*2.1. Thrombolytic Activity*

Thromboembolic disease is a main cause of mortality and disability. For instance, stroke is responsible for 5.2% of all mortalities in the world [26]. Most of them are ischemic strokes, which would trigger transient or permanent occlusion of cerebral vessels causing brain infarcts, cerebral tissue death, and focal neuronal damage after blocking for 6 h [27]. Therefore, the key to saving stroke patients is solving thromboembolism in an efficient way. It is needed to search for more potent and safer drugs for the inhibition and treatment of ischemic symptoms.

In 1999, Hu et al. isolated compound **1** from the fungus *S. microspora* IFO 30018, with a preliminary determination for its plasminogen activation and fibrinolysis activity at 80–150 μM in vitro [20]. In 2010, Hashimoto et al. established a novel cerebral infarction model for predicting cerebral infarction, in which generated embolus transferred to the brain in the right common carotid artery of Mongolian gerbils, induced by acetic acid [28]. In the same year, they assessed the therapeutic effect of compound **1** and t-PA in the cerebral infarction model [29]. The fibrinolytic activity of compound **1** (<20% of positive control) was lower than t-PA (>140% of positive control) in the 3 h after administration, but the activity of compound **1** increased about 3.5-fold during 1–3 h and was higher than t-PA after 3 h. It was attributed that the activity of compound **1** gradually increased. Meanwhile, compound **1** extended the therapeutic time window. Compared with a clear infarct in the cerebral hemisphere by t-PA (10 mg/kg) treatment, there was no visible infarction in the group of compound **1** at 3 h after ischemia. More importantly, there was little hemorrhagic region with 10 mg/kg of compound **1**, suggesting that it is a latent safe cerebral infarction therapy method. Hu et al., (2012) disclosed that compound **1** enhanced plasmin generation in vivo [20]. The level of plasmin-α2-antiplasmin complex, an indicator of plasmin formation, increased by 1.5-fold in male ICR mice after the treatment with 5 and 10 mg/kg of compound **1**. In 2014, the antithrombotic activity of compound **1** was further demonstrated in the male cynomolgus monkey model [30,31], which achieved excellent effects (Table 1). In addition, Ito et al. found that the combination therapy with warfarin and compound **1**, in the middle cerebral artery occlusion, improved the treatment safety and reduced hemorrhagic transformation [32]. Compound **1**, as a safe thrombolytic agent, relieved the side effects, such as severe infarction, edema, and hemorrhage, induced by warfarin in the middle cerebral artery occlusion model. All mice treated with compound **1** survived and the hemorrhagic severity score (1.3 ± 0.5) indicated decreased hemorrhagic transformation.


**Table 1.** The antithrombotic effect of compound **1** in the severe embolic stroke monkey model.

Then, Wang et al. isolated compound **1** from a rare marine fungus *Stachybotrys longispora* FG216 and evaluated its fibrinolysis activity [33]. Additionally, 0.1–0.4 mmol/L of compound **1** increased the Glu-plasminogen and Lys-plasminogen activation by 2.05–11.44 times in vitro. Meanwhile, 10 mg/kg compound **1** dissolved most pulmonary thrombus in the Wistar rat in vivo. Yan et al. further researched the thrombolysis and hemorrhagic activities of compound **1**, from *S. longispora* FG216, in vitro and on acute pulmonary embolism Wistar rat model in vivo [34]. Compound **1**, from 5 to 25 μM, induced fibrin hydrolysis in vitro; moreover, its thrombolytic activity was evaluated with fluorescence lung tissues in vivo. It was observed that compound **1**, of 5 and 10 mg/kg, displayed effective dissolving capacity (less fluorescence halo). Meanwhile, the euglobulin lysis time (ELT) was shortened for 30 s by the treatment of compound **1** in the Wistar rat model. Shortening ELT was related to the activation of the fibrinolytic system. Therefore, compound **1** exhibited fibrinolytic activity in vivo. Compound **1** (5, 10, and 25 mg/kg), especially, did not induce fibrinogenolysis at 30 min and 2 h after administration, which suggested that compound **1** reduced the risk of hemorrhage. Thus, compound **1** was a potential thrombolytic agen<sup>t</sup> without hemorrhage [34]. In 2021, Gao et al. detected that compound **1**, with low concentration (0.096 mM), enhanced fibrinolytic activity by 2.2-fold in vitro; however, it inhibited fibrinolytic activity at excess doses (above 0.24 mM) [35].

Congeners **3**, **5**, and **7** enhanced fibrinolysis activities at 0.25 mM, in the 125I-Fibrin degradation experiment, by 2.3-fold, 1.9-fold, and 2.7-fold, respectively [36]. Congener **8** (80 μM) also increased fibrinolysis activity by eight-fold in the fibrin binding of 125Iplasminogen [20]. In 2003, Hu et al. isolated congeners **4**, **6,** and **9** with the activation effect on the urokinase-catalyzed plasminogen in vitro [24]. In 2012, congeners **11**–**13**, isolated from *S. microspore*, showed similar plasminogen activation activities when compared with compound **1** [37]. In 2018, Shibata et al. evaluated fibrinolysis activities of congeners **12** and **17** in an acetic acid-induced cerebral infarction mouse model [38]. Compared with compound **1**, of 10 mg/kg, congeners **12** and **17** reduced the size of the infarction area, neurological score, and edema percentage (Table 2).


**Table 2.** The fibrinolysis activities of compound **1**, as well as congeners **12** and **17**.

To gain a deep insight into the antithrombotic effect, many studies attempted to illustrate the detailed mechanism for compound **1**. Hashimoto et al., firstly, confirmed excellent thrombolytic activity of compound **1** (no visible infarction area after treatment with 10 mg/kg for 3 and 6 h) in an acetic acid-induced novel embolic cerebral infarction model in vivo. They hypothesized that compound **1** could relieve cerebral infarction by combined effects, giving rise to studies on other activities of compound **1** [28]. In 2010, they demonstrated that compound **1** possessed thrombolytic and anti-inflammatory activities [29]. By the treatment with compound **1**, at 3 h after ischemia, mRNA expression of interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), and interleukin-6 (IL-6) did not increase significantly. Therefore, compound **1** ameliorated hemorrhage and neurologic deficits, with a

wide therapeutic time window in thrombolysis, by inhibiting inflammation. One year later, Akamatsu et al. further completed the anti-inflammatory mechanism of compound **1** in fibrinolysis [39]. Matrix metalloproteinase-9 (MMP-9) was significantly inhibited (92 kDa band) with compound **1** in transient focal cerebral ischemia, suggesting a cerebral neuroprotective effect of compound **1,** on ischemia/reperfusion injury, and a reduction risk for hemorrhagic transformation. Moreover, compound **1** inhibited the expression of an early superoxide anion and nitrotyrosine for 2 h after ischemia/reperfusion, so it showed anti-oxidative activity to reduce ischemia/reperfusion damage [39]. In 2014, Hanshimoto et al. observed that reactive oxygen species could cause overexpression of proinflammatory cytokines [40]. Compound **1** inhibited the overexpression of a signal transducer and activator of transcription 3, to extend the therapeutic time window in thrombolysis therapy, by exhibiting its anti-oxidative effect. In addition, Huang et al. observed the inhibitory activity of compound **1** on pro-MMP-9, which inhibited the degradation of the basal membrane and the blood–brain barrier, reducing the risk of hemorrhage [41]. Moreover, Koyanagi et al., (2014) found compound **1** performed better plasminogen activation activity with the presence of physiological cofactors [42]. Compound **1**, of 20–60 μmol/L, promoted the activation of Glu-plasminogen, by 10-fold, with phosphatidylcholine and phosphatidylserine. Meanwhile, compound **1** also showed promoted plasminogen activation activity, with the presence of gangliosides and oleic acid released from thrombus in the process of clot lysis. These endogenous cofactors might change the fifth kringle domain conformation of plasminogen to induce the interactions between compound **1** and the plasminogen, whose mechanism could be elucidated in detail in future.

Wu group also investigated the thrombolytic mechanism of compound **1**. Compound **1** of 0.1–0.4 mmol/L activated the Glu-plasminogen and Lys-plasminogen (2.05–11.44 folds), but it had no fibrinolytic activity in the absence of u-PA or a plasminogen in vitro [33]. Meanwhile, the treatment, with 10 mg/kg of compound **1** after 24 h, performed efficiently in the pulmonary embolism Wistar rat model, meaning that u-PA and plasminogen mediated its thrombolytic effect. Furthermore, Wu group detected enzymatic kinetic parameters of compound **1** by chromogenic-substrate associated with *p*-nitroaniline from the enzymatic reaction [43]. The results indicated that the increase in *k*cat and *k*cat*/K*m activity was related to the concentration of compound **1**, which exhibited 26.5-fold and 22.8-fold activity at 40 μg/mL. Moreover, the affinity of plasminogen and pro-uPA to the enzyme substrate presented a faint decrease with an increasing concentration of compound **1**, as the *Km* increased (from 0.413 to 0.484 μmol/L) along with the increasing concentration of compound **1** (0–40 μg/mL). The results further proved that reciprocal activation of pro-uPA and plasminogen was critical to the fibrinolysis activity of compound **1**, which enhanced the maximum catalytic efficiency and total catalytic activity of fibrinolysis. The fibrinolysis activity of compound **1** featured an enzymatic kinetic characteristic.

Wu group further studied the interaction mechanism between compound **1** and plasminogen [35]. The Glu-plasminogen, in the bloodstream, contains a Pan-apple domain (PAp), five kringle domains (KR1-KR5), and a serine protease domain (SP). Lysine-binding sites (LBS), in kringle domains of plasminogen, were essential for the interaction of plasminogen and compound **1** [44–46]. Compound **1**, firstly, bound the LBS of KR1, while KR1 activated and mediated the interaction between plasminogen and the C-terminal lysine moiety on fibrin. KR5 dropped from PAp and was exposed to closed plasminogen temporarily [36]. Then, compound **1** formed hydrogen bonds with Asp518 and Asp534 in KR5, which induced conformational change and structural rearrangement. After that, additional LBSs on kringle domains were exposed, leading the movement of PAp. Therefore, these additional LBSs interacted with compound **1**, causing an open conformation of plasminogen, which could be easily activated by u-PA (Figure 3). The theoretical binding mode showed that compound **1** formed a stable complex with Glu39, Thr41, and Arg43 in plasminogen with hydrogen bonds (Figure 4).

**Figure 3.** The mechanism of plasminogen activation by compound **1**.

**Figure 4.** The interactions between compound **1** and plasminogen: (**a**) binding site; (**b**) the 3D docking model of compound **1** with plasminogen.

The structure–activity relationships of the plasminogen modulator compound **1** and its congeners have been studied in detail. Most congeners contain the same geranylmethyl side-chain, but they bear different *N*-linked side-chains. Congener **2** possessed a hydroxylated geranylmethyl side-chain [17] and could not enhance plasminogen binding to the activated plasminogen. Thus, the side-chain of geranylmethyl plays a key role in promoting plasminogen activation. In 2016, Otake et al. found that the geranylmethyl side-chain of congeners was critical to inhibitory activity of soluble expoxide hydrolase (sEH), an enzyme mediating anti-inflammatory action [47]. sEH lost inhibitory activity with the increasing number of hydroxyl groups on geranylmethyl side-chains or missing geranylmethyl side-chains, and the terminal hydroxy group of side-chain led to the damage of cellular localization. For the *N*-linked side-chain, Hasumi et al. isolated the simplest congener (SMTP-0) without an *N*-linked side-chain, which had no plasminogen activation effect [48] (Figure 5). It can be concluded that the *N*-linked side-chain was essential for plasminogenmodulating activity. In 2010, Hasegawa et al. confirmed the crucial role of the *N*-linked side chain in modulating plasminogen [21]. The congeners, without ionizable groups in *N*-linked side-chain, were inactive in plasminogen activation, such as with congener **10**. Moreover, the congeners, with an aromatic group and a negatively ionizable group on the side-chain, were more active for the enhancement of plasminogen activation than those with an aliphatic group and a negatively ionizable group, such as congener **3** (*E*max = 15-fold). Koide et al. further isolated a series of SMTP congeners with different *N*-linked side-chains and evaluated their bioactivities. Congeners **11** (*E*max = 126-fold) and **12** (*E*max = 159-fold) were as potent as compound **1** (*E*max = 102-fold) in plasminogen-modulating activity [37]. Only these congeners could express higher plasminogen-modulating and anti-oxidative activities, and other isomers or phenolic hydroxy groups, at different positions, did not

present satisfactory activities in plasminogen activation. Moreover, the congeners with *N*-linked side-chains showed anti-oxidative activities too. All the congeners bearing a phenolic hydroxy group and a carboxylic acid group displayed higher anti-oxidative activities. Congener **12**, especially, possessed more than 1.7 times the anti-oxidative activity in comparison to compound **1** (Figure 5).

**Figure 5.** Structure–activity relationships of congeners **11** and **12**.

Pharmacokinetics is an essential evaluation system for the development of new drugs,which finally determines the metabolism and efficacy of drugs in vivo [49]. In 2013, Su et al. observed the pharmacokinetics and tissue distribution of compound **1** in Wistar rats [50]. Compound **1** had a half-life (*t*1/2) of ca. 22.37 min, and it was suitable for twocompartment models, by intravenous administration, for 10 and 20 mg/kg. From the viewpoint of tissue distribution, compound **1** was present in the highest concentration in the liver, but it had low or undetectable concentrations in the brain, suggesting that compound **1** did not cross the blood–brain barrier (the results were wrong, and compound **1** could cross the blood–brain barrier). In 2019, Ma et al. evaluated pharmacokinetic properties in beagle dogs and permeability characterization in Caco-2 cells [51]. *<sup>t</sup>*1/2 of compound **1** was determined in dogs' brains, and *<sup>t</sup>*1/2, in beagle dogs, was about two times longer than in Wistar rats (48.7 min in average). Moreover, compound **1** performed low penetrability in a human Caco-2 cell's monolayer model and the rapid distribution into organs, suggesting intravenous injection was more appropriate than oral. In addition,absorption and transportation characteristics of compound **1** had been studied [52]. In Caco-2 cells model, compound **1** expressed passive diffusion of the absorption pattern, and it was not the substrate of P-gp, indicating that compound **1** could cross the blood–brain barrier.Therefore, compound **1** had the potential to be a thrombolytic agen<sup>t</sup> for the treatment of occluded cerebral vessels.

The modification of compound **1** enhanced fibrinolytic activity to access more efficient and safer thrombolytic agents. In 2021, Wang et al. synthesized a series of compound **1** derivatives through the modification of phenyl groups, at the C2-OH and C2--OH positions on compound **1**, and evaluated their fibrinolytic activities (Figure 6) [53] (The compound, modified by Wang et al., was the enantiomer of compound **1** (8S, 9S)). Derivative **a**, with methyl, and derivative **b,** with *para*-bromobenzyl, presented significant fibrinolytic activity with the EC50 values of 59.7 μM and 42.3 μM, respectively. Derivative **b** showed rapidly increasing fibrinolytic activity in the early stage (0–40 min), dose-dependently. Furthermore, derivative **b** displayed weak activities of inducing apoptosis and anti-inflammation on HeLa cells, suggesting that derivative **b** was a potential antithrombotic agent.

**Figure 6.** The modification of compound **1**.

#### *2.2. Effects on Inflammation and Oxidant Related Damage: In Reperfusion of Occluded Vessels*

A large portion of tissue damage in diseases is caused by inflammation. In 2000, it had already been confirmed that inflammation could affect the coagulation system and regulation, which was responsible for thrombotic complications in vivo [54]. In 2005, it was found that patients with inflammatory diseases were more likely to develop thrombosis. For instance, the patients with inflammatory bowel disease suffered from a three-fold risk of pulmonary embolism or vein thrombosis [55]. In 2008, the unique role of inflammation was focused in the formation of venous thrombus [56]. Thus, researchers investigated the relationship between the reduced damage of ischemia/reperfusion and anti-inflammatory activity by the treatment of compound **1**.

Mammalian sEH contributes to inflammatory response through hydrolyzing lipid signaling molecules, and it has been developed as a potential therapeutic target [57,58]. More than 100 sEH inhibitor patents have been published for the treatment of diabetes, hypertension, pain, and cardiovascular diseases [59,60]. The geranylmethyl side-chain of compound **1** is crucial to the inhibitory activity of sEH [47]. Therefore, compound **1** and other staplabin congeners possessed grea<sup>t</sup> anti-inflammatory potential.

sEH is a bifunctional enzyme with a C-terminal domain (Cterm-EH) and an N-terminal domain (Nterm-phos). Cterm-EH catalyzes hydrolysis of epoxyeicosatrienoic acids (EETs, an endogenous signaling molecule involved anti-inflammation), and Nterm-phos hydrolyzes lipid phosphates [60–62]. Therefore, the inhibition to Cterm-EH is the key to inhibit inflammation. Mastsumoto et al. performed sEH inhibition kinetic analysis of congeners [63]. Congeners **14** (IC50 = 12 ± 1) and **15** (IC50 = 5 ± 2) had better Cterm-EH inhibitory activities in comparison to congeners **10** (IC50 > 100) and **16** (IC50 > 100). Therefore, although the geranylmethyl side-chain was essential to the inhibitory activity, the nature of *N*-linked side chains also affected the inhibitory potency of sEH (Figure 7). Compound **1** inhibited the hydrolysis of EETs with IC50 of 6.5 μM. SMTP-0 (IC50 = 1.2 μM) and congener **18** (IC50 = 9.2 μM) were also efficient for inhibiting the hydrolysis of EETs [63]. Meanwhile, both 10 mg/kg of compound **1** and congener **18** improved neuritis symptoms in a rat Guillain–Barré syndrome model, and they alleviated symptoms of ulcerative colitis and Crohn's disease in mice. The results proved that compound **1** and staplabin congeners possessed grea<sup>t</sup> anti-inflammatory potential [63].

**Figure 7.** The structure of congeners **14**, **15,** and the SARs study of inhibitory effect on Cterm-EH and Nterm-phos.

Occluded vessels reperfusion could produce reactive oxygen species (ROS), which would stimulate ischemic cells, secreting excessive pro-inflammatory and inflammatory cytokines, such as IL-6, IL-1β, and TNF-<sup>α</sup>. The overexpressed cytokines cause damage, hemorrhage, and even inflammation in cerebral vessels, which is a major factor in ischemic brain injury [64]. Shibata et al. observed little hemorrhagic region with compound **1** (10 mg/kg), in the model of cerebral infarction mice, in comparison with 10 mg/kg t-PA treatment; moreover, they investigated the involved mechanism [29]. mRNA expression of IL-6, IL-1β, and TNF-α were not increased by the treatment of compound **1**, in comparison to t-PA treatment, at 3 h after ischemia. Hashimoto et al. found that compound **1** decreased expression of IL-6, the signal transducer and activator of transcription 3, S100 calcium binding protein A8, and MMP-9 by microarray and RT-PCR analysis [40]. Therefore, compound **1** inhibited the secretion of pro-inflammatory and inflammatory cytokines to improve the hemorrhage and ischemic brain injury. Meanwhile, Akamatsu et al. detected that superoxide anions (one ROS in cerebral ischemia) were observed by hydroethidine signals at 2 h after reperfusion [39]. Hydroethidine signal was reduced by the treatment of compound **1** in comparison to the vehicle group, meaning that compound **1** inhibited the production of ROS to decrease reperfusion damage. In addition, the treatment of compound **1** reduced the expression of nitrotyrosine and MMP-9, causing attenuated ischemic neuronal damage. Moreover, inflammatory tissue could release proteolytic enzymes of MMP-9, which is associated with blood–brain barrier breakdown and hemorrhagic complications in cerebral infarction [65]. Ito et al. indicated that compound **1** inhibited the activation of MMP-9 to protect the blood–brain barrier from destruction and hemorrhagic transformation (pro-MMP-9: 88.9 ± 34.2; MMP-9: 5.0 ± 1.6) in mice [32]. Besides, Huang et al. suggested that compound **1** could inhibit oxidative stress to reduce ischemia/reperfusion injury [41]. Compound **1** decreased the expression of 4-hydroxy-2-nonenal (4-NHE), 3-nitrotyrosine, and 8-hydroxy-2--deoxyguanosine (8-OHdG) significantly, which provided therapeutic benefits for ischemic stroke. Therefore, compound **1** possessed anti-inflammatory and anti-oxidative activities in the reperfusion of occluded vessels (Figure 8).

**Figure 8.** The anti-inflammatory and anti-oxidative mechanisms of compound **1** in ischemia– reperfusion damage.
