AU2010200718A1 - Intravenous injection of non-neurotoxic plasminogen activators for the treatment of stroke - Google Patents

Description

AUSTRALIA Patents Act COMPLETE SPECIFICATION (ORIGINAL) Class Int. Class Application Number: Lodged: Complete Specification Lodged: Accepted: Published: Priority Related Art: Name of Applicant: PAION Deutschland GmbH Actual Inventor(s): Robert Medcalf, Wolf-Dieter Schleuning, Mariola S6hngen, Wolfgang S6hngen Address for Service and Correspondence: PHILLIPS ORMONDE FITZPATRICK Patent and Trade Mark Attorneys 367 Collins Street Melbourne 3000 AUSTRALIA Invention Title: INTRAVENOUS INJECTION OF NON-NEUROTOXIC PLASMINOGEN ACTIVATORS FOR THE TREATMENT OF STROKE Our Ref: 880179 POF Code: 1282/472115 The following statement is a full description of this invention, including the best method of performing it known to applicant(s): - 1- "Intravenous injection of non-neurotoxic plasminogen activators for the treatment of stroke" The present application is a divisional application from Australian patent application number 2004233601, the entire disclosure of which is incorporated herein by reference. The invention refers to the intravenous application of non-neurotoxic plasmi nogen activators, in particular of genetically modified plasminogen activators and plasminogen activators from the saliva of Desmodus rotundus (DSPA) for the treatment of stroke in humans. The treatment of stroke with these plasminogen activators is known from the international patent application PCT/EP02/12204, the disclosure of which is completely taken into reference. Clinical characteristics and biochemistry of stroke Different clinical pictures are summarized under the term "stroke" which cor relate in their clinical symptoms. According to the respective pathogenesis a first differentiation between these clinical pictures in so called ischaemic and haemorrhagic insults is possible. Ischaemic insults (ischaemia) are characterized in a reduction or interruption of the blood circulation in the brain due to a lack of arterial blood supply. Often this is caused by thrombosis of an arteriosclerotic stenosed vessel or by arterio arterial, respecitively, cardial embolisms. Haemorrhagic insults are based inter alia on the perforation of brain supply ing arterias damaged by arterial hypertonia. However, only approximately 20 % of all cerebral insults are caused by haemorrhagic insults. Thus, stroke due to thrombosis is much more relevant. In comparison to other tissue ischaemias the ischaemia of the neuronal tis sue is widely accompanied by necrosis of the effected cells. The higher inci dence of necrosis in neuronal tissue can be explained with the new under standing of the phenomenon "excitotoxicity" which is a complex cascade comprising a plurality of reaction steps. The cascade is initiated by ischae- -2 mic neurons affected by a lack of oxygen which then lose ATP instantane ously and depolarize. This results in an increased postsynaptic release of the neurotransmitter glutamate which activates membrane bound glutamate receptors regulating cation channels. However, due to the increased gluta mate release glutamate receptors become over activated. Glutamate receptors regulate voltage dependent cation channels which are opened by a binding of glutamate to the receptor. This results in a Na* and Ca2+ influx into the cell massively disturbing the Ca2+ dependent cellular metabolism. Especially the activation of the Ca2+ dependent catabolic enzymes could give reason to the subsequent cell death (Lee, Jin-Mo et al., "The changing landscape of ischaemic brain injury mechanisms"; Dennis W. Zhol "Glutamate neurotoxicity and diseases of the nervous system"). Although the mechanism of glutamate mediated neurotoxicity is not yet entirely understood it is agreed upon that it contributes in a large extent to the neuronal cell death following cerebral ischaemia (Jin-Mo Lee, et al.). Therapy of stroke Besides safeguarding vital functions and stabilizing physiological parameter the re-opening of the closed vessel has priority in the therapy of acute cere bral ischaemia. The re-opening can be performed by different means. The mere mechanical re-opening, as e.g. the PTCA after heart attack, so far has not yet led to satisfying results. Only with a successful fibrinolysis an acceptable improvement of the physical condition of patients can be achieved. This can be accomplished by a local application using a catheter (PROCAT, a study with pro-urokinase). However, despite first positive results this method has not yet been officially approved as a pharmaceutical treatment.

-3 The naturally occurring fibrinolysis is based on the proteolytic activity of the serine protease plasmin which originates from its inactive precursor by catalysis (activation). The natural activation of plasminogen is catalyzed by the plasminogen activators u-PA (urokinase type plasminogen activator) and t-PA (tissue plasminogen activator) occurring naturally in the body. In contrast to u-PA, t-PA forms a so called activator complex together with fibrin and plasminogen. Thus, the catalytic activity of t-PA is fibrin dependent and is enhanced in its presence approximately 550-fold. Besides fibrin also fibrinogen can stimulate t-PA mediated catalysis of plasminogen to plasmin even though to a smaller extent. In the presence of fibrinogen the t-PA activity is only increases 25-fold. Also the cleavage products of fibrin (fibrin degradation products (FDP)) are stimulating t-PA. Known therapeutic approaches a) Streptokinase Early attempts of thrombolytic treatment of acute stroke go back to the 1950s. First extensive clinical trials with streptokinase, a fibrinolytic agent from beta-haemolysing streptococci, started only in 1995. Together with plasminogen streptokinase forms a complex which catalyzes other plasmi nogen molecules into plasmin. The therapy with streptokinase has severe disadvantages since it is a bacte rial protease and therefore can provoke allergic reactions in the body. Fur thermore, due to a former streptococci infection including a production of antibodies the patient may exhibit a so called streptokinase resistance making the therapy more difficult. Besides this, clinical trials in Europe (Multicenter Acute Stroke Trial of Europe (MAST-E), Multicenter Acute Stroke Trial of Italy (MAST-)) and Australia (Australian Streptokinase Trial (AST)) indicated an increased mortality risk and a higher risk of intracerebral -4 bleeding (intracerebral haemorrhage, ICH) after treating patients with strep tokinase. These trials had to be terminated early. b) Urokinase Alternatively, urokinase - also a classical fibrinolytic agent - can be appli cated. In contrast to streptokinase it does not exhibit antigenic characteris tics since it is an enzyme naturally occurring in various body tissues. It is an activator of plasminogen and independent of a co-factor. Urokinase is pro duced in kidney cell cultures. c) recombinant t-PA (rt-PA) Extensive experience on therapeutic thrombolysis is available for the tissue type plasminogen activator - the so called rt-PA - (see EP 0 093 619, US 4,766,075), which is produced in recombinant hamster cells. In the 90s sev eral clinical trials were performed world-wide using t-PA - with acute myo cardial infarction as the main indication - leading to partially non-understood and contradictory results. In the so called European Acute Stroke Trial (ECASS) patients were treated within a time frame of 6 hours after the onset of the symptoms of a stroke intravenously with rt-PA. After 90 days the mortality rate as well as the Barthel-Index were examined as an Index for the disability or the independent viability of patients. No significant improvement of the viability was reported but an - even though not significant - increase of mortality. Thus, it could be concluded, a thrombolytic treatment with rt-PA of patients being individually selected according to their respective case history immediately after the beginning of the stroke could possibly be advantageous. However, a general use of rt-PA within the time frame of 6 hours after the onset of stroke was not recommended since an application during this time increases the risk of intracerebal haemorrhage (ICH) (C. Lewandowski C and Wiliam Barsan, 2001: Treatment of Acute Stroke; in: Annals of Emergency Medicine 37:2; S. 202 ff.).

-5 The thrombolytic treatment of stroke was also subject of a clinical trial con ducted by the National Institute of Neurologic Disorder and Stroke (so called NINDS rtPA Stroke Trial) in the USA. This trial concentrated on the effect of intravenous rt-PA treatment within only three hours after the onset of the symptoms. Patients were examined three months after the treatment. Due to the observed positive effects of this treatment on the viability of patients, rt PA treatment within these limited time frame of three hours was recom mended although the authors found a higher risk for ICH. Two further studies (ECASS 11 Trial: Alteplase Thrombolysis for Acute Non interventional Therapy in Ischaemic Stroke (ATLANTIS)) examined whether the positive effects of rt-PA treatment within three hours after the onset of stroke could be repeated even with a treatment within six hours time. How ever, this question could not be answered affirmatively since no improve ment of the clinical symptoms or any decrease in mortality was observed. The higher risk for ICH remained. According to a review of all stroke trials first published in 1997 and updated in March 2001, all treatments with thrombolytics (urokinase, streptokinase, rt-PA or recombinant urokinase) resulted in a significant higher mortality within the first 10 days after the stroke while the total number of either dead or disabled patients was reduced when the thrombolytica where applied within six hour after stroke onset. This effects were mainly due to ICH. The broad use of thrombolytica for the treatment of stroke was therefore not rec ommended. Even before, such results gave reason to some other authors mere sarcastic statement that stroke patients had the choice to either die or to survive dis abled (SCRIP 1997: 2265, 26).

-6 Nevertheless, so far the therapy with rt-PA is the only treatment of acute cerebral ischaemia approved by the Food and Drug Administration (FDA) in the USA. However, it is restricted to an application of rt-PA within three hours after the onset of stroke. Recombinant plasminogen activator is currently marketed under the desig nation Alteplase or Reteplase for the analogous drug. The latter is a thera peutically active t-PA fragment with a lower half-life period. The therapeutic dose for Alteplase is about 70-100 mg, for Reteplase 2 x 560 mg, whereat Alteplase is mainly applied via a drip infusion and Reteplase via a bolus injection repeated twice in an interval of about 30 min (Mutschler: "Arzneimittelwirkungen", 8 th Edition, pages 512-513). Side effects of t-PA Neurotoxicity and Excitotoxicity The approval of rt-PA was reached in 1996. Before, in the year 1995, first announcements about negative side effects of t-PA became known, which provide an explanatory basis for its dramatic effects when applied in stroke treatment outside the three hour time frame. Accordingly, microglia cells and neuronal cells of the hippocampus produce t-PA which contributes to the glutamate mediated excitotoxicity. This is concluded from a comparative study on t-PA deficient and wild type mice when glutamate agonists were injected in their hippocampus, respectively. The t-PA deficient mice showed a significant higher resistance against external (inthrathecal) applicated glutamate (Tsirka SE et al., Nature, Vol. 377, 1995, "Excitoxin-induced neu ronal degeneration and seizure are mediated by tissue plasminogen activa tor"). These results were confirmed in 1998 when Wang et al. could prove nearly a double quantity of necrotic neuronal tissue in t-PA deficient mice when t-PA was injected intravenously. This negative effect of external t-PA on wild type mice was only approximately 33 % (Wang et al., 1998, Nature, -7 "Tissue plasminogen activator (t-PA) increases neuronal damage after focal cerebral ischaemia in wild type and t-PA deficient mice".) Further results on the stimulation of excitotoxicity by t-PA were published by Nicole et al. in the beginning of 2001 (Nicole 0., Docagne F Ali C; Margaill I; Carmeliet P; MacKenzie ET, Vivien D and Buisson A, 2001: The proteolytic activity of tissue-plasminogen activator enhances NMDA receptor-mediated signaling; in: Nat Med 7, 59-64). They could prove that t-PA being released by depolarized cortical neurons could interact with the so called NR1 sub unit of the glutamate receptor of the NMDA type leading to a cleavage of NR1. This increases the receptor's activity resulting in a higher tissue dam age after glutamate agonist NMDA was applied. The NMDA agonist induced excitotoxicity. Thus, t-PA exhibits a neurotoxic effect by activating the glutamate receptor of the NMDA type. Since the blood-brain barrier breaks down in the affected tissue area during a stroke, soluble plasma proteins like fibrinogen and as well therapeutically applied t-PA get into contact with the neural tissue, where the t-PA, being stimulated by the fibrinogen, displays its excitotoxic effect via the activation of the glutamate receptor. Despite its neurotoxic side effect and its increasing effect on the mortality t PA was approved by FDA. This can only be explained by the lack of harm less and effective alternatives - thus it is due to a very pragmatic cost benefit analysis. Therefore, there is still a need for safe therapies. However, if they were still based on thrombolytica - in case it is not possible to find alterna tives to thrombolysis - the problem of neurotoxicity has to be considered (see for example Wang et al. a.a.O.; Lewandowski and Barson 2001 a.a.0.). Therefore, further examination of known thrombolytica including DSPA (Desmodus rotundus Plasminogen Activator) in order to develop new drugs for stroke was terminated although principally all thrombolytica are poten tially suitable. Especially in case of DSPA its potential suitability for this medical indication was pointed out earlier (Medan P; Tatlisumak T; Takano -8 K; Carano RAD; Hadley SJ; Fisher M: Thrombolysis with recombinant Des modus saliva plasminogen activator (rDSPA) in a rat embolic stroke model; in: Cerebrovasc Dis 1996:6; 175-194 ( 4 th International Symposium on Thrombolic Therapy in Acute Ischaemic Stroke). DSPA is a plasminogen activator with a high homology (resemblance) to t-PA. Therefore - and in addition to the disillusionment resulting from the neurotoxic side effects of t PA - there were no further expectations, for DSPA being a suitable drug for stroke treatment. Alternative therapeutic approaches The investigation of alternative therapeutic approaches is currently e.g. focused on anticoagulants like heparin, aspirin or ancrod, the active sub stance derived from the poison of the Malayan pit viper. Two further clinical trials examining the effects of heparin (Intemational Stroke Trial (IST) and Trial of ORG 10172 in Acute Stroke Treatment (TOAST)) however, do not indicate a significant improvement of mortality or a prevention of stroke. A further new treatment focuses neither on thrombus nor on blood thinning or anti coagulation but attempts to increase the vitality of cells damaged by the interruption of blood supply (WO 01/51613 Al and WO 01/51614 Al). To achieve this antibiotics from the group of quinons, aminoglycosides or chloramphenicol are applied. For a similar reason it is further suggested to begin with the application of citicholin directly after the onset of stroke. In the body, citicholin is cleaved to cytidine and choline. The cleavage products form part of the neuronal cell membrane and thus support the regeneration of damaged tissue (US 5,827,832). Recent research on safe treatment is based on the new finding that a part of the fatal consequences of stroke is caused only indirectly by interrupted blood supply but directly to the excito- or neurotoxicity including over acti vated glutamate receptors. This effect is increased by t-PA (see above). A

-

9 concept to reduce excitotoxicity is therefore to apply so called neuroprotectives. They can be used separately or in combination with fibrinolytic agents in order to minimize neurotoxic effects. They can lead to a reduced excitotoxicity either directly e.g. as a glutamate receptor antagonist or indirectly by inhibiting voltage dependent sodium or calcium channels (Jin-Mo Lee et al. a.a.O.). A competitive inhibition (antagonistic action) of the glutamate receptor of NMDA type is possible e.g. with 2-amino-5-phosphonovalerate (APV) or 2 amino-5-phosphonoheptanoate (APH). A non competitive inhibition can be achieved e.g. by substances binding to the phencyclidine side of the channels. Such substances can be phencyclidine, MK-801, dextrorphane or cetamine. So far, treatments with neuroprotectives have not shown the expected success, possibly because neuroprotectives had to be combined with thrombolytic agents in order to exhibit their protective effects. This applies to other substances (Fig. 10). Even a combination of t-PA and neuroprotective agents results only in a limited damage. Nevertheless, the disadvantageous neurotoxicity of the fibrinolytic agent as such is not avoided. Non-neurotoxic plasminogen activators Plasminogen activators for the treatment of stroke, the enzymatic activity of which is highly selectively increased by fibrin many times over, i.e. by a factor greater than 650, are known from the international patent application PCT/EPO2/12204, the disclosure of which is completely taken into reference. The character and administration of these plasminogen activators are based on the knowledge, that the neurotoxicity of the tissue-plasminogen activator -10 (t-PA) is mainly due to the fact, that the blood-brain barrier is impaired or breaks down in consequence of the tissue destruction in the brain caused by the stroke and that the fibrinogen circulating in the blood can thus permeate into the neural tissue of the brain. There it activates t-PA, which indirectly leads to further tissue damage by activating the glutamate receptor or by the activation of plasminogen. (see above). In order to prevent this effect, a plasminogen activator is applied, which dis plays an increased selectivity for fibrin and - as a reversal conclusion - a decreased potential of activation by fibrinogen. This has the consequence, that these plasminogen activators will not be activated or, in comparison to t-PA, will be activated to a much lower extent when fibrinogen permeates from the blood into the neural tissue in consequence of the damaged blood brain barrier, since their activator fibrin cannot enter the neural tissue for reason of its size and insolubility. These plasminogen activators are thus non-neurotoxic. a) genetically modified plasminogen activators According to a preferred embodiment of the invention, non-toxic plasmino gen activators are used, which comprise at least one element of the so called cymogene triade. A comparable triade is known from the catalytic center of serine proteases of the chymotrypsine family consisting of three interacting amino acids aspartate 194, histidine 40 and serine 32. However, this triade does not exist in t-PA which belongs also to the family of chymo trypsine like serine proteases. Nevertheless, it is known, that the directed mutagenesis of native t-PA for the purpose of introducing at least one of the above amino acids at a suitable position results in a reduced activity of the pro-enzyme (single chain t-PA) and to an increased activity of the mature enzyme (double chain t-PA) in the presence of fibrin. Therefore, the intro duction of at least one amino acid of the triade - or of an amino acid with the respective function in the triade - can increase the cymogenity of t-PA (i.e.

- 11 the ratio between the activity of the mature enzyme an the activity of the pro enzyme). As a result the fibrin specificity is remarkably increased. This is due to conformational interaction between the introduced amino acid residue and/or amino acid residues of the wild type sequence. It is known that the mutagenesis of the native t-PA with substitution of Phe305 by His (F305H) and of Ala 292 by Ser (A292S) leads to a 20-fold increase of the cymogenity, whereas the variant F305H alone already leads to 5 times higher cymogentity (EL Madison, Kobe A, Gething M-J; Sambrook JF, Goldsmith EJ 1993: Converting Tissue Plasminogen Activator to a Zymogen: A regulatory Triad of Asp-His-Ser; Science: 262, 419-421). In the presence of fibrin these t-PA mutants show an activity increase of 30.000 times (F305H) and 130.000 times (F305H, A292S) respectively. In addition these mutants comprise a substitution of Arg275 to R275E in order to pre vent cleavage by plasmin at the cleavage site Aug275-lle276, thereby con verting the single chain t-PA to the double chain form. The mutant site R275E alone leads to a 6.900 fold increase of the fibrin specificity of t-PA (K Tachias, Madison EL 1995: Variants of Tissue-type Plasminogen Activator Which Display Substantially Enhanced Stimulation by Fibrin, in: Journal of Biological Chemistry 270, 31: 18319-18322). The positions 305 and 292 of t-PA are homologous to the positions His40 and Ser32 of the known triade of the chymotryptic serine proteases. By the corresponding substitutions introducing histidine or respectively serine, these amino acids can interact with the aspartate477 of t-PA resulting in a functional triade in the t-PA mutants (Madison et al., 1993). These t-PA mutants can be used for the treatment of stroke according to the invention because they show no or - compared to wild type t-PA - a signifi cantly reduced neurotoxicity due to their increased fibrin specificity. For the purpose of disclosure of the mentioned t-PA mutants F305H; F305H; A292S alone or in combination with R275E we incorporate the publications of Madi- -12 son et al., (1993) and Tachias and Madison (1995) hereby are fully incorpo rated by reference. The increase of fibrin specificity of plasminogen activators can alternatively be achieved by a point mutation of Aspi 94 (or an aspartate at a homologous position). Plasminogen activators belong to the group of serine proteases of the chymotrypsin family and therefore comprise the conserved amino acid Asp194, which is responsible for the stability of the catalytic active conformation of the mature proteases. It is known that Asp194 interacts with His40 in the cymogenic form of serine proteases. After the cymogene is activated by cleavage this specific interaction is interrupted and the side chain of the Asp194 rotates about 1700 in order to form a new salt bridge with 1le16. This salt bridge essentially contributes to the stability of the oxyanione pocket of the catalytic center of the mature serine proteases. It is also present in t-PA. The introduction of a point mutation replacing Asp194 prima facie impedes the formation or respectively the stability of the catalytic confirmation of ser ine proteases. Despite this the mutated plasminogen activators show a sig nificant increase of activity in the presence of their co-factor fibrin - especially in comparison to the mature wild type form - which can only be explained in a way that the interaction with fibrin , allows a conformational change promoting catalytic activity (L Strandberg, Madison EL, 1995: Variants of Tissue-type Plasminogen Activator with Substantially Enhanced Response and Selectivity towards Fibrin co-factors, in: Journal of Biological Chemistry 270, 40: 2344-2349). In conclusion, the Asp194 mutants of the plasminogen activators show a high increase of activity in presence of fibrin which enables their use according to the invention.

-13 In a preferred embodiment according to the invention, a mutant t-PA is used, in which Asp194 is substituted by glutamate (D194E) or respectively by asparagine (D194N). In these mutants the activity of t-PA is reduced 1 to 2000 fold in the absence of fibrin, whereas in the presence of fibrin, an increase of activity by a factor of 498.000 to 1.050.000 can be achieved. These mutants can further comprise a substitution of Arg15 to R15E, which prevents the cleavage of the single chain t-PA at the peptide bond Arg15 Ile16 by plasmin, leading to the double chain form of t-PA. This mutation alone increases the activation of t-PA by fibrin by the factor 12.000. For rea sons of disclosure of the t-PA mutations at positions 194 and 15, the publi cations of Strandberg and Madison (1995) are fully incorporated by refer ence. An increase of the fibrin dependency of plasminogen activators can also be achieved by the introduction of point mutations in the so called "autolysis loop". This element is known from trypsine; it can also be found as a homologous part in serine proteases and is especially characterized by three hydrophobic amino acids (Leu, Pro and Phe). The autolysis loop in plasminogen activators is responsible for the interaction with plasminogen. Point mutations in this area can have the effect that the protein-protein interaction between plasminogen and plasminogen activators cannot be effectively formed any longer. These mutations are only functionally relevant in the absence of fibrin. In the presence of fibrin, they, in contrast, are responsible for an increased activity of the plasminogen activators (K Song Hua, Tachias K, Lamba D, Bode W, Madison EL, 1997: Identification of a Hydrophobic exocite on Tissue Type Plasminogen Activator That Modulates Specificity for Plasminogen, in: Journal of Biological Chemistry 272; 3, 1811 1816). In a preferred embodiment t-PA is used showing point mutations in the posi tions 420 to 423. If these residues are substituted by directed mutagenesis this increases the fibrin dependency of t-PA is increased by a factor up to -14 61.000 (K Song-Hua et al.). Song-Hua et al. examined the point mutations L420A, L420E, S421G, S421E, P422A, P422G, P422E, F423A and F423E. These publications are fully incorporated by reference for disclosure of the use according to the invention. According to a further advantageous embodiment a modified tissue plasminogen activator with an amino acid sequence according to SEQ ID No. 1 (figure 14) is used. This modified t-PA differs from the wild type t-PA by the exchange of the hydrophobic amino acids in the position 420 to 423 in the autolysis loop as follows: His420, Asp421, Ala422 and Cys423. This t-PA preferentially contains a phenyl alanine at the position 194. Further the position 275 can be occupied by glutamate. Advantageously the position 194 is occupied by phenyl alanine. Further, a modified urokinase can be used according to the invention. The urokinase according to the invention can comprise the amino acid sequence according to SEQ ID No. 2 (figure 13) in which the hydrophobic amino acids of the autolysis loop are substituted by Val420, Thr421, Asp422 and Ser423. Advantageously the urokinase is carrying an lle275 and a Glu194. This mutant shows - in comparison to wild type urokinase - a 500-fold increased fibrin specificity. Both mutants - urokinase as well as t-PA - were analyzed in semi quantitative tests and showed an increased fibrin specificity in comparison to the wild type t-PA. b) plasminogen activator from Desmodus rotundus (DSPA) The plasminogen activator (DSPA) from the saliva of the vampire bat (Desmodus rotundus) also shows a highly increased activity in the presence of fibrin - in specific a 100.000-fold increase. Thus it can be used preferentially according to the invention. The term DSPA comprises four different -15 proteases, which fulfill an essential function for the vampire bat, namely an increased duration of bleeding of the wounds of pray (Cartwright, 1974). These four proteases (DSPAalphal, DSPAalpha2, DSPAbeta, DSPAgamma) display a high similarity (homology) to each other and to the human t-PA. They also show similar physiological activities, leading to a common classification under the generic term DSPA. DSPA is disclosed in the patents EP 0 352 119 Al and of US 6,008,019 and 5,830,849 which are hereby fully incorporated by reference for purpose of disclosure. DSPAalpha so far is the best analyzed protease from this group. It has an amino acid sequence with a homology greater than 72 % in comparison to the known human t-PA amino acid sequence (Krstzschmar et al, 1991). However, there are two essential differences between t-PA and DSPA. Firstly all DSPA has full protease activity as a single chain molecule, since it is - in contrast to t-PA - not converted into a double chain form (Gardell et al., 1989; KrAtzschmar et al., 1991). Secondly, the catalytic activity of DSPA is nearly absolutely dependent on fibrin (Gardell et al., 1989; Bringmann et al., 1995; Toschie et al., 1998). For example the activity of DSPAalphal is increased 100.000 fold in the presence of fibrin whereas the t-PA activity is only increased 550 fold. In contrast, DSPA activity is considerably less strongly induced by fibrinogen, since it only shows a 7 to 9 fold increase (Bringmann et al., 1995). In conclusion, DSPA is considerably more dependent of fibrin and much more fibrin specific as wild type t-PA which is only activated 550-fold by fibrin. Because of its fibrinolytic characteristics and the strong similarity to t-PA, DSPA is an interesting candidate for the development of a thrombolytic agent. Despite this, the therapeutic use of DSPA as a thrombolytic agent was restricted to the treatment of myocardinal infarction in the past, because - due to the contribution of t-PA to the glutamate induced neurotoxicity - no justified hopes existed, that a plasminogen activator which is related to t-PA could reasonably be used for a treatment of acute stroke.

- 16 Surprisingly it has been shown that DSPA has no neurotoxic effects even though it shows a high resemblance (homology) to t-PA and even though the physiological effects of the molecules are comparable to a large extent. The above conclusion led to the idea that DSPA after all may be successfully used as a thrombolytic agent for the therapy of stroke without causing severe risks of neuronal tissue damage. Especially interesting is the fact, that DSPA can also be used later than 3 hours after the onset of stroke symptoms. Experimental proof for the missing neurotoxicity of DSPA The new teaching is based on in vivo comparative examinations of the neurodegenerative effect of t-PA on one side and of DSPA on the other side which are performed by using the so called kainic acid model and a model for the examination of NMDA induced lesion of the striatum. The kainic acid model (also kainic acid injury model) is based on the stimu lation of the neurotoxic glutamate cascade by the external application of kainic acid (KA) as an agonist of the glutamate receptor of the kainic acid type (KA type) and of the NMDA and AMPA glutamate receptors. Using a t PA deficient mouse stem as an experimental model it was possible to show that the sensitivity of the laboratory animals against kainic acid only reached the level of wild type mice after a supplementary application of external t-PA. In contrast, an infusion of an equimolar concentration of DSPA under the same experimental conditions does not restore the sensitivity to kainic acid (KA). It was concluded that the neurotoxic effect of t-PA was not induced by DSPA. A summary of these results is shown in figure 15 (table 1). Quantitative examinations based on this model revealed that even a 10-fold increase of the DSPA concentration could not restore the sensitivity of the t PA deficient mice to the KA treatment whereas already a 10-fold lower t-PA - 17 concentration led to KA induced tissue damages. This leads to the conclu sion that DSPA possesses an at least 100 fold lower neurotoxic potential as t-PA with respect to the stimulation of the neurodegeneration after KA treat ment (see also figures 11 and 12). In the second model of neurodegeneration, the possible effects of t-PA as well as DSPA on the stimulation of the NMDA dependent neurodegeneration were compared to wild type mice. For this purpose, NMDA (as an agonist of the glutamate receptor of the NMDA type) was injected in wild type mice alone or in combination with either t-PA or DSPA. This model allows the comparison of the effects of these proteases under conditions, which always lead to a neurodegeneration and to an influx of plasma proteins due to the breake down of the blood brain barrier (Chen et al., 1999). While working on this model the injection of NMDA led to reproducible lesions in the striatum of mice. The volume of lesions was increased by a combined injection of t-PA and NMDA by at least 50 %. The co-injection with DSPAa1 in contrast did not lead to an increase or extension of the lesions caused by NMDA. Even in the presence of plasma proteins which can freely diffuse in the region of the lesion induced by NMDA, DSPA did not result in an increase neurodegeneration. A summary of these results is given in figure 16 (table 2). First results from clinical trials show the transferability of these results also for the treatment of stroke in humans. It was found that significant improve ments can be achieved in patients after a successful perfusion (improvement by 8 points NIHSS or NIHSS score 0 to 1). This is shown in figure 17 (table 3). In a further experiment, it was tested, weather t-PA and DSPA were able to permeate the damaged blood-brain barrier when being intravenously administered and in consequence increase the tissue lesions in the brain. To - 18 address this question, mice were stereotactically injected with NMDA in order to produce tissue lesions in the striatum, followed by an intravenous application of t-PA or DSPA 6 or 24 h after the NMDA injection. In compari son to a negative control, the experimental animals showed an increase of the NMDA induced damaged tissue area of about 30% when t-PA was given as an infusion 24 hours after the NMDA injection; in contrast, the same treatment with DSPA did not produce such an increase of tissue damage, although its penetration into the damaged tissue area was positively detected by means of an antibody staining (see Fig. 18, 19). When t-PA or DSPA were intravenously applied in a corresponding manner 6 hours after the NMDA-injection, an increase of the damaged tissue area was not yet to be detected. This may be thereby explained, that the blood-brain barrier still provided a sufficient barrier function at this moment of the t-PA or DSPA injection. These results show, that DSPA constitutes a mostly inert protease in the central nervous system of a mammal (and thus also of a human) and, in contrast to t-PA, does not produce a potentiation of the neurotoxicity induced by KA or NMDA. This lack of neurotoxicity - against the general expectation makes DSPA a suitable thrombolytic for the treatment of acute stroke. Therapeutic potential of the non-neurotoxic plasminogen activators The lacking neurotoxicity of DSPA and of the other non-neurotoxic plasmi nogen activators (see above) offer the special advantage in stroke treatment that the use of these plasminogen activators - in contrast to the wild type t PA - is not limited to a short maximum period of only 3 hours after the onset of stroke. In contrary, the treatment can be initiated later - for example after 6 hours or even later, since there is nearly no risk of stimulating excitotoxic responses. First clinical trials with DSPA prove a safe treatment of patients even in a time range of over 6 to 9 hours after the onset of stroke symptoms.

-19 This option of a timely unlimited treatment with non-neurotoxic activators is of special importance, since it allows for the first time to treat patients with acute stroke symptoms safely even when diagnosis is delayed or the onset of the stroke cannot be determined with sufficient security. In the prior art, this group of patients was excluded from thrombolytic therapy with plasminogen activators due to unfavorable risk estimation. Consequently, an essential contra-indication for the authorized use of a thrombolytic agent for stroke is eliminated. Application of plasminogen activators In contrast to the already established stroke therapeutic rt-PA, no validated information is yet available for a possible mode of application for the stroke treatment with the non-neurotoxic plasminogen activators. It is thus the objective of the invention to provide an advantageous mode of application for these non-neurotoxic plasminogen activators. According to the invention, plasminogen activators, the activity of which is increased in the presence of fibrin by a factor greater than 650, are intravenously applied for the treatment of stroke. Thus, the present invention provides a method of treating stroke in a mammal, comprising administering to the mammal a composition comprising DSPAa1 in an amount from 60 to 230 pg/kg body weight of the mammal. The intravenous administration of these non-neurotoxic plasminogen activators for the treatment of stroke has already been evaluated in clinical trials, in which DSPA - as an example for this group of fibrinolytics - was intravenously applied in these patients, thereby causing only minor side effects. These results of the clinical studies were unexpected, since it was well known, that the intravenous application of t-PA and other common fibrinolytics is associated with a severe risk of cerebral bleedings (see above).

-20 In order to reduce intra-cerebral bleedings, recent efforts were made to develop strategies to apply these substances no more intravenously, but via the intra-arterial route into the direct vicinity to the intravascular thrombus by means of a catheter. Practical experience for this is already available for the recombinantly produced urokinase (PROKAT as a study with pro-urokinase). Since this mode of application allows for a significant reduction of the total dose, a decrease in the dose-dependent side effects - thus also of the cere bral bleedings - is realised. These significant advantages of the intra-arterial application however are contravened by two possible drawbacks. First, this treatment prerequisites a time consuming preparation of the patient, which in stroke treatment is not possible to be realized within the given window in time of just 3 hours. On the one hand one sure can achieve a lower total dose. A higher concentra tion of the drug however will locally reach the terminal arterial vessels. In consequence of the impaired barrier function of the vascular endothelium in case of a stroke, the pharmaceutical substance will also reach the sur rounding tissues in locally high concentrations. There it may then exert undesired side effects. In contrast thereto, the concentration of the pharmaceutical substance will be diluted by the venous blood flow in case of an intravenous application. Thus, the intra-arterial injection is problematic in case of drugs having a tis sue destructing potential like t-PA (Forth, Henschler, Rummel, Starke: "Pharmakologie und Toxikologie", 6th Edition, 1992, page 29). However, the limitations complicating the intra-arterial application - the nar row window in time and the tissue damaging side effects - are not valid for the plasminogen activators applied according to the invention. Thus, the intra-arterial application for reason of its undeniable advantages (see above) in principle constitutes a very promising mode of application for non-neuro- - 21 toxic plasminogen activators. It would thus have been obvious to follow this path when searching for a preferable mode of application for these drugs. Nevertheless, the applicant also for the non-neurotoxic plasminogen activa tors has chosen to rely on the rather problematic intravenous application, which then surprisingly has proven to be advantageous. Moreover, the mode of application according to the invention also departs from the common therapeutic practice to mostly use an intramuscular injec tion or an intravenous drip infusion in order to administer proteins with an increased immunogenic potential, which has the aim to reduce the risk of an anaphylactic shock (Mebs: "Gifttiere", 2th Edition, 2000). In contrast to the t-PA as a natural bodily substance, the plasminogen acti vators applied according to the invention are either foreign proteins derived from animals (like e.g. DSPA) or genetically modified bodily proteins dis playing novel epitopes in consequence of their structural differences. The accompanying problem of anaphylactic reactions - in particular when apply ing high therapeutic doses, like it is commonly necessary in case of intrave nous application - applies also for other fibrinolytics consisting of foreign proteins like e.g. streptokinase. In a particularly advantageous embodiment, the plasminogen activators ap plied according to the invention are administered by means of a bolus injec tion (intravenous quick injection), which can also be administered as a single intravenous quick injection containing the complete therapeutic dose, In the context of clinical studies it was found, that even the intravenous application of surprisingly low therapeutic doses lead to advantageous results. Preferable therapeutic results were e.g. achieved with doses between 90 pg/kg and 230 pg/kg. Particularly preferable therapeutic results here were achieved with doses between 62,5 to 90 pg/kg. In the examined -22 patients, the periods between the stroke and the application of the drug were between 3 and 9 hours. By means of suitable examination methods it was possible to determine the onset of a therapeutic effect (see Figures 20 to 29). DSPA as well as further non-neurotoxic plasminogen activators do not show tissue damaging side effects. However, it can be advantageous to apply them in combination with a neuroprotective agent for the treatment of stroke in order to limit the tissue damages induced by the glutamate occurring naturally in the human body. Neuroprotective agents inhibiting the glutamate receptor competitively or non-competitively can be used. Useful combina tions are e. g. with the known inhibitors of the glutamate receptors of the NMDA type, the kainic acid type or the quisqualate type, as for example APV, APH, phencyclidine, MK-801, dextrorphane or cetamine. Further a combination with cations can be advantageous since cations, especially Zn-ions, block the cation channel regulated by the glutamate receptor and can therefore reduce neurotoxic effects. In a further advantageous embodiment, non-neurotoxic plasminogen acti vators can be combined with at least one further therapeutic agent or with a pharmaceutically tolerable carrier. The combination with a therapeutic agent which supports the reduction of tissue damage by vitalizing the cells is especially advantageous, since it contributes to the regeneration of already damaged tissue or serves for the prevention of further stroke incidents. Advantageous examples are combinations with antibiotics as quinones, anticoagulants as heparin or hirudin as well as with citicholine or acetylsali cylic acid. A combination with at least one thrombin inhibitor can also be advanta geous. Preferentially, thrombomodulin and thrombomodulin analogs like for example solulin, triabin or pallidipin can be used. Further combinations with -23 anti- inflammatory substances are advantageous, since they influence the infiltration by leucocytes. Below the mode of application and formulation according to the invention is outlined by means of distinct therapeutic examples. -> Examples Comparative examination of t-PA and DSPA: A. Methods 1. Animals Wild-type mice (c57/Black 6) and t-PA deficient mice (t-PA-/-mice) (c57/Black 6) (Carmeliet et al., 1994) were supplied by Dr. Peter Carmeliet, Leuven, Belgium. 2. Protein extraction from brain tissue The assessment of proteolytic activity in brain tissue following infusion of either t-PA or DSPAa1 was performed by zymographic analysis (Granelli Piperno and Reich, 1974). After an infusion over a period of seven days into the hippocampus, mice were anaesthetised, then transcardially perfused with PBS and the brains removed. The hippocampus region was removed, transferred to eppendorf tubes and incubated in an equal volume (w/v) (approx. 30-50 pm) of 0,5 % NP-40 lysis buffer containing no protease inhibitors (0,5 % NP-40, 10 mM Tris-HCI pH 7,4, 10 mM NaCL, 3 mM MgCI2, 1 mM EDTA). The brain extracts were homogenized by means of a hand held glass homogeniser and left on ice for 30 minutes. The samples were then centrifuged and the supernatant was removed. The amount of proteins present was determined (Bio-Rad-reagent).

- 24 3. Zymographic analysis of the proteases The proteolytic activity in the samples and the brain tissue extracts was determined by zymographic analysis according to the method of Granelli, Piperno and Reich (1974). The samples with recombinant proteins (up to 100 nM) or the brain tissue extracts (20 pg) were subjected to a (10 %) SDS PAGE under non-reducing conditions. The gels were removed from the plates, washed in 1 % triton X 100 for 2 hours and then overlaid onto an agarose gel containing polymerized fibrinogen and plasminogen (Granelli, Piperno and Reich, 1974). The gels were incubated at 37 0 C in a humified chamber until proteolysed zones appeared. 4. Intra-hippocampal infusion of t-PA, DSPA and subsequent injec tion of kainic acid The kainic acid injury model was based on studies of Tsirka et al. (1995). The animals were injected intraperitoneally (i. p.) with atropine (4 mg/kg) and then anaesthetised with an i. p. injection of sodium pentobarbitol (70 mg/kg). Afterwards mice were placed in a stereotaxic frame and a micro-osmotic pump (Alzet model 1007D, Alzet CA. USA) containing 100 pl of either PBS or recombinant human t-PA (0,12 mg/ml, 1,85 pM) or DSPAa1 (1,85 pM) was implanted subcutaneously between the shoulder blades. The pumps were connected via sterile tubes to a brain cannula and inserted through a burr opening made through the skull at coordinates bregma -2,5 mm, midiolateral 0,5 mm and dorsoventral 1,6 mm in order to introduce the liquid near the midline. The cannula was fixed at the desired position and the pumps were allowed to infuse the respective solutions at a rate of 0,5 pl per hour for a total of 7 days. Two days after infusion of the proteases the mice were reanaesthetised and again placed in the stereotaxic frame. Afterwards 1,5 nmol of kainic acid - 25 (KA) in 0,3 pl PBS was injected unilaterally into the hippocampus. The coor dinates were: bregma -2,5 mm, medial-lateral 1,7 mm and dorsoventral 1,6 mm. The excitotoxin (KA) was delivered for a duration of 30 seconds. After the kainic acid treatment the injection needle remained at these coordinates for further 2 minutes in order to prevent a reflux of the liquid. 5. Brain processing procedure Five days after KA injection, the animals were anaesthetised and transcar dially perfused with 30 ml PBS followed by 70 ml of a 4 % paraformaldehyd solution, post fixed in the same fixative followed by incubation in 30 % sucrose for further 24 hours. Coronal sections (40 pm) of the brain were then cut on a freezing microtome and either counter-stained with thionin (BDH, Australia) or processed for immunohistochemical examination as described below. 6. Quantification of neuronal loss within the hippocampus The quantification of neuronal loss in the CA1-CA3 hippocampal subfields was performed as previously described (Tsirka et al., 1995; Tsirka et al., 1996). Five consecutive parts of the dorsal hippocampus from all treatment groups were prepared taking care that the parts indeed comprised the place of the CA-injection and lesion area. The hippocampal subfields (CA1-CA3) of these sections were traced by means of camera lucida drawings of the hippocampus. The entire lengths of the subfields was measured by com parison to 1 mm standards traced under the same magnification. The lengths of tissue with viable pyramidal neurons (having normal morphology) and lengths of tissue devoid of neurons (no cells present, no thionin staining) was determined. The lengths, representing intact neurons and neuronal losses over each hippocampal subfield were averaged across sections and the standard deviations were determined.

- 26 7. Intra-striatal NMDA excitotoxic lesions with or without t-PA or DSPA Wild type mice (c57/Black 6) were anaesthetised and placed in a sterertaxic frame (see above). Mice then received an unilateral injection of 50 nmol NMDA in the left stratum, injected alone or in combination with either 46 pM rt-PA or 46 pM DSPAa1. As controls t-PA and DSPA were also injected alone (both at a concentration of 46 pM). The injection coordinates were: bregma -0,4 mm, midiolateral 2,0 mm and dorsoventral 2,5 mm. The solu tions (1 pl total volume for all treatments) were transferred over a period of 5 minutes at a rate of 0,2 p1/min and the needle was left in place for further 2 minutes after the injection in order to minimize the reflux of fluid. After 24 hours the mice were anaesthetised and perfused transcardially with 30 ml PBS followed by 70 ml of a 4 % paraformaldehyd solution, post fixed in the same fixative for 24 hours with followed by incubation in 30 % sucrose for further 24 hours. Brains were then cut (40 pm) on a freezing microtome and mounted onto gelatin coated glass slides. 8. Quantification of the lesion volume following NMDA injection The quantification of the striatal lesion volume was performed using the method described by Callaway et al. (2000). Ten consecutive coronal sec tions spanning the lesioned area were prepared. The lesioned area was visualised using the Callaway method and the lesion volume was quantified by the use of a micro computer imaging device (MCID, Imaging Research Inc., Brock University, Ontario, Canada). 9. Immunohistochemistry Immunohistochem istry was performed using standard methodologies. Coronal sections were immersed in a solution of 3 % H 2 0 2 and 10 % metha nol for 5 minutes followed by an incubation in 5 % normal goat serum for 60 - 27 minutes. The sections were incubated over night either with an anti-GFAP antibody (1:1.000; Dako, Carpinteria, CA, USA) for the detection of astro cytes, with an anti-MAC-1 antibody (1:1.000; Serotec, Raleigh, NC, USA) for the detection of microglia or with polyclonal anti-DSPA antibodies (Schering AG, Berlin). After rinsing, the sections were incubated with the appropriate biotinylated secondary antibodies (Vector Laboratories, Burlingame, CA, USA). This was followed by a final incubation with avidin/biotin-complex (Vector Laboratories, Burlingame, CA, USA) for 60 minutes before visualisa tion with 3,3'-diaminebebcidine/0,03 % H 2 0 2 . Sections were then mounted on gelatin coated slides, dried, dehydrated and coverslipped with permount. 10. Enhancement of the tissue lesions induced by NMDA injection by intravenous administration of t-PA or DSPA In order to induce tissue lesions within the striatum, mice were stereotacti cally injected with NMDA. Six hours after NDMA injection t-PA or DSPA (100 pl; 10 mg/kg) were injected via the tail vein. As control 100 pl 0,9% NaCI were injected and subsequently PBS infused. After 28 hours the animals were killed and the lesion volume was determined. In a second trial sets of animals including up to 15 mice were injected accordingly, with an infusions of t-PA or DSPA 24 hours after NMDA injec tion and a subsequent determination of the tissue damage. As proof of the presence of DSPA within the brain, coronal sections were stained with an anti-DSPA antibody according to standard methods. B. Results 1. Infusion of t-PA or DSPA disperses into the hippocampus of t-PA /- mice and retains proteolytic activity - 28 The initial experiments were designed to confirm that both DSPA and t-PA retain their proteolytic activity for the 7 day period of the infusion. To this end, aliquots of t-PA and DSPA (100 nmol) were incubated at 37 *C and at 30 0 C for 7 days in a water bath. In order to determine the proteolytic activity, 5 fold serial dilutions of the probes were subjected to SPS-PAGE under non reducing conditions and proteolytic activity was assessed by zymographic analyses. An aliquot of t-PA and DSPA which had been kept frozen for a period of 7 days was used as a control. As can be seen in figure 1 there was only a minor loss of DSPA or t-PA activity at an incubation with either 25 0 C or 37*C over this period of time. 2. t-PA and DSPA activity is recovered in hippocampal extracts pre pared from t-PA -- mice following infusion First it had to be confirmed that the infused proteases were present in the brain of the infused animals and also retained their proteolytic activity while being in this compartment. To address this point, t-PA -/- were infused for seven days with either t-PA or DSPA (see above). Mice were then transcar dially perfused with PBS and the brains removed. The ipsilateral and con tralateral hippocampal regions were isolated as well as a region of the cere bellum (taken as a negative control). Tissue samples (20 pg) were subjected to SDS-PAGE and zymographic analysis according to the description in the methods section. As can be seen in figure 2, both t-PA and DSPA activities were detected in the ipsilateral region of the hippocampus, while some activity was also detected on the contralateral side. This indicates that the infused proteases not only retained their activity in the brain but had also diffused within the hippocampal region. As a control, no activity could be detected in the extract prepared from the cerebellum. 3. Immunohistochemical assessment of DSPA - 29 To further confirm that DSPA had indeed diffused into the hippocampal region, coronal brain sections of t-PA -/- mice were analysed immunohisto chemically after DSPA infusion. DSPA-antigen was detected in the hip pocampal region with the most prominent staining in the area of the infusion site. This result confirms that the infused DSPA is soluble and is indeed pre sent in the hippocampus. 4. DSPA infusion does not restore kainic-acid mediated neurode generation in vivo t-PA -/- mice are characteristically resistant to kainic acid (KA) mediated neurodegeneration. However, intrahippocampal infusion of rt-PA completely restores the sensitivity to KA-mediated injury. To determine whether DSPA could be substituted for t-PA in this model, t-PA -/- mice were infused intra hipocampically with either t-PA or DSPA using a mini-osmotic pump. For both groups 12 mice were tested. 2 days later the animals were injected with kainic acid and left to recover. 5 days later the animals were killed and the brains removed and prepared (see above). As controls, t-PA -/- mice were also infused with PBS prior to KA treatment (N = 3). Coronal brain sections were prepared and the neurons detected by Nissl staining. As shown in figures 4a and 4b, t-PA -/- mice infused with PBS were resistant to subsequent challenge with KA. However, infusion of recombinant t-PA restored sensitivity to KA treatment. In contrast, infusion of the same concentration of DSPA into the hippocampal region did not alter the sensitivity of the animals to KA. A quantitation of those results was based on data obtained from 12 mice in each group. In 2 of the 12 mice infused with DSPA a small extend of neurodegeneration was observed. The reason for that in unclear and pos sibly not related to the presence of DSPA. The combined data consider this minor effect that was observed in the case of these 2 animals. All 12 mice - 30 treated with t-PA were sensitive against the KA treatment. These results show that in case of an infusion of t-PA or DSPAa1 in equimolar concentra tions only the administering of t-PA led to the restoration of sensitivity to KA induced neurodegeneration. 5. DSPA infusion does not result in microglial activation The restauration of the KA sensitivity of the t-PA -/- mice caused by a t-PA infusion also results in a microglia activation (Rogove et al., 1999). To as sess the degree of microglial activation following t-PA or DSPA infusion and subsequent KA treatment, coronal sections of mice were subjected to an immunohistochemical staining for activated microglia cells using the Mac-1 antibody. The restauration of KA sensitivity following t-PA infusion resulted in a clear increase in Mac-I positive cells. This was not observed in mice infused with DSPA. Hence, the presence of DSPA does not result in the activation of microglia cells following KA treatment. 6. Titration of DSPA and t-PA in the mice hippocampus region. The concentration of t-PA used for the infusion was based on the concen tration described by Tsirka et al. (1995) (100pl of 0,12 mg/mI [1,85 pM]). The KA-injury experiments were repeated using a 10-fold lower of t-PA (0,185 pM) and a 10-fold higher amount of DSPA (18,5 pM). The lower t-PA con centration was still able to restore the sensitivity to KA treatment (n=3). Of special interest was the finding that the infusion of 10 fold increased DSPA concentration only caused a little neuronal loss following KA treatment. These data strongly point out that DSPA does not lead to an increase of sensitivity to KA.

- 31 7. Effect of t-PA and DSPA on NMDA-dependent neurodegeneration in wild type mice The effects of t-PA and DSPA were also examined in a model of neurodegeneration in wild type mice. The injection of t-PA in the striatum of these mice provably led to an increase of the neurodegenerative effects caused by the glutamate analogue NMDA (Nicole et al., 2001). NMDA was injected into the striatal region of wild type mice in the presence of t-PA or DSPA (each 46 pM) with a total volume of 1 pl. After 24 hours the brains were removed and the size of the lesions was quantified according to the Callaway method (Callaway et al., 2000) (see above). As can be seen in figure 7, injection of NMDA alone caused a reproducible lesion in all treated mice (N=4). When t-PA and NMDA were applied together, the size of the lesions was increased about 50 % (P<0,01, n=4). In a clear contrast the co-injection of NMDA and the same concentration of DSPA did not lead to an increase in lesion size compared to NMDA alone. Injection of t-PA or DSPA alone did not lead to a detectable neurodegeneration. The lacking effect of t-PA when being administered alone is consistent with the results of Nicole et al. (2001). These data show that the presence of DSPA does not increase neurodegeneration even during a neurodegenerative event. In order to confirm that the injection of DSPA had indeed spread into the hippocampal region, immunohistochemistry was performed on coronal sections by use of the DSPA antibody. The examination showed that DSPA did indeed enter the striatal region. Kinetic analysis of the plasminogen activation by indirect chromogen test -32 Indirect chromogen tests of the t-PA activity were performed using the sub strate Lys-plasminogen (American Diagnostica) and spectrocyme PL (American Diagnostica) according to Madisan E.L., Goldsmith E.J., Gerard R.D., Gething M.-J., Sambrook J.F. (1989) Nature 339 721-724; Madison E.LO., Goldsmith E.J., Gething M.J., Sambrook J.F. and Bassel-Duby R.S. (1990) Proc. Nati. Acad. Sci U.S:A 87, 3530-3533 as well as Madison E.L., Goldsmith E.J., Gething M.J., Sambrook J.F. and Gerard R.D. (1990) J. Biol. Chem 265, 21423-21426. Tests were performed both in the presence and absence of the co-factor DESAFIB (American Diagnostica). DESAFIB is a preparation of soluble fibrin monomeres gained by the cleavage of highly pure human fibrinogen with the protease batroxobin. Batroxobin cleaves the Arg 16 -Gly"-binding in the Aa-chain of fibrinogen and thereby releases fibrinopeptid A. The resulting des-AA-fibrinogen representing fibrin I mono mers is soluble in the absence of the peptide Gly-Pro-Arg-Pro. The concen tration of Lys-plasminogen was varied from 0,0125 up to 0,2 pM in the pres ence of DESAFIB and from 0,9 to 16 pM in absence of the co-factor. Indirect chromogen tests in the presence of different stimuli. Indirect chromogen standard tests were performed according to the publica tions cited above. Probes of 100 pl total volume containing 0,25-1 ng enzyme, 0,2 pM Lys-plasminogen and 0,62 mM spectrocyme PL were used. The tests were performed either in the presence of buffer, 25 pg/mI DESAFIB, 100 pg/mI cyanogen bromide fragments of fibrinogen (American Diagnostica) or 100pg/ml of the stimulatory 13 amino acid peptide P368. The analysis were performed in microtiter-plates and the optic density was determined at a wave length of 405 nm every 30 seconds for 1 hour in a "Molecular Devices Thermomax". The reaction temperature was 37 *C.

- 33 8. DSPA also in case of intravenous application does not cause an increase of neural tissue damages In the striatum of mice, tissue lesions were induced by the injection of NMDA, followed by the intravenous application of t-PA or DSPA 6 or 24 hours thereafter. In comparison to a negative control, the experimental animals showed an increase of about 30% of the damaged tissue area produced by the NMDA-injection, when t-PA was given as an intravenous infusion 24 hours after the NMDA-injection; this was in contrast to DSPA, which did not cause such an increase of the tissue damages (see Fig. 18). By the staining of coronal sections with an anti-DSPA-antibody, it was possible to detect, that the DSPA intravenously administered 24 hours after the NMDA-injection had permeated into the damaged tissue areas (see Fig. 19). In case of a corresponding intravenous application of t-PA or DSPA 6 hours after the NMDA injection, an increase of the damaged tissue area was not yet to be detected. This may have its reason in that the blood-brain barrier at this moment of the t-PA or DSPA application still provided a sufficient barrier function. DSPA thus also in case of intravenous application does not display neurotoxic side effects. The discussion of documents, acts, materials, devices, articles and the like is included in this specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all of these matters formed part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed before the priority date of each claim of this application. Where the terms "comprise", "comprises", "comprised" or "comprising" are used in this specification (including the claims) they are to be interpreted as specifying the presence of the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components, or group thereof.

- 34 References: Baranes D, Lederfein D, Huang YY, Chen M, Bailey CH, Kandel ER. 1998. Tissue plasminogen activator contributes to the late phase of LTP and to synaptic growth in the hippocampal mossy fiber pathway. Neuron 21:813 25. Bringmann P, Gruber D, Liese A, Toschi L, Kratzchmar J, Schleuning WD, Donner P. 1995. Structural features mediating fibrin selectivity of vampire bat plasminogen activators. J Biol Chem 270:25596-25603. Callaway JK, Knight MJ, Watkins DJ, Beart PM, Jarrott B, Delaney PM. 2000. A novel, rapid, computerized method for quantitation of neuronal damage in a rat model of stroke. J Neurosci Methods 102:53-60 Carmeliet P, Schoonjans L, Kieckens L, Ream B, Degen J, Bronson R, De Vos R, van den Oord JJ, Collen D, Mulligan RC. 1994. Physiological conse quences of loss of plasminogen activator gene function in mice. Nature 368:419-424. Cartwright T. 1974. The plasminogen activator of vampire bat saliva. Blood 43:317-326. Chen ZL Strickland S. 1997. Neuronal death in the hippocampus is promoted by plasmin-catalysed degradation of laminin. Cell 91:917-925. Chen, Z-L, Indyk, J.A., Bugge, T.H., Kombrinck, K.W., and Strickland S. 1999. Neuronal Death and blood-brain barrier breakdown after excitotoxic injury are independent processes. J. Neuroscience 19:9813-9820 Frey U, Muller M, Kuhl D. 1996. A different form of long-lasting potentiation revealed in tissue plasminogen activator mutant mice. J Neurosci 16:2057-2063. Gardell SJ, Duong LT, Diehl RE, York JD, Hare TR, Register RB, Jacobs JW, Dixon RA, Friedman PA. 1989. Isolation, characterization, and cDNA cloning of a vampire bat salivary plasminogen activator. J Biol Chem 264:17947-17952. Gardell SJ, Ramjit DR, Stabilito 11, Fujita T, Lynch JJ, Cuca GC, Jain D, Wang SP, Tung JS, Mark GE, et al. 1991. Effective thrombolysis without marked - 35 plasminemia after bolus intravenous administration of vampire bat salivary plasminogen activator in rabbits. Circulation 84:244-253. Granelli-Piperno, A and Reich, E. 1974. A study of proteases and protease inhibitor complexes in biological fluids. J. Exp. Med. 148: 223-234. Huang YY, Bach ME, Lipp HP, Zhuo M, Wolfer DP, Hawkins RD, Schoonjans L, Kandel ER, Godfraind JM, Mulligan R, Collen D, Carmeliet P. 1996. Mice lacking the gene encoding tissue-type plasminogen activator show a selective interference with late-phase long-term potentiation in both Schaffer collateral and mossy fiber pathways. Proc Natl Acad Sci U S A. 93:8699-86704. Kratzschmar J, Haendler B, Langer G, Boidol W, Bringmann P, Alagon A, Donner P, Schleuning WD. 1991. The plasminogen activator family from the salivary gland of the vampire bat Desmodus rotundus: cloning and ex pression. Gene 105:229-237. Madani R, Hulo S, Toni N, Madani H, Steimer T, Muller D, Vassalli JD. (1999) Enhanced hippocampal long-term potentiation and learning by increased neuronal expression of tissue-type plasminogen activator in transgenic mice. EMBO J.18:3007-3012 Mellott MJ, Stabilito 11, Holahan MA, Cuca GC, Wang S, Li P, Barrett JS, Lynch JJ, Gardell SJ. 1992. Vampire bat salivary plasminogen activator pro motes rapid and sustained reperfusion concomitant systemic plasminogen activation in a canine model of arterial thrombosis. Arterioscler. Thromb. 12:212-221. Muschick, P., Zeggert D., Donner, P., Witt, W. 1993. Thrombolytic properties of Desmodus (vampire bat) plasminogen activator DSPAa1, Alteplase and streptokinase following intravenous bolus injection in a rabbit model of ca rotid artery thrombolysis. Fibrinolysis 7: 284-290. MOller, C.M., Griesinger, C.B. 1998. Tissue plasminogen activator mediates reverse occlusion plasticity in visual cortex. Nature Neuroscience. 1: 47 53. National Institute of Neurological Disorders and stroke rt-PA study group. 1995. New. Engl. J. Med. 333: 1581-1587 - 36 Nicole, 0., Docagne, F., Ali, C., Margaill, I., Carmeliet, P., MacKenzie, E.T., Vivien, D. & Buisson, A. (2001) The proteolytic activity of tissue-plasmino gen activator enhances NMDA receptor-mediated signaling. Nat Med 7, 59-64. Rogove AD, Siao C, Keyt B, Strickland S, Tsirka SE. 1999. Activation of micro glia reveals a non-proteolytic cytokine function for tissue plasminogen ac tivator in the central nervous system. J Cell Sci 112:4007-4016. Schleuning WD, Alagon A, Boidol W, Bringmann P, Petri T, Kratzschmar J, Haendler B, Langer G, Baldus B, Witt W, et al. 1992. Plasminogen activa tors from the saliva of Desmodus rotundus (common vampire bat): unique fibrin specificity. Ann N Y Acad Sci 667:395-403. Seeds, N.W., Williams, B.L., Bickford, P.C. 1995. Tissue plasminogen activator induction in purkinje neurons after cerebellar motor learning. Science. 270:1992-1994. Stewart RJ, Fredenburgh JC, Weitz JI. 1998. Characterization of the interac tions of plasminogen and tissue and vampire bat plasminogen activators with fibrinogen, fibrin, and the complex of D-dimer noncovalently linked to fragment E. J Biol Chem 273:18292-18299. Traynelis, SF, Lipton, SA. 2001. Is tissue plasminogen activator a threat to neurons? Nature Medicine, 7:17-18. Toschi L, Bringmann P, Petri T, Donner P, Schleuning WD. 1998. Fibrin se lectivity of the isolated protease domains of tissue-type and vampire bat salivary gland plasminogen activators. Eur J Biochem 252:108-112. Tsirka SE, Gualandris A, Amaral DG, Strickland S. 1995. Excitotoxin-induced neuronal degeneration and seizure are mediated by tissue plasminogen activator. Nature. 377:340-344. Tsirka, S., Rogove, A. D., and Strickland, S. 1996. Neuronal cell death and tPA. Nature 384: 123-124 Tsirka SE, Rogove AD, Bugge TH, Degen JL, Strickland S. 1997. An extracel lular proteolytic cascade promotes neuronal degeneration in the mouse hippocampus. J Neurosci 17:543-552 - 37 Tsirka SE, Bugge TH, Degen JL, Strickland S. 1997. Neuronal death in the central nervous system demonstrates a non-fibrin substrate for plasmin. Proc Natl Acad Sci USA 94:9779-9781. Wang YF, Tsirka SE, Strickland S, Stieg PE, Soriano SG, Lipton SA, 1998. Tissue plasminogen activator (tPA) increases neuronal damage after focal cerebral ischemia in wild-type and tPA-deficient mice. Nat Med. 4:228 231. Walker JB, Nesheim ME. 2001. A kinetic analysis of the tissue plasminogen activator and DSPAalphal cofactor activities of untreated and TAFla treated soluble fibrin degradation products of varying size. J Biol Chem 276:3138-3148. Witt W, Maass B, Baldus B, Hildebrand M, Donner P, Schleuning WD. 1994. Coronary thrombolysis with Desmodus salivary plasminogen activator in dogs. Fast and persistent recanalization by intravenous bolus administration. Circulation. 90:421-426.

Claims (4)

1. Pharmaceutical composition, comprising DSPA alpha 1 with a dose of 62.5 pg/kg to 230 pg/kg in humans.

2. Pharmaceutical composition, comprising DSPA alpha 1 for the treatment of stroke with a dose of 62.5 pg/kg to 230 pg/kg in humans.

3. Pharmaceutical composition according to claim 1 or 2 wherein the dosage is from

62.5 to 90 pg/kg. 4. Pharmaceutical composition, comprising 5 mg to 25 mg of DSPA alpha 1. 5. Pharmaceutical composition according to one of the above claims wherein the DSPA alpha 1 comprises an amino acid sequence according to Seq. ID No. 1. 6. Pharmaceutical composition according to one of the above claims for the therapeutic treatment of stroke in humans at least 3 hours after stroke onset. 7. Pharmaceutical composition according to at least one of the above claims for the therapeutic treatment of stroke in humans at least 6 hours after stroke onset. 8. Pharmaceutical composition according to at least one of the above claims for the therapeutic treatment of stroke in humans at least 9 hours after stroke onset. 9. Pharmaceutical composition according to at least one of the above claims for the therapeutic treatment of stroke patients in which the onset of stroke is temporally not exactly determined. 10. Pharmaceutical composition according to one of the above claims characterized in a drip infusion. 11. Pharmaceutical composition according to one of the above claims characterized in a bolus injection. 12. Pharmaceutical composition according to claim 11 characterized in a single bolus injection. 13. Pharmaceutical composition containing DSPA according to at least one of the above claims and at least one additional pharmaceutically active component or its - 39 pharmaceutically acceptable salt. 14. Pharmaceutical composition according to claim 13 characterized in a neuroprotective agent. 15. Pharmaceutical composition according to claim 13 characterized in a glutamate receptor antagonist. 16. Pharmaceutical composition according to claim 15 characterized in a competitive or non-competitive antagonist. 17. Pharmaceutical composition according to claim 13 characterized in at least one thrombin inhibitor, preferentially selected from the group of the following substances: thrombomodulin, thrombomodulin analogues, triabin, pallidipin or solulin. 18. Pharmaceutical composition according to claim 13 characterized in at least one anticoagulant agent, preferentially selected from the group of the following anticoagulant agents: hirudin, heparin, acetylsalicylic acid or ancrod. 19. Pharmaceutical composition according to claim 13 characterized in anti inflammatory substances. 20. Pharmaceutical compositions according to claim 13 characterized in antibiotic agent. 21. Pharmaceutical composition according to claim 13 characterized in citicholine.