Recombinant Staphylokinase Variants With Altered Immunoreactivity
I: Construction and Characterization
Background Recombinant staphylokinase offers promise for thrombolytic therapy in acute myocardial infarction, but it is immunogenic. Although reduced immunogenicity of heterologous proteinaceous drugs by protein engineering has not previously been reported, an attempt was made to achieve this in staphylokinase by site-specific mutagenesis.
Methods and Results Biospecific interaction analysis of a panel of 17 murine monoclonal antibodies against recombinant staphylokinase (SakSTAR variant) identified three nonoverlapping immunodominant epitopes, two of which could be eliminated by substitution mutagenesis of clusters of two or three charged amino acids with alanine. Circulating anti-staphylokinase antibodies elicited in patients by treatment with SakSTAR were incompletely (<90%) absorbed by these mutants. Therefore, the combination variants K35A,E38A,K74A,E75A,R77A (SakSTAR.M38) and K74A,E75A,R77A,E80A,D82A (SakSTAR.M89) were constructed, expressed in Escherichia coli, highly purified by ion-exchange and hydrophobic interaction chromatography, and characterized. These variants had specific activities that were approximately half that of SakSTAR, and they combined the reduced reactivity with the panels of monoclonal antibodies of their parent molecules. Absorption of circulating antibodies elicited in patients by treatment with SakSTAR was incomplete in 13 of 16 patients (median values, 68% and 65% with SakSTAR.M38 and SakSTAR.M89, respectively).
Conclusions SakSTAR contains three immunodominant epitopes, two of which were eliminated by site-directed mutagenesis, yielding combination mutants with relatively maintained specific activities that were not recognized by a significant fraction of the antibodies elicited in patients by treatment with wild-type SakSTAR. These mutants appear to be suitable for more detailed investigation of their thrombolytic and antigenic properties.
Staphylokinase, a 136-amino-acid protein produced by certain strains of Staphylococcus aureus, was shown to have profibrinolytic properties more than four decades ago.1 The staphylokinase gene has been cloned from the bacteriophages sakφC2 and sak42D3 as well as from the genomic DNA (sakSTAR) of a lysogenic S. aureus strain.4 Only four nucleotide differences were found in the coding regions of the sakφC, sak42D, and sakSTAR genes, one of which constituted a silent mutation.3 4 5 6 Although there are differences in thermostability between these natural variants,7 the thrombolytic potencies of SakSTAR and Sak42D in patients with thromboembolic disease are comparable (unpublished data). The protein sequence of wild-type SakSTAR5 is represented in Fig 1⇓.
In patients with acute myocardial infarction8 9 10 or with peripheral arterial occlusion,11 recombinant staphylokinase emerged as a potent and highly fibrin-specific thrombolytic agent, with full preservation of plasma fibrinogen, α2-antiplasmin, and plasminogen levels at therapeutic doses. SakSTAR was initially found to induce less antibody formation than streptokinase in dogs12 and baboons,13 but this could not be extended to patients given an intravenous infusion of up to 20 mg SakSTAR. Indeed, although neutralizing antibody titers against SakSTAR were low at baseline and up to 6 days after infusion, high titers of antibodies (neutralizing on average ≈30 μg SakSTAR/mL plasma) were consistently demonstrated in plasma beyond 2 weeks and remained elevated for more than 1 year.8 9 10 11
Staphylokinase is a heterologous protein, and it is not obvious that its immunoreactivity can be reduced by protein engineering. Furthermore, deletion of 17 NH2-terminal amino acids or of 2 COOH-terminal amino acids or substitution of Met26 abolishes the activity of staphylokinase,14 15 which suggests that it is very sensitive to inactivation by site-directed mutagenesis.
In the present study, we found that SakSTAR contains three nonoverlapping immunodominant epitopes. Two of these can be eliminated, without inactivation of the molecule, by specific site-directed mutagenesis of two or three charged amino acids to alanine. The combination mutants SakSTAR.M38, in which Lys35, Glu38, Lys74, Glu75, and Arg77 were substituted with Ala, and SakSTAR.M89, with Lys74, Glu75, Arg77, Glu80, and Asp82 substituted with Ala, were found to combine the reduced reactivity of their parent molecules with a panel of MAbs. These variants also appeared to be less reactive with antibodies elicited in patients by treatment with SakSTAR.
The low-molecular-weight calibration kits for SDS-PAGE, Sepharose 4B, SP-Sepharose, and Q-Sepharose fast flow were purchased from Pharmacia; restriction enzymes from Pharmacia or from Boehringer Mannheim; T4 DNA ligase, Klenow fragment of Escherichia coli DNA polymerase I, and alkaline phosphatase from Boehringer Mannheim; and the expression vector pMEX6 from Medac GmbH. M13KO7 helper phage was purchased from Promega, yeast extract and tryptone from Difco Laboratories, and Luria broth growth medium from Life Technologies. All other plasmids as well as the E. coli host strain TG1 are commonly used in molecular cloning experiments and have been described in detail elsewhere. Taq DNA polymerase, PCR buffer, and 2′-deoxynucleotides were purchased from Boehringer Mannheim. The chromogenic substrate (S2403) l-pyroglutamyl-l-phenylalanyl-l-lysine-p-nitroaniline hydrochloride was purchased from Chromogenix. 125I-Labeled fibrinogen was purchased from Amersham. Plasminogen was purified from human plasma as described elsewhere.16
In the “clustered charge–to-alanine” scan, clusters of hydrophilic charged amino acids were targeted. SakSTAR contains 45 charged amino acids (2 His, 14 Glu, 8 Asp, 1 Arg, and 20 Lys), which were mutagenized to Ala in clusters of two or three amino acids, as summarized in Fig 1. A total of 21 variants in which the indicated charged amino acids were replaced by alanine were prepared by site-directed mutagenesis and expressed in E. coli as detailed elsewhere.17 Variants D5A,K6A (SakSTAR.M20) through K86A,E88A (SakSTAR.M10) in Fig 1 were obtained by use of the pMa/c system with the repair-deficient E. coli strain WK6MutS.18 The variants D93A,K94A (SakSTAR.M11) through E134A,K135A,K136A (SakSTAR.M19) were constructed in the Institut fu¨r Molekulare Biotechnologie, Jena, Germany, as previously described.9 17
The oligonucleotides used for the construction of variants SakSTAR.M38 and SakSTAR.M89, 5′-ATAGCAATGCA-TTTCCTGCACTATCAAC-3′ (sakSTAR.M3), 5′-CTAATT-CAACTACTGCAAACGCTGCATATGCTGTCGCATC-3′ (sakSTAR.M8), and 5′-TTTGCGCTTGGCGCCAATGCAACTACTCTAAACTCTTTATAT-3′ (sakSTAR.M9), were custom synthesized by Pharmacia Biotech. The oligonucleotide-directed mutagenesis system and the pMa/c plasmids were kindly provided by Corvas.18
Plasmid pMEX602SakB, used as a source of the sakSTAR gene, has previously been described.9 17 It contains the coding region of the sakSTAR gene in which the signal peptide region is replaced by a methionine start codon, an appropriately spaced Shine-Dalgarno sequence, and several unique restriction sites and in which the 3′ end is modified to generate restriction sites close to the end of the coding region. The sequence of the 453-bp EcoRI-HindIII fragment used for the construction of SakSTAR.M38 and SakSTAR.M89 is represented in Tables 1⇓ and 2⇓.
MAbs Against Wild-type SakSTAR
MAbs against SakSTAR were produced essentially by the method of Galfre´ and Milstein.19 BALB/c mice were immunized by injection of 10 μg SakSTAR SC in complete Freund's adjuvant, which was followed 2 weeks later by injection of 10 μg SakSTAR IP in incomplete Freund's adjuvant. After an interval of at least 6 weeks, the mice were boosted with 10 μg SakSTAR IP in saline on days 4 and 2 before the cell fusion. Spleen cells were isolated and fused with P3X63-Ag.8-6.5.3 myeloma cells (obtained from Dr O. Scho¨nherr, Organon) according to Fazekas and Scheidegger.20 After selection in hypoxanthine/aminopterine/thymidine medium, the supernatants were screened for specific antibody production with a one-site noncompetitive micro-ELISA with microtiter plates coated with a 4-μg/mL solution of SakSTAR in PBS. Coating was allowed to proceed for 48 hours at 4°C, and the wells were saturated with a 1% BSA solution for 1 hour at room temperature. The bound immunoglobulins were detected with horseradish peroxidase–conjugated rabbit anti-mouse IgG.21 Positive clones were used for the production of ascitic fluid in pristane-primed BALB/c mice.22 The IgG fraction of the MAbs was purified from ascites by affinity chromatography on protein A-Sepharose.23
Epitope Mapping of Wild-type SakSTAR and SakSTAR Variants
The epitope specificity of a panel of 17 murine MAbs raised against SakSTAR was determined by real-time biospecific interaction analysis with the BIAcore instrument (Pharmacia, Biosensor AB), which allows direct measurement of interactions in real time.24 SakSTAR or variants were immobilized on the surface of Sensor Chip CM5 by use of an amine coupling kit (Pharmacia Biosensor AB) as recommended by the manufacturer. This procedure links primary amino groups in the ligand to the carboxymethylated dextran surface of the sensor chip.25 Immobilization was performed from protein solutions at a concentration of 10 μg/mL in 10 mmol/L sodium acetate at pH 5.0 at a flow rate of 5 μL/min for 6 minutes. This resulted in covalent attachment of 1000 to 1500 RU of SakSTAR (corresponding to ≈0.07 pmol/mm2).26 The MAb solutions were passed by continuous flow at 20°C past the sensor surface. At least four concentrations of each analyte (range, 50 nmol/L to 50 μmol/L) in 10 mmol/L HEPES, 3.4 mmol/L EDTA, 0.15 mol/L NaCl, and 0.005% surfactant P20, pH 7.2, were injected at a flow rate of 5 μL/min for 6 minutes in the association phase. Then the sample was replaced by buffer, also at a flow rate of 5 μL/min for 6 to 30 minutes. After each cycle, the surface of the sensor chip was regenerated by injection of 5 μL of 15 mmol/L HCl. Apparent association (kass) and apparent dissociation (kdiss) rate constants were derived from the sensorgrams as described in detail elsewhere.27
The epitope specificity of the panel of MAbs was determined by testing the ability of pairs of MAbs to bind simultaneously to the antigen. The analysis of competitive binding of pairs of MAbs was carried out by use of staphylokinase linked to the sensor chip surface as described in application note 101 (Pharmacia Biosensor AB).
DNA Preparation, Manipulation, and Sequencing
Plasmid DNA was isolated with the Nucleobond AX purification kit (Macherey-Nagel). Digestion of DNA with restriction enzymes, ligation, filling in of protruding 5′-ends with Klenow fragment of DNA polymerase I, and 5′-phosphorylation with polynucleotide kinase were carried out according to standard protocols28 or the instructions of the manufacturers of the enzymes. Transformations of E. coli were performed by the calcium phosphate coprecipitation procedure. DNA sequencing was performed by the dideoxy chain termination reaction method and the automated laser fluorescent A.L.F. method (Pharmacia). Site-directed mutagenesis, propagation of the pMa/c plasmids or derivatives, preparation of ssDNA, and expression in E. coli WK6MutS were carried out as previously described.18
Construction, Production, and Purification of SakSTAR.M38
The 453-bp EcoRI-HindIII fragment (Table 1) containing the entire coding region for SakSTAR was cut out of the plasmid pMEX602SakB (ampicillin resistant) and cloned into the EcoRI-HindIII site of the pMc5-8 plasmid (chloramphenicol resistant) to yield pMc-STAR. Single-strand pMc-STAR and the EcoRI-HindIII fragment of pMa5-8 were used to prepare a gapped-duplex DNA molecule, which was hybridized with the 28-base synthetic oligonucleotide sakSTAR.M3 (containing an Nsi I restriction site). Extension reactions were carried out with the Klenow fragment of DNA polymerase and ligase. After transformation of E. coli WK6MutS cells and selection on ampicillin, 81 colonies were grown on nitrocellulose membranes and denatured in situ, and DNA was hybridized overnight at room temperature with radiolabeled sakSTAR.M3 oligonucleotide (1.5×108 cpm of [γ32P]ATP for T4 polynucleotide kinase labeling of 20 to 30 ng of oligonucleotide). Filters were washed at 42°C with solutions containing 0.1% SDS and 2×SSC, 1×SSC, 0.2×SSC, and 0.1×SSC. DNA of 1 selected clone out of 16 positives was prepared from 150 mL bacterial cultures and analyzed by restriction enzyme digestion with Nsi I and Pvu I. The SakSTAR.M3 mutation (pMa-STAR3) was confirmed by sequencing of the complete coding region. Then ssDNA was prepared by transformation of pMa-STAR3 in E. coli as described above. Single-strand pMa-STAR3 was hybridized with pMc (EcoRI-HindIII) and with the 40-base synthetic oligonucleotide sakSTAR.M8, which contains an Nde I restriction site, as described above. After transformation of E. coli WK6MutS cells and selection on chloramphenicol, 100 colonies were grown on nitrocellulose membranes and hybridized with labeled sakSTAR.M8 oligonucleotide. Positive clones (2 of 7) were grown and analyzed by restriction enzyme digestion (Nsi I-HindIII and Nde I), yielding 1 positive clone. The double mutant sakSTAR.M38 was then ligated back into the pMEXSAK602B expression vector. Of 12 minipreparations of DNA, 6 had the correct restriction pattern. One of these clones (pMEXSakSTAR.M38) was sequenced and used for preparation of SakSTAR.M38 under control of the IPTG-inducible tac promoter and two Shine-Dalgarno sequences in tandem.
One hundred microliters of a suspension of E. coli WK6 cells transformed with the recombinant plasmid pMEXSakSTAR.M38 was incubated in 100 mL LB medium (GIBCO/BRL) containing 100 μg/mL ampicillin. The mixture was incubated overnight at 37°C while being shaken at 200 rpm, resulting in a cell density of ≈5 absorbance units at 600 nm. Aliquots of 20 mL were transferred to 2-L volumes (in 5-L flasks) of LB medium containing 100 μg/mL ampicillin. The mixtures were incubated for 3 hours at 37°C while being shaken before addition of 200 μmol/L IPTG for induction of SakSTAR.M38 expression, which was allowed to proceed for 4 hours. The cells were pelleted by centrifugation at 4000 rpm for 20 minutes, resuspended in 1/10 volume of 0.01 mol/L phosphate buffer, pH 6.5, and disrupted by sonication at 0°C. Cell debris was removed by centrifugation for 30 minutes at 20 000 rpm, and the supernatant was stored at −20°C until used.
Pooled cleared cell lysates (2-L volumes) from 20 to 30 L of bacterial cultures were adjusted to pH 5.9, sterilized by filtration through a 0.22-μm Sartorius filter, and applied to a 5×25-cm column of SP-Sepharose, preconditioned with 0.5 mol/L NaOH and with sterilized 0.01 mol/L phosphate buffer, at a flow rate of 12 mL/min and at 4°C in a laminar flow. The column was washed with 2 to 3 L of buffer and eluted with a salt gradient from 0 to 1 mol/L for 500 mL and from 1 to 2 mol/L for 250 mL at a flow rate of 10 mL/min and at 4°C. The SakSTAR.M38-containing fractions localized by SDS-PAGE were pooled (≈200 mL) and dialyzed against 15 L of sterilized 0.01 mol/L phosphate buffer, pH 8.0, at 4°C. The dialyzed material was centrifuged at 4000 rpm for 30 minutes, sterilized again by filtration, and applied to a 2.5×12-cm column of Q-Sepharose fast flow preconditioned with 0.5 mol/L NaOH and with sterilized 0.01 mol/L phosphate buffer, pH 8.0, at a flow rate of 3 mL/min at 4°C. The column was washed with ≈600 mL 0.01 mol/L phosphate buffer, pH 8.0, at a flow rate of 8 mL/min and eluted with a salt gradient from 0 to 0.17 mol/L for 30 mL, from 0.17 to 0.2 mol/L for 100 mL, and from 0.2 to 1.5 mol/L for 200 mL at a flow rate of 4 mL/min. The SakSTAR.M38-containing fractions localized by SDS-PAGE were pooled, the protein concentration was adjusted to 1 mg/mL, and the material was sterilized by filtration through a 0.22-μm Millipore filter.
Construction, Production, and Purification of SakSTAR.M89
The 453-bp EcoRI-HindIII fragment (Table 2) containing the entire coding region for SakSTAR was cut out of the plasmid pMEX602SakB (ampicillin resistant) and cloned into the EcoRI-HindIII site of the pMc5-8 plasmid (chloramphenicol resistant) to yield pMc-STAR. Single-strand pMc-STAR and the EcoRI-HindIII fragment of pMa5-8 were used to prepare a gapped-duplex DNA molecule, which was hybridized with the 42-base synthetic oligonucleotide sakSTAR.M9 containing a Nar I restriction site. Extension reactions were carried out with the Klenow fragment of DNA polymerase and ligase. After transformation of E. coli WK6MutS and selection on ampicillin, colonies were grown on nitrocellulose membranes and denatured in situ, and DNA was hybridized overnight at room temperature with radiolabeled sakSTAR.M9 oligonucleotide (1.5×108 cpm of [γ32P]ATP for T4 polynucleotide kinase labeling of 20 to 30 ng of oligonucleotide). Filters were washed at 42°C with solutions containing 0.1% SDS and 2×SSC, 1×SSC, 0.2×SSC, and 0.1×SSC. DNA from 2 selected clones of 4 positives was prepared, and 1 (pMA-STAR9) was characterized by nucleotide sequence analysis. The EcoRI-HindIII insert from pMa-STAR9 was then ligated back into the pMEXSAK602B expression vector. The 58 clones were screened by in situ hybridization with radiolabeled sakSTAR.M9 oligonucleotide as a probe. One clone, pMEXSakSTAR.M9, was characterized by nucleotide sequence analysis and subsequently used for the construction of sakSTAR.M89.
To construct sakSTAR.M89, mutation 8 was introduced in sakSTAR.M9 by PCR. PCR was performed in a total volume of 100 μL with 5 U enzyme and 1 μg of each of the following primers: oligonucleotide II, 5′-CAGGAAACAGAATTCAGGAG, oligonucleotide III, 5′-TATATAATATTCGACATAGTATTCAATTTTT-3′, oligonucleotide IV, 5′-TATCCCGGGCATTAGATGCGACAGCATATGCAGCGTTTGCAGTA-3′, and oligonucleotide V, 5′-CAAAACAGCCAAGCTTCATTCATTCAGC-3′ (see Table 2 for their respective positions on the sakSTAR sequence). The concentrations of dATP, dCTP, dGTP, and dTTP were 200 μmol/L. Denaturation was carried out for 1 minute at 94°C, annealing for 2 minutes at 55°C, and extension for 1.5 minute at 72°C. After 30 cycles, samples were incubated for 10 minutes at 72°C and cooled to 4°C. In a first PCR, 2 ng of pMEXSakSTAR.M9 was amplified with oligonucleotides IV and V as primers. The PCR amplicon was digested with Sma I and HindIII and purified after electrophoresis in a 1.5% agarose gel with a Prep-A-gene kit (Bio-Rad Laboratories). The resulting fragment was cloned into the Sma I-HindIII sites of pUC18 (Pharmacia BioTech) by use of the rapid DNA ligation kit (Boehringer Mannheim). After transformation in E. coli WK6 cells, DNA from 12 colonies was prepared, and all 12 generated a fragment of ≈230 bp when digested with EcoRI and HindIII. One of these DNAs (pUC18-M89Δ) was used for cloning of a second PCR product (see below). A second PCR was performed on 2 ng of pMEXSakSTAR.M9 with oligonucleotides II and III as primers. The PCR product was digested with Ssp I and EcoRI and further purified as described above. The resulting fragment was ligated into the Sma I-EcoRI sites of pUC18-M89Δ. After transformation in E. coli WK6 cells, six clones were selected for DNA preparation. Five of the six generated a fragment of ≈453 bp after digestion with EcoRI and HindIII. This fragment coding for the entire SakSTAR.M89 was cloned into the EcoRI-HindIII sites of the expression vector pMEXSak602B. After transformation of E. coli WK6 cells, DNA from six colonies was analyzed by digestion with EcoRI and HindIII generating a fragment of ≈453 bp in all cases. One of these DNAs was further characterized by nucleotide sequence analysis using two fluorescent primers, as described above.
One hundred milliliters of LB medium containing 100 μg/mL ampicillin was inoculated with 100 μL of a suspension of E. coli WK6 cells transformed with the recombinant plasmid pMEXSakSTAR.M89. The culture was incubated overnight at 37°C while being shaken at 140 rpm to a cell density of ≈5 absorbance units at 600 nm. Production of SakSTAR.M89 after IPTG induction under the conditions described for SakSTAR.M38 above resulted in low yields. Therefore, aliquots of 4 mL were used to inoculate 2 L of cultures (in SL baffled flasks) in “Terrific Broth” medium28 containing 150 μg/mL ampicillin. The cultures were incubated for ≈20 hours at 30°C and at 140 rpm, resulting in a final cell density of ≈4×109 cells/mL (A1 cm600 nm=8). The cells were pelleted by centrifugation at 4000 rpm for 20 minutes, resuspended in 1/5 volume of 0.01 mol/L phosphate buffer, pH 6.5, and disrupted by sonication at 0°C. The pH was then adjusted to 5.8, and the cell debris was removed by 30 minutes of centrifugation at 20 000 rpm. The supernatant was stored at −20°C until it was processed further.
Pooled cleared cell lysates (1800 mL) were adjusted to pH 5.8 and applied to a 2.5×20-cm column of SP-Sepharose preconditioned with 0.5 mol/L NaOH and fresh 0.01 mol/L phosphate/2.5 mol/L NaCl buffer, pH 7.5, at a flow rate of 2 mL/min at 4°C. The column was washed with 500 mL buffer and eluted with a salt gradient from 0 to 1 mol/L for 200 mL at a flow rate of 6 mL/min. The pooled SakSTAR.M89 fractions, identified with SDS-PAGE, were adjusted to 2.5 mol/L with solid NaCl and subjected to hydrophobic interaction chromatography on a 2.5×20-cm column of phenyl-Sepharose, preconditioned with 0.5 mol/L NaOH and fresh 0.01 mol/L phosphate/2.5 mol/L NaCl buffer, pH 7.5, at a flow rate of 2 mL/min and 4°C. The column was washed with ≈500 mL buffer and eluted with 0.01 mol/L phosphate buffer, pH 6.5. The SakSTAR.M89-containing fractions localized by SDS-PAGE were pooled and dialyzed against 2 L of 0.01 mol/L phosphate buffer, pH 9.0. The dialyzed material was centrifuged at 4000 rpm for 30 minutes and applied to a 1.6×5-cm column of Q-Sepharose fast flow, preconditioned with 0.5 mol/L NaOH and with fresh 0.01 mol/L phosphate buffer, pH 9.0, at a flow rate of 2 mL/min at 4°C. The column was washed with ≈150 mL 0.01 mol/L phosphate buffer, pH 9.0, and eluted with a salt gradient from 0 to 1 mol/L NaCl over 100 mL at a flow rate of 4 mL/min. The SakSTAR.M89 fractions localized by SDS-PAGE were pooled, the protein concentration was adjusted to 1 mg/mL, and the material was sterilized by filtration through a 0.22-μm Millipore filter.
Physicochemical Analysis of SakSTAR Variants
Protein concentration was measured by the method of Bradford,29 using an in-house standard of purified natural staphylokinase (STAN5) as a primary standard and BSA as a secondary reference.4 NH2-terminal amino acid sequence analysis was performed on an Applied Biosystems 477A protein sequencer with identification of amino acids by high-performance liquid chromatography (courtesy of Dr J. Van Damme and Dr P. Proost, Rega Institute, KU Leuven, Belgium).
The specific activities of the purified staphylokinase moieties were determined with a plasminogen-coupled chromogenic substrate assay and expressed in HU by comparison with the in-house standard, which was assigned an activity of 100 000 HU/mg protein. The plasminogen-coupled chromogenic substrate assay for staphylokinase was carried out in microtiter plates with a mixture of 80 μL staphylokinase-containing solution, 100 μL Glu-plasminogen solution (final concentration, 0.5 μmol/L), and 30 μL 0.1 mol/L phosphate buffer, pH 7.3, containing 0.01% Tween 80. After incubation for 30 minutes at 37°C, generated plasmin was quantified by addition of 30 μL S-2403 (final concentration, 1 mmol/L; Chromogenix), and the absorbance at 405 nm was measured after 5 or 10 minutes.
To test the temperature stability, purified preparations of SakSTAR variants were diluted to a concentration of 1 mg/mL in 0.15 mol/L NaCl/0.05 mol/L Tris-HCl buffer, pH 7.5, and incubated at 4°C, 22°C, 30°C, 37°C, 56°C, and 70°C. Aliquots were removed at different time intervals (1 hour to 5 days) and frozen at −20°C until assayed. The residual activity was determined by the plasminogen-coupled chromogenic substrate assay described above.
Absorption With SakSTAR Variants of Antibodies Elicited in Patients by Treatment With SakSTAR
Plasma samples from 16 patients with acute myocardial infarction obtained several weeks after treatment with SakSTAR8 9 10 were absorbed with 100 μg/mL (at least a 1.5-fold molar excess over the staphylokinase-neutralizing activity determined as described above) of SakSTAR, SakSTAR.M3, SakSTAR.M8, SakSTAR.M9, SakSTAR.M38, and SakSTAR.M89. Absorption was carried out for 10 minutes before determination of residual binding to SakSTAR by biospecific interaction analysis as described above. Briefly, the dilution of patient plasma (1/30 to 1/200) for which binding to SakSTAR-substituted chips amounted to ≈2000 RU was determined, and a calibration curve for antibody binding was constructed by use of further serial twofold dilutions. Then the residual binding of the plasma sample preincubated with the SakSTAR moieties was determined and expressed as a percentage with the patient's calibration curve.
Epitope Mapping of SakSTAR
The equilibrium association constants (Ka), calculated as the ratio of kass and kdiss, for the binding to wild-type staphylokinase of the panel of 17 MAbs ranged between 0.6×109 and >25×109 (mol/L)−1 (median value, 1010 [mol/L]−1) (Table 3⇓).
Pairwise binding tests divided the 17 MAbs into three groups representing three nonoverlapping epitopes on the antigen, as illustrated in Fig 2⇓. The independence of these epitopes was confirmed by the direct demonstration of additive binding of the MAbs 26A2, 28H4, and 24C4. The antibodies were aligned according to their epitope specificity as illustrated in Table 3⇑.
Epitope Mapping of ‘Clustered Charge–to-Alanine’ Variants of SakSTAR
Of the 21 variants, designed as illustrated in Fig 1, E99A,E100A (SakSTAR.M13) and E99A,E100A,E102A (SakSTAR.M14) could not be obtained in purified form, whereas K11A,D13A,D14A (SakSTAR.M1), E46A,K50A (SakSTAR.M4), and E65A,D69A (SakSTAR.M7) were ≤20% active. K35A,E38A (SakSTAR.M3) and E80A,D82A (SakSTAR.M9) reacted poorly with the antibody cluster 7H11, 25E1, and 40C8, whereas K74A,E75A,R77A (SakSTAR.M8) reacted poorly with the cluster 26A2, 30A2, 2B12, and 3G10. These results formed the basis for the design of the combination mutants SakSTAR.M38 and SakSTAR.M89. The other 13 variants reacted with the MAbs in a manner similar to SakSTAR.
Purification and Characterization of SakSTAR.M38 and SakSTAR.M89
The results of three preparations from batches of 20 to 30 L of E. coli–conditioned culture broth are summarized in Table 4⇓. For SakSTAR.M38, the average recovery of protein was 80±25 mg, with a yield based on the recovery of staphylokinase activity of 18±5%, and the final materials were obtained at a concentration of ≈3 mg/mL, with a specific activity of 45 000±5200 HU/mg. For SakSTAR.M89, the average recovery of protein was 73±17 mg, the yield of the purification was 44±7%, and the final material was obtained at a concentration of ≈2 mg/mL with a specific activity of 51 000±3500 HU/mg.
SDS-PAGE under nonreducing or reducing conditions (Fig 3⇓) revealed the presence of single components with Mr of ≈16 000. NH2-terminal amino acid sequence analysis of purified SakSTAR.M38 revealed a single main sequence corresponding to Ser-Ser-Ser-Phe-Asp-Lys-Gly-Lys-Tyr-Lys and a second minor sequence (≈20%) corresponding to Gly-Lys-Tyr-Lys-Lys-Gly-Asp-Asp-Ala, which is derived from the main sequence by removal of the six NH2-terminal amino acids.5
The temperature stability of the preparations dissolved to a concentration of 1.0 mg/mL in 0.15 mol/L NaCl/0.05 mol/L Tris-HCl buffer, pH 7.5, at various temperatures is illustrated in Fig 4⇓. At temperatures up to 30°C, SakSTAR, SakSTAR.M38, and SakSTAR.M89 remained fully active for up to 5 days. At 37°C, 56°C, and 70°C, however, SakSTAR.M38 and SakSTAR.M89 were less stable than SakSTAR.
Epitope Mapping of SakSTAR.M38 and SakSTAR.M89
Determination of the equilibrium association constants for the binding of MAbs to SakSTAR, SakSTAR.M38, and SakSTAR.M89 (Table 5⇓) revealed that SakSTAR.M38 and SakSTAR.M89 did not react with antibody cluster 26A2, 30A2, 2B12, and 3G10 (like SakSTAR.M8) or with antibody cluster 7H11, 25E1, and 40C8 (like SakSTAR.M3 and SakSTAR.M9).
Absorption With SakSTAR Variants of Antibodies Elicited in Patients by Treatment With Wild-type SakSTAR
Whereas wild-type SakSTAR absorbed >95% of the binding antibodies from all plasma samples of 16 patients, incomplete absorption (<90%) was observed with variant K35A,E38A (SakSTAR.M3) in 4 patients, with variant K74A,E75A,R77A (SakSTAR.M8) in 12 patients, and with variant E80A,D82A (SakSTAR.M9) in 5 patients, whereas, as anticipated, a mixture of SakSTAR.M8 with either SakSTAR.M3 or SakSTAR.M9 absorbed >90% of the antibodies (Table 6⇓).
With the combination mutants, incomplete absorption was observed with SakSTAR.M38 in 13 patients (median value, 68%) and with SakSTAR.M89 in 13 patients (median value, 65%) (Table 6).
The present study was initiated by our observations that recombinant staphylokinase may constitute a potent and highly fibrin-specific but immunogenic thrombolytic agent in patients with acute myocardial infarction or peripheral arterial occlusion.8 9 10 11 Although immunogenicity has not constituted an insurmountable obstacle for the use of streptokinase, it clearly hampers its unrestricted use and precludes its repeated administration.
It is now generally accepted that most of the entire surface of a protein is potentially antigenic, that epitopes usually consist of an area comprising 15 to 22 amino acids, and that most antigenic sites are conformational in nature and discontinuous, ie, composed of amino acids that are located on several peptide loops that are brought close together by folding into the three-dimensional structures (see, eg, References 30 and 31). The three-dimensional structure of staphylokinase, however, is unknown, and no homologous proteins that could serve as a template have been described to date. A rational approach to the reduction of the immunoreactivity of staphylokinase (in terms of both immunogenicity and antigenicity) by site-specific mutagenesis would therefore seem elusive. However, a comprehensive analysis of the antigenicity of human growth hormone against a panel of 21 murine monoclonal antibodies with a combination of homologue and alanine scanning mutagenesis revealed that only a few amino acid side chains in the contact area of an epitope dominated the binding to a reactive antibody.31 These amino acids tended to be Arg>Pro>Glu∼Asp∼Phe-Ile. Thus, apart from Pro, which may act via conformational perturbations, charged amino acids (except Lys) are particularly important for binding. However, the low antigenic importance of Lys in human growth hormone may not be general, since it was found to be of functional importance in the N-1 epitope of staphylococcal nuclease.32 On the basis of these considerations, it appeared worthwhile to evaluate the immunoreactivity of site-specific mutants in which charged amino acids are substituted with alanine, first in terms of immunoreactivity and subsequently, possibly, in terms of antibody induction.
Our observation that a panel of 17 MAbs could be mapped to only 3 nonoverlapping epitope clusters fueled the hope that a limited number of substitutions might alter the epitope specificity.
A set of 21 mutants in which 2 or 3 closely spaced charged amino acids were replaced by alanine (charged cluster–to-alanine scan) was designed, as illustrated in Table 1. Of these, 3 variants with unaltered specific activity reacted poorly with antibodies specific for 1 of the 3 nonoverlapping immunodominant epitopes: variant SakSTAR.M8 had a more than 10-fold reduced affinity for 4 of the 5 MAbs specific for epitope cluster 1, whereas SakSTAR.M3 had a reduced affinity for 3 and SakSTAR.M9 for 4 of the 5 MAbs specific for epitope cluster 3 (Table 3).
To verify whether treatment with SakSTAR would elicit antibodies against the epitopes removed from SakSTAR.M8 on the one hand (cluster 1) and of SakSTAR.M3 or SakSTAR.M9 on the other hand (cluster 3), plasma samples of 16 patients treated with SakSTAR were absorbed with an excess of wild-type or variant SakSTAR before determination of residual binding to SakSTAR. As anticipated, SakSTAR absorbed all antibodies, but absorption was incomplete in 4 patients with SakSTAR.M3, in 5 patients with SakSTAR.M9, and in 12 patients with SakSTAR.M8.
Therefore, combination variants of wild-type staphylo-kinase (SakSTAR) were constructed, with substitution, on the one hand, of Lys35, Glu38, Lys74, Glu75, and Arg77 with alanine (SakSTAR.M38) and, on the other hand, of Lys74, Glu75, Arg77, Glu80, and Asp82 with alanine (SakSTAR.M89). Highly purified material was obtained by expression of the combination variants in E. coli and purification from the cytosolic fraction by ion exchange and hydrophobic interaction chromatography. These variants combined the reduced reactivity of the parent molecules with the panel of MAbs and on average absorbed only two thirds of the antibodies elicited in patients by treatment with SakSTAR. However, these variants had a specific activity that was only ≈50% of that of wild-type staphylokinase and a lower temperature stability at 56°C and 70°C. Nevertheless, the residual specific activities and thermostabilities remained within limits acceptable for in vitro and subsequent in vivo investigation of their comparative thrombolytic potency and immunogenicity (see accompanying article).
In aggregate, the present data indicate that the wild-type staphylokinase variant SakSTAR contains three nonoverlapping immunodominant epitopes, at least two of which can be eliminated, without excessive loss of functional activity, by site-directed mutagenesis. This study thus shows that, in principle, it will be possible to produce engineered variants of staphylokinase that are functional but less antigenic than the wild-type molecule. Such variants, if also less immunogenic, could constitute preferred alternatives to streptokinase and wild-type staphylokinase for thrombolytic therapy in patients with thromboembolic disease.
Selected Abbreviations and Acronyms
|HU||=||in-house units of activity|
|PCR||=||polymerase chain reaction|
|SSC||=||standard saline citrate|
This study was supported in part by a sponsored research agreement between the University of Leuven (Leuven Research and Development VZW) and Thromb-X NV, a spin-off company of the University of Leuven in which Dr Collen has an equity interest.
- Received October 9, 1995.
- Revision received January 8, 1996.
- Accepted January 22, 1996.
- Copyright © 1996 by American Heart Association
Sako T, Tsuchida N. Nucleotide sequence of the staphylokinase gene from Staphylococcus aureus. Nucleic Acids Res.. 1983;11:7679-7693.
Collen D, Van de Werf F. Coronary thrombolysis with recombinant staphylokinase in patients with evolving myocardial infarction. Circulation.. 1993;87:1850-1853.
Vanderschueren S, Barrios L, Kerdsinchai P, Van den Heuvel P, Hermans L, Vrolix M, De Man F, Benit E, Muyldermans L, Collen D, Van de Werf F. A randomized trial of recombinant staphylokinase versus alteplase for coronary artery patency in acute myocardial infarction. Circulation.. 1995;92:2044-2049.
Vanderschueren S, Stockx L, Wilms G, Lacroix H, Verhaeghe R, Vermylen J, Collen D. Thrombolytic therapy of peripheral arterial occlusion with recombinant staphylokinase. Circulation.. 1995;92:2050-2057.
Collen D, De Cock F, Stassen JM. Comparative immunogenicity and thrombolytic properties toward arterial and venous thrombi of streptokinase and recombinant staphylokinase in baboons. Circulation.. 1993;87:996-1006.
Gase A, Hartmann M, Gu¨hrs KH, Ro¨cker A, Collen D, Behnke D, Schlott B. Functional significance of NH2− and COOH-terminal regions of staphylokinase in plasminogen activators. Submitted.
Deutsch DG, Mertz ET. Plasminogen: purification from human plasma by affinity chromatography. Science.. 1970;170:1095-1096.
Silence K, Hartmann M, Gu¨hrs KH, Gase A, Schlott B, Collen D, Lijnen HR. Structure-function relationships in staphylokinase as revealed by ‘clustered charge-to-alanine’ mutagenesis. J Biol Chem.. 1995;270:27192-27198.
Stanssens P, Opsomer C, McKeown YM, Kramer W, Zabeau M, Fritz HJ. Efficient oligonucleotide-directed construction of mutations in expression vectors by the gapped duplex DNA method using alternating selectable markers. Nucleic Acids Res.. 1989;17:4441-4454.
Nakane PK, Kawaoi A. Peroxidase-labeled antibody: a new method of conjugation. J Histochem Cytochem.. 1974;22:1084-1091.
Jo¨nsson U, Malmqvist M. Real time biospecific interaction analysis: the integration of surface plasmon resonance detection, general biospecific interface chemistry and microfluidics into one analytical system. Adv Biosensors.. 1992;2:291-336.
BIAcore System Manual, 5-2. Uppsala, Sweden: Pharmacia Biosensor AB; 1991.
Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989.
Stern P. Predicting antigenic sites on proteins. Tibtech.. 1991;9:163-169.
Smith AM, Benjamin DC. The antigenic surface of staphylococcal nuclease, II: analysis of the N-1 epitope by site-directed mutagenesis. J Immunol.. 1991;146:1259-1264.