APPLIED AND ENVIRONMENTAL MICROBIOLOGY,0099-2240/98/$04.0010

Mar. 1998, p. 824–829 Vol. 64, No. 3

Copyright © 1998, American Society for Microbiology

Engineering of Plasmin-Resistant Forms of Streptokinase andTheir Production in Bacillus subtilis: Streptokinase

with Longer Functional Half-LifeXU-CHU WU, RUIQIONG YE, YANJUN DUAN, AND SUI-LAM WONG*

Division of Cellular, Molecular and Microbial Biology, Department ofBiological Sciences, University of Calgary, Calgary, Alberta, Canada

Received 4 August 1997/Accepted 4 December 1997

The short in vivo half-life of streptokinase limits its efficacy as an efficient blood clot-dissolving agent. Duringthe clot-dissolving process, streptokinase is processed to smaller intermediates by plasmin. Two of the majorprocessing sites are Lys59 and Lys386. We engineered two versions of streptokinase with either one of the lysineresidues changed to glutamine and a third version with both mutations. These mutant streptokinase proteins(muteins) were produced by secretion with the protease-deficient Bacillus subtilis WB600 as the host. The puri-fied muteins retained comparable kinetics parameters in plasminogen activation and showed different degreesof resistance to plasmin depending on the nature of the mutation. Muteins with double mutations had half-lives that were extended 21-fold when assayed in a 1:1 molar ratio with plasminogen in vitro and showed betterplasminogen activation activity with time in the radial caseinolysis assay. This study indicates that plasmin-mediated processing leads to the inactivation of streptokinase and is not required to convert streptokinase toits active form. Plasmin-resistant forms of streptokinase can be engineered without affecting their activity, andblockage of the N-terminal cleavage site is essential to generate engineered streptokinase with a longer in vitrofunctional half-life.

The formation of pathologic blood clots that block the cir-culation to heart muscle can result in acute myocardial infarc-tion (heart attack). Several blood clot-dissolving agents includ-ing streptokinase and tissue-specific plasminogen activator arecommonly applied to treat these patients. Many large-scaleclinical trials (10, 14, 17) have demonstrated both the short-and long-term benefits of these agents in saving lives. To gainmaximum benefits of thrombolytic therapy in restoring bloodflow, limiting damage to heart muscles, and preserving heartfunctions, early treatment is particularly important (14). Therelationship between early treatment with thrombolysis andlower mortality has been well established (6, 11, 11a, 16, 27, 30,33, 34, 44). However, one of the limitations of these bloodclot-dissolving agents is their short in vivo half-lives. Withhalf-lives of 30 min for streptokinase (15) and 5 min for tPA(19), these agents are commonly introduced into the patientsby a 30- to 90-min infusion. If the half-lives of these agents canbe prolonged, the agents could possibly be administered as asingle bolus intravenous injection and patients could be treatedupon the arrival of the medical personnel. This would helpminimize the time delay in the transportation of patients tohospital. The reocclusion rate could also be reduced by usingthese long-half-life clot-dissolving agents.

Streptokinase is a 47-kDa (414-amino-acid) protein frompathogenic strains of the Streptococcus family (25). To dissolvea blood clot, streptokinase forms a 1:1 molar complex withplasminogen (1). The resulting complex (8) has the ability toconvert plasminogen to plasmin, the active protease that de-grades fibrin in the blood clot. However, plasmin also rapidly

processes streptokinase to smaller fragments. This can be amajor factor contributing to the short half-life of streptokinase.The processing pathway of streptokinase has been well char-acterized (28, 38, 39). Several intermediates, including a fewproducts with molecular masses of 37 to 44 kDa, are transientlyaccumulated (38, 39). The 42- to 44-kDa intermediates appearfirst and are generated by C-terminal processing, since theyhave identical N-terminal residues to those observed in theintact streptokinase (28). Isolation of a short C-terminal pep-tide with the N-terminal sequence corresponding to Tyr402(38) indicates that one of the C-terminal cleavage events takesplace between Arg401 and Tyr402. The 37-kDa product ap-pears later and is relatively stable (4, 28, 38, 39). N-terminalsequencing and composition analysis suggest that this fragmenthas the sequence corresponding to Ser60 to Lys386 from theauthentic streptokinase (38). This product is therefore gener-ated through both N- and C-terminal processing events. Al-though this 37-kDa product has high affinity to both plasmin-ogen and plasmin, it retains only 16% of the activity of theintact streptokinase in plasminogen activation (38). Furtherprocessing of the 37-kDa product at a series of cleavage sites(38) results in the complete degradation of streptokinase intosmall fragments. Since plasmin is a trypsin-like serine proteasethat specifically cleaves the peptide bond after lysine or argi-nine (45), it would be interesting to see whether selectivelychanging lysine and arginine residues at these processing sitesto other amino acids (e.g., glutamine) would generate plasmin-resistant streptokinase that may have a longer functional half-life. The successful generation of these new versions of strep-tokinase depends critically on whether processing at these sitesis a necessary event. Currently, it is uncertain whether theseprocessing events are simply a consequence of positioning ly-sine and arginine residues at the flexible and surface-exposedregions or are essential events in the conversion of streptoki-nase to the active form. It has been observed that the 7-kDaN-terminal peptide can associate with the 37-kDa intermediate

* Corresponding author. Mailing address: Division of Cellular, Mo-lecular and Microbial Biology, Department of Biological Sciences,University of Calgary, 2500 University Dr., N.W., Calgary, AlbertaT2N 1N4, Canada. Phone: (403) 220-5721. Fax: (403) 289-9311. E-mail: slwong@acs.ucalgary.ca.

824

on May 12, 2019 by guest

http://aem.asm

.org/D

ownloaded from

to form a functional plasminogen activator that has almost thefull activity of the intact streptokinase (38). In a similar situa-tion, N-terminal processing that results in the removal of thefirst 10 amino acids from staphylokinase, another plasminogenactivator from lysogenic strains of Staphylococcus, by plasmin isdemonstrated to be essential in the generation of the activestaphylokinase (37). To determine the significance of the N-terminal processing event and to explore the possibility ofdeveloping engineered forms of streptokinase with longerfunctional half-lives, we report the development of variousforms of streptokinase through site-directed mutagenesis.These mutant proteins (abbreviated as muteins) of streptoki-nase were produced by the Bacillus subtilis secretory produc-tion system (48). Biochemical characterization and in vitroprocessing studies of these purified streptokinase muteins il-lustrate that plasmin-resistant streptokinase can be developed.Some of these engineered muteins have longer in vitro func-tional half-lives and show better activity with time.

MATERIALS AND METHODS

Bacterial strains and culture conditions. A six-extracellular-protease-deficientB. subtilis strain, WB600 (trpC2 nprA apr epr bfp mpr::ble nprB::ery) (49), was usedfor routine transformation and expression studies. Transformed cells were platedon tryptose blood agar base (TBAB; Difco, Detroit, Mich.) plates containing 10mg of kanamycin per ml. Cells carrying expression vectors were cultivated insuperrich medium (12) with kanamycin. All the expression vectors are pUB18derivatives (43). Therefore, WB600(pUB18) serves as a negative control for thestudy of streptokinase production. The initial cell density in the culture wasadjusted to 10 Klett units (1 Klett unit is equivalent to approximately 106 cells/ml). When the cell density reached 100 Klett units, sucrose was added at a finalconcentration of 2% (wt/vol) to induce the expression. The culture supernatantwas collected by centrifugation 5 h after induction.

Site-directed mutagenesis of the streptokinase production plasmid. PlasmidpSK3 (48) is an expression vector in B. subtilis that is used to produce streptoki-nase with the sucrose-inducible regulatory region from B. subtilis sacB, encodinglevansucrase, to control the expression. In this expression vector, secretion ofstreptokinase is directed by the sacB signal sequence (41, 47). To change Lys59to either glutamine or glutamic acid, site-directed mutagenesis based on theinverse PCR method described by Hemsley et al. (13) was used. Two primers,SKMF [59 CAAGGCTTAAGTCCA(C/G)AATCAAAACC 39] and SKMB (59CTCTGTCTTTCCTCCATGAGCAGG) were used for PCR by using the super-coiled pSK3 plasmid as the template. The amplified fragment was then endrepaired, kinase treated, and recircularized by ligation. Plasmid DNA was trans-formed to WB600. This site-directed mutagenesis method allows direct intro-duction of mutations to B. subtilis vectors without using E. coli vectors as theintermediate. The restriction enzymes and DNA-modifying enzymes used in thisstudy are from New England Biolabs Canada, Ltd. (Mississauga, Ontario, Can-ada), Pharmacia Biotech Inc. (Baie d’Urfe, Quebec, Canada), and GIBCO BRLCanada (Burlington, Ontario, Canada).

Purification of streptokinase and its derivatives. Streptokinase (or its deriva-tives) from the culture supernatant was precipitated and concentrated by addingammonium sulfate to 60% saturation. After dialysis, the sample was applied topreparative nondenaturing polyacrylamide gels (7.5%, wt/vol) that have the samecomposition as that for the standard sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE) except for the omission of SDS. To minimize thepresence of free radicals that may modify functional groups of proteins, the gelswere prerun for 2 h with the addition of reduced glutathione, at a final concen-tration of 50 mM, to the electrophoresis buffer in the upper reservoir. Thepreelectrophoresis buffer was decanted, and fresh buffer containing 0.1 mMsodium thioglycolate was used for the actual run. At the end of the electro-phoretic run, a strip of gel was excised and briefly stained with 1% (wt/vol)Coomassie blue R-250 for 5 min. The location of the major protein band wasdetermined and used as the reference point to locate streptokinase in the non-stained gel. The excised gel was cut into small pieces, and the protein waselectroeluted under a constant current (5 mA per tube) with Tris-glycine buffer(25 mM Tris base, 192 mM glycine [pH 8.3]) for 8 h at 4°C with the electroelutionsystem from Bio-Rad Laboratories Canada Ltd. (Mississauga, Ontario, Canada).The eluted protein sample was collected and dialyzed against the streptokinaseassay buffer (50 mM Tris-HCl [pH 7.2], 0.1 M NaCl, 0.001% [wt/vol] Tween 80).

Preparation of the 37-kDa processing intermediate from streptokinase. Todetermine the N-terminal sequence of the 37-kDa processing intermediate fromstreptokinase, streptokinase was mixed with plasminogen in a 1:1 molar ratio inthe assay buffer for 10 min. The reaction was terminated by adding the sample-loading buffer for SDS-PAGE, and the sample was loaded onto a 12% polyacryl-amide gel containing SDS. The resolved protein bands were then electroblottedto Immobilon membrane as previously described (26). These protein bands werebriefly stained, and the protein band corresponding to the 37-kDa protein was

excised. The first five amino acid residues from this protein was determined atthe Microchemistry Center, University of Victoria.

Determination of the activity of streptokinase and its kinetic parameters forplasminogen activation. The activity of streptokinase was determined by twomethods (48): the colorimetric method (7) with tosyl-glycyl-prolyl-lysine-4-ni-troanilide acetate (Chromozym PL; Boehringer Mannheim Canada, Laval, Que-bec, Canada) as the substrate and the radial caseinolysis method (36) withagarose containing both plasminogen and skim milk. To determine the kineticparameters for the activation of plasminogen by streptokinase and its muteins,the conditions described by Shi et al. (38) were used except that Chromozym PLwas used as the substrate. In these assays, streptokinase or its muteins weremixed with plasminogen at various concentrations (0.02 to 0.4 mM) and thechange in absorbance at 405 nm was monitored at 37°C by using a BeckmanDU65 spectrophotometer equipped with a constant-temperature cuvette cham-ber. The final concentration of streptokinase or its muteins was 0.003 mM. Thekinetic data were analyzed with the mathematic model presented by Wohl et al.(46) and graphed as a Lineweaver-Burk plot. This one-stage assay allows thedetermination of the apparent Michaelis constant (Km) of streptokinase and itsderivatives to plasminogen and the catalytic rate constant (kp) of plasminogenactivation.

Half-life determination. To determine the half-lives of various forms of strep-tokinase in the plasminogen activation process, streptokinase was mixed withplasminogen in a 1:1 molar ratio and samples were collected at different timepoints up to 60 min and added to microcentrifuge tubes containing sampleapplication buffer for SDS-PAGE in a boiling-water bath. SDS-PAGE and West-ern blotting with antibodies against streptokinase from rabbit were performed asdescribed previously (48). To ensure that all the proteins were completely trans-ferred to the nitrocellulose filter, the electroblotted gels were restained withCoomassie blue. Blots with complete protein transfer were used for quantitativeanalysis. Pictures of Western blot were taken with the GDS 7500 gel documen-tation system from UVP, Inc. (San Gabriel, Calif.). The intensity of the 47-kDaprotein which represents the intact form of streptokinase (see Fig. 2) on theWestern blot was quantified with a Fuji bioimaging analyzer (BAS 1000, FujiMedical Systems U.S.A., Stamford, Conn.) and the MacBAS software.

Other methods. Protein concentrations were determined by the Bradfordmethod (2) with reagents from Bio-Rad Laboratories Canada Ltd. Glu-plasmin-ogen was prepared from human plasma by using essentially the lysine-Sepharosemethod (9, 42). To identify colonies that show streptokinase activity, cells wereplated on TBAB agar plates overlaid with a thin layer of agarose (0.5% [wt/vol]agarose in physiological buffered saline with 0.5 mg of plasminogen and 0.1 g ofskim milk in a final volume of 10 ml). Other general chemicals and reagents arefrom Sigma-Aldrich Canada Ltd. (Oakville, Ontario, Canada).

RESULTS

Streptokinase muteins with mutations in the N-terminalregion. To determine whether the plasmin-mediated process-ing of streptokinase at the N-terminal region is an essentialstep in the generation of active streptokinase, the nucleotidesequence AAA, which corresponds to Lys59 in the naturalstreptokinase, was changed to CAA or GAA by site-directedmutagenesis involving inverse PCR. From 200 transformants,28 were randomly selected and spotted on to TBAB agar platesthat had been overlaid with a thin layer of agarose containingboth plasminogen and skim milk. Based on the halo size, thesetransformants were divided into three groups. The first group(14 colonies) had the largest halos. The second group (13colonies) had halos smaller than those in group 1 but stillslightly larger than that for the positive control strain, WB600(pSK3), which produces the wild-type streptokinase. The thirdgroup (1 colony) did not show any halo surrounding the colony.The nucleotide sequence of a 253-bp ClaI-BstEII region whichcovers the predicted mutation was determined from five group1 mutants. They all carried the A-to-C mutation, which con-verts Lys59 to glutamine. No other mutations could be ob-served within the sequenced 253-bp region. Five randomlyselected group 2 mutants also carried the A-to-G mutation,which converts Lys59 to glutamic acid. The only mutant fromgroup 3 was found to carry the same mutation as that observedin the group 1 mutants except for the presence of an unex-pected 1-nucleotide deletion at the 59 end region of the mu-tagenic primer. This introduced a frameshift mutation andprovided an explanation for the failure to observe the strep-tokinase activity from this clone. Since Taq polymerase does

VOL. 64, 1998 PLASMIN-RESISTANT STREPTOKINASE FROM B. SUBTILIS 825

on May 12, 2019 by guest

http://aem.asm

.org/D

ownloaded from

not have the proofreading function (35), it could possibly in-troduce extra mutations to DNA fragments during amplifica-tion. To eliminate the possibility of the presence of extra mu-tations in both the group 1 and group 2 mutants, the 253-bpClaI-BstEII fragment was isolated from mutants in bothgroups and each of these fragments was ligated to the 6-kbClaI-BstEII-digested pSK3, which has never been subjectedto inverse PCR-based mutagenesis. ClaI and BstEII sites wereselected for this fragment exchange reaction because each ofthese sites is unique on pSK3 and they flank the predictedmutation. To confirm the successful introduction of the group1 and group 2 mutations to pSK3, the 253-bp ClaI-BstEII re-gion in the resulting transformants was sequenced. PlasmidspSKN460 and pSKN461 were shown to carry group 1 andgroup 2 mutations, respectively. Consistent with the previ-ous observation, the halos of WB600(pSKN460) and WB600(pSKN461) are larger than that of WB600(pSK3) (data notshown). Since these muteins were produced at a comparablelevel relative to the wild-type streptokinase (data not shown),this observation indicated that these muteins retain relativelygood activity in plasminogen activation. SKN460 was selectedfor further characterization because of its high activity. Thesecretory production of SKN-460 is shown (Fig. 1a, lane 5).

Streptokinase muteins with mutations in the C-terminalregion. To block the plasmid-mediated processing of streptoki-nase at the C-terminal region, Lys386 in streptokinase shouldbe changed to glutamine. As reported previously (48), residualproteases from WB600 could also degrade wild-type streptoki-nase at the C-terminal region and generated a low percentageof degraded streptokinase (Fig. 1a, lane 3). To eliminate theC-terminal degradation, hydrophobic residues at positions 380to 384 were changed to either polar or charged residues andLys386 was changed to glutamine. The resulting streptokinasemuteins can be produced in WB600 in intact form and retainalmost the full activity of the authentic streptokinase (48). Oneof the WB600 strains carrying the mutated streptokinase genein the expression vector (pSKC32) was used here to study the

C-terminal processing event mediated by plasmin. Figure 1a(lane 4) shows the production of this mutein in intact formfrom WB600.

Streptokinase muteins with mutations in both the N- andC-terminal regions. To generate a streptokinase mutein thatshows resistance to the plasmin-mediated processing, the 1.3-kb BstEII-PstI fragment encoding the C-terminal portion ofstreptokinase in pSKN460 was replaced by the one frompSKC32 to generate pSKN460-C32. The successful exchangeof this fragment was confirmed by nucleotide sequencing.WB600(pSKN460-C32) produced this mutein in intact form(Fig. 1a; lane 6). This mutein retains biological activity inplasminogen activation (see Table 1 and Fig. 3a).

Plasmin-mediated processing of streptokinase and its deriv-atives. Natural streptokinase (SK3) and three other streptoki-nase muteins (SKN460, SKC32, and SKN460-C32) producedfrom WB600 strains were purified from the culture superna-tant by electrophoresis on a native polyacrylamide gel. Purifiedstreptokinase proteins were found to be homogeneous (Fig.1b) and were used to study the plasmin-mediated processing bymixing streptokinase with plasminogen in a 1:1 molar ratio.The processing reaction was conducted at 37°C. To avoid thecomplication for the presence of plasminogen and its deriva-tives in the reaction mixture, streptokinase and its processedintermediates were identified by Western blotting with strepto-kinase-specific polyclonal antibodies. As shown in Fig. 2a, nat-ural streptokinase was rapidly converted to various processedforms with molecular masses around 44 kDa. The 37-kDaintermediate could also be observed after 1 min of reactionand became the major product after 10 min of reaction. N-terminal sequencing of the first five amino acid residues fromthe electroblotted 37-kDa protein showed the sequence Ser-Lys-Pro-Phe-Ala. This sequence matched that at positions 60to 64 in the natural streptokinase and confirmed that an N-terminal processing event took place between Lys59 and Ser60.For the streptokinase mutein SKN460, the change of Lys59 toglutamine indeed blocked the major N-terminal processingevent mediated by plasmin. Accumulation of the 44- to 46-kDaprocessing intermediates was observed (Fig. 2b). The 44-kDaproduct was relatively stable and could be observed even after60 min of reaction. This is not the case for the wild-typestreptokinase. At least one new intermediate was detected. Ona relative scale, it migrated faster than the stable 37-kDa in-termediate generated in the reaction with the natural strep-tokinase. For the streptokinase mutein SKN460-C32, this pro-tein showed resistance to plasmin (Fig. 2c). Typical processingintermediates (i.e., the 44- and 37-kDa products) observed withthe natural streptokinase were not detected here. The half-livesof natural streptokinase, SKN460, and SKN460-C32 in thepresence of plasmin generated during the plasminogen activa-tion process were found to be 2, 6.4, and 43 min, respectively.

Steady-state kinetic parameters of plasminogen activationby streptokinase and its muteins. Although the half-life of thestreptokinase mutein SKN460-C32 was extended 21-fold underthe in vitro condition, it is important to examine whether theamino acid changes at the N- and C-terminal regions affect thebinding affinity and the ability of SKN460-C32 to activate plas-minogen. Kinetic parameters for the activation of plasminogenby purified streptokinase and its derivatives (SKN460, SKC32,and SKN460-C32) were determined in three independent deter-minations. As shown in Table 1, the apparent Michaelis con-stant Km and the catalytic rate constant kp for these streptoki-nase proteins were comparable, indicating that these aminoacid changes in streptokinase affect neither the binding nor theactivation of plasminogen.

FIG. 1. Production and purification of streptokinase and its muteins. (a)Western blot of secreted streptokinase. Lanes: 1, prestained markers, withthe molecular mass in kilodaltons shown on the left; 2 to 6, 60 ml of culturesupernatant from WB600(pUB), WB600(pSK3), WB600(pSKC32), WB600(pSKN460), and WB600(pSKN460-C32), respectively. (b) SDS-PAGE analysisof purified streptokinase. Lanes: 1, molecular mass markers; 2 to 5, 5 ml ofpurified SKN460, SKC32, SKN460-C32 and natural streptokinase, respectively.

826 WU ET AL. APPL. ENVIRON. MICROBIOL.

on May 12, 2019 by guest

http://aem.asm

.org/D

ownloaded from

Biological activity of streptokinase and its engineered de-rivatives as determined by radial caseinolysis. In the determi-nation of the steady-state kinetic parameters of streptokinaseand its derivatives, the initial velocity of the reaction was mea-sured. The effects on the extension of the half-lives of theseengineered streptokinase derivatives as plasminogen activatorswill not be reflected in this analysis. Plasmin-resistant strep-tokinase derivatives with longer half-lives would be expected tofunction as plasminogen activators for a longer period, and thisshould be reflected in the radial caseinolysis assay by showinga bigger clearing zone. In this assay, the culture supernatantwith streptokinase or its derivatives was applied in equal quan-tity (confirmed by Western blotting) to individual wells in anagarose gel containing skim milk and plasminogen. Relative to

natural streptokinase as the reference, the engineered deriva-tives SKN460 and SKN460-C32 showed better total activity asplasminogen activators. This was reflected by a 2.2- to 2.5-foldincrease in halo size (Fig. 3). However, the streptokinase mu-tein, SKC32 has a halo size similar to that of the naturalstreptokinase.

DISCUSSION

There are several approaches to prolonging the half-life ofblood clot-dissolving agents. These include the preparationof the streptokinase-acylated plasminogen complex known asAPSAC (40), attachment of polyethylene glycol (5) or maltosebinding protein to streptokinase (15), chemical coupling ofhuman serum albumin to urokinase (3), and site-directed mu-tagenesis of glycosylation sites and domains in tissue plasmin-ogen activator (19, 24). While some of these agents have shownpromising results, others have lower activity or become heter-ogeneous nature because of the chemical modification. As thefirst step to the development of streptokinase with a longerfunctional half-life, we genetically engineered plasmin-resis-tant streptokinase. Since streptokinase is processed N-termi-nally between Lys59 and Ser60 and C-terminally betweenLys386 and Asp387 to generate the 37-kDa intermediate whichretains only 16% of the intact streptokinase activity during theplasminogen activation process (38), lysine residues in thesesites are the logical targets for site-directed mutagenesis. Threeversions of streptokinase were developed in this study. Theyeither carried a single mutation that led to the conversion oflysine to glutamine (SKN460 and SKC32) or a double mutationthat changed both lysine residues to glutamine (SKN460-C32).Glutamine was selected to replace lysine because the length ofits side chain is comparable to that of lysine and so it shouldnot significantly perturb the three-dimensional structure ofstreptokinase. It also does not introduce a positive charge tostreptokinase. Therefore, plasmin with the trypsin-like sub-strate specificity should not cut the engineered streptokinase at

FIG. 2. Processing of streptokinase and its muteins by plasmin. Streptokinaseand plasminogen were mixed in a 1:1 molar ratio and incubated at 37°C. Sampleswere collected at different time points (in minutes) and analyzed by Western blotwith streptokinase-specific polyclonal antibodies. (a) Wild-type streptokinase;(b) SKN460; (c) SKN460-C32. An asterisk marks a new form of intermediategenerated during the processing of SKN460.

FIG. 3. Activity of various forms streptokinase based on the radial caseinoly-sis. Each form of streptokinase, in equal quantities, was loaded into individualwells and incubated at 37°C for 12 h. Numbers indicate the relative activity ofeach form of streptokinase.

TABLE 1. Steady-state kinetic parameters of streptokinaseand its muteins for plasminogen activation

Streptokinase Km (mM) kp (min21)

Wild type 0.20 6 0.02 34.36 6 1.2SKN-460 0.22 6 0.03 32.16 6 1.3SKC-32 0.21 6 0.03 34.02 6 1.2SKN460-C32 0.28 6 0.04 35.17 6 1.4

VOL. 64, 1998 PLASMIN-RESISTANT STREPTOKINASE FROM B. SUBTILIS 827

on May 12, 2019 by guest

http://aem.asm

.org/D

ownloaded from

these sites. This prediction was supported by our processingstudy (Fig. 2) and the observed increase in biological activity ofSKN460 and SKN460-C32 on radial caseinolysis assays (Fig.3). SKN460 with the change of Lys59 to glutamine allowedC-terminal processing events to proceed. The appearance ofthe 46-kDa (Fig. 2b, lane 1 min) and 44-kDa (Fig. 2b, 5 min to60 min) intermediates was consistent with processing atArg401 and Lys386, respectively. Both intact SKN460 andthese intermediates were more stable and could be observedeven after 60 min of reaction. This could be explained by theabolition of the rapid N-terminal processing at Lys59. These44- to 46-kDa intermediates are expected to retain good activ-ity for plasminogen activation since C-terminal deletion of 31amino acid residues from streptokinase does not significantlyaffect the activity for plasminogen activation (18, 20). Thisexpectation is supported by the observation of a 2.2-fold in-crease in the total activity of SKN460 in the radial caseinolysisassay. The faint protein band with a molecular mass around 36kDa (Fig. 2b) could possibly be a fragment with a sequencecorresponding to Ile1 to Lys332 of the intact SKN460. It wasformed by processing of the 44-kDa intermediate at Lys332and could be observed as a transiently accumulated interme-diate because of the blockage of the N-terminal processing site.If it is really the case, this intermediate is unlikely to be activesince residues 244 to 352 (31) and 332 to 386 (38) in streptoki-nase play an important role in mediating tight binding to plas-minogen and residues 332, 334, and 369 to 373 are importantfor plasminogen activation (20, 23, 32, 50).

To block the C-terminal processing of streptokinase by plas-min at Lys386, not only was this lysine residue in SKC32changed to glutamine but also hydrophobic amino acids lo-cated between residues 380 and 384 were converted to aminoacids with hydrophilic side chains. These modifications elimi-nate the proteolytic cleavages within the region of 382 to 384by residual proteases from B. subtilis WB600 during the secre-tory production of streptokinase. As demonstrated previously(48), these modifications do not significantly affect the activityof streptokinase. This is further supported by the determina-tion of the steady-state kinetic parameters observed in thepresent study (Table 1).

The design of SKN460-C32 allows the generation of plas-min-resistant streptokinase and its production in intact formfrom the B. subtilis secretory production system. When bothcritical lysine residues were changed to glutamine, the half-lifeof intact streptokinase during the plasminogen activation pro-cess was greatly extended and SKN460-C32 was apparentlyprocessed at other minor processing sites without generatingany transiently stable intermediates. Although both the appar-ent Michaelis constant and catalytic rate constant were un-changed in this mutein in reference to those of the naturalstreptokinase, radial caseinolysis indicated that SKN460-C32was a better plasminogen activator. This can be explained bythe prolonged half-life of SKN460-C32 as the functional plas-minogen activator. The kinetic parameter measurement didnot reflect any effect of prolonged half-life of SKN460-C32,since only the initial rate was determined in this type of anal-ysis. Although replacement of Lys59 and Lys386 with glu-tamine does not affect either the binding or the catalytic ac-tivity of streptokinase, some lysine residues in streptokinaseare essential for its function. Lysine residues at positions 256and 257 of streptokinase are important for binding to plasmin-ogen, and lysine residues 332 and 334 are required for catalyticactivity (23).

Our results showed that only the conversion of Lys59 toglutamine was important in extending the functional half-lifeof streptokinase. This is consistent with the idea that C-termi-

nal processing at Lys368 does not affect the activity of the44-kDa intermediate to function as an efficient plasminogenactivator. Many pieces of evidence indicate that the first 59amino acids have multiple functional roles for streptokinase.Site-directed mutagenesis of Val19 (22) and Gly24 (21) inac-tivates streptokinase. Residues between Phe37 and Lys51 aresuggested to function as a plasminogen binding site (29). Thefirst 59 residues are also suggested to be required for stabiliz-ing the conformation of streptokinase (38, 50). Without theseN-terminal amino acids, the streptokinase fragment (residues60 to 414) has much lower activity and shows a disorderedsecondary structure (50). Our study also illustrates that theplasmin-mediated proteolytic degradation of streptokinaseleads only to the inactivation of streptokinase as the plasmin-ogen activator. These cleavages are not required to convertstreptokinase to the active form to mediate the plasminogenactivation process. This is opposite to the case for staphyloki-nase, another bacterial plasminogen activator. In that situa-tion, removal of the first 10 amino acid residues from the Nterminus of staphylokinase is essential to generate the activeplasminogen activator (37). Our next target is to examine thein vivo half-life and biodistribution of SKN460-C32 in theexperimental animal system and its efficiency in clot lysis.

ACKNOWLEDGMENTS

We thank Canada Red Cross at Calgary for heparinized blood andDavid A. Hart (Department of Microbiology and Infectious Diseases,University of Calgary) for advice in the preparation of human Glu-plasminogen. Sequence determination for some of the mutated strep-tokinase genes by Louise Tran is greatly appreciated.

This work was supported by a strategic grant from the NaturalSciences and Engineering Research Council of Canada. S.-L. Wong isa senior medical scholar of the Alberta Heritage Foundation for Med-ical Research.

REFERENCES

1. Bajaj, A. P., and F. J. Castellino. 1977. Activation of human plasminogen byequimolar levels of streptokinase. J. Biol. Chem. 252:492–498.

2. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein dye bind-ing. Anal. Biochem. 72:248–254.

3. Breton, J., N. Pezzi, A. Molinari, L. Bonomini, J. Lansen, G. Gonzalez DeBuitrago, and I. Prieto. 1995. Prolonged half-life in the circulation of achemical conjugate between a pro-urokinase derivative and human serumalbumin. Eur. J. Biochem. 231:563–569.

4. Brockway, W. J., and F. J. Castellino. 1974. A characterization of nativestreptokinase and altered streptokinase isolated from a human plasminogenactivator complex. Biochemistry 13:2063–2070.

5. Brucato, F. H., and S. V. Pizzo. 1990. Catabolism of streptokinase andpolyethylene glycol-streptokinase: evidence for transport of intact formsthrough the biliary system in the mouse. Blood 76:73–79.

6. Cannon, C. P., and S. Z. Goldhaber. 1995. The importance of rapidly treat-ing patients with acute myocardial infarction. Chest 107:598–600.

7. Castellino, F. J., J. M. Sodetz, W. J. Brockway, and G. E. Siefring, Jr. 1976.Streptokinase. Methods Enzymol. 45:244–257.

8. Davidson, D. J., D. L. Higgins, and F. J. Castellino. 1990. Plasminogenactivator activities of equimolar complexes of streptokinase with variantrecombinant plasminogens. Biochemistry 29:3585–3590.

9. Deutsch, D. G., and E. T. Mertz. 1970. Plasminogen: purification fromhuman plasma by affinity chromatography. Science 170:1095–1096.

10. Gruppo Italiano Per Lo Studio Della Sopravvivenza Nell’Infarto Mio-cardico. 1990. GISSI-2: a factorial randomised trial of alteplase versus strep-tokinase and heparin versus no heparin among 12,490 patients with acutemyocardial infarction. Lancet 336:65–71.

11. GUSTO Angiographic Investigators. 1993. The effects of tissue plasminogenactivator, streptokinase, or both on coronary-artery patency, ventricularfunction, and survival after acute myocardial infarction. N. Engl. J. Med. 329:1615–1622.

11a.GUSTO Angiographic Investigators. 1993. An international randomizedtrial comparing four thrombolytic strategies for acute myocardial infarction.N. Engl. J. Med. 329:673–682.

12. Halling, S. M., F. J. Sanchez-Anzaldo, R. Fukuda, R. H. Doi, and C. F.Meares. 1977. Zinc is associated with the beta subunit of DNA dependentRNA polymerase of Bacillus subtilis. Biochemistry 16:2880–2884.

828 WU ET AL. APPL. ENVIRON. MICROBIOL.

on May 12, 2019 by guest

http://aem.asm

.org/D

ownloaded from

13. Hemsley, A., N. Arnheim, M. D. Toney, G. Cortopassi, and D. J. Galas. 1989.A simple method for site-directed mutagenesis using the polymerase chainreaction. Nucleic Acids Res. 17:6545–6551.

14. Hennekens, C. H., C. J. O’Donnell, P. M. Ridker, and V. J. Marder. 1995.Current issues concerning thrombolytic therapy for acute myocardial infarc-tion. J. Am. Coll. Cardiol. 25:18S–22S.

15. Houng, A., S. Quen, L.-F. Jean, and G. L. Reed. 1995. Construction of arecombinant streptokinase that resists cleavage and inactivation by plasmin.Thromb. Haemostasis 73:1130. (Abstract.)

16. Huber, K. and G. Maurer. 1996. Thrombolytic therapy in acute myocardialinfarction. Semin. Thromb. Hemostasis 22:15–26.

17. ISIS-3 Collaborative Group. 1992. ISIS-3: A randomised comparison ofstreptokinase vs tissue plasminogen activator vs anistreplase and of aspirinplus heparin vs aspirin alone among 41 299 cases of suspected acute myo-cardial infarction. Lancet 339:753–770.

18. Jackson, K. W., H. Malke, D. Gerlach, J. J. Ferretti, and J. Tang. 1986.Active streptokinase from the cloned gene in Streptococcus sanguis is withoutthe carboxyl-terminal 32 residues. Biochemistry 25:108–114.

19. Keyt, B. A., N. F. Paoni, C. J. Refino, L. Berleau, H. Nguyen, A. Chow, J. Lai,L. Pena, C. Pater, and J. Ogez. 1994. A faster-acting and more potent formof tissue plasminogen activator. Proc. Natl. Acad. Sci. USA 91:3670–3674.

20. Kim, I. C., J. S. Kim, S. H. Lee, and S. M. Byun. 1996. C-terminal peptideof streptokinase, Met369-Pro373, is important in plasminogen activation.Biochem. Mol. Biol. Int. 40:939–945.

21. Lee, B. R., S. K. Park, J. H. Kim, and S. M. Byun. 1989. Site-specificalteration of Gly-24 in streptokinase: its effect on plasminogen activation.Biochem. Biophys. Res. Commun. 165:1085–1090.

22. Lee, S. H., S. T. Jeong, I. C. Kim, and S. M. Byun. 1997. Identification of thefunctional importance of valine-19 residue in streptokinase by N-terminaldeletion and site-directed mutagenesis. Biochem. Mol. Biol. Int. 41:199–207.

23. Lin, L. F., S. Oeun, A. Houng, and G. L. Reed. 1996. Mutation of lysines ina plasminogen binding region of streptokinase identifies residues importantfor generating a functional activator complex. Biochemistry 35:16879–16885.

24. Madison, E. 1994. Probing structure-function relationships of tissue-plas-minogen activator by site-specific mutagenesis. Fibrinolysis 8:221–236.

25. Malke, H., and J. J. Ferretti. 1984. Streptokinase: cloning, expression, andexcretion by Escherichia coli. Proc. Natl. Acad. Sci. USA 81:3557–3561.

26. Matsudaira, P. 1987. Sequence from picomole quantities of proteins elec-troblotted onto polyvinylidene difluoride membranes. J. Biol. Chem. 262:10035–10038.

27. Maynard, C., R. Althouse, M. Olsufka, J. L. Ritchie, K. B. Davis, and J. W.Kennedy. 1989. Early versus late hospital arrival for acute myocardial infarc-tion in the Western Washington Thrombolytic Therapy Trials. Am. J. Car-diol. 63:1296–1300.

28. McClinock, D. K., M. E. Englert, C. Dziobkowski, E. H. Snedeker, and P. H.Bell. 1974. Two distinct pathways of the streptokinase-mediated activation ofhighly purified human plasminogen. Biochemistry 13:5334–5344.

29. Nihalani, D., and G. Sahni. 1995. Streptokinase contains two independentplasminogen-binding sites. Biochem. Biophys. Res. Commun. 217:1245–1254.

30. Rawles, J. 1994. Halving of mortality at 1 year by domiciliary thrombolysis inthe Grampian Region Early Anistreplase Trial (GREAT). J. Am. Coll.Cardiol. 23:1–5.

31. Reed, G. L., L.-F. Lin, B. Parhami-Seren, and P. Kussie. 1995. Identificationof a plasminogen binding region in streptokinase that is necessary for thecreation of a functional streptokinase-plasminogen activator complex. Bio-chemistry 34:10266–10271.

32. Rodrıguez, P., P. Fuentes, M. Barro, J. G. Alvarez, E. Munoz, D. Collen, andH. R. Lijnen. 1995. Structural domains of streptokinase involved in theinteraction with plasminogen. Eur. J. Biochem. 229:83–90.

33. Rozenman, Y., M. Gotsman, T. Weiss, C. Lotan, M. Mosseri, D. Sapoznikov,S. Welber, H. Nassar, Y. Hasin, and D. Gilon. 1994. Very early thrombolysisin acute myocardial infarction—a light at the end of the tunnel. Isr. J. Med.Sci. 30:99–107.

34. Rozenman, Y., M. S. Gotsman, A. T. Weiss, C. Lotan, M. Mosseri, D.Sapoznikov, S. Welber, Y. Hasin, and D. Gilon. 1995. Early intravenousthrombolysis in acute myocardial infarction: the Jerusalem experience. Int.J. Cardiol. 49(Suppl.):S21–S28.

35. Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn,K. B. Mullis, and H. A. Erlich. 1988. Primer-directed enzymatic amplifica-tion of DNA with a thermostable DNA polymerase. Science 239:487–491.

36. Saksela, O. 1981. Radial caseinolysis in agarose: a simple method for detec-tion of plasminogen activators in the presence of inhibitory substances andserum. Anal. Biochem. 111:276–282.

37. Schlott, B., K. H. Guhrs, M. Hartmann, A. Rocker, and D. Collen. 1997.Staphylokinase requires NH2-terminal proteolysis for plasminogen activa-tion. J. Biol. Chem. 272:6067–6072.

38. Shi, G.-Y., B.-I. Chang, S.-M. Chen, D.-H. Wu, and H.-L. Wu. 1994. Functionof streptokinase fragments in plasminogen activation. Biochem. J. 304:235–241.

39. Siefring, G. E., Jr. and F. J. Castellino. 1976. Interaction of streptokinasewith plasminogen: isolation and characterization of a streptokinase degra-dation product. J. Biol. Chem. 251:3913–3920.

40. Smith, R. A., R. J. Dupe, P. D. English, and J. Green. 1981. Fibrinolysis withacyl-enzymes: a new approach to thrombolytic therapy. Nature 290:505–508.

41. Steinmetz, M., D. LeCoq, S. Aymerich, G. Gonzy-Treboul, and P. Gay. 1985.The DNA sequence of the secreted Bacillus subtilis enzyme levansucrase andits genetic control sites. Mol. Gen. Genet. 200:220–228.

42. Summaria, L., L. Arzadon, P. Bernabe, K. C. Robbins, and G. Barlow. 1973.Characterization of the NH2-terminal glutamic acid and NH2-terminal lysineforms of human plasminogen isolated by affinity chromatography and iso-electric focusing methods. J. Biol. Chem. 248:2984–2991.

43. Wang, L.-F., and R. H. Doi. 1987. Promoter switch during development andthe termination site of the s43 operon of Bacillus subtilis. Mol. Gen. Genet.207:114–119.

44. Weaver, W. D., M. Cerqueira, A. P. Hallstrom, P. E. Litwin, J. S. Martin,P. J. Kudenchuk, and M. Eisenberg. 1993. The Myocardial Infarction Triageand Intervention Project Group: prehospital-initiated vs hospital-initiatedthrombolytic therapy. The Myocardial Infarction Triage and InterventionTrial. JAMA 270:1211–1216.

45. Weinstein, M. J. and R. F. Doolittle. 1972. Differential specificities of throm-bin, plasmin and trypsin with regard to synthetic and natural substrates andinhibitors. Biochim. Biophys. Acta 258:577–582.

46. Wohl, R. C., L. Summaria, and K. C. Robbins. 1980. Kinetics of activation ofhuman plasminogen by different activator species at pH 7.4 and 37 degreesC. J. Biol. Chem. 255:2005–2013.

47. Wong, S.-L. 1989. Development of an inducible and enhancible expressionand secretion system in Bacillus subtilis. Gene 83:215–223.

48. Wong, S.-L., R. Ye, and S. Nathoo. 1994. Engineering and production ofstreptokinase in a Bacillus subtilis expression-secretion system. Appl. Envi-ron. Microbiol. 60:517–523.

49. Wu, X.-C., W. Lee, L. Tran, and S.-L. Wong. 1991. Engineering a Bacillussubtilis expression-secretion system with a strain deficient in six extracellularproteases. J. Bacteriol. 173:4952–4958.

50. Young, K.-C., G.-Y. Shi, Y.-F. Chang, B.-I. Chang, L.-C. Chang, M.-D. Lai,W.-J. Chuang, and H.-L. Wu. 1995. Interaction of streptokinase and plas-minogen—studied with truncated streptokinase peptides. J. Biol. Chem. 270:29601–29606.

VOL. 64, 1998 PLASMIN-RESISTANT STREPTOKINASE FROM B. SUBTILIS 829

on May 12, 2019 by guest

http://aem.asm

.org/D

ownloaded from