Jang. ApplEnviron Microbiol 62:2482–2488.

19. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a labora-tory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold SpringHarbor, NY.

20. Guo X, Xiong Z, Jia S, Xu Y. 1991. The construction of multifunctionalshuttle vectors of Bacillus subtilis-Escherichia coli. Chin J Biotechnol7:224 –229.

21. An Y, Lv A, Wu W. 2010. A Quikchange-like method to realize efficientblunt-ended DNA directional cloning and site-directed mutagenesis si-multaneously. Biochem Biophys Res Commun 397:136 –139. http://dx.doi.org/10.1016/j.bbrc.2010.05.042.

22. Bradford MM. 1976. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dyebinding. Anal Biochem 72:248 –254. http://dx.doi.org/10.1016/0003-2697(76)90527-3.

23. King J, Laemmli UK. 1971. Polypeptides of the tail fibres of bacteriophage T4.J Mol Biol 62:465–477. http://dx.doi.org/10.1016/0022-2836(71)90148-3.

24. Park SG, Kho CW, Cho S, Lee DH, Kim SH, Park BC. 2002. A functionalproteomic analysis of secreted fibrinolytic enzymes from Bacillus subtilis168 using a combined method of two-dimensional gel electrophoresis andzymography. Proteomics 2:206 –211. http://dx.doi.org/10.1002/1615-9861(200202)2:2206::AID-PROT206�3.0.CO;2-5.

25. Blumentals II, Robinson AS, Kelly RM. 1990. Characterization of so-dium dodecyl sulfate-resistant proteolytic activity in the hyperthermo-philic archaebacterium Pyrococcus furiosus. Appl Environ Microbiol 56:1992–1998.

26. Adams LD, Weaver KM. 1990. Detection and recovery of proteins fromgels following zinc chloride staining. Appl Theor Electrophor 1:279 –282.

27. Cheng G, Zhao P, Tang X-F, Tang B. 2009. Identification and charac-terization of a novel spore-associated subtilase from Thermoactinomycessp. CDF. Microbiology 155:3661–3672. http://dx.doi.org/10.1099/mic.0.031336-0.

28. Madern D, Ebel C, Zaccai G. 2000. Halophilic adaptation of enzymes.Extremophiles 4:91–98. http://dx.doi.org/10.1007/s007920050142.

29. Zhu H, Xu BL, Liang X, Yang YR, Tang X-F, Tang B. 2013. Molecularbasis for auto- and hetero-catalytic maturation of a thermostable subtilasefrom thermophilic Bacillus sp. WF146. J Biol Chem 288:34826 –34838.http://dx.doi.org/10.1074/jbc.M113.498774.

30. Tanaka S, Matsumura H, Koga Y, Takano K, Kanaya S. 2007. Four newcrystal structures of Tk-subtilisin in unautoprocessed, autoprocessed, andmature forms. Insight into structural changes during maturation. J MolBiol 372:1055–1069.

31. Sloma A, Rufo GA, Theriault KA, Dwyer M, Wilson SW, Pero J. 1991.Cloning and characterization of the gene for an additional extracellularserine protease of Bacillus subtilis. J Bacteriol 173:6889 – 6895.

32. Kho CW, Park SG, Cho S, Lee DH, Myung PK, Park BC. 2005.Confirmation of Vpr as a fibrinolytic enzyme present in extracellular pro-teins of Bacillus subtilis. Protein Expr Purif 39:1–7. http://dx.doi.org/10.1016/j.pep.2004.08.008.

33. Sloma A, Ally A, Ally D, Pero J. 1988. Gene encoding a minor extracel-lular protease in Bacillus subtilis. J Bacteriol 170:5557–5563.

34. Bruckner R, Shoseyov O, Doi RH. 1990. Multiple active forms of a novelserine protease from Bacillus subtilis. Mol Gen Genet 221:486 – 490.

35. Voorhorst WGB, Eggen RIL, Geerling ACM, Platteeuw C, Siezen RJ, deVos WM. 1996. Isolation and characterization of the hyperthermostableserine protease, pyrolysin, and its gene from the hyperthermophilic ar-chaeon Pyrococcus furiosus. J Biol Chem 271:20426 –20431. http://dx.doi.org/10.1074/jbc.271.34.20426.

36. Dai Z, Fu H, Zhang Y, Zeng J, Tang B, Tang X-F. 2012. Insights into thematuration of hyperthermophilic pyrolysin and the roles of its N-terminalpropeptide and long C-terminal extension. Appl Environ Microbiol 78:4233– 4241. http://dx.doi.org/10.1128/AEM.00548-12.

37. Kawamura F, Doi RH. 1984. Construction of a Bacillus subtilis double

Meng et al.

326 aem.asm.org January 2016 Volume 82 Number 1Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from

mutant deficient in extracellular alkaline and neutral proteases. J Bacteriol160:442– 444.

38. Sloma A, Rudolph CF, Rufo GA, Sullivan BJ, Theriault KA, Ally D,Pero J. 1990. Gene encoding a novel extracellular metalloprotease in Ba-cillus subtilis. J Bacteriol 172:1024 –1029.

39. Yang MY, Ferrari E, Henner DJ. 1984. Cloning of the neutral proteasegene of Bacillus subtilis and the use of the cloned gene to create an invitro-derived deletion mutation. J Bacteriol 160:15–21.

40. Gupta M, Rao KK. 2009. Epr plays a key role in DegU-mediated swarm-ing motility of Bacillus subtilis. FEMS Microbiol Lett 295:187–194. http://dx.doi.org/10.1111/j.1574-6968.2009.01596.x.

41. Connelly MB, Young GM, Sloma A. 2004. Extracellular proteolytic activityplays a central role in swarming motility in Bacillus subtilis. J Bacteriol 186:4159–4167. http://dx.doi.org/10.1128/JB.186.13.4159-4167.2004.

42. Morikawa M, Kagihiro S, Haruki M, Takano K, Branda S, Kolter R,Kanaya S. 2006. Biofilm formation by a Bacillus subtilis strain that pro-

duces gamma-polyglutamate. Microbiology 152:2801–2807. http://dx.doi.org/10.1099/mic.0.29060-0.

43. Corvey C, Stein T, Düsterhus S, Karas M, Entian KD. 2003. Activationof subtilin precursors by Bacillus subtilis extracellular serine proteases sub-tilisin (AprE), WprA, and Vpr. Biochem Biophys Res Commun 304:48 –54. http://dx.doi.org/10.1016/S0006-291X(03)00529-1.

44. Veening JW, Igoshin OA, Eijlander RT, Nijland R, Hamoen LW,Kuipers OP. 2008. Transient heterogeneity in extracellular protease pro-duction by Bacillus subtilis. Mol Syst Biol 4:184. http://dx.doi.org/10.1038/msb.2008.18.

45. Okamoto H, Fujiwara T, Nakamura E, Katoh T, Iwamoto H, Tsu-zuki H. 1997. Purification and characterization of a glutamic-acid-specific endopeptidase from Bacillus subtilis ATCC 6051; application tothe recovery of bioactive peptides from fusion proteins by sequence-specific digestion. Appl Microbiol Biotechnol 48:27–33. http://dx.doi.org/10.1007/s002530051010.

Maturation of Bacillopeptidase F

January 2016 Volume 82 Number 1 aem.asm.org 327Applied and Environmental Microbiology

on March 26, 2018 by guest

http://aem.asm

.org/D

ownloaded from