Mutated tissue plasminogen activators and uses thereof

Abstract

The present invention relates to mutated tissue plasminogen activators, and their use for treating thrombotic diseases.

Claims

1. A polynucleotide encoding for a protein selected from the group consisting of: i) a protein comprising sequence SEQ ID NO: 2 or SEQ ID NO:25, wherein said sequence comprises: a mutation A′ consisting of the replacement of at least one of an amino acid selected from the group consisting of the aspartic acid at position 236, the aspartic acid at position 238, and the tryptophan at position 253 of SEQ ID NO: 2 or SEQ ID NO:25 by a hydrophilic amino acid selected from the group consisting of arginine, glutamic acid, lysine, asparagine, glutamine, serine, threonine, tyrosine and histidine, or the tryptophan at position 253 is replaced by aspartic acid, or a mutation B consisting of the replacement of arginine in position 275 of SEQ ID NO: 2 or SEQ ID NO:25 by serine, or a double mutation A′ and B consisting of the replacement of at least one of an amino acid selected from the group consisting of the aspartic acid at position 236, the aspartic acid at position 238, and the tryptophan at position 253 of SEQ ID NO: 2 or SEQ ID NO:25 by a hydrophilic amino acid chosen from arginine, glutamic acid, lysine, asparagine, glutamine, serine, threonine, tyrosine and histidine, or the tryptophan at position 253 is replaced by aspartic acid, and the replacement of arginine in position 275 of SEQ ID NO: 2 or SEQ ID NO:25 by serine, ii) a protein comprising a sequence having at least 80% identity with SEQ ID NO: 2 over its whole length or SEQ ID NO:25 over its whole length, said protein comprising mutation A′, mutation B, or mutation A′ and B, and iii) a protein consisting of a fragment of SEQ ID NO:2, said fragment consisting of the Kringle 2 domain, the catalytic domain, and mutation A′, mutation B, or mutation A′ and B.

2. An expression vector comprising a polynucleotide according to claim 1.

3. An isolated host cell comprising an expression vector according to claim 2.

4. An isolated host cell comprising a polynucleotide according to claim 1.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) Important Note:

(2) In the following figures which are related to the example, amino acids are numbered from the N-terminal serine of the mature Rattus norvegicus tPA sequence (UniProtKB: P19637).

(3) FIG. 1A-B. Comparison table of three plasminogen activators. A) Human tPA, Rat tPA and Desmodus rotundus plasminogen activator (DSPA) exhibit an almost similar sequence of domains ranging from a finger domain at the amino terminal extremity to the protease domain at the carboxyl terminal extremity. The kringle 2 domain of tPA is absent in DSPA. Moreover human and rat tPA and DSPA share strong homologies (>80% between rat and human tPA; >65% between human tPA and DSPA). B) Map of the plasmid pcDNA5/FRT used in the experiments for the expression of rat wt tPA.

(4) FIG. 2A-B. Primary structure comparison of human and rat tPA and DSPA. A) Sequence analysis of the kringle domain of DSPA reveals naturally occurring amino acid substitutions leading to a non-functional lysine-binding site: the anionic charges in position D237 and D239 (black box 1) and the hydrophobic amino acid W254 (black box 2) are missing (SEQ ID NOs: 40-42). B) DSPA (SEQ ID NO: 44) is a specific protease in that it exists only in a single-chain form whereas proteases such as human (SEQ ID NO: 27) or rat tPA (SEQ ID NO: 43) may be processed into a two-chain form.

(5) FIG. 3A-C. Biochemical characterization of the tPA-related muteins. A) Equal amounts (100 ng) of wt tPA, ΔK2-tPA, K2*-tPA and sc*-tPA muteins were subjected to immunoblotting. (B-C) Activity of the tPA-related muteins measured either on a fluorogenic substrate (B) or by plasminogen-casein zymography assays (C).

(6) FIG. 4A-B. K2-related muteins have improved fibrinolytic properties. A) ΔK2-tPA and K2*-tPA reveal a fibrinolytic activity as wt-tPA when subjected to fibrin agarose zymography following non reduced SDS PAGE electrophoresis. B) In vitro evaluation of fibrinolytic activity using a platelet-poor plasma clot (PPP-clot). K2*-tPA and ΔK2-tPA muteins express improved global fibrinolytic efficiency compared to wt-tPA (26% and 51% respectively). Fibrinolytic activity was normalized to rat wt-tPA, using the half-time for clot lysis.

(7) (**p<0.02; ***p<0.01).

(8) FIG. 5A-C. Invalidation of the constitutive lysine-binding site of the tPA kringle 2 domain abolishes tPA-mediated neurotoxicity. (A-C) Neuronal death was assessed by measuring LDH release in the bathing media 24 hours after an 1 hour exposure of primary cultured cortical neurons (14 days in vitro—DIV) to 50 μM NMDA alone or supplemented with either (A) human tPA or rat wt-tPA (0.3 μM; n=12, 3 independent experiments) (B) wt-tPA, ΔK2-tPA or K2*-tPA (0.3 μM; n=12, 4 independent experiments) or (C) human tPA in the presence or not of 0.1 mM of the lysine analogue ε-ACA (ε-amino caproic acid) (n=19, 5 independent experiments). Data are presented as the mean value±SD of neuronal death in percent relative to control.

(9) (***p<0.01; ns: not significant).

(10) FIG. 6A-B. Fibrin partially restores plasminogen activation function of the inactive sc*-tPA. Whereas sc*-tPA is not able to promote alone the conversion of plasminogen into plasmin, its fibrin cofactor partially brings back its plasminogen activation function as detected (A) by fibrin agarose zymography following non reduced SDS PAGE electrophoresis and (B) in a platelet-poor plasma clot (PPP-clot) lysis assay. Fibrin clots restore the activity of sc*-tPA to a higher level than half the fibrinolytic activity of wt-tPA (n=3). Fibrinolytic activity was normalized to rat wt-tPA, using the half-time for clot lysis.

(11) (***p<0.01).

(12) FIG. 7. Restoring sc-tPA zymogenicity rescues neurons from tPA potentiation of NMDA-mediated neurotoxicity. Neuronal death was assessed by measuring LDH release in the bathing media 24 hours after a 1 hour exposure of primary cultured cortical neurons (14 days in vitro—DIV) to 50 μM NMDA alone or supplemented with either rat wt-tPA or rat sc*-tPA (0.3 UM; n=12, 4 independent experiments). Data are presented as the mean value±SD of neuronal death in percent relative to control.

(13) (***p<0.01; ns: not significant).

(14) FIG. 8. Characterization of the human tPA variants. Equal amounts (200 ng) of the human tPA variants were subjected to immunoblotting and compared to the commercially available forms of tPA (actilyse) and reteplase (rapilysin).

(15) FIG. 9. Biochemical characterization of the tPA variants. A), intrinsic activity of the tPA variants determined by measuring the increase in absorbance of the free chromophore (AMC) generated, in comparison to the original substrate, per unit time at λ440 nm. The fluorogenic substrate used is Spectrofluor tPA (formula: CH.sub.3SO.sub.2-D-Phe-Gly-Arg-AMC.AcOH, American Diagnostica). Measurements were performed in duplicate using 5 different doses, in three independent experiments. B), Fibrinolytic activity of the double or triple tPA mutants normalized to the commercially available tPA (actilyse) using the half-time for clot lysis toward euglobulin-derived clots. C), Fibrinolytic activity of the double or triple tPA mutants normalized to the commercially available tPA (actilyse) using the half-time for clot lysis toward whole plasma-derived clots.

(16) FIG. 10. In vitro proof of concept of the non-neurotoxic effect of the human tPA variants. Neuronal cell death was assessed by measuring lactate dehydrogenase release in the bathing media as described in the methods section. Human tPA (actilyse), hutPAsc* or Opt-PA2 (0.3 μM; 4 independent experiments) (for hutPAsc* and Opt-PA2 definitions, see table of the sequences above). Data are presented as the mean value±SD of neuronal death in percent relative to control; ns: not significant.

(17) FIG. 11. In vitro proof of concept of the non-neurotoxic effect of the human tPA variants. Neuronal cell death was assessed by measuring lactate dehydrogenase release in the bathing media as described in the methods section. Human tPA (actilyse), hutPAK2* or Opt-PA (0.3 μM; 4 independent experiments) (for hutPAK2* and Opt-PA definitions, see table of the sequences above). Data are presented as the mean value±SD of neuronal death in percent relative to control; ns: not significant.

(18) FIG. 12. Opt-PA does not promote NMDA-induced neurotoxicity in vivo. NMDA-induced excitotoxic brain lesions were measured by Magnetic Resonance Imaging (MRI) as described in the methods section, 24 hours after intrastriatal injection of NMDA (12.5 mM) alone or in combination with either actilyse (5 Opt-PA (5 μM), or hutPA K2* (5 μM) (for hutPAK2* and Opt-PA definitions, see table of the sequences above). Data are presented as the mean values±SD of lesion volumes in mm3.

(19) FIG. 13. Summary of the tPA-related muteins produced in the study.

(20) FIG. 14. Biochemical characteristics of the tPA variants. (1): sequence available in the UniProt Database, accession number P19637; (2): fibrinolytic activities obtained from euglobulin clot lysis time assay by reference to the International Reference Preparation (IRP 98/714) using the time to obtain 50% clot lysis; (3-4): Kd for fibrin (3) and Km and kcat for plasminogen in the presence of fibrin (4) obtained from 3 independent experiments (12 tested doses).

EXAMPLE 1: RAT TPA MUTANTS

(21) Important Note:

(22) In the following study, amino acids are numbered from the N-terminal serine of the mature Rattus norvegicus tPA sequence (UniProtKB: P19637).

(23) Material and Methods

(24) Chemicals.

(25) N-methyl-D-aspartate (NMDA) was purchased from Tocris (Bristol, United Kingdom). Spectrofluor 444FL was purchased from American Diagnostica (Stamford, USA). 6-aminocaproic acid (E-ACA), Dulbecco's modified Eagle's medium (DMEM), poly-D-lysine, cytosine fl-D-arabinoside and hygromycin B were from Sigma-Aldrich (L'Isle d'Abeau, France). The QuickChange XL site-directed mutagenesis kit was from Stratagene (La Jolla, Calif., USA). Plasminogen was purchased from Calbiochem (Nottingham, United Kingdom). Lipofectamine 2000, Opti-MEM RSM, foetal bovine and horse sera, laminin were from Invitrogen (Cergy Pontoise, France). tPA (Actilyse®) came from Boehringer-Ingleheim (Germany)

(26) Construction of Wild-Type tPA and ΔK2-tPA Muteins in pcDNA5/FRT Vector.

(27) The full-size rat wild-type tPA coding sequence was amplified by PCR using an upstream primer 5′ CCGGGATCCTCCTACAGAGCGACC 3′ (SEQ ID NO:17) and a downstream primer 5′ GGCAAGCTTTTGCTTCATGTTGTCTTGAATCCAGTT 3′ (SEQ ID NO:18). A 6×His tag was placed at the N-terminal position of the mature protein. Digested PCR products were then inserted into a pcDNA5/FRT vector (Invitrogen, Cergy-Pontoise, France). Fusion PCR was performed to obtain ΔK2-tPA from wt-tPA coding sequence using the same protocol with the following fusion primers: upstream 5′ CAGGCCGCACGTGGAGTCCTGAGTTGGTCCCTTAGG 3′ (SEQ ID NO:19) and downstream 5′ TCCACCTGCGGCCTG 3′ (SEQ ID NO:20). Final constructs were checked using an automated sequence analysis.

(28) Site-Directed Mutagenesis.

(29) Mutagenesis of full-length tPA wt (tPA W254R) has been performed by using QuikChange® XL Site-Directed Mutagenesis Kit purchased from Stratagene (Agilent Technologies, Massy, France) and the following primers 5′ GGACCGAAAGCTGACACGGGAATATTGCGACATGTCC 3′ (SEQ ID NO:21) and 5′ GGACATGTCGCAATATTCCCGTGGTCAGCTTTCGGTCC 3′ (SEQ ID NO:22). Non-cleavable tPA (tPA R276S) has been obtained using 5′ TACAAACAGCCTCTGTTTCGAATTAAAGGAGGA 3′ (SEQ ID NO:23) and 5′ TCCTCCTTTAATTCGAAACAGAGGCTGTTTGTA 3′ (SEQ ID NO:24) primers. Mutations have been confirmed using an automated sequence analysis.

(30) Human Embryonic Kidney (HEK)-293 Cell Cultures and Stable Transfection.

(31) Human embryonic kidney 293 cells already stable transfected with the pFRT/lacZeo vector (HEK-FlpIn, Invitrogen) were grown in RPMI-1640 medium supplemented with 10% fetal bovine serum and 2 mM glutamine. Cells at high confluence were transfected using lipofectamine 2000 reagent according to manufacturer protocol (Invitrogen) with a mixture containing the tPA-related plasmids and the plasmid helper pOG44. After 24 hours, cells were washed. 48 hours after transfection positive clones were isolated by hygromycine B selection. The quality of the transfection was assessed by RT-PCRq.

(32) Conditioned Media-Containing the tPA-Related Muteins.

(33) High confluency cells stable transfected with the different tPA-related plasmids were incubated for 24 hours in minimal medium composed of Opti-MEM RSM (Invitrogen) added of 2 mM glutamine et containing 10 IU/ml aprotinin and 200 μg/ml hygromycin B. Supernatant were harvested in 0.01% azide, 2 mM EDTA, 0.01% tween 20, centrifuged 15 minutes at 10.000 g and finally stored at −20° C.

(34) Bioreactor Production of the tPA-Related Muteins.

(35) To produce high level of muteins, stable transfected HEK cells were grown in a laboratory-scale bioreactor CELLine AD 1000. Two weeks after a 1×10.sup.6 viable cells/ml inoculation, cell compartment is harvested twice a week during four months. Each harvested supernatant is controlled in terms of pH, turbidity, centrifuged 15 minutes at 10.000 g and stored at −20° C. prior to 6×his purification.

(36) 6×his Muteins Purification.

(37) Purification was processed using nickel-nitrilotriacetic acid (Ni-NTA) metal-affinity chromatography matrice (Qiagen, Courtaboeuf, France) according to manufacturer protocol. Muteins were then conditioned in a NH.sub.4HCO.sub.3 0.5 M buffer, quantified and stored.

(38) tPA Immunoblotting.

(39) Immunoblottings were performed using a monoclonal mouse antibody raised against a penta-histidine sequence (1/1000eme), followed by incubation with the appropriate biotinilated-conjugated secondary antibody. Signal was amplified using the Extravidine (Sigma) biotin-peroxydase conjugate (1/5000). Immunoblots were revealed with an enhanced chemoluminescence ECL Plus immunoblotting detection system (Perkin Elmer-NEN, Paris, France).

(40) SDS-PAGE Plasminogen-Casein Zymography.

(41) Zymography assay was performed by addition of plasminogen (4.5 μg/ml) and casein (1%) in 10% SDS-polyacrylamide gels. Electrophoresis was performed at 4° C. Gels were washed with Triton X-100 (2.5%) and incubated for 2 hours at 37° C. Caseinolytic bands were visualized after Coomassie staining.

(42) Amidolytic Activity Assay.

(43) tPA-related muteins were incubated in the presence of a fluorogenic substrate (5 μM) (Spectrofluor® FL444). The reaction was carried out at 37° C. in 50 mM Tris (pH 8.0) containing 150 mM NaCl in a total volume of 100 μL. The amidolytic activity was measured as the change in fluorescence emission at 440 nm (excitation at 360 nm). Using Spectrozyme®, an amidolytic substrate (Spectrozyme tPA, SptPA)), wt-tPA and sc*-tPA (0.3 nM) were incubated with increasing concentrations of the SptPA (0-1 mM) in a microplate (200 μL per well) and OD405 nm recorded every minute using a microplate spectrophotometer (ELx 808, Biotek, USA). Then, the maximal velocity (Vmax) of the reaction was calculated and the data were plotted as follows:

(44) $\frac{1}{V}=f\left(\frac{1}{\left[\mathrm{SptPA}\right]}\right)=\frac{\mathrm{Km}}{V_m}.Math.\left(\frac{1}{\left[\mathrm{SptPA}\right]}\right)+\frac{1}{V_m}.$
Fibrin Agarose Zymography.

(45) Proteins (10 μg) and reference proteins (10 μL of tPA 0.06 iu/mL, uPA 0.25 iu/mL and plasmin 200 nM) were electrophoresed in a 8% polyacrylamide gel under non-reducing conditions. SDS was then exchanged with 2.5% Triton X-100. After washing-off excess Triton X-100 with distilled water, the gel was carefully overlaid on a 1% agarose gel containing 1 mg/mL of bovine fibrinogen, 100 nM plasminogen and 0.2 NIH unit/mL of bovine thrombin. Zymograms were allowed to develop at 37° C. during 12 h and photographed at regular intervals using dark-ground illumination. Active proteins in cell lysates were identified by reference to the migration of known markers (uPA, tPA, plasmin). To verify the activator identity, zymograms were made on a fibrin-agarose gel containing a polyclonal antibody directed against tPA or a non immun IgG.

(46) Clot Lysis Time.

(47) Human plasma was collected and the euglobulin fractions, containing β- and γ-globulins were separated by dilution of one volume of chilled plasma in 20 volumes of chilled acetic acid 2.9 mM. After incubation at 4° C. for 15 minutes and centrifugation at 3000 g for 10 minutes, the euglobulin fraction was precipitated, the supernatant discarded and the precipitate dissolved in HEPES buffer (10 mM HEPES pH 7.4, 150 mM NaCl). The euglobulin solution (100 μL) was supplemented with 15 mM calcium chloride and 5, 10, 15, 20, 25 or 30 I.U. of the tPA muteins. The time to clot lysis was recorded by optical density (405 nm absorbance) at 37° C. Tests were performed in duplicate. Results are expressed as the time to 50% clot lysis.

(48) Neuronal Cell Culture.

(49) Neuronal cultures were prepared from foetal mice (embryonic day 15-16) as previously described (Nicole et al., 2001). Briefly cortices were dissected and dissociated in DMEM, and plated on 24-well plates previously coated with poly-D-Lysine (0.1 mg/mL) and laminin (0.02 mg/mL). Cells were cultured in DMEM supplemented with 5% fetal bovine serum, 5% horse serum and 2 mM glutamine. Cultures were maintained at 37° C. in a humidified 5% CO/atmosphere. Cytosine β-D-arabinoside (10 μM) was added after 3 days in vitro (DIV) to inhibit glial proliferation. Various treatments were performed after 14 DIV.

(50) Excitotoxic Neuronal Death.

(51) Excitotoxicity was induced by exposure of cortical neurons to NMDA (50 μM) in serum-free DMEM supplemented with 10 μM of glycine, for 1 hour. Recombinant human tPA and rat tPA-related muteins were applied with NMDA when indicated. Neuronal death was quantified 24 hours later by measuring the activity of lactate dehydrogenase (LDH) released from damaged cells into the bathing medium by using a cytotoxicity detection kit (Roche Diagnostics; Mannheim, Germany). The LDH level corresponding to the maximal neuronal death was determined in sister cultures exposed to 200 μM NMDA (LDH.sub.max). Background LDH levels were determined in sister cultures subjected to control washes (LDH.sub.min). Experimental values were measured after subtracting LDH.sub.min and then normalized to LDH.sub.max−LDH.sub.min in order to express the results in percentage of neuronal death relative to control.

(52) Kinetics of Plasminogen Activation in the Presence of Fibrin.

(53) Kinetics of the activation of plasminogen on a fibrin surface were determined for each of the tPA mutants as previously describe by Angles-cano et al. Briefly, fibrinogen (0.3 μM) was immobilized on PVC plates previously activated by glutaraldehyde. Then, thrombin (10 NIH U/mL) was added for 2 h at 37° C. to convert fibrinogen into fibrin. The plates are then washed with 9 nM PPACK-containing binding buffer (50 mM PO.sub.4 pH 6.8, 80 mM NaCl, 0.4% BSA, 0.01% Tween 20, 0.01% azide and 2 mM EDTA). tPA variants were then incubated on fibrin surfaces for 1 h at 37° C. with 50 pd., of binding buffer. Unbound proteins were eliminated by washing with a buffer (50 mM PO.sub.4 pH 7.4, 80 mM NaCl, 0.2% BSA, 0.01% Tween 20, 0.01% azide) and the reaction started by adding 50 μL of assay buffer (50 mM PO.sub.4 pH 7.4, 80 mM NaCl, 0.2% BSA) containing increasing amounts of plasminogen (0-500 nM) and a fixed concentration (0.75 mM) of the plasmin-selective chromogenic substrate (CBS0065, Diagnostica STAGO, Asnieres, France). The absorbance at 405 nm was recorded for 18 h using a spectrophotometer (ELx 808, Biotek, USA), and data were plotted as follows: ([Pn]=f(t)). The maximal velocity (M.sub.Pn.Math.s.sup.−1) was measured for each activator concentration and was plotted against activator concentrations (Vi=f([Pg]). Kinetic parameters were determined by fitting data to the Lineweaver-Burk equation:

(54) $\frac{1}{\mathrm{Vi}}=\frac{\mathrm{Km}}{V_M}\left(\frac{1}{\left[\mathrm{Pg}\right]}\right)+\frac{1}{V_M}$
The kcat was calculated by using the following equation:

(55) $\mathrm{kcat}=\frac{V_M}{\left[\mathrm{tPA}\right]}$
Statistical Analysis.

(56) All the statistical analyses were performed by the two-tailed Kruskall-Wallis' test, followed by post-hoc comparisons, with the two-tailed Mann-Whitney's test. Results are expressed as mean±SD relative to control. Statistical significance is considered for p<0.05.

(57) Results

(58) Generation of New Thrombolytics Originated from tPA.

(59) Structural differences between human tPA (UniProtKB: P00750), rat tPA (UniProtKB: P19637) and DSPAα1 (named DSPA) (UniProtKB: P98119) were studied using multiple alignments. Rat tPA shares 81% amino acids identity and 89% conserved substitutions with the human tPA (FIG. 1A). DSPA shares 67% amino acids identity and 79% conserved substitutions with the rat tPA. DSPA contains a single kringle domain having a high degree of amino acid sequence homology with the tPA's kringle 1 domain (FIG. 2A), including the absence of a constitutive lysine-binding site (FIG. 2A—black boxes). On the other hand the tPA's kringle 2 domain contains a constitutive lysine-binding site formed by the pair of aspartic acid in position 237 and 239 and the tryptophane in position 254. A second point of interest is that in contrast to tPA, DSPA is an exclusive single-chain serine protease (Schleuning et al., 1992). Indeed, analysis of the primary sequence of DSPA reveals the lack of the cleavage site present in tPA, Arg276-Iso277 (FIG. 2B). All these features of DSPA when compared to tPA are interestingly associated with an increased affinity for fibrin (Schleuning et al., 1992) and a lack of neurotoxicity (Liberatore et al., 2003).

(60) Thus, based on these observations, the inventors have designed and generated three muteins derived from the rat tPA (Rattus norvegicus) (rat wild type tPA named wt-tPA): (i) a rat tPA genetically engineered with complete deletion of its K2 domain (deletion of the amino acids 181 to 262), named ΔK2-tPA; (ii) a rat tPA containing a tryptophan to arginine point mutation at position 254 (W254R), named K2*-tPA; (iii) an exclusive rat single-chain tPA obtained by an arginine to serine point mutation at position 276 (R276S), named sc*-tPA (FIG. 13). After PCR-induced appropriate deletion/mutation as described above, the corresponding 6× histidine-tagged cDNAs were inserted into a mammalian expression vector pcDNA5/FRT (FRT: Flp Recombination Target) (FIG. 1B) and stable transfected in HEK-293 cells expressing the Flp-In system (Invitrogen) for stable production of the corresponding recombinant proteins, as described in the methods section. Once purified using nickel affinity chromatography, the muteins were subjected to SDS-PAGE electrophoresis and immunoblotting. wt-tPA, sc*-tPA and K2*-tPA displayed similar molecular weights, whereas the K2 deleted tPA, ΔK2-tPA, showed a 15 kDa reduced molecular weight (FIG. 3A). Interestingly, sc*-tPA is present under its exclusive single-chain form whereas wt-tPA, K2*-tPA and ΔK2-tPA present two-chain forms (at 35 kDa and 25 kDa for ΔK2-tPA). Thus the R276S point mutation (sc*-tPA) leads to the generation of a non-cleavable form of tPA Because tPA binds and cleaves several substrates beyond plasminogen, such as the PDGF-C or the NR1 subunit of the NMDAR with no identified allosteric regulator, the inventors have first evaluated the intrinsic proteolytic activity of each of these muteins. Thus, plasminogen-containing zymography assays (FIG. 3B) and amidolytic activity assays toward a fluorogenic substrate (Spectrofluor) (FIG. 3C) were performed for the different tPA-related muteins cited above. Our data reveal that although wt-tPA and kringle 2-related mutants (ΔK2-tPA and K2*-tPA) display amydolytic activity comparable to that observed for wt-tPA, sc*-tPA does not. Hereafter, muteins concentrations are normalised to their intrinsic proteolytic activity.

(61) The inventors measured the ability of each of the tPA mutants to bind fibrin with Kd's of 0.26 nM, 1.2 nM, 0.5 nM and 0.82 nM for wt-tPA, sc*-tPA, K2*-tPA and ΔK2-tPA, respectively (FIG. 14).

(62) tPA is known to bind and cleave several substrates beyond plasminogen (such as the GluN1 subunit) with no identified allosteric regulator. Therefore, the inventors evaluated the intrinsic proteolytic activity of each of the tPA variants. As such, amidolytic activity assays toward a fluorogenic substrate (Spectrofluor) and plasminogen-containing zymography assays were performed for the different tPA-related mutants cited above. The data reveal that, although wt-tPA and kringle 2-related mutants (ΔK2-tPA and K2*-tPA) display an amidolytic activity comparable to that observed for wt-tPA, sc*-tPA does not. To further investigate the behavior of the sc*-tPA variant when compared to the wt-tPA, the inventors determined the Km of both plasminogen activators by using the amidolytic Spectrozyme®, as the substrate. The data showed that, the point mutation within the cleavage site of tPA leads to a 3-fold increase of the Km value when compared to the wt-tPA (2.83E-04 and 9.12E-05 M, respectively). Hereafter, concentrations of the tPA mutants are normalised to their intrinsic amidolytic activity, unless otherwise mentioned.

(63) Kringle 2-Related Muteins (ΔK2-tPA and K2*-tPA) Display a Higher Fibrinolytic Activity and Failed to Promote NMDA Receptors Mediated Neurotoxicity.

(64) K2-related muteins were characterized toward their ability to initiate fibrinolysis on fibrin-agar plates as described in the methods section. ΔK2-tPA and K2*-tPA trigger activation of plasminogen into plasmin in the presence of fibrin as wt-tPA does (FIG. 4A). In vitro clot lysis assay, performed on platelet-poor human plasma clot (PPP-clot) as substrate, revealed that K2*-tPA and ΔK2-tPA displayed an enhanced fibrinolytic activity when compared to wt-tPA (+26% and +51% respectively) (FIG. 4B). To estimate their effect on NMDA receptor mediated neurotoxicity, pure cultures of cortical neurons (14 days in vitro) were subjected to 1 hour exposure of 50 μM NMDA either alone or in combination with either purified ΔK2-tPA or K2*-tPA (0.3 μM equivalent of their respective amidolytic activity) prior measure of the neuronal death 24 hours later. Although the rat wt-tPA leads to a 39% potentiation of NMDAR-mediated excitotoxicity (71% of neuronal death when compared to 51% with NMDA alone), an effect similar to what is observed for Actilyse®-containing human tPA (FIG. 5A; n=3, p<0.01), ΔK2-tPA and K2*-tPA (FIG. 5B; n=4, p<0.01) have no pro-neurotoxic profiles. Thus, the tryptophan 254, a constitutive amino-acid of the kringle 2 LBS of tPA is critical to mediate the pro-neurotoxicity of tPA. Accordingly, same experiments performed in the presence of ε-amino caproic acid (8-ACA), a lysine analog known to compete with the LBS of tPA, show that blockage of the LBS function prevented wild type tPA-induced potentiation of NMDAR-mediated neurotoxicity (FIG. 5C; n=5, p<0.01).

(65) A Zymogenic tPA (sc*-tPA) Displays a Non Pro-Neurotoxic Profile.

(66) The inventors have tested both the fibrinolytic activity and the pro-neurotoxicity of the non-cleavable form of rat tPA, sc*-tPA, generated and purified as described above. In contrast to its lack of intrinsic amidolytic activity (FIG. 3B-C), sc*-tPA remains fibrinolytic in the presence of fibrin (FIG. 6A) despite a lower activity to that of wt-tPA (−39%) (FIG. 6B). Then, this mutein was tested for its ability to influence NMDAR-mediated neurotoxicity in primary cultures of cortical neurons. Interestingly, sc*-tPA fails to potentiate NMDA receptors-dependent excitotoxicity when compared to wt-tPA (n=3, p<0.01) (FIG. 7).

(67) Altogether, the inventors have generated and characterized a set of original fibrinolytics derived from tPA: a K2*-tPA (SEQ ID NO: 4) characterized by a higher fibrinolytic activity and a lack of pro-neurotoxicity and a sc*-tPA (SEQ ID NO: 8) characterized by both a lack of amydolytic activity and pro-neurotoxicity despite a conserved fibrinolytic activity. These in vitro data provide the bases of further studies to evaluate the efficacy of this new generation of fibrinolytics in experimental models of thrombosis, prior possible transfer to clinical applications.

EXAMPLE 2: HUMAN TPA MUTANTS

(68) Material and Methods

(69) Chemicals

(70) N-methyl-D-aspartate (NMDA) was purchased from Tocris (Bristol, United Kingdom); Spectrofluor 444FL from American Diagnostica (ADF Biomedical, Neuville-sur-oise, France); 6-aminocaproic acid (s-ACA), Dulbecco's modified Eagle's medium (DMEM), poly-D-lysine, cytosine β-D-arabinoside and hygromycin B from Sigma-Aldrich (L'Isle d'Abeau, France). Lipofectamine 2000, foetal bovine and horse sera, laminin and the GeneArt® Site-Directed Mutagenesis System were from Invitrogen (Cergy Pontoise, France). tPA (Alteplase®) came from Boehringer-Ingleheim (Paris, France). Reteplase (Rapilysin) came from Actavis (Paris, France).

(71) Construction of Wild-Type tPA in pcDNA5/FRT Vector.

(72) The human tPA was amplified by PCR using primers: 5′ GGCGCTAGCATGGATGCAATGAAGAGAGGGC 3′ (SEQ ID NO:32) and 5′ CCGGGCAAGCTTTTGCTTCATGTTGTCTTGAATCCAGTT 3′ (SEQ ID NO:33) (with a 6×His tag at the N-terminal position of the mature protein). PCR products were inserted into a pcDNA5/FRT vector (Invitrogen, Cergy-Pontoise, France). Final construct was automatically sequenced.

(73) Site-Directed Mutagenesis

(74) Mutagenesis of hutPAwt was performed using GeneArt® Site-Directed Mutagenesis System and the following primers:

(75) TABLE-US-00002 tPA K2* (W253R) of SEQ ID NO: 28: (SEQ ID NO: 34) 5′ GCCAAGCCCCGGTGCCACGTGC 3′ and (SEQ ID NO: 35) 5′ GCACGTGGCACCGGGGCTTGGC 3′. tPA sc* (R275S) of SEQ ID NO: 29: (SEQ ID NO: 36) 5′ GTACAGCCAGCCTCAGTTTAGCATCAAAGGAGGGC 3′ and (SEQ ID NO: 37) 5′ AAACTGAGGCTGGCTGTACTGTCTCAGGCCGC 3′. P125R point mutation of tPA of SEQ ID NO: 31: (SEQ ID NO: 38) 5′ GCAGCGCGTTGGCCCAGAAGCGCTACAGCGGGC 3′ and (SEQ ID NO: 39) 5′ CTTCTGGGCCAACGCGCTGCTGTTCCAGTTGG 3′.

(76) Mutations were confirmed by sequence analysis.

(77) Human Embryonic Kidney (HEK)-293 Cell Cultures and Stable Transfection Stable human embryonic kidney 293 cells transfected with the pFRT/lacZeo vector (HEK-FlpIn, Invitrogen) were grown in RPMI-1640 medium supplemented with 10% foetal bovine serum and 2 mM glutamine. Cells were transfected using lipofectamine 2000. Positive clones were isolated by hygromycine B selection. The quality of the transfection was assessed by RT-PCRq.
Bioreactor Production of the tPA-Related Mutants

(78) To produce high yields of mutant genes, stable transfected HEK cells were grown in a laboratory-scale bioreactor CELLine AD 1000 (Dominique Dutscher SAS, Brumath, France).

(79) Purification of 6×his Mutants

(80) Purification was processed using nickel-nitrilotriacetic acid (Ni-NTA) metal-affinity chromatography matrice (Qiagen, Courtaboeuf, France). tPA mutants were then conditioned in a NH.sub.4HCO.sub.3 0.5 M buffer and stored.

(81) tPA Immunoblotting

(82) Immunoblottings were performed using a polyclonal sheep antiserum raised against human tPA (1:5000) prepared at the National institute for agronomic research (INRA, Clermont-Theix, France) and a polyclonal rabbit antiserum raised against murine tPA (125 ng/μl), followed by incubation with the appropriate peroxidase-conjugated secondary antibody. Immunoblots were revealed with an enhanced chemoluminescence ECL Plus immunoblotting detection system (Perkin Elmer-NEN, Paris, France).

(83) Amidolytic Activity Assay

(84) tPA variants were incubated in the presence of a fluorogenic substrate (5 μM) (Spectrofluor® FL444). The reaction was carried out at 37° C. in 50 mM Tris (pH 8.0) containing 150 mM NaCl in a total volume of 100 μL. The amidolytic activity was measured as the change in fluorescence emission at 440 nm (excitation at 360 nm).

(85) Clot Lysis Time

(86) Human plasma was obtained from citrated blood. Plasma was supplemented with 15 mM of calcium chloride and each of the tPA mutants at 400, 420, 440, 460, 480 and 500 I.U. The euglobulin fraction was recovered as described above, supplemented with 15 mM calcium chloride and 15, 20, 25, 30, 35 or 40 I.U. of the tPA muteins. The time to clot lysis was recorded by optical density measurements (A405 nm) at 37° C. by reference to the commercially available form of tPA (actilyse). Tests were performed in duplicate (from 3 independent experiments). Results are expressed as the time to obtain 50% clot lysis.

(87) Neuronal Cell Culture

(88) Neuronal cultures were prepared from foetal mice (embryonic day 15-16). Cortices were dissected and dissociated in DMEM, and plated on 24-well plates previously coated with poly-D-Lysine (0.1 mg/mL) and laminin (0.02 mg/mL). Cells were cultured in DMEM supplemented with 5% foetal bovine serum, 5% horse serum and 2 mM glutamine. Cultures were maintained at 37° C. in a humidified 5% CO.sub.2 atmosphere. Cytosine β-D-arabinoside (10 μM) was added after 3 days in vitro (DIV) to inhibit glial proliferation. Various treatments were performed after 14 DIV.

(89) Excitotoxic Neuronal Death

(90) Excitotoxicity was induced by exposure of cortical neurons to NMDA (50 μM) in serum-free DMEM supplemented with 10 μM of glycine, for 1 hour. The different tPA variants were applied with NMDA when indicated. Neuronal death was quantified 24 hours later by measuring the activity of lactate dehydrogenase (LDH) released from damaged cells into the bathing medium by using a cytotoxicity detection kit (Roche Diagnostics; Mannheim, Germany). The LDH level corresponding to the maximal neuronal death was determined in sister cultures exposed to 200 μM NMDA (LDHmax). Background LDH levels were determined in sister cultures subjected to control washes (LDHmin). Experimental values were measured after subtracting LDHmin and then normalized to LDHmax−LDHmin in order to express the results in percentage of neuronal death relative to control.

(91) Excitotoxic Lesion

(92) Excitotoxic lesions were performed under isoflurane-induced anaesthesia in male swiss mice (25-30 g; CURB, Caen, France). Striatal injections (coordinates: 0.5 mm posterior, +2.0 mm lateral, −3.0 mm ventral to the bregma; Paxinos & Watson, 1995) of 12.5 nmol NMDA versus either NMDA/actilyse, NMDA/Opt-PA or NMDA/hutPA K2* (12.5 mM NMDA and 5 μM equivalent amidolytic activity of tPA; total volume of 1 μl) were performed after placing the animals under a stereotaxic frame. Injections were made using adapted needles (calibrated at 15 mm/μL; assistant ref 555/5; Hoecht, Sodheim-Rhoen, Germany) and removed 5 minutes later. After 24 hours, brains were MRI analysed.

(93) Magnetic Resonance Imaging (MRI)

(94) Experiments were carried out at 24 hours following excitotoxic lesions on a Pharmascan 7T (Bruker, Germany). T2-weighted images were acquired using a Multi-Slice Multi-Echo (MSME) sequences: TE/TR 51.3 ms/1700 ms with 70×70×350 μm3 spatial resolution. Lesion sizes were quantified on these images using ImageJ software (v1.45r).

(95) Results

(96) Generation of New Thrombolytics Originated from tPA.

(97) The inventors have designed and generated six tPA mutants derived from the human tPA: (i) a human wild-type tPA named hutPA wt (SEQ ID NO: 27); (ii) a human tPA genetically engineered with complete deletion of its K2 domain (deletion of the amino acids 180 to 261), named hutPA ΔK2; (iii) a human tPA containing a tryptophan to arginine point mutation at position 253 (W253R, SEQ ID NO: 28), named hutPA K2*; (iv) an exclusive human single-chain tPA obtained by an arginine to serine point mutation at position 275 (R275S, ID SEQ NO: 29), named hutPA sc*; (v) a human tPA containing the double mutation W253R R275S, named Opt-PA (SEQ ID NO: 30); (vi) a human tPA containing the triple mutation P125R W253R R275S (SEQ ID NO: 31), named Opt-PA2. After PCR-induced appropriate deletion/mutation as described above, the corresponding 6× histidine-tagged cDNAs were inserted into a mammalian expression vector pcDNA5/FRT and stable transfected in HEK-293 cells expressing the Flp-In system (Invitrogen) for stable production of the corresponding recombinant proteins, as described in the methods section. Once purified using nickel affinity chromatography, the tPA mutants were subjected to SDS-PAGE electrophoresis and immunoblotting. Reteplase and activase were used as standards (FIG. 8). Interestingly, tPA mutants carrying the R275S point mutation are present under their exclusive single-chain form.

(98) Biochemical Characterization of the Human Derived tPA Mutants.

(99) The inventors have first evaluated the intrinsic proteolytic activity of each of these mutants. Thus amidolytic activity assay toward a fluorogenic substrate (Spectrofluor) (FIG. 9—LEFT) was performed. Our data reveal that sc*-tPA, Opt-PA and Opt-PA 2 show an amidolytic activity decreased by 10, 2.5 and 2 respectively. The mutants were characterized toward their ability to initiate fibrinolysis in models of in vitro clot assays performed on platelet-poor human plasma clot (PPP-clot) as substrate. These assays reveal that both Opt-PA and Opt-PA show similar potentiality to trigger fibrinolysis (FIG. 9—CENTER), even in the presence of the tPA′ inhibitors (differences are no more than order of magnitude, FIG. 9—RIGHT).

(100) R275S Point Mutation is not Sufficient to Abolish tPA-Related NMDA Receptors Mediated Neurotoxicity.

(101) To estimate the effect of the tPA mutants hutPA sc* and Opt-PA 2 on NMDA receptor mediated neurotoxicity, pure cultures of cortical neurons (14 days in vitro) were subjected to 1 hour exposure of 50 μM NMDA either alone or in combination with the purified mutants (0.3 μM) prior measure of the neuronal death 24 hours later. Although actilyse leads to a 61% potentiation of NMDAR-mediated excitotoxicity (59% of neuronal death when compared to 37% with NMDA alone), a similar effect is observed for hutPA sc* (62% of neuronal cell death, FIG. 10; n=4, p<0.05). Thus the R275S point mutation is not sufficient by itself to abolish tPA-related NMDA receptors mediated neurotoxicity. The inventors also tested the triple mutant Opt-PA in the above experimental setting up. They observed a marked tendency (but not significant) to abolish tPA-related NMDA receptor mediated neurotoxicity (51% of neuronal cell death, n=4, p=0.08).

(102) The Kringle 2-Related Human tPA Mutants Show a Non-Neurotoxic Profile.

(103) hutPA K2* and Opt-PA were used in place of hutPA sc* and Opt-PA2 in the excitotoxic neuronal death assay (FIG. 11). Here, although actilyse leads to a 41% potentiation of NMDAR-mediated excitotoxicity (75% of neuronal death when compared to 53% with NMDA alone), hutPA K2* and Opt-PA do not promote NMDAR-mediated neurotoxicity (64% and 62% of neuronal cell death respectively, FIG. 11; n=4, p<0.05). Thus, the tryptophan 253, a constitutive amino-acid of the kringle 2 LBS of tPA is critical to mediate the pro-neurotoxicity of tPA. The two tPA mutants hutPA K2* and Opt-PA have an interesting non-neurotoxic profile.

(104) Opt-PA does not Increase Neurotoxicity in an In Vivo Model of Striatal Lesion.

(105) The inventors have then tested the neurotoxicity of both hutPA K2* and Opt-PA in a model of striatal lesion in vivo. As described in the method section, 12.5 mM of NMDA and 5 μM of the tPA variants are injected into the striatum of swiss mice. 24 hours after injection the lesion volume is measured using non-invasive MRI imaging (FIG. 12). Whereas actilyse leads to a 48% potentiation of NMDA-mediated excitotoxicity (5.36 mm.sup.3 lesion volume % when compared to 3.62 mm.sup.3 with NMDA alone), hutPA K2* has an heterogeneous neurotoxic effect (5.15 mm.sup.3 lesion volume) and Opt-PA does not promote NMDA-mediated neurotoxicity (4.00 mm.sup.3 lesion volume, FIG. 12; n=11, p<0.05)

(106) Altogether, the inventors have generated and characterized a set of original fibrinolytics derived from human tPA. From this set of mutants, Opt-PA (SEQ ID NO: 30) is characterized by a fibrinolytic activity similar to actilyse and a lack of pro-neurotoxicity in vitro and in vivo.

(107) These data provide the bases of further studies to evaluate the efficacy of this new fibrinolytic in experimental models of thrombosis, prior possible transfer to clinical applications.