Efficacy of SMTP‐7, a small‐molecule anti‐inflammatory thrombolytic, in embolic stroke in monkeys

Abstract SMTP‐7 (Stachybotrys microspora triprenyl phenol‐7) is a small molecule that promotes thrombolysis and suppresses inflammation possibly through plasminogen modulation and soluble epoxide hydrolase (sEH) inhibition, respectively. Here, we demonstrate an efficacy of SMTP‐7 in a severe embolic stroke model in monkeys. The middle cerebral artery was embolized by an autologous blood clot. Saline, SMTP‐7, or tissue‐type plasminogen activator (t‐PA) (n = 5 in each group) was given after 3 hours, and neurologic deficit scoring and infarct characterization were performed after 24 hours. Hemorrhagic infarct‐accompanied premature death was observed for two animals in t‐PA group. SMTP‐7 treatment significantly reduced the sizes of infarct by 65%, edema by 37%, and clot by 55% compared to saline treatment. Plasma levels of the products of plasminogen activation (plasmin‐α2‐antiplasmin complex) and sEH reaction (dihydroxyeicosatrienoic acid) in SMTP‐7 group were 794% (P < 0.05) and 60% (P = 0.085) compared to saline group, respectively. No significant changes in the plasma levels of MMP‐9, CRP, MCP‐1, and S100B were found. There was an inverse correlation between plasmin‐α2‐antiplasmin complex level and infarct volume (r = 0.93, P < 0.05), suggesting a role of thrombolysis in the SMTP‐7 action to limit infarct development. In conclusion, SMTP‐7 is effective in treating severe embolic stroke in monkeys under conditions where t‐PA treatment tends to cause hemorrhagic infarct‐associated premature death.

hemorrhage. 4 The major limitation is the time window of its use: within 3 to 4.5 hour following symptom onset, depending on regional regulation or guideline. 4 Good outcome declines, and bad outcome increases along with the time to the t-PA treatment, and the 3-to 4.5-hour time window is the critical lines to benefit patients who receive t-PA. 4 The major cause of the time-dependent drawback is postulated to be ischemia-reperfusion injury, which leads to neuronal cell death, cerebral inflammation, and eventually bloodbrain barrier disruption. 5 SMTP, named after Stachybotrys microspora triprenyl phenol, is a family of small molecules derived from the fungus S. microspora. 6 SMTP was first discovered by a screening of microbial cultures for a compound that promoted plasminogen binding to fibrin, a crucial step of physiological fibrinolysis. 6  activator-catalyzed activation of plasminogen, [7][8][9] leading to a promotion of thrombus clearance in vivo. 8,10 On the other hand, SMTP-7 exhibits anti-inflammatory activities in vivo in disease models that are irrelevant to thromboembolism. 11 Inhibition of soluble epoxide hydrolase (sEH) may account for the anti-inflammatory mechanism. 11,12 SMTP-7 is effective in treating several types of ischemic stroke models in rodents. These include thrombotic and embolic strokes as well as transient or permanent mechanical ischemia of the brain. [13][14][15][16][17][18] The efficacy of SMTP-7 in the mechanical ischemia models suggests a significant role of its anti-inflammatory action. Furthermore, experiments using several analogs of SMTP suggest that anti-inflammatory and anti-oxidative properties, as well as the profibrinolytic action, contribute to the therapeutic activity of SMTP. 19,20 The risk of SMTP-7 to compromise hemostasis is far low compared to that of t-PA. 17 The difference in the mechanism of action of the two agents may account for this result. SMTP-7 enhances endogenous fibrinolytic process through modulating plasminogen conformation, 6 whereas t-PA directly activates plasminogen. Although t-PA exacerbated ischemia-reperfusion-associated cerebral hemorrhage, SMTP-7 ameliorated cerebral hemorrhage in a mouse model where animals were pretreated with the anticoagulant warfarin. 16 In addition, time window of SMTP-7 treatment is broader than that of t-PA treatment as observed in embolic stroke models in rodents. 14,18 Stroke is heterogeneous both in its cause and clinical course. 21 Therefore, animal testing using various models in higher order species that resemble clinical setting is beneficial to predict clinical efficacy of a drug candidate. 22,23 Regarding thrombotic stroke, we observed an efficacy of SMTP-7 in a photochemical thrombosis model in cynomolgus monkeys. 17 Along with thrombosis, another major cause of ischemic stroke is thrombotic embolization. 24 We, therefore, attempted to test SMTP-7 in an embolic stroke model in monkeys to confirm its efficacy. The timing of drug treatment was set at 3 h after thrombus embolization where t-PA failed to yield good outcome. We also explored clinically relevant biomarkers that predict outcome based on the mechanism of action, such as plasminogen activation and inhibition of sEH.

| Drugs
SMTP-7 was produced by S. microspora 25 and converted to sodium salt (lot N01YHN, purity 99.19%). Recombinant t-PA (ACTIVACIN for Injection) was obtained from Kyowa Hakko Kirin, Tokyo, Japan. SMTP-7 (10 mg/kg), t-PA (3 mg/kg), and vehicle (saline) were administered intravenously (5 mL/kg) as follows: bolus injection of 10% of the total dose for 5 seconds followed by infusion of the remaining dose over 30 minutes (for SMTP-7 or saline) or bolus injection of 10% of the total dose for 1 minute followed by infusion of the remaining dose over 60 minutes (for t-PA). The decision to set the doses of SMTP-7 and t-PA were based on the results from preliminary experiments using the cynomolgus monkey embolic stroke model. The t-PA dose setting was supported by the fact that its fibrinogenolytic activity in monkeys at 3 mg/kg (reducing plasma fibrinogen level to 60% of control) 26 was comparable to that in humans at a clinically relevant dose (60 mg/body; reducing plasma fibrinogen level to 57% of control). 27

| Animal experiment
The animal study was performed in compliance with the Standards Animals received an antibiotic mixture once daily (12.5 mg/kg dihydrostreptomycin sulfate and 10 000 U/kg benzylpenicillin procaine) and buprenorphine hydrochloride (10 g/kg) twice daily for 3 days.
On the third day, blood (3 mL) was drawn from the femoral vein and  (Table S1) and clinical sign observed 1 and 3 hours after the embolization; and (2) any animal judged unsuitable to continue the study during the observation period (until 24 hours after embolization). Details of the exclusion criteria are shown in Table 1. We excluded four ineligible animals based on the exclusion criterion 1.
The results of the exclusion are summarized in Table 1.
The eligible animals (n = 24) were randomly assigned to three groups (n = 8 for each group). The drug dosing was made after 3 hours of the thrombus embolization. The treatment time point was set based on our preliminary experiments and the knowledge from the literature that showed irreversible brain damages after 2-3 hours of severe ischemia in a monkey transient ischemia model. 29 We scored neurologic deficits at 5, 7, 18, and 24 hours following the embolization. The observer was not aware of any information on individual animals including treatment assignment. Nine animals that met the exclusion criterion 2 during the observation period were excluded from the study (Table 1). Macroscopic appearance of the origin of the MCA in the excluded animals is shown in Figure S1.
Premature death with hemorrhagic infarct in t-PA or SMTP-7 groups was not to be excluded but to be counted as a "dosing-related death." This settlement was due to the fact that t-PA treatment caused hemorrhagic infarct-associated premature death in some animals in the previous experiments conducted at the Shin Nippon Biomedical Laboratories using this embolic stroke model. In this study, there were two cases that met the dosing-related death criterion in the t-PA group (Table 1).

Exclusion criteria Excluded animal
(1) Any animal judged unsuitable as a model 3 h after embolization based on: Total: n = 4 (i) neurologic deficit score of 6 or below for consciousness or skeletal muscle coordination, or clinical sign symptomatic of anterior cerebral artery occlusion (unable to maintain sitting position due to right lower limb extension)

#2, #27
(ii) neurologic deficit score of 16 or greater in the consciousness category (iii) neurologic deficit score of 16 or greater in the skeletal muscle coordination category #1, #13 (iv) total neurologic deficit score of 47 or greater, and any two of the following abnormalities: vomitus, severe salivation, and muscle weakness in ipsilateral limb (left upper and lower limbs) (v) exclusion is judged necessary by the study director (2) Any animal judged unsuitable to continue the study during the observation period (after dosing until 24 h after embolization) based on: Total: n = 9 Saline group: n = 3 SMTP-7 group: n = 3 t-PA group: n = 3 (i) neurologic deficit score of 16 or greater in the consciousness category (ii) neurologic deficit score of 16 or greater in the skeletal muscle coordination category Saline group: #12 t-PA group: #4 (iii) total neurologic deficit score of 47 or greater, and any two of the following abnormalities: vomitus, severe salivation, and muscle weakness in ipsilateral limb (left upper and lower limbs) (iv) exclusion is judged necessary by the study director There was no statistical significance in the differences in the number of excluded animals among the 3 groups when analyzed by χ 2 test. The animal number corresponds to the picture number in Figure S1. Any animal to which one of the following criteria conforms were to be excluded from the study. Twenty-eight animals (number #1 to #28, ascending order from the lowest body weight) were subjected to embolization. After excluding 4 animals (#1, #2, #13, and #27), 24 animals were randomized to saline (#3, #9, #12, #14, #17, #19, #21, and #23), SMTP-7 (#5, #7, #10, #15, #20, #22, #25, and #28), and t-PA (#4, #6, #8, #11, #16, #18, #24, and #26) groups. For humane reason, the excluded animals were euthanized by an intravenous injection of sodium pentobarbital (64.8 mg/mL, 0.4 mL/kg), followed by exsanguination. a There were two more prematurely died animals in the t-PA group (#8 and #18). These had massive hemorrhagic infarct and were regarded as "dosingrelated death," which were not excluded from the study.
The survived animals (n = 13) were anesthetized by an intravenous injection of sodium pentobarbital (26 mg/kg) and perfused with cold saline from the left ventricle to draw blood 24 hours after the thrombus embolization. Subsequently, the brain was removed, and its external appearance including the origin of the MCA was photographed, followed by preparation of successive 14 coronal brain slices (4-mm thickness). The sections were inspected macroscopically for hemorrhage and stained with 1% 2,3,5-triphenyltetrazolium chloride (TTC). Digital images of the stained sections were recorded to quantitate infarct area using an image analysis software

| Data analysis
Data for neurologic deficits scores and bleeding time are indicated as the median and the interquartile range (IQR). Infarct size, and other parameters are expressed as the mean ± SD. Since the primary objective of this study was to assess the advantage of SMTP-7 over the control (saline treatment), statistical analyses were planned to examine differences between SMTP-7 and saline groups at a one-

| Dosing-related death
Two prematurely died animals in the t-PA group had massive hemorrhagic infarct ( Figure S2) and were counted as a dosing-related death. Embolized thrombi were cleared from the MCA in these animals, suggesting an occurrence of thrombolysis by the drug treatment ( Figure S2). There were no animals that died with intracranial hemorrhage in the saline and SMTP-7 groups. between the saline and SMTP-7 group was significant (P < 0.05).

| Infarct and edema development
The distribution of infarct area in each of the 14 brain slices  Figure 2B. The infarct volume (% of total ipsilateral brain volume), calculated from all the TTC stained sections, were 29.1% to 13.4% in the saline group, 10.1% to 4.5% in the SMTP-7 group, and 17.4% to 7.4% in the t-PA group ( Figure 2C). The difference between the saline and SMTP-7 group was statistically significant (P < 0.05). Figure 2D shows the relative volume of edema, which were 12.3% to 4.1% in the saline group, 7.7% to 2.1% in the SMTP-7 group, and 10.4% to 7.0% in the t-PA group. The difference between the saline and SMTP-7 group was significant (P < 0.05).

| Neurologic deficits
Neurologic deficits were monitored before (

| Bleeding
We measured bleeding time just after the thrombus embolization and just before the termination of the drug infusion, since the drugs used could be categorized as thrombolytics. We found no significant intragroup dosing-related change in the bleeding time in each of the three groups (by Friedman test) as well as intergroup differences in the post-dosing values (by Wilcoxon rank sum test) ( Figure S4).

| Blood parameters
To seek a biomarker that predicts efficacy of SMTP-7, we tested the following substances for their plasma levels after the drug treatment: PAP, DHET, MMP-9, CRP, MCP-1, and S100B ( Figure 4A). PAP is formed from plasmin and its physiological inhibitor α 2 -antiplasmin, representing plasminogen activation. 30 DHET is a product of sEHmediated hydrolysis of the anti-inflammatory fatty acid epoxide epoxyeicosatrienoic acid, representing sEH inhibition. 31 MMP-9, a zinc-binding protease that degrades components of the basal lamina around blood vessels, can be a marker of ischemic stroke development as well as its hemorrhagic transformation. 32 The acute-phase reactant CRP can be a marker of ischemic stroke development and a predictor of stroke prognosis. 33 A high level of the chemokine MCP-1 is found in ischemic stroke patients, and it can mirror the development of inflammatory response in the brain. 34 The level of S100B can predict a course of infarction in ischemic stroke patients. 35 The average levels of MMP-9, CRP, MCP-1, S100B, and DHET in the SMTP-7 group were 36%, 50%, 86%, 69%, and 60% of the respective levels in the saline group, and the respective levels in the group (98.1 ± 93.7 ng/mL) was significantly higher than that in the saline group (12.4 ± 10.5 ng/mL; P < 0.05) ( Figure 4G).

| Correlations among stroke pathologies and blood parameters
To characterize relationships among the stroke pathologies and blood biochemical parameters, we analyzed possible correlations using the data obtained from individual animals of all of the treatment groups (Table S2). The results demonstrated a network covering infarct, edema, DHET, MMP-9, S100B, and MCP-1 ( Figure 5A,B).
The correlations between MCP-1 and PAP, PAP, and CRP, and infarct and neurologic deficits extended this network. These results suggest a relationship between the developments of stroke pathology and inflammation in this model. When analyzed for each treatment group (Table S2), a significant inverse correlation between PAP level and infarct volume (r = 0.93; P < 0.05) was found in the SMTP-7 group ( Figure 5C). The PAP level in the t-PA group was profoundly high (1459 ± 63.2 ng/mL for the three survived animals) ( Figure 4G), whereas neither significant correlation was found between PAP and infarct volume in the t-PA group nor saline group ( Figure 5C).

| DISCUSSION
We investigated efficacy of SMTP-7 in a monkey model of embolic stroke, which was severe in that a large thrombus (10 cm in length) was introduced to occlude the MCA. Moreover, the timing of drug treatment was set at 3 hours following embolization, which was too late to achieve good outcome by the treatment with t-PA. Indeed, two animals of five tested in the t-PA group prematurely died in this experiment, and none of the pathological parameters improved in the remaining three survived animals as compared with the salinetreated animals in post-hoc analyses, setting a one-sided significance level of 5%. The animals that prematurely died developed hemorrhagic infarction, while the injected thrombus was cleared ( Figure S2).
A negative correlation was found between clot size and infarct volume in the survived animals in the t-PA group (Table S2) There may be some differences in the actions of SMTP-7 and t-PA that lead to the different outcomes. One possibility is the difference in the thrombolytic mechanism that may affect severity of ischemiareperfusion injury. t-PA treatments cause reductions in plasma levels of the hemostatic factors such as α 2 -antiplasmin, plasminogen, and fibrinogen. 27,36,37 These declines can be the consequences of systemic activation of plasminogen. The action of SMTP-7 is to enhance endogenous fibrinolytic system; it is not a direct plasminogen activator but a modulator that relaxes plasminogen conformation. 8,9 This could lead to plasminogen activation only where, when, and to the extent endogenous plasminogen activators are supplied. 6 Indeed, SMTP-7 does not reduce the levels of α 2 -antiplasmin and fibrinogen in normal monkeys. 17 In consistent with this notion, we observed a moderate elevation of PAP level in the SMTP-7 group, whereas the elevation of PAP in the t-PA group was strikingly high ( Figure 4G).
As for ischemia-reperfusion injury in the brain, a gradual reperfusion can be safer than a rapid flow restoration. 38 These findings suggest the benefit of physiological thrombolysis compared to inordinate thrombolysis arising from t-PA treatment.
Another possibility is the involvement of the anti-inflammatory and anti-oxidative actions of SMTP-7. 11,19,39,40 Inflammatory and oxidative reactions play crucial roles in the development of ischemiareperfusion injury. 41 Indeed, there are several correlation networks that tie stroke pathologies with inflammatory responses in the embolic stroke model used in this study ( Figure 5A), suggesting a role of inflammation in stroke development in this model. Although t-PA is proinflammatory to cerebral tissues, 42 SMTP-7 suppresses inflammatory and oxidative responses in stroke models. 10,13,16,19,20 As shown in Figure 4, our results are in line with this idea.
In conclusion, our results demonstrate an efficacy of SMTP-7 in a severe embolic stroke model where recanalization therapy may cause ischemia-reperfusion injury that impairs outcome. There are, however, some issues that require considerations: (1)