Potential clinical applications of phytopharmaceuticals for the in‐patient management of coagulopathies in COVID‐19

Abstract Thrombotic complications occur in many cardiovascular pathologies and have been demonstrated in COVID‐19. The currently used antithrombotic drugs are not free of adverse reactions, and COVID‐19 patients in particular, when treated with a therapeutic dose of an anticoagulant do not receive mortality benefits. The clinical management of COVID‐19 is one of the most difficult tasks for clinicians, and the search for safe, potent, and effective antithrombotic drugs may benefit from exploring naturally bioactive molecules from plant sources. This review describes recent advances in understanding the antithrombotic potential of herbal drug prototypes and points to their future clinical use as potent antithrombotic drugs. Although natural products are perceived to be safe, their clinical and therapeutic applications are not always apparent or accepted. More in‐depth studies are necessary to demonstrate the clinical usefulness of plant‐derived, bioactive compounds. In addition, holistic approaches in systematic investigations and the identification of antithrombotic mechanisms of the herbal bioactive molecule(s) need to be conducted in pre‐clinical studies. Moreover, rigorous studies are needed to compare the potency of herbal drugs to that of competitor chemical antithrombotic drugs, and to examine their interactions with Western antithrombotic medicines. We have also proposed a road map to improve the commercialization of phytopharmaceuticals.


| THE COVID-19 PANDEMIC: A COMMUNITY HEALTH DISASTER
Billions of people worldwide have been infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which was initially identified in December 2019. By the end of March 2020, the infections had been rapidly transmitted in several countries, and to date, more than 175 million cases have been confirmed worldwide, with at least 3.8 million deaths. The World Health Organization (WHO) declared that coronavirus disease 2019 (COVID-19) was a global pandemic and a great concern for community health. COVID-19 symptoms can be extremely variable (Table 1) and patients with COVID-19 usually display several physiological complications and distress. Seriously ill patients, however, show an alarming risk of thrombotic complications, including microvascular thrombosis, venous thromboembolic disorder, and stroke that can lead to multi-organ failure, which contributes to the high mortality rate in COVID-19 (Acharya, Alameer, Calpin, Alkhattab, & Sultan, 2021;McFadyen, Stevens, & Peter, 2020). Importantly, thrombosis-associated cardiovascular diseases (CVDs) are also the leading causes of death globally (Jackson, 2011;Montrief, Davis, Koyfman, & Long, 2019;Raskob et al., 2014). Consequently, approaches for curbing the thrombotic complications in COVID-19 and CVDs, in general, have received much attention in clinical research.
The highly regulated physiological hemostatic coordination, which encompasses platelet aggregation, blood coagulation, and subsequent fibrinolysis maintains the closed circulatory system that operates under high pressure in mammals post vascular injury (Montrief et al., 2019;Raskob et al., 2014). As shown in Figure 1, the thrombus (blood clot) formation inside the blood vessels, which is temporary and spatial under normal physiological conditions, is under strict control of the regulatory system (Raskob et al., 2014). When runaway processes (like those observed in  override the hemostasis or a swing in the hemostatic equilibrium occurs toward pro-coagulation, the result is the development of a thrombus (clot) in the vein and artery. In microvascular circulation, the ultimate result is a clinical condition known as thrombosis (Day et al., ;Lippi, Franchini, & Targher, 2011;McFadyen et al., 2020;Raskob et al., 2014). The anticipated mechanism of SARS-CoV-2-induced coagulopathies in COVID-19 is shown in Figure 2.

| CLINICAL THERAPY AND PROPHYLAXIS OF THROMBOTIC COMPLICATIONS IN COVID-19: THE KEY ISSUES AND CHALLENGES
Currently, three major classes of antithrombotic drugs: blood thinners (anticoagulants), antiplatelet drugs (inhibitors of platelet aggregation), and clot-busting drugs (fibrinolytic drugs) are widely applied to treat T A B L E 1 The most common clinical symptoms are shown by patients suffering from COVID-19 1. Complications of respiratory system • Acute lower respiratory infection characterized by the following symptoms-congestion, running nose, mild fever, sore throat, and dry fever. • Acute respiratory distress syndrome (ARDS), where lungs are severely damaged, is characterized by severe breathing trouble, sometimes confusion, and fatigue. • Pneumonia is characterized by cough maybe with bloody mucus, rapid and shallow breathing, occasional chest pain, fever, and loss of appetite. • Pneumothorax, also known as a collapsed lung, shows the clinical symptoms of chest pain, shortness of breath, increased heart rate, dizziness, and the patient may undergo a coma in severe conditions. • Respiratory failure.

Circulatory and hematological disorders
• Acute myocarditis and myocardial injury • Cardiac arrest • Disseminated intravascular coagulation (DIC) results in stroke and sometimes death. Symptoms are bold clots, fall in blood pressure, bleeding, and confusion. • Lymphocytopenia due to low lymphocyte count in blood is characterized by joint pains, skin rash, weight loss, night sweat, enlarged lymph node, cough, fever, and running nose. • Pulmonary embolism is characterized by a rapid and irregular heartbeat, anxiety and sweating, dizziness, and swelling of legs due to deep vein thrombosis. • Thrombocytopenia due to low circulatory platelet count is characterized by bleeding of gums, blood in urine, stool, or vomit, and rectal bleeding. • Venous thromboembolism (VTE) is characterized by swelling(oedematous) of legs with intense pain, tenderness of the thigh or calf, and reddish discoloration.

Disorders of liver and kidney
• A liver function test can assess an acute liver injury. • Acute kidney injury may lead to kidney failure.

Some uncommon symptoms
• The appearance of rash and discoloration of fingers or toes • Aches and pains.
• Complete or partial loss of taste and smell. • Diarrhea.
• Headache. arterial and venous thrombosis in thrombotic complications McFadyen et al., 2020;Wijaya, Andhika, & Huang, 2020). Table 2 shows some of the modern Western drugs used to treat or prevent thrombosis in CVDs and COVID-19. The oral anticoagulants, which act at different levels in the blood coagulation cascade, have short-term and long-term goals for preventing venous thromboembolism (VTE), stroke, and systemic embolism and controlling arterial and venous thrombosis (Bielecki, Lee, & Hamad, 2018). These antithrombotic drugs are not free of adverse reactions, however, as indicated by the following issues: 1. Despite extensive research to develop potent antithrombotic cardiovascular drugs, a significant proportion of CVD-related mortality cannot be prevented (Ibrahim et al., 2018).
F I G U R E 1 A schematic diagram shows a fine balance between fibrin clot formation (blood coagulation) and fibrinolysis (clot lysis) processes under normal physiological conditions. However, a deviation from this delicate balance due to physiological disorder causes a hemostatic disturbance, leading to thrombosis and other cardiovascular complications F I G U R E 2 The proposed antithrombotic mechanism of SARS-CoV-2-induced coagulopathies in COVID-19. Direct SARS-CoV-2-platelet interaction results in high levels of platelet activation, promoting a pro-thrombotic state. Direct viral trauma and resultant inflammation lead to fibrinogen elevations through IL-6, leukocyte activation, NETosis, endothelial cell activation, and inflammatory mediator release. Subsequent activation of both the tissue factor and contact activation pathways of the coagulation cascade further potentiates a hyper-coagulable state, which leads to the development of thromboembolic complications in patients. CRP, C-reactive protein; FXII, coagulation factor XII; FXIIa, activated coagulation factor XII; FX, coagulation factor X; FXa, activated coagulation factor X; IL-6, interleukin 6; IL-8, interleukin 8; TNF-α, tumour necrosis factor α; TF, tissue factor. (Reprinted Figure 1 from Page & Ariëns, 2021 with permission from the publisher) 2. Some limiting factors (e.g., resistance or poor efficacy of the drugs in some patients, drug-food or drug-drug interactions, adverse effects including increased risk of bleeding, gastrointestinal dysfunctions, and low therapeutic index) are significant obstacles to the success of these life-saving drugs Schurgers, Aebert, Vermeer, Bültmann, & Janzen, 2004;Vranckx, Valgimigli, & Heidbuchel, 2018).

Interactions between antiplatelet and anticoagulant drugs can lead
to serious complications in patients.
Recent studies have shown that plant-derived active compounds and some natural compounds can find practical therapeutic applications in treating various complications of COVID-19 (Brendler et al., 2021;Ganguly & Bakhshi, 2020;Islam et al., 2020;Kalita, Saviola, Samuel, & Mukherjee, 2021). Consequently, as discussed below, a far-reaching drug discovery program is needed to develop superior and safe antithrombotic drugs by exploring natural resources for the hospital management of thrombotic complications in COVID-19 and other CVDs (Cordier, Cromarty, Botha, & Steenkamp, 2012;Li et al., 2020).

| PLANT-DERIVED NATURAL PRODUCTS AS POTENTIAL ANTITHROMBOTIC DRUGS: THE PROS AND CONS
The innovative advancements made in combinatorial chemistry and the drugs obtained from natural resources may have a tremendous impact on drug discovery, and as a result, nearly half of the commercial pharmaceuticals are from the exploration of natural resources (Cragg & Newman, 2013;Maiti, Nagori, & Singh, 2017;Newman & Cragg, 2020;Rastelli, Pellati, Pinzi, & Gamberini, 2020;Rijo & Mori, 2020). Natural resources display a significant chemical diversity in terms of the structure and function of their biomolecules, and they serve as an essential reservoir of bioactive compounds to improve medications by providing unique prototypes. Such compounds can also serve to create new configurations and structural modifications for more powerful and safer medicines (Cragg & Newman, 2013;Maiti et al., 2017;Mukherjee, Bahadur, Harwansh, Biswas, & Banerjee, 2017). Plants are currently envisaged to be the chief source of novel drugs, new compounds, and new chemical molecules (Atanasov et al., 2015;Mukherjee, 2012;Newman & Cragg, 2020). In a recent study, about 85% of the conventional herbal medicines were shown to cater to the drugs of approximately 65% and 80% of the world's populations and developing countries, respectively (Newman & Cragg, 2020).
Recent studies have focused on discovering natural products as active supplements or as replacements for ongoing synthetic antithrombotic drugs (Zhao et al., 2020). The active ingredients of natural products (e.g., extracts of traditional herbs and medicinal plants), the traditional system of medicine used in several countries (e.g., Chinese medicines), the Indian system of ancient medicine (Ayurveda), and functional foods have shown remarkable antithrombotic properties in vitro and in pre-clinical studies (Cordier et al., 2012;Gogoi, Ramani, Bhartari, Chattopadhyay, & Mukherjee, 2019;Li, Liang, & Sun, 2019;Memariani, Moeini, Hamedi, Gorji, & Mozaffarpur, 2018;O'Kennedy, Raederstorff, & Duttaroy, 2017;Sen & Chakraborty, 2017;Yamamoto et al., 2018). The significant advantages of using plant-derived natural products for therapeutic treatment have been realized because several of these compounds (as crude extracts or as partially purified plant extracts) are comprised of multiple components. Each component may target different factors in the coagulation cascade and their synergistic interactions may enhance their therapeutic efficacies (Gogoi et al., ,b, 2020. Moreover, numerous intriguing properties of natural products (e.g., rapid absorption when ingested by the oral route, marginal side-effect(s) in the gastrointestinal tract, and their non-immunogenicity) suggest their strong efficacy and safety (Fuentes & Palomo, 2014).
7. Antithrombotic herbal compounds may modulate a patient's gut microbiota and influence a number of physiological functions.
Some natural antithrombotic compounds may be harmful to the gut microbiota, or even more readily absorbed in the gut (Vamanu & Gatea, 2020), but again, little is known about such interactions.  Table 3. The anticoagulant and antiplatelet mechanism(s) of crude/ partially purified extracts and their bioactive constituents are described below (and shown in Figure 4a,b). In most instances, researchers have not compared the activities of plant extracts or the purified active components to the antithrombotic drugs that are widely prescribed by clinicians. Moreover, the dose-dependent efficacies of several plant extracts/active compounds have not been determined, so this pre-requisite data are still needed before further preclinical or clinical studies can be conducted. The lack of such studies is a significant impediment to the successful clinical application of antithrombotic phytopharmaceuticals. The following sections describe the antithrombotic mechanisms of these phytopharmaceuticals.
The analysis of structure-activity relationships (SARs) has revealed the crucial role played by the hydroxyl group at C-3 in these flavonoid compounds for their thrombin-inhibiting activity, which can be improved by increasing the number of OH groups in the B-ring of the flavonoids (Liu et al., 2010). Generally, the mechanism of thrombin inhibition by plant compounds is achieved by direct inhibition of the catalytic site of thrombin via hydrogen bonding or indirect inhibition by binding to exosite-I (binds to fibrinogen and fibrin) and/or anion-  (Gogoi, Arora, et al., 2018). Such dual inhibitors may be in great demand by the pharmaceutical industry for preparing herbal antithrombotic drugs (Gogoi, Arora, et al., 2018).
Some of the herbal compounds tested under in vitro conditions demonstrated low anticoagulant potency compared to the commercial synthetic direct inhibitors of thrombin (i.e., dabigatran and bivalirudin); however, the preclinical studies suggest that these compounds are safe to administer (Gogoi, Arora et al., 2018. In some instances, the purified active compounds demonstrated less anticoagulant potency than the crude or partially purified plant extract, which suggests that the active components act synergistically to enhance their antithrombotic activity (Gogoi et al., 2019). Clinical studies are still needed to evaluate the therapeutic efficacies of these herbal compounds.

Hyperoside In vitro and in vivo
Prolongs APTT and PT of platelet-poor plasma Inhibits thrombin-and collagen-induced platelet aggregation in vitro, and adenosine diphosphateinduced platelet aggregation in vivo.  Hyperfibrinogenemia also induces the proliferation of lipids in the blood vessel wall, which can lead to atherosclerosis and ischemic pathologies (Singh, Mengi, Xu, Arneja, & Dhalla, 2002). Therefore, reducing the elevated fibrinogen level in plasma is essential for regulating the clinical progression to thrombus formation.
A class of enzymes (proteases) can degrade fibrin, fibrinogen, and/or fibrin and fibrinogen, and are designated as fibrinolytic, fibrinogenolytic, and fibrin(ogen)olytic enzymes, respectively. The plant-derived fibrin(ogeno)lytic enzymes also demonstrate defibrinogenation activity (lowering the fibrinogen content of blood plasma) ( Table 3). The fibrin(ogen)olytic enzymes that have an affinity and can subsequently catalyze αand β-chains of fibrin/fibrinogen are classified as α and/or β fibrinogenases. Some fibrinogenases, purified from medicinal plants, demonstrate preferential hydrolysis of the Aαsubunit of fibrinogen (Choi et al., , 2014Kim et al., 2013), and purified αβ-fibrinogenase from plants has also been reported (Gogoi et al., 2020;Gogoi, Arora, et al., 2018). Besides having fibrin(ogeno) lytic activity, some of these enzymes in vitro have shown the twin inhibition of thrombin and FXa, antiplatelet activity, and the inhibition of in vivo thrombus formation in the mouse tail. They did not show any adverse effects in mice, indicating their safety and therapeutic efficacy (Gogoi et al., 2020;Gogoi, Arora, et al., 2018). Further preclinical and clinical studies are necessary to demonstrate their therapeutic potency.

| Antiplatelet activity inhibiting blood coagulation
Platelet activation is one of the most critical events in initiating the coagulation cascade and maintaining cellular hemostasis (Tomaiuolo, Brass, & Stalker, 2017). Likewise, inhibiting platelet aggregation by crude extracts or active constituents of plants is a crucial mechanism to delay the onset of blood coagulation (Figure 4b). In vitro antiplatelet activity has been shown for several herbal extracts and purified herbal compounds (e.g., hydroxycinnamaldehyde, methoxycinnamaldehyde, coniferaldehyde, eugenol, amygdalactone, and cinnamic alcohol (Figure 6i-   commercially as antiplatelet agents for decades, even with their limitations (Fitzpatrick et al., 1986). The ratio of TXB2 and 6-keto-PGF1 to the stable metabolites of TXA2 and prostaglandin I2 (PGI2), respectively, also regulates thrombus formation. The higher the ratio of TXA2/PGI2, the more thrombus formation, and the lower the ratio of TXA2/PGI2, the less platelet aggregation and thrombus formation, with a greater risk of bleeding (Rucker & Dhamoon, 2021).
In an in vitro study by Chang et al. (2013),

| Clot-busting activity (thrombolysis)
The decisive coagulation and thrombotic cascade are controlled by the catalytic transformation of plasma fibrinogen to fibrin, with the resulting development in a steady fibrin mass (Standeven, Ariëns, & Grant, 2005). Clot-buster drugs (i.e., fibrinolytic drugs) such as the plasmin-like proteases (e.g., nattokinase and lumbrokinase) and F I G U R E 6 Chemical structure of antithrombotic herbal compounds showing antiplatelet activity. The figures were drawn using ChemSketch software. (i) hydroxycinnamaldehyde, (ii) methoxycinnamaldehyde, (iii) coniferaldehyde, (iv) eugenol, (v) amygdalactone, (vi) cinnamic alcohol, and (vii) oligoporin A plasminogen activators (e.g., tissue-type plasminogen activator tPA and streptokinase) dissolve the fibrin clot(s) inside the blood vessels via direct and indirect mechanisms, to restore blood flow in the affected area (Standeven et al., 2005). As shown in Table 3 have been investigated in an experimental animal model (Ritschel, Kastner, Hussain, & Koch, 1990) and in humans (Duchateau et al., 2012). In a clinical study, the absolute oral bioavailability, plasma clearance, distribution volume, and BS turnover were determined to be 0.41%, 85 ml/h, 46 L, and 5.8 mg/day, respectively, in healthy human volunteers (Duchateau et al., 2012). Preclinical studies showed the tissue distribution of BS in the ovaries, adrenal glands, brain, testicles, and skin; however, it is metabolized to different compounds in the liver and other tissues, and the excretion of a significant proportion (80%) of the absorbed BS is reported to occur via feces (reviewed by Bin Sayeed, Karim, Sharmin, & Morshed, 2016). In vivo animal studies have also shown that the bioavailability of BS post oral intake is approximately 9%, and its distribution half-life and terminal distribution half-life is 3 and 129 hr, respectively (Ritschel et al., 1990). In a preclinical model, BS demonstrated an in vivo anticoagulant activity via inhibition of thrombin and inhibition of thrombin-catalyzed platelet aggregation (Gogoi, Pal, et al., 2018). According to the FAO/WHO Expert Committee on Food Additives, the No-Observed-Adverse-Effect-Level for BS is 4,200 mg/kg/HBW/day, though the in vivo oral anticoagulant dose of BS in mice was determined to be 50 mg/kg. BS did not show in vivo defibrinogenation in the plasma of mice (Gogoi, Pal, et al., 2018). This data suggests a high therapeutic index of BS F I G U R E 7 A schematic diagram is showing the antiplatelet mechanism of oligoproin A (Park et al., 2012) that would be suitable for its application in treating thrombosisassociated disorders.

| Borneol
Borneol (C 10 H 18 O), a terpene derivative with a molecular mass of 54.253 g/mol (PubChem CID: 1201518), is widely used to relieve pain, reduce inflammation, and treat CVDs (Figure 8i). Preclinical studies in mice have shown the absolute bioavailability of borneol after intranasal and oral administrations to be 90.68 and 42.99%, respectively though injection of borneol via the parental route was found to have a higher bioavailability, and fast distribution and metabolism compared to oral supplementation (Zhao et al., 2012). Natural borneol was detected in mice brain post 5 min of oral intake with a maximum concentration of 86.52 μg/g . Studies have also shown that intravenous and intranasal administration of borneol in mice resulted in its distribution in blood-supply tissues, particularly in the heart, brain, and kidney, but less in the liver, spleen, and lung (Zhao, Du, Lu, Wu, & Li, 2013). The oral toxicity (LD 50 ) of borneol was found to be 5,800 mg/kg in rats, though no information is available about its effects on reproductive organs, mutagenicity, teratogenicity, and neurotoxicity. Such studies would be essential if borneol is to be developed as an oral natural antithrombotic agent.

| Butein
Butein (C 15 H 12 O 5 ), has a molecular mass of 272.25 g/mol, and is a chalcone of chalconoids, which is represented by an (E)-chalcone moiety that has four supplementary hydroxy substituents at positions 2 0 , 3, 4, and 4 0 (PubChem CID: 5281222) (Figure 8 ii). A pharmacokinetic study in rats showed that 53 and 20% of the total administered dose of butein is excreted in urine and feces, respectively, within 24 hr following administration via the parental and oral routes (Brown & Griffiths, 1983). Butein at an oral dose of 2000 mg/kg was safe in rats and devoid of cytotoxicity against cultured mammalian cells under in vitro conditions (reviewed by Semwal, Semwal, Combrinck, & Viljoen, 2015). The pharmacokinetic parameters of butein (5 mg/kg) post i.v. administration in male Sprague-Dawley rats are shown in Table 4. These pharmacokinetic and antithrombotic properties, determined by preclinical studies are inspiring for developing a phytopharmaceutical-based oral antithrombotic agent following further in-depth clinical studies.  (Touil et al., 2011). The liver rapidly biotransforms the fisetin to sulphates and glucuronides (Shia, Tsai, Kuo, Hou, & Chao, 2009). A nanoemulsion of fisetin was shown to have an improved pharmacokinetic and therapeutic efficiency (Ragelle et al., 2012).
Although rat is considered an excellent animal model for biotransformation and pharmacokinetic studies, clinical trials are necessary to prove the therapeutic efficiency of fisetin as an antithrombotic herbal drug.  (Cao et al., 2009). In another pharmacokinetic study, the GLA concentration in Sprague-Dawley rats post intravenous or oral administration was determined .

| Glaucocalyxin A
In addition, a preclinical study showed that the GLA concentration reached the maximum concentration in rat plasma at approximately 0.71-0.75 hr post-oral feeding of Rabdosia japonica, which was rapidly eliminated from the plasma (t1/2 = 1.1 hr). GLA could still be detected in rat plasma post 6 hr by liquid chromatography-mass spectrometry (LC-MS)/MS analysis (Huang, Guan, & Lv, 2018). Further, by the UHPLC-MS/MS method, GLA was shown to infiltrate the blood-brain barrier in rats, though the concentrations of GLA in rat brain and lung tissues were substantially lower compared to the plasma concentrations at the same dose. A comparison showed that the Cmax of GLA in plasma was 1.4-fold and 2.6-fold greater than the Cmax of GLA in lung and brain tissue, respectively, and the mean tissue:plasma ratio of GLA (AUC (0-t, tissue)/AUC (0-t, plasma) was 0.74 and 0.47, for lungs and brain, respectively (Deng, Liu, He, . In vivo studies in rats showed that GLA was biotransformed to 32 phase I metabolites with different structures and 6 phase II metabolites comprising 25, 18, 17, and 7 structures, in rat urine, feces, and bile plasma, respectively (Sun et al., 2020). The preclinical study data support further clinical trials of GLA as a potential oral antithrombotic agent.
F I G U R E 8 Structure of antithrombotic herbal drug prototypes whose pharmacokinetics and pharmacodynamics properties were determined. The figures were drawn using ChemSketch software.
(i) borneol, (ii) butein, (iii) fisetin, (iv) glaucocalyxin A, (v) hyperoside, (h ng/ml) and 8.8 ± 1.3 hr, respectively (Yin et al., 2013). The tissue distribution of HP post-oral feeding in rats showed a rapid and extensive distribution throughout the whole body. The maximum distribution of HP occurred in the stomach, and it was followed by the kidney at 0.5 hr post-oral administration, which was correlated to the metabolism in the liver and clearance by the kidneys . A toxicity study of HP post-6-month oral administration was followed by a 1-month recovery period in Wistar rats, which showed kidney damage. The damage was reversible after withdrawal of HP treatment, which suggests the importance of regular monitoring of kidney function and hematological parameters of patients undergoing HP treatment (Ai, Huang, Wang, & Zhang, 2012).

| Polyphenols
Polyphenols (e.g., flavonoids, tannic acid, and ellagitannin) represent nation phase (t1/2 gamma, 7.5 hr) (Hao et al., 2006). By LC-MS/ MS analysis, the tissue distribution of TS was found in the following tissues, in descending order: stomach > small intestine > lung > liver > fat > muscle > kidneys > spleen > heart >plasma > brain > testes; however, most of the rat tissues post 20 hr oral administration at a dose of 60 mg/kg showed the presence of TS (Bi et al., 2007). Approximately 99.2% of TS is bound to plasma proteins in plasma, including 77.5% lipoprotein (Hao et al., 2006). TS's low aqueous solubility and partial membrane permeability resulted in its poor absorption and low bioavailability (<3.5%). Nevertheless, at a concentration of 6 μM, TS demonstrated pericardial edema, spinal curvature, and missing tails in zebrafish embryos, indicating its potential cardiotoxicity at a higher dose . In vivo dose optimization and clinical studies of TS are warranted to understand its potential as an antithrombotic cardiovascular drug.

| A PROPOSED ROADMAP FOR THE DEVELOPMENT AND COMMERCIALIZATION OF ANTITHROMBOTIC HERBAL DRUGS
Natural compounds are often considered to be the primary sources of novel drug candidates, and the idea that "natural is safe" has prevailed in society; however, for the reasons discussed above, the commercialization and mainstream medical use of herbal products as potent, safe medicines for treating or managing thrombosis-associated CVDs requires proper strategic planning. The production and quality assess- A herb or its parts must be carefully selected for the production of an antithrombotic drug, preferably based on monographs of traditional medicinal plants prepared by the WHO or from a list of herbal plants from the country of interest. The choice of herb for preparing an antithrombotic drug must be done cautiously to avoid using restricted or endangered species from a particular region.
The practice of collecting a herb should include authenticating the starting material to avoid adulterations and identifying the geographical locale, specific harvesting time, and season(s) to obtain the optimal quality of the active antithrombotic compound(s). To accurately identify herbs and plant parts (roots, barks, and powders) that are sold commercially, modern molecular biology approaches (i.e., DNA barcoding of different genomes (Jiang et al., 2018) and high-resolution melting should be used (Yu et al., 2021)).
Metabarcoding technology that uses next-generation sequencing is a highly proficient method that could be used to identify mixed samples.
The purification and molecular mass determination of bioactive components from herbal plants could help to elucidate their antithrombotic activity, mechanism of action, and chemical structure.
Importantly, the efficacies of herbal drugs should be compared to those from drugs that are already available on the market, in preclinical studies. Ligand-based, computer-aided drug designing techniques (in silico analysis) and target-based drug discovery (i.e., reversed pharmacology) should be used to explore the active compounds from herbal plants as potential inhibitors of the coagulation cascade (e.g., antithrombin or -FXa or other coagulation factors) (Ibrahim et al., 2020). Network pharmacology that integrates pharmacology and information technologies (i.e., bioinformatics, system biology, and high-throughput histology) are new and promising research strategies that could be used for evaluating traditional herbal medicines (Hopkins, 2007;Zhou et al., 2020). Similar computer-based approaches that take advantage of high-throughput screening could be used with different herbs to discover next-generation antithrombotic drugs .
The absorption, distribution, metabolism, excretion, and toxicity properties of bioactive components from plants with antithrombotic activities can also be predicted using the pkCSM online server (http:// biosig.unimelb.edu.au/pkcsm) (Pires, Blundell, & Ascher, 2015) or similar software programs. The drug properties can be used to reveal possible therapeutic applications of a given antithrombotic herbal drug. In particular, Lipinski's "rule of five" is an ideal tool for assessing the connection between predicted structures and drug-like properties (Lipinski, Lombardo, Dominy, & Feeney, 1997) and any predicted structure of an antithrombotic herbal drug/compound should not violate this rule. Predicted inhibitors can be chemically synthesized or isolated from the medicinal plant (natural inhibitor), and their antithrombotic potencies determined in vitro before being evaluated under in vivo conditions (i.e., the classic or forward pharmacological approach).
Machine learning is another advanced tool that can predict the biochemical properties and antithrombotic effects of plant-derived natural products (Jeon, Kang, & Kim, 2021). Machine learning can also predict the toxicity and drug-drug interactions of novel antithrombotic molecules solely based on their molecular structures (Jeon et al., 2021).
T A B L E 5 Pharmacokinetic parameters of sulphuretin and its conjugated metabolites in rat plasma determined by LC-MS/MS analysis (Jin et al., 2015) Pharmacological (i) Regional documentation on the use of ethnomedicines and herbal remedies as antithrombotic drugs will be available. This strategy will safeguard the interest and traditional knowledge of the indigenous people and communities from exploitation.
(ii) A list of such plants in different geographical locales of the country will be made available to each country's national regulatory agency to ensure the exploitation and habitat destruction of these plants.
(iii) A country-specific monograph of antithrombotic herbal medicinal plants, including the list of restricted or endangered species of herbs, should be made available to the World Health Organization.
(iv) This action will encourage the proper selection of herb/plant or parts for antithrombotic drug production, preferably from the monograph of traditional medicinal plants available in each country.
(v) This effort will lead to modern molecular biological approaches for good collection practice, including authenticating starting material (herb) to avoid adulteration.
(vi) Purification of active constituents from plants. Characterization of chemical and biochemical properties of active constituents for discovering novel antithrombotic compounds from plants and assessment of their potency as compared to their competitors in the market.
(vii) The ligand-based computer-aided drug designing (in silico analysis) or targetbased drug discovery program will be encouraged.
(viii) It will augment the research on network pharmacology that integrates pharmacology and information technology (bioinformatics, system biology, and high-throughput histology).
(ix) Using online software, this strategy will help predict adsorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of antithrombotic phytomedicines.
(x) This strategy will also help apply machine learning to predict biochemical properties, including antithrombotic effects, toxicity, drug-drug interaction, and plant-derived natural products, to shed light on their possible therapeutic application as a cardiovascular drug.
2. Augmentation of the pre-clinical and clinical research on antithrombotic herbal drug prototypes for translation to a therapeutic agent ready for commercialization.
(i) This effort will help explore in vivo efficacy, mechanism of antithrombotic action, safety, storage stability, and therapeutic index of plant-derived natural compounds, preferably in a GLP-compliance laboratory for their dose optimization.
(ii) Synergistic interaction of two or more herbal compounds to significantly enhance their antithrombotic activity in vivo conditions to develop a more potent and effective drug will be explored.
(ii) Determination of pharmacodynamics parameters of antithrombotic herbal drugs in rodent models to analyze their ADEMT properties and the potency of the herbal drugs with contender synthetic drugs will be known.
(iv) Understanding the risk: benefit ratio will attract the pharmacological companies' interest to take forward the lead molecule(s) in the following stages of drug development.
(v) Knowledge on the interaction of herbal drugs with Western medicines and food components, including their mechanism(s) of interaction, significances, and severity of such interactions, will be beneficial to determine their safety postadministration.
(vi) Knowledge of the influence of human gut microbiota on the bioavailability and bioactivity of antithrombotic herbal compounds will help design effective ways(s) of oral delivery of antithrombotic herbal drugs.
3. Advancement of globalization and commercialization of antithrombotic herbal medicine.
(i) Good manufacturing practice (GMP) of cardiovascular herbal drugs will be boosted to maintain customers' and clinicians' quality and acceptance.
(ii) Phytochemical markers-based quality control and quality assurance (QA and QC) of herbal antithrombotic drugs will ascertain the quality and prevent batch-tobatch variation which will augment their acceptability and enhance the commercialization of antithrombotic herbal medicines in Western countries. Strategies Impact (iii) Setting QA and QC laboratories by government, private industries, or publicprivate partnership (PPP) programs will enhance drugs' trust and acceptability and provide employability to technicians and scientists.
(iv) The tremendous increase in physicians' choice to prescribe the phytochemical marker-assisted quality assured, formulated herbal drug composition.
(iv) The above steps would help toward international cooperation on the use and proposition of antithrombotic herbal drugs fulfilling all the regulatory compliances of each nation.
4. Conservation and cultivation strategies of medicinal plants demonstrating antithrombotic activity.
(i) A well-planned national policy on the conservation of medicinal plants will emerge to prevent the extinction of plants.
(ii) Prevent the commercialization of antithrombotic herbal drugs prepared from wildly grown plants/herbs.
(iii) Establishment of new research centers. Launching of the new programs on in situ and ex situ conservation of medicinal plants.
(iv) Training of local farmers on scientific ways of cultivation and organic farming and sustainable use of medicinal plants will tremendously increase the production of antithrombotic herbs and plants and will be a source of income to farmers.
(v) Enhanced production and cultivation of antithrombotic medicinal plants will also boost the establishment of herbal-drug-based bio-industrial sectors for that region's income generation and economic development.
F I G U R E 9 A roadmap is proposed to show the developing and improved globalization of antithrombotic herbal medicine discovered from indigenous herbs or plants. From the safety assessments of herbal drug prototypes, pharmacological companies may be encouraged to further explore the commercialization of herbal compounds and move them along the drug development process.

The interactions between herbal antithrombotic drugs and
Western medicines could lead to unwanted side-effects (e.g., bleeding complications), and consequently, rigorous evaluation of the literature and pharmacological or clinical reports is needed to understand the mechanism(s) of interaction, and the significance and severity of any interactions between allopathic (Western) anticoagulant/antiplatelet drugs and herbal antithrombotic drugs (Tsai et al., 2013;Zuo et al., 2020). Because the antithrombotic efficiency of any formulation from two or more purified natural products could surpass the treatment efficacy of the individual components, potential herb--drug interactions need to be carefully scrutinized (Zuo et al., 2020).
The bioavailability and bioactivity of some natural compounds can be influenced by the patient's gut microbiota, and some probiotic strains can enhance intestinal and colon absorption and bioavailability (Vamanu & Gatea, 2020). This area of research is particularly active to understand how herbal compounds can have enhanced oral bioavailability and avoid being degraded in the colon, leading to improved clinical efficacy.
6.3 | Good manufacturing practices, quality assurance, and international cooperation can enhance the commercialization and globalization of antithrombotic herbal drugs Good manufacturing practices (GMP) for cardiovascular herbal drugs must be used to maintain their quality and gain the acceptance of consumers and clinicians. GMP can be monitored by analyzing the quality of herbal drugs, to further support their broad acceptance and commercialization. The active constituent(s) of herbal antithrombotic drugs can be monitored and accurately quantified with phytochemical markers that use biochemical and biophysical techniques (i.e., biochemical assays, molecular mass determinations by gas-chromatography-MS (GC-MS) and high-resolution MS, and purity determinations by RP-HPLC and MS). Such combination analyses are known as analytical fingerprinting techniques, which can involve both targeted and untargeted approaches. The result is better quality control (quality assurance) and less chance of batch-to-batch variations in the product quality (Kharbach, Marmouzi, El Jemli, Bouklouze, & Vander, 2020).
Large-scale herbal drug manufacturers should also establish in- Finally, the optimal production of natural components can be facilitated by enhancing the growth of herbal plants and encouraging organic farming practices to produce medicinal plants with higher biomass yields (Chen et al., 2016).

| CONCLUSION AND FUTURE PROSPECTS
Recent evidence on the therapeutic value of antithrombotic herbal drugs has led to envisaging their great promise as an effective natural therapy for treating thrombosis complications in cardiovascular and COVID-19 diseases, though several drawbacks associated with the production and quality assurance of antithrombotic herbal drugs need to be adequately addressed. One of the most critical issues is applying the scientific parameters for assessing pharmaceuticals to herbal products. Consumers often use herbal products based on the herbalism approach (using only herbal components) rather than using the phytotherapy approach (using standardized herbal products). Accordingly, the pharmacological efforts to produce herbal products must first be evidence-based before the products are accepted by clinicians.
The challenging tasks can be achieved by herbal drug manufacturers by following standard guidelines for production and quality assurance, but without QA and QC certification, these drugs should not be sold on the market. Rigorous studies are needed to establish the clinical efficacy of plant-derived compounds, and indepth studies must be carried out to elucidate the antithrombotic mechanism of herbal drugs, and their safety and potency, compared to the properties of Western medicines. International cooperation for the standardized use and quality assurance of therapeutic herbal drugs must also be supported by research scientists, governments, and NGOs. Modern research methods, involving network pharmacology and machine learning will likely lead to the discovery and formulation of new antithrombotic herbal drugs, and specific herbal components with potent antithrombotic activity. Great care must also be taken to prevent the overexploitation or habitat destruction of possibly rare plants and their sustainable use and cultivation must be facilitated by specific farming practices.
By encouraging academic-industry partnerships, fundamental research on antithrombotic drug discovery may be translated into improved clinical products. The PPP model can link academicians, clinicians, research scientists, and governmental organizations with pharmacological companies to procure additional resources and escalate the discovery and development of next-generation natural antithrombotic herbal drugs. By improving the production quality of herbal medicines, products can be sustainable, clinically efficacious, and less costly, ultimately reducing the global burden of CVDs and mortality associated with COVID-19.

ACKNOWLEDGMENTS
The authors thank Ms. U. Puzari for drawing the chemical structure of antithrombotic molecules, Ms. Bhabna Das for technical help in preparing the manuscript, and Glen Wheeler, Canda for editing the manuscript. AKM received financial assistance from the core fund of the IASST, Guwahati.

CONFLICT OF INTEREST
The authors have no conflicts of interest to declare.

AUTHOR CONTRIBUTIONS
Ashis K. Mukherjee concenived the idea, wrote and revised the manuscript, and Dhruba J. Chattopadhyay reviewed and edited the manuscript.