Functional properties of recombinant staphylokinase variants obtained by site-specific mutagenesis of methionine-26

BH Biochimica et Biophysica Acta 1204 (1994) 235-242

ELSEVIER

etBiochi~i Biophysicat~ta

Functional properties of recombinant staphylokinase variants obtained by site-specific mutagenesis of methionine-26 Bernard Schlott a,1, Manfred Hartmann a, Karl-Heinz Giihrs a, Eckehart Birch-Hirschfeld Ariane Gase a, Stephan Vettermann a, D6sir6 Collen b,,, Henri R. Lijnen b a

a,

Institut fiir Molekulare Biotechnologie, Beutenbergstrasse 11, 0-7745 Jena, Germany, b Center for Molecular and Vascular Biology, University ofLeuven, Leuven, Belgium

(Received 20 July 1993)

Abstract Variants of recombinant staphylokinase (Sak) were produced by site-specific mutagenesis of the unique Met-26 residue and purified to homogeneity from the cell extract of transformed E. coli. The desired mutations were confirmed by cDNA and amino-acid sequence analysis. Sak-M26L, Sak-M26C, Sak-M26R, Sak-M26V and Sak-M26A were selected for further analysis on the basis of their plasminogen activating activity. The specific fibrinolytic activities of Sak-M26L, Sak-M26C and Sak were comparable (76000_+ 10000, 75000_+2400 and 78000+9700 H U / m g , respectively; mean_+ S.E., n = 3 or 4). Active site exposure in equimolar (4.5/,M) mixtures with plasminogen at room temperature was more rapid with Sak-M26L than with Sak (quantitative exposure within 4 min and 8 min, respectively). Activation of 1 /,M plasminogen by catalytic amounts (5 nM) of Sak-M26L initially appeared to be somewhat faster, but comparable 50 to 60% activation was obtained within 30 min. In contrast, Sak-M26R and Sak-M26V were virtually inactive, did not form active complexes with plasminogen and did not activate plasminogen. The catalytic efficiencies for plasminogen activation were comparable for plasmin-Sak-M26L, plasmin-Sak-M26C and plasmin-Sak (0.14/,M -1 s - I , 0 . 1 6 / , M -1 s -1 or 0.12 IzM -1 s -1, respectively). Comparable dose-dependent lysis of 0.06 ml 125I-fibrin labeled human plasma clots submerged in 0.3 ml human plasma was obtained with Sak-M26L, Sak-M26C and Sak (concentration required for 50% lysis in 2 h, ECs0, of 17 + 1.6 nM, 19 -+ 1.4 nM and 14 + 2.5 nM, respectively), whereas Sak-M26R or Sak-M26V were inactive. Sak-M26A did not form a stable complex with plasminogen, as shown by gel filtration. These data establish that substitution of the unique Met residue in position 26 of the Sak sequence with Leu or Cys has little or no influence on its plasminogen activating or fibrinolytic potential. In contrast, substitution of Met-26 with either Arg or Val results in total loss of the functional activity. Thus, the amino acid in position 26 of Sak appears to be of crucial importance for the activation of plasminogen by staphylokinase. Key words: Plasminogen activation; Staphylokinase; Site-specific mutagenesis

I. Introduction R e c o m b i n a n t s t a p h y l o k i n a s e (Sak) i n d u c e s fibrinspecific clot lysis in a h u m a n p l a s m a m i l i e u [1-3]. It forms a 1 : 1 s t o i c h i o m e t r i c c o m p l e x with p l a s m i n ( o g e n ) t h a t activates o t h e r p l a s m i n o g e n m o l e c u l e s [2,4,5]. Exp o s u r e o f t h e active site in t h e c o m p l e x o f Sak with p l a s m i n o g e n r e q u i r e s c o n v e r s i o n o f p l a s m i n o g e n to

* Corresponding author. Fax: + 32 16 345990. 1 Address reprint requests to this author. 0167-4838/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0167-4838(93)E0200-L

Abbreviations: Sak, recombinant staphylokinase translated from the Sak42D gene; Sak-M26A, Sak with Met-26 mutagenized to Ala; Sak-M26C, Sak with Met-26 mutagenized to Cys; Sak-M26G, Sak with Met-26 mutagenized to Gly; Sak-M26H, Sak with Met-26 mutagenized to His; Sak-M26I, Sak with Met-26 mutagenized to Ile; Sak-M26K, Sak with Met-26 mutagenized to Lys; Sak-M26L, Sak with Met-26 mutagenized to Leu; Sak-M26R, Sak with Met-26 mutagenized to Arg;Sak-M26S, Sak with Met-26 mutagenized to Ser; Sak-M26T, Sak with Met-26 mutagenized to Thr; Sak-M26V, Sak with Met-26 mutagenized to Val; S-2251, D-VaI-Leu-Lys-NH-PhNO2;ECs0, concentration of plasminogen activator required for 50% lysis in 2 h of a human plasma clot submerged in human plasma; HU, home units of activity determined by comparison with an in-house standard of natural staphylokinase, which was assigned an activity of 100000 HU/mg; Pig, plasminogen; S.E., standard error of the mean.

236

B. Schlott et al. / Biochimica et Biophysica Acta O0 (1994) 000-000

plasmin [6] and the plasmin-Sak complex is very rapidly neutralized by a2-antiplasmin [2,7]. Sak has recently been shown to offer promise for in vivo thrombolysis [8-11]. It has been shown previously that wild-type Sak and a variant lacking the 10 NH2-terminal amino acids have the same plasminogen activating potential in purified systems and a comparable fibrinolytic activity in human plasma [3,12]. In contrast, removal of the COOH-terminal lysine was found to yield an inactive Sak mutant, indicating that the COOH-terminal end is sensitive to deletion mutagenesis (Schlott et al., unpublished results). More detailed information on the structure-function relations which determine the interaction of Sak with plasminogen is, however, not available. The Sak molecule has been shown to consist of two folded domains of similar size [13]. Sak has a unique Met-residue at position 26 of its primary structure [14-16]. Substitution of Met-26 by another amino acid and introduction of a new Met residue closer to the middle of the structure, might therefore allow to separate the two halves of the molecule by treatment with CNBr. In a first step towards this goal, we have in the present study prepared mutants of Sak by site-specific mutagenesis of the unique Met residue in position 26 and evaluated their functional properties in purified systems and in human plasma in vitro. Surprisingly, we have found that the amino acid in position 26 of Sak plays a crucial role in the interaction with plasminogen.

2. Materials and methods Production o f Sak mutants. Recombinant staphylokinase (Sak) was purified from the cell extract of E. coli transformed with an expression vector containing the Sak42D gene cloned from bacteriophage ~b42D [15,17]. Site-specific mutagenesis of Met-26 to other amino acids (Ala, Arg, Cys, Gly, His, Ile, Leu, Lys, Ser, Thr, Val) was performed using three multicloning steps with appropriate oligonucleotide linker couples. In the first step the coding sequence of amino acids 1 to 22 of the Sak42D gene including the portable translation signal sequence was constructed. The second step included the construction of the complete coding sequence downstream from amino-acid position 32 of the Sak42D gene. Simultaneously, silent mutations creating restriction sites for NaeI, SalI, SacI, SacII and Sty I were introduced at the codons 22/23, 32/33, 38/39, 42/43 and 129/130, respectively. In a final step the mutations at amino-acid position 26 were performed by insertion of two types of randomized oligonucleotide spacer couples between the NaeI and SalI sites. At codon 26 the spacers contained the randomized sequences (AGC)(AGCT)A or (GCT)(AGCT)C

coding for the amino acids Cys, Ala, Ser, His, Asp, Tyr and Phe or for lie, I_eu, Lys, Thr, Arg, Glu, Gly, Ala, Val, Gin and Pro, respectively. For the present study the genes coding for the Sak42D mutants Sak-M26A, Sak-M26C, Sak-M26G, Sak-M26H, Sak-M26I, SakM26K, Sak-M26L, Sak-M26R, Sak-M26S, SakM26T and Sak-M26V were introduced into the expression vector pMEX6 (Medac, Hamburg, FRG). The mutant proteins were produced intracellularly and in soluble form in E. coli TG1 cells transformed with these vectors [18]. The desired mutations were confirmed by double stranded plasmid-DNA sequencing according to Sanger et al. [19] of the complete Sak42D gene. Purification of the mutants was performed as follows. Sak-M26 mutants, as well as the wild-type protein Sak 42D were purified from sonicated bacterial extracts using cation exchange- and hydrophobic interaction chromatography, as described elsewhere [13]. The protein concentrations were determined by the method of Bradford [20], the plasminogen activating activity by a plasminogen-coupled chromogenic substrate assay, and the specific fibrinolytic activities by a fibrin clot lysis assay, as described previously [12]. Peptide mapping of tryptic digests of Sak-M26 mutant proteins was performed using the HPLC-system LC 10a from Shimadzu. Tryptic digests were prepared at an enzyme/protein ratio of 1:2.5 in 0.1 M (NH4)HCO3-buffer (pH 8.3), containing 3 M urea. Incubation was carried out at 37°C for 4 h. Tryptic peptides were separated on a 0.46 × 25 cm C-18 Vydac HPLC-column (The Separation Group, Hesperia, CA, USA), using an acetonitrile gradient in trifluoroacetic acid. Peptides were freeze dried and sequenced on an Applied Biosystems protein sequencer model 476a. Other proteins and reagents. Native human plasminogen (Glu-plasminogen) was purified from plasma and characterized, as described elsewhere [21,22]. It was treated with aprotinin-Sepharose to remove traces of contaminating plasmin (detection limit < 0.005% with S-2251). Plasma was pooled citrated human plasma obtained from at least five healthy blood donors. 1251fibrinogen was purchased from Amersham (Buckinghamshire, UK), the chromogenic substrate S-2251 from Chromogenix (Antwerp, Belgium) and the active site titrant p-nitrophenyl-p'-guanidinobenzoate (NPGB) from E. Merck (Darmstadt, Germany). Aprotinin-Sepharose was prepared by coupling aprotinin (Trasylol®, Bayer, Leverkusen, Germany) to CNBr-activated Sepharose 4B. Analytical techniques. SDS-PAGE was performed with the Phast-SystemT M (Pharmacia, Uppsala, Sweden) using 10 to 15% gradient gels for the analysis of the chain composition of plasminogen or 20% highdensity gels for the analysis of the molecular size of Sak. Reduction of samples was performed by heating

B. Schlott et al. / Biochimica et Biophysica Acta O0 (1994) 000-000

at 100°C for 3 min in the presence of 1% SDS and 1% dithioerythritol. Fibrinogen levels in plasma samples were monitored with a clotting rate assay [23] and Sak-related antigen was measured by ELISA [9]. Complex formation with plasminogen. The generation of an active site in complexes of plasminogen with Sak moieties was determined by titration with NPGB in 1 ml cuvettes in a Pye Unicam SP 1800 spectrophotometer at 410 nm. A 0.1 M Veronal buffer (pH 8.3), containing 0.1 M arginine, was used at room temperature, following the instructions of Chase and Shaw [24] and McClintock and Bell [25]. Concentrated stock solutions were diluted in buffer in the cuvette (final volume of 1 ml) to final concentrations of 4.5 ~M for plasminogen and 5 ~M for the Sak moieties and NPGB was added to a final concentration of 100 /~M at different time points after mixing of plasminogen and Sak (0-20 min). The concentration of active site was determined from the 'burst' of p-nitrophenol, using a molar extinction coefficient of 16700 [24]. Samples were removed 5 min after addition of NPGB and subjected to reduced SDS-gel electrophoresis.

Amidolytic activity of the plasmin-Sak complexes. Plasminogen (final concentration 1/zM) was incubated with the Sak moieties (final concentration 1.1 /~M) at 37°C in 0.1 M phosphate buffer (pH 7.4), containing 0.01% Tween 80. At different time intervals (0-10 min) generation of an active site in the plasmin-Sak complex was measured with S-2251 (final concentration 1 mM) after 100-fold dilution of samples in 0.05 M Tris.HCl buffer (pH 7.4), containing 0.038 M NaCI and 0.01% Tween 80. The kinetic parameters for the hydrolysis of S-2251 (final concentration 0.1 to 2.0 mM) at 37°C by equimolar plasmin-Sak complexes (final concentration 10 or 25 nM in 0.05 M Tris-HCl buffer (pH 7.4), containing 0.038 M NaCI and 0.01% Tween 80), were determined by Lineweaver-Burk analysis. In additional experiments, plasminogen (final concentration 250 nM) was preincubated for 10 min at 37°C with Sak-M26V (final concentration 250 nM-1 /~M), followed by addition of wild-type Sak (final concentration 250 nM) and measurement of active-site exposure as described above. Activation of plasminogen by Sak moieties. Activation of plasminogen (final concentration 1 /zM) by catalytic amounts of Sak moieties (final concentration 5 nM) was monitored at 37°C in 0.1 M phosphate buffer (pH 7.4), containing 0.01% Tween 80. Therefore, at different time intervals (0-30 min), generated plasmin was measured with S-2251 (final concentration 1 mM) after 50-fold dilution of samples. For kinetic analysis, equimolar plasmin-Sak complexes (final concentration approx. 2 /.tM) were prepared by incubation of plasminogen with Sak moieties at 37°C for 10 min in 0.1 M phosphate buffer (pH 7.4),

237

containing 25 percent glycerol; the mixtures were then stored on ice. Plasmin-Sak complexes (final concentration 5 nM) were incubated with human plasminogen (final concentration 1 to 33 /xM) at 37°C in 0.1 M phosphate buffer (pH 7.4). Generated plasmin was measured at different time intervals (0-4 min) with S-2251 (final concentration 1 raM) after at least 20-fold dilution of the sample. Initial activation rates were obtained from plots of the concentration of generated plasmin versus time.

Formation of Sak-plasminogen complex as monitored by gel filtration chromatography. Mixtures of Sak42D or Sak-M26A with excess plasminogen (6.5/zM Sak moiety) were incubated for 10 min at 37°C. The samples were loaded on a Superdex T M 16/60 column (Pharmacia) and chromatograped at a flow rate of 0.5 m l / min, using a FPLC-system (Pharmacia). Protein elution was monitored by continous measurement of UV-light absorption at 280 nm.

Fibrinolytic and fibrinogenolytic properties of Sak moieties in human plasma in vitro. 125I-fibrin labeled plasma clots were prepared from human plasma following addition of 500000 c p m / m l of 125I-labeled human fibrinogen and coagulation with CaC12 (final concentration 25 mM) and thrombin (final concentration 2 NIH U/ml). Lysis of 125I-fibrin labeled plasma clots (0.06 ml volume) by addition of different concentrations of Sak moieties (final concentration range 0 to 250 nM) in 0.3 ml human plasma, was measured after 2 h as previously described [3]. The concentration of Sak required to obtain 50 percent clot lysis in 2 h (EC50), was determined from plots of percent lysis versus the concentration of Sak. Residual fibrinogen levels after 2 h at ECs0 were determined from plots of residual fibrinogen versus the concentration of Sak.

3. Results

Production of Sak mutants Sak mutants were purified from the cell extract of

E. coli and obtained with yields ranging between 4 and 30 mg/l, representing recoveries of 30 to 50% of the starting material. The plasminogen activating activities of the Sak mutants, determined with a plasminogen-coupled chromogenic assay (using 40/zM plasminogen and varying concentrations of Sak moiety in the nM range), ranged between 0.3 and 35/.~mol plasmin//xmol Sak per min, as compared to 16 /zmol plasmin/~mol Sak per min for wild-type Sak (Fig. 1). On the basis of these results, Sak-M26L, Sak-M26C, Sak-M26R, Sak-M26V and in some experiments Sak-M26A, were selected for further functional analysis. Wild-type Sak and all Sak-M26 mutants described were obtained as homogeneous pro-

238

B. Schlott et aL / Biochimica et Biophysica Acta O0 (1994) 000-000

40

1

35

2

3

4

5

6

7

8

9

10

, 25

15 10

0

~

~

Fig. 1. Plasminogen activating activities of Sak moieties, determined in a plasminogen-coupledchromogenicsubstrate assay, using 40/xM plasminogen and varying concentrations Sak moieties in the nM range. The data are expressed as /zmol plasmin generated per min per/xmol Sak.

teins, migrating with an apparent M r of 18 000 on 20% S D S - P A G E (Fig. 2). Peptides containing amino-acid residue 26 were isolated by peptide mapping on H P L C (Fig. 3) and the sequence at position 26 was verified, to confirm the desired mutation.

Fig. 2. SDS-PAGE under non-reducing conditions of purified Sak moieties using 20% high density gels and silver staining. Lane h Sak42D; lane 2: Sak42D cleaved with CNBr at Met 26; lane 3: Sak-M26L; lane 4: Sak-M26C; lane 6: Sak-M26A;lane 7: Sak-M26H; lane 8: Sak-M26V; lane 9: SakM26T; lane 10: Sak-M26G. Lane 5 represents a protein calibration mixture consisting of bovine serum albumin (66 kD), egg ovalbumin (45 kD), glyceraldehyde-3-phosphate dehydrogenase(36 kD), carbonic anhydrase (29 kD), trypsinogen (PMSF treated) (24 kD), trypsin inhibitor (20.1 kD) and tzlactalbumin (14.4 kD). All the mutants are purified to homogeneity and migrate with an apparent Mr of 18000. 8 min) and with Sak-M26L (within 4 min), whereas no significant conversion within 20 min was observed in the experiments with Sak-M26R or Sak-M26V (Fig. 4,

Specific activity The specific fibrinolytic activities (expressed in HU, determined by comparison with an in house standard of natural staphylokinase, per mg protein, as determined with the Bradford assay) were (mean + S.E.; n = 3 or 4) 78000_+9700 H U / m g , 76000_+10000 H U / m g or 75000 _+ 2400 H U / m g for Sak, Sak-M26L or Sak-M26C, as compared to 1500 H U / m g and 400 H U / m g for Sak-M26R or Sak-M26V, respectively.

X-~g ~

J

x

i

x

x

Complex formation with plasminogen When the Sak moieties (final concentration 5 /~M) were added to mixtures of plasminogen (final concentration 4.5 /zM) and NPGB (final concentration 100 /xM) in 0.1 M Veronal buffer (pH 8.3), containing 0.1 M arginine, titratable active sites were not generated in any of the mixtures. However, when NPGB was added after preincubation of plasminogen and Sak for different time periods (0-480 s), a progressive generation of active sites was observed with a close to 1:1 stoichiometric recovery of active site in the mixtures with Sak and with Sak-M26L (Fig. 4). Active site exposure was more rapid with Sak-M26L than with wild-type Sak (quantitative exposure within 4 min or 8 min, respectively). In contrast, preincubation of plasminogen with Sak-M26R or Sak-M26V for up to 20 min did not result in exposure of measurable active sites. SDSP A G E using 10-15% gradient gels under reducing conditions, revealed quantitative conversion of plasminogen to plasmin in the experiments with Sak (within

]PVal ~

X" Leu ~

--

i

= ~ e t -,-...., 20

t0

(~

mill

SSSFDK GK YK K GDDASYFEPTGPYLXVNVTGVDGK R NELLSPR T1

T2

T3

T4

T5

YVEFPIKPGTTLTK EK IEYYVEWALDATAYK EFR VVELDPSAK 1"6 T7 T8 1"9 T10

IEV'WYDK NK K K EETK SFPITEK GFVVPDLSEHIK NPGFNLITK T11

T12

T13

T14

T15

T16

VVlGK K "1"17 Fig. 3. HPLC-patterns of tryptic digests of Sak and Sak-M26 mutants with identification of the peptides containing amino-acid 26.

B. Schlott et al. / Biochimica et Biophysica Acta O0 (1994) 000-000

239

// 5 s

4

/

v

OJ .4.J °--

G) >

2

A

B

C

D

s

2 8 2 4

3 20 3 20

min

A B 2 8 2 4

C D s 3 2 0 3 20

min

/ s

,i

0 1

2.

0

0

I

I

I

I

100

200

300

I

//

A _

,

4.00 500

1200

Time Fig. 4. Active-site titration of mixtures of plasminogen and Sak with NPGB at room temperature in 0.1 M Veronal buffer (pH 8.3), containing 0.1 M arginine. Plasminogen (final concentration 4.5 ~ M ) was incubated with Sak moieties (final concentration 5/zM) for different time periods (0-20 rain) before addition of NPGB (final concentration 100/~M). The symbols used are: ( • ) , Sak; ( , ) , Sak-M26L; (o), Sak-M26R; and ( • ) , Sak-M26V. The inset shows SDS-PAGE under reducing conditions, using 10-15% gradient gels (I) or 20% homogeneous gels (II), of samples taken from the cuvettes after addition at different time points of NPGB to mixtures of plasminogen with Sak (lanes A), Sak-M26L (lanes B), Sak-M26R (lanes C) or Sak-M26V (lanes D). Lanes S represent a protein calibration mixture consisting of phosphorylase b ( M r 97000), albumin ( M r 67000), ovalbumin ( M r 45 000), carbonic anhydrase ( M r 30000), trypsin inhibitor (M r 20100) and a-lactalbumin (M r 14400).

inset, panel I). SDS-PAGE using 20% gels under reducing conditions, revealed quantitative conversion of Sak to a lower M r derivative in the experiments with Sak or Sak-M26L and partial conversion in the experiment with Sak-M26R, but no significant conversion in the experiment with Sak-M26V(Fig. 4, inset, panel II).

Amidolytic activity of plasmin-Sak complexes In equimolar mixtures of plasminogen with Sak, Sak-M26L or Sak-M26C (1/zM each), an active site is rapidly and progressively exposed, as monitored with the chromogenic substrate S-2251. In contrast, no amidolytic activity was generated in equimolar mixtures of

I::

E 100

:,

~-i--~---,--,---i--m

?

200 150

D/~

~ v,,.~v ~ v ' ' ' - ' ' ' ' ~

.

.

.

.

.

.

..=

>

o

•~°

loo

_~ o .'2_

so

E .,£

0

.Z'I

.

~

4o

E m _~

2o

c

/ .

.

.

.

.

-------i------

'i--

2

4 Time

.

.

.

--

.

I 6

.

.

.

. --I

. a

'1"-

o.

10

(rnin)

Fig. 5. Generation of active sites in equimolar mixtures of plasminogen with Sak ( • ) , Sak-M26L (.), Sak-M26C ( v ) , Sak-M26R (e) or Sak-M26V ( • ) (final concentration 1 /zM). Amidolytic activity was measured with S-2251 (final concentration 1 raM) after 50-fold dilution of the samples. The data are mean + S.E. of three experiments.

0

i 0

l

5

10

15

20

25

30

Time (rnin) Fig. 6. Activation of plasminogen (final concentration 1 ~tM) as a function of time by Sak ( l ) , Sak-M26L ( , ) , Sak-M26C (*), SakM26R (*) or Sak-M26V ( • ) (final concentration 5 nM each). The data are mean :t: S.E. of three experiments.

B. Schlott et al. / Biochimica et Biophysica Acta O0 (1994) 000-000

240

did not impair subsequent active site exposure following addition of wild-type Sak at equimolar concentration to plasminogen (data not shown). The kinetic constants for the hydrolysis of S-2251 by plasmin-Sak, plasmin-Sak-M26L or plasmin-Sak-M26C, as determined by Lineweaver-Burk analysis, were (mean of two determinations) K m = 0.48 mM and k 2 = 28 S - 1 , K m = 0.46 mM and k 2 = 32 S - 1 o r K m = 0 . 5 8 mM and k 2 = 23 s -l, respectively. The catalytic efficiency of the plasmin-Sak-M26L complex for the hydrolysis of S-2251 thus is slightly higher than that of the plasmin-Sak or the plasmin-Sak-M26C complex (70 mM -~ s -1, as compared to 58 mM -~ s -l, or to 40 mM-1 s-l).

pig + SAK M26A _

_

(~)

pig + SAK 42D

(2)

E

¢

o

f

SAK M26A

(a)

Activation of plasminogen by Sak moieties

0 o c q 0 w rJ

f~

<

~(~

SAK 42D

(4)

pig

~ _ _

(5)

I

I

I

I

I

I

20

40

60

80

1O0

120

Elutlon

Following an initial lag phase of approx. 4 min, Sak, Sak-M26L and Sak-M26C caused comparable progressive plasminogen activation (Fig. 6). Plasminogen activation appeared to be slightly more rapid with SakM26L than with Sak or Sak-M26C, but comparable results were obtained after 25 to 30 rain (50 to 60% activation). No significant plasminogen activation was obtained with either Sak-M26R or Sak-M26V. Kinetic analysis revealed that activation of plasminogen to plasmin by plasmin-Sak, plasmin-Sak-M26L and plasmin-Sak-M26C obeyed Michaelis-Menten kinetics, as shown by lineair double-reciprocal plots of the initial activation rate versus the plasminogen concentration (not shown). The kinetic constants, obtained by lineair regression analysis (mean of two determinations with r _> 0.98) were K m = 20 /xM and k 2 = 2.5 s -1 for plasmin-Sak, as compared to K m = 17/zM and k 2 = 2.4 s -1 for plasmin-Sak-M26L and to K m = 14

volume (ml)

Fig. 7. Gel filtration on Superdex T M 16//60 of mixtures of plasminogen and Sak-M26A (i) or Sak42D (2). The elution patterns of Sak-M26A (3), Sak42D (4) and Pig (5) are also shown.

plasminogen with Sak-M26R or Sak-M26V (Fig. 5). Furthermore, preincubation of plasminogen with up to a 4-fold molar excess of Sak-M26V at 37°C for 10 min,

~jv

A

100

80

8O

can o

:

,°

.c

6o

J~

40

-6

40

2O

~

20

........

10 0

B

100

qt---7.~.l

101 Concentration

, • .......

10 2 (riM)

o

10 a

10 0

10 1 Concentrat;on

102

10 3

(nM)

Fig. 8. Lysis of h u m a n 125I-fibrin labeled plasma clots submerged in h u m a n plasma (panel A) 2 h after addition of different concentrations of Sak (Ill), Sak-M26L (e), Sak-M26C ( v ) , Sak-M26R (e) or Sak-M26V ( * ) . Clot lysis was monitored from the release of 125I-labeled fibrin degradation products and is expressed in percent. Residual fibrinogen levels after 2 h (panel B) are expressed in percent of the baseline value. The data are m e a n 4- S.E. of three experiments.

B. Schlott et al. / Biochimica et Biophysica Acta O0 (1994) 000-000

/zM and k 2 = 2.2 s -~ for plasmin-Sak-M26C. The catalytic efficiencies of plasmin-Sak-M26L, plasmin-SakM26C and plasmin-Sak for plasminogen activation thus were comparable (0.14 ~M -1 s -1, 0.16/~M -1 S - 1 and 0.12 ~M -~ s -1, respectively). In contrast, equimolar mixtures of plasminogen and Sak-M26V or Sak-M26R (final concentration 2 tzM), preincubated for 10 min at 37°C, did not activate plasminogen under the same experimental conditions (data not shown).

Complex formation between Sak moieties and plasminogen as shown by gel filtration chromatography The formation of a Sak-plasmin(ogen) complex was demonstrated by gel filtration of mixtures of wild-type Sak42D and plasminogen. This complex is characterized by a shift of the elution volume towards lower values as compared to the elution volume of plasminogen in control experiments. In case of mixtures of Sak-M26A and plasminogen both components were eluted from the column at different positions (Fig. 7). Fibrinolytic and fibrinogenolytic properties of Sak moieties in human plasma in vitro Comparable dose-dependent lysis of 125I-fibrin labeled human plasma clots submerged in human plasma was obtained with Sak, with Sak-M26L and with SakM26C (Fig. 8A). Equi-effective concentrations, causing 50% clot lysis in 2 h (ECs0), were (mean + S.E.; n = 3) 14 + 2.5 nM for Sak, 17 ± 1.6 nM for Sak-M26L and 19 ± 1.4 nM for Sak-M26C. Residual fibrinogen levels after 2 h at ECs0 were 88 + 8%, 85 ± 7% or 92 + 1% respectively (Fig. 8B). In contrast, with Sak-M26R or Sak-M26V no detectable clot lysis occurred within 2 h at concentrations up to 250 nM (Fig. 8A) and fibrinogen levels did not decrease significantly (Fig. 8B).

4. Discussion

Staphylokinase (Sak) is a 136 amino-acid protein composed of two folded domains of similar size with a mean distance between the centres of gravity of both domains of 3.7 nm [13]. Little information is, however, available on structure-function relations in Sak which determine its interaction with plasminogen. One approach to study structure/function relationships consists of separation of the two domains of the molecule. This could be achieved by substitution of the unique Met residue in position 26, followed by introduction of a new Met residue near the middle of the molecule, allowing cleavage with CNBr. Therefore, in the present study, we have constructed variants of Sak in which Met-26 was substituted but, surprisingly, the nature of the exchanged amino acid

241

dramatically influenced the potency for plasminogen activation. The functional properties of four of these mutants (Sak-M26L, Sak-M26C, Sak-M26R and SakM26V), selected on the basis of initial screening of the plasminogen activating activity, were studied in more detail. Our data show that Sak-M26L and Sak-M26C are comparable to wild-type Sak, with respect to the rate and extent of complex formation with plasminogen, specific fibrinolytic activity, catalytic efficiency for plasminogen activation and fibrinolytic potency in a human plasma milieu in vitro. In contrast, Sak-M26R and Sak-M26V were virtually inactive in these functional assays, both in purified systems containing plasminogen and in human plasma. At present, it is not possible to determine the correlation between the nature of the amino-acid exchanges and their effects on the capability of the mutant proteins to activate plasminogen. Indeed, the introduction of polar as well as apolar amino acids led to inactive molecules. Furthermore, the amino-acid exchanges described here caused no significant changes of the solution structure of mutant proteins as compared to the wild-type, as revealed by dynamic light scattering- and CD-measurements (Damashun et al., personal communication). Therefore, only minor changes of the protein structure seem to occur, possibly in the micro-environment of the amino-acid position Met-26. A detailed knowledge of the 3-dimensional structure of Sak may be required to explain the observed effects at the molecular level. It is at present unclear whether substitution of amino acid 26 alters the initial complex formation between staphylokinase and plasminogen, or a subsequent step in the activation pathway, which may include conformational changes in either of the moieties, or altered conversion of plasminogen to plasmin. The inactive Sak-M26A mutant is not able to form a stable complex with plasminogen. This finding supports the hypothesis of an impact of the amino-acid side chain at position 26 on complex formation. Our present study is of a descriptive nature and does not provide mechanisms. Our finding that preincubation of plasminogen with Sak-M26V does not impair subsequent active site exposure upon addition of Sak, nevertheless indicates that Sak-M26V does not bind with high affinity to sites on the plasminogen molecule which are required for binding of wild-type Sak. In any case, the critical role of amino acid 26 in staphylokinase for the activation of plasminogen appears well established.

5. References [1] Matsuo, O., Okada, K., Fukao, H., Tomioka, Y., Ueshima, S., Watanuki, M. and Sakai, M. (1990) Blood 76, 925-929.

242

B. Schlott et al. / Biochimica et Biophysica Acta O0 (1994) 000-000

[2] Lijnen, H.R., Van Hoef, B., De Cock, F., Okada, K., Ueshima, S., Matsuo, O. and Collen, D. (1991) J. Biol. Chem. 266, 11826-11832. [3] Lijnen, H.R., Van Hoef, B., Vandenbossche, L. and Collen, D. (1992) Fibrinolysis 6, 214-225. [4] Kowalska-Loth, B. and Zakrzewski, K. (1975) Acta Biochim. Pol. 22, 327-339. [5] Lijnen, H.R., Van Hoef, B. and Collen, D. (1993) Eur. J. Biochem. 211, 91-97. [6] Collen, D., Schlott, B., Engelborghs, Y., Van Hoef, B., Hartmann, M., Lijnen, H.R. and Behnke, D. (1993) J. Biol. Chem. 268, 8284-8289. [7] Sakai, M., Watanuki, M. and Matsuo, O. (1989) Biochem. Biophys. Res. Commun. 162, 830-837. [8] Lijnen, H.R., Stassen, J.M., Vanlinthout, I., Fukao, H., Okada, K., Matsuo, O. and Collen, D. (1991) Thromb. Haemost. 66, 468-473. [9] Collen, D., De Cock, F., Vanlinthout, I., Declerck, P.J., Lijnen, H.R. and Stassen, J.M. (1992) Fibrinolysis 6, 232-242. [10] Collen, D., De Cock, F. and Stassen, J.M. (1993) Circulation 87, 996-1006. [11] Collen, D. and Van de Werf, F. (1993) Circulation 87, 18501853. [12] Collen, D., Silence, K., Demarsin, E., De Mol, M. and Lijnen, H.R. (1992) Fibrinolysis 6, 203-213.

[13] Damaschun, G., Damaschun, H., Gast, K., Misselwitz, R., Zirwer, D., Giihrs, K.-H., Hartmann, M., Schlott, B., Triebei, H. and Behnke, D. (1993) Biochim. Biophys. Acta 1161, 244-248. [14] Sako, T. and Tsuchida, N. (1983) Nucleic Acids Res. 11, 76797693. [15] Behnke, D. and Gerlach, D. (1987) Mol. Gen. Genet. 210, 528-534. [16] Collen, D., Zhao, Z.A., Holvoet, P. and Marynen P. (1992) Fibrinolysis 6, 226-231. [17] Gerlach, D., Kraft, R. and Behnke, D. (1988) Zentralbl. Bakteriol. Mikrobiol. Hyg. A 269, 314-322. [18] Schlott, B., Hartmann, M., Giihrs, K.H., Birch-Hirschfeld, E., Pohl, D., Vanderschueren, S., Van de Werf, F., Michael, A., Collen, D. and Behnke D. (1993) Biotechnology, in press. [19] Sanger, F., Nicklen, S. and Coulson, A.R. (1972) Proc. Natl. Acad. Sci. USA 64, 5463-5467. [20] Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. [21] Deutsch, D.G. and Mertz, E.T (1970) Science 170, 1095-1096. [22] Lijnen, H.R., Van Hoef, B. and Collen, D. (1986) Biochim. Biophys. Acta 884, 402-408. [23] Clauss, A. (1957) Acta Haematol. 17, 237-246. [24] Chase, T., Jr and Shaw, E. (1969) Biochemistry 8, 2212-2224. [25] McClintock, D.K. and Bell, P.H. (1971) Biochim. Biophys. Res. Commun. 43, 694-702.