Investigation on the Essential Oils of the Achillea Species: From Chemical Analysis to the In Silico Uptake against SARS-CoV-2 Main Protease

In this study, phytochemicals extracted from three different Achillea genera were identified and analyzed to be screened for their interactions with the SARS-CoV-2 main protease. In particular, the antiviral potential of these natural products against the SARS-CoV-2 main protease was investigated, as was their effectiveness against the SARS-CoV-1 main protease as a standard (due to its high similarity with SARS-CoV-2). These enzymes play key roles in the proliferation of viral strains in the human cytological domain. GC-MS analysis was used to identify the essential oils of the Achillea species. Chemi-informatics tools, such as AutoDock 4.2.6, SwissADME, ProTox-II, and LigPlot, were used to investigate the action of the pharmacoactive compounds against the main proteases of SARS-CoV-1 and SARS-CoV-2. Based on the binding energies of kessanyl acetate, chavibetol (m-eugenol), farnesol, and 7-epi-β-eudesmol were localized at the active site of the coronaviruses. Furthermore, these molecules, through hydrogen bonding with the amino acid residues of the active sites of viral proteins, were found to block the progression of SARS-CoV-2. Screening and computer analysis provided us with the opportunity to consider these molecules for further preclinical studies. Furthermore, considering their low toxicity, the data may pave the way for new in vitro and in vivo research on these natural inhibitors of the main SARS-CoV-2 protease.


Introduction
Iran is home to a significant share of plant species and endless natural habitats characterized by numerous special plants and centers of local endemic species [1]. A. millefolium, with the common name yarrow, is a member of the Asteraceae family and is a flowering plant [2] comprising 130 species, 19 of which are grow wild in Iran [1]. As far as the Achillea genus has existed across the world (from Asia and Europe to North America), it has been used in ancient life in traditional or folk medicine. Bumadaran is a common name for some species of Achillea in Persian culture [3].
The genus Achillea has a long history of use in traditional medicine as an antiinflammatory, diaphoretic, anti-spasmodic, tonic, diuretic, and emmenagogic agent [3], and it has been used as a natural remedy (in Iranian traditional medicine) for the treatment of bleeding, headaches, respiratory infections, inflammation, spasmodic diseases, flatulence, dyspepsia [4,5], pneumonia, hemorrhaging, rheumatic pain, and wounds, and it is useful for liver disease and acts as a mild sedative [6]. Currently, the different medicinal functions of yarrow, such as its use as a spasmolytic, a choleretic, a wound-healing treatment, and an Life 2023, 13, 378 3 of 14 patients (≥12 years of age) suffering from mild to moderate COVID-19 who are at high risk of progression to severe COVID-19 and/or hospitalization. However, this authorization was revoked on April 16, 2021 owing to the resistance of variants to bamlanivimab [44]. Remdesivir has been authorized for use in pediatric and adult patients (≥12 years of age) that require hospitalization, though it requires administration by injection or infusion in a therapeutic setting, with frequent monitoring [44,45]. In December 2021, two new oral antiviral agents, molnupiravir and nirmatrelvir/ritonavir (Paxlovid), gained emergency use authorization from the FDA for adult patients with mild to moderate COVID-19 who were at high risk of progression to severe COVID-19 under certain limitations [44,46,47]. Tocilizumab [48,49] and baricitinib [50,51] (both anti-inflammatory agents) have also gained EAU from the FDA for people of ≥ 18 years of age with COVID-19 who were admitted to a hospital [52].

Samples (Plant Materials)
The plants (three species of the Achillea genus: A. millefolium, A. wilhelmsii C. Koch, and A. Tenuifolia Lam) were collected during June and July 2021 from their natural habitat in the village of Homayoun in Zanjan province, Iran (elevation 1868 m, latitude east 48°28′35.36″, longitude 36°45′5.28″). A voucher specimen was maintained at the Research Institute of Modern Biological Techniques (RIMBT), University of Zanjan. Mr. Angourani deposited the botany herbarium voucher specimens under the following numbers: A. millefolium, number 1518; A. tenuifolia Lam, number 1522; and A. wilhelmsii C. Koch, number 1528. All samples were prepared from the whole flowers and leaves of the plants that were harvested at the full bloom stage of growth and then air-dried at room temperature in the shade. The most promising drug target is the main protease of SARS-CoV-2 (M pro or 3CL pro ) because it plays a vital role in viral replication and transcription. Thus, the main axis of recent research has focused on the rapid development of SARS-CoV-2 M pro inhibitors as drug candidates. Several techniques have been used for the discovery of SARS-CoV-2 M pro inhibitors, including the high-throughput screening of structurally diverse compound libraries [53,54], drug repurposing [55,56], structure-based drug design [57,58], and in silico studies [59,60]. Despite considerable efforts, only one compound, nirmatrelvir/ritonavir (Paxlovid), has gained EUA by the FDA for adult patients with mild to moderate COVID-19 who are at a high risk of hospitalization [46], and a new compound (S-217622) is undergoing clinical trials (NCT05305547) [59]. Owing to their long history of folk use in Iranian traditional medicine and because of their low toxicity and excellent pharmacokinetics, the Achillea genera, alongside the reported biological activities of three Achillea species (A. millefolium, A. wilhelmsii C. Koch, and A. tenuifolia Lam), were the subject of this research. We wanted to address the question of whether the extracted compounds from three different EOs of the Achillea species can be the inhibitors, or have the potency to be a possible source of strong and/or effective inhibitors, of the SARS-CoV-2 3CL protease. The antiviral potential of the extracted compounds (from their EOs) against SARS-CoV-2 3CL pro (and SARS-CoV-1 3CL pro, as a standard) was investigated for the first time. Initially, the EOs of A. millefolium, A. wilhelmsii C. Koch, and A. tenuifolia Lam were obtained and their compounds were identified using GC-MS. The identified compounds were selected to investigate their antiviral potential against viral targets using AutoDcock 4.2.6. These compounds also showed low toxicity and very good pharmacokinetic properties in the SwissADME and ProTox-II evaluations. The compounds with the lowest binding energies (with the viral targets) were the best antiviral candidates against the viral protease and are worthy of further in vitro and in vivo analyses to clarify their antiviral potential.

Samples (Plant Materials)
The plants (three species of the Achillea genus: A. millefolium, A. wilhelmsii C. Koch, and A. Tenuifolia Lam) were collected during June and July 2021 from their natural habitat in the village of Homayoun in Zanjan province, Iran (elevation 1868 m, latitude east 48 • 28 35.36 , longitude 36 • 45 5.28 ). A voucher specimen was maintained at the Research Institute of Modern Biological Techniques (RIMBT), University of Zanjan. Mr. Angourani deposited the botany herbarium voucher specimens under the following numbers: A. millefolium, number 1518; A. tenuifolia Lam, number 1522; and A. wilhelmsii C. Koch, number 1528. All samples were prepared from the whole flowers and leaves of the plants that were harvested at the full bloom stage of growth and then air-dried at room temperature in the shade.

Essential Oil Extraction
The essential oils from the flowers and leaves of the plants (A. millefolium, A. wilhelmsii C. Koch, and A. tenuifolia Lam) were extracted by hydrodistillation for approximately 4 h using a Clevenger apparatus in accordance with the British Pharmacopeia (1988) [61]. The finely cut plant material was fully submerged in distilled water (600 mL) in a round-bottom flask, which was then placed in a heating mantle. The condensed vapors were collected in a separating funnel and the resulting EO was separated, dehydrated with an appropriate amount of Na 2 SO 4 , filtered, and stored in brown vials at 4 • C.

GC-MS Analysis of the Essential Oil Components
The EOs were dehydrated with Na 2 SO 4 , and the samples were injected into the device under the thermal planning of the column. The compounds of the EOs were identified using the retention time (R.T) and the Kovats index (R.I) and by studying the mass spectra and checking these parameters with the standard compounds in the GC/MS library.
The EO extraction yields were estimated for 100 g of fresh plant material. The EOs were analyzed using gas chromatography-mass spectrometry (GC-MS). This device includes a gas chromatograph (model 7890B) and mass spectrometer (model 5977A) made by Agilent Company of America that was equipped with a split/splitless injection system and an ion bombardment ionization model (possessing the NIST and WILEY mass libraries). To analyze the fatty acids, an HP5-MS column with a length of 60 m, an inner diameter of 0.25 mm, and a thickness of 0.25 µm was used. The injection site, interface, and ionization site temperatures were set to 280, 290, and 250 • C, respectively. The temperature program of the column was triggered at an initial temperature of 60 • C, which remained constant for 5 min. The temperature reached 180 • C (with a slope of 15 • C/min) for 2 min before increasing to 280 • C (with a slope of 20 • C/ min), and it remained at this temperature for 10 min. The split ratio was set to 1 to 20 and the injection volume was half of a microliter.

Molecular Docking
The molecular and 3D structures of the targets (SARS-CoV-2 3CL pro : PDB code 6LU7 and SARS-CoV-1 3CL pro : PDB code 2H2Z) were extracted from the Protein Data Bank (www.rcsb.org (accessed on 1 January 2022)). To prepare the molecular structures of the natural compounds, their 2D chemical conformations were first drawn using the ChemSketch tool of the ACD/LAB package (www.acdlabs.com (accessed on 1 December 2022)) before transferring them to the Avogadro package to minimize energy and optimize their conformation via a steep algorithm.

Docking Studies
Molecular docking is a highly important sub-technique of molecular modelling that plays a significant role in computer-based drug design. The docking protocol was divided into two general stages using the AutoDock Autogrid package. Initially, blind docking was performed for all viral targets. Structures that bound to the active site of proteases with binding energies of >5.5 kcal/mol were chosen for the second stage (targeted docking). First, gastiger charges and polar hydrogen atoms were added to all ligands using the MGLtools package [62]. All bonds were set as active for ligands, and energetic maps were estimated for each atom type using Autogrid 4. For blind docking, the search space was set as large as all proteins were accessible for ligand binding. The active sites of the viral targets were set as search spaces for targeted docking. Finally, 250 runs of molecular docking according to the Lamarckian genetic algorithm were performed for ligands that met the conditions of the first step of the docking protocol [63]. Discovery studio 4.5 was used for the 3D visualizations of the results, and LigPlot + V.2.2 was applied for the 2D presentations.

Achillea millefolium (Yarrow)
The essential oils of the flowering and leafy branches of the dried vegetative body of A. millefolium were obtained via water distillation and a Clevenger apparatus, and the oil yield was 0.74% (content (% v/w)). The GC-MS results showed that the EO of the plant was composed of 105 compounds, of which 18 compounds accounted for 60.5% of the essential oil (Table 1 and Figure 2).

A. wilhelmsii C. Koch
The essential oil of A. wilhelmsii C. Koch was yellow in color, with a yield of 0.89% (% v/w). The GC-MS results revealed that the essential oil of this plant in the target area comprised 106 substances, of which 21 compounds represented 66.93% of the total essential oil (Table 2 and Figure 3).

A. tenuifolia Lam
According to the GC-MS data, the essential oil of this plant contained 88 compounds. The essential oil of A. tenuifolia Lam was orange-yellow in color, with a yield of 0.82% (% v/w). The main chemical composition of the desert yarrow essential oil is presented in Table 3 and Figure 4. We identified 21 compounds that constituted 73.48% of the essential oil.

A. tenuifolia Lam
According to the GC-MS data, the essential oil of this plant contained 88 compounds. The essential oil of A. tenuifolia Lam was orange-yellow in color, with a yield of 0.82% (% v/w). The main chemical composition of the desert yarrow essential oil is presented in Table 3 and Figure 4. We identified 21 compounds that constituted 73.48% of the essential oil.

Docking Studies of SARS-CoV-2 3CL pro
Molecular docking is a potent approach used to elucidate protein and small-molecule interactions [24,25,64]. Among the identified compounds from the EOs of the three different types of yarrow tested against SARS-CoV-2 3CL pro , four natural inhibitors (Kessanyl acetate, Chavibetol (m-Eugenol), Farnesol, and 7-epi-β-Eudesmol) extracted from the EO of A. millefolium exhibited strong inhibitory effects against the viral target ( Table 4). The binding energies of the proteases are listed in Table 4. It was clear that 3CL pro (or the main protease) catalyzed the most important maturation cleavage phenomenon and also played Life 2023, 13, 378 8 of 14 a significant role in the viral replication of SARS-CoV-2, signifying its importance as a drug target [65].

Docking Studies of SARS-CoV-2 3CL pro
Molecular docking is a potent approach used to elucidate protein and small-molecule interactions [24,25,64]. Among the identified compounds from the EOs of the three different types of yarrow tested against SARS-CoV-2 3CL pro , four natural inhibitors (Kessanyl acetate, Chavibetol (m-Eugenol), Farnesol, and 7-epi-β-Eudesmol) extracted from the EO of A. millefolium exhibited strong inhibitory effects against the viral target ( Table 4). The binding energies of the proteases are listed in Table 4. It was clear that 3CL pro (or the main protease) catalyzed the most important maturation cleavage phenomenon and also played a significant role in the viral replication of SARS-CoV-2, signifying its importance as a drug target [65].  As shown in Figure 5A (for the 2D version, see Figure S4), kessanyl acetate formed three H-bonds with the viral target: two with Ser 144 and one with the important residue Cys 145 (located in domain II and acting as a nucleophile in the initial step of the hydrolysis reaction in the catalytic region of the enzyme with the assistance of His 41 as a base catalyst) [65]; therefore, the interaction with this residue may have remarkably caused a malfunction in its enzymatic activities. Moreover, the hydrophobic interactions between kessanyl acetate and the important residues located in domains I and II [65], including His 41, Met 49, Leu 141, Asn 142, Gly 143, and Gln 189, can ratify the potent inhibitory effect of this natural product on the active site of the enzyme.
Cys 145 is located in the active site positioned in the oxyanion loop (residues 138-146), which, along with Gly 143, defines an oxyanion hole (via interactions between their amide groups and the carbonyl group of peptides), and, together with the β-strand segment (His163-Pro168), they are highly important for the preparation of active sites and substrate binding. Furthermore, residues near the loop segment play a significant role in expanding the active site [65]. Based on this information, it is worth pointing out that chavibetol (m-eugenol) binds to the active site of 3CL pro through five strong hydrogen bonds with the residues Leu 141, Gly 143, Ser 144, and His 163, showing that this natural inhibitor may have high impairment and disruptive effects on the active site, and it may potentially disturb active site formation and accurate substrate binding ( Figure 5B).
These disruptive effects on the active site formation and substrate binding of 3CLpro can be seen for the other two remaining natural inhibitors (Farnesol and 7-epi-β-Eudesmol) because each of them form H-bonds with Leu 141 and Ser 144, although the former shows an additional H-bond with His 163. Interestingly, both had hydrophobic interactions with Cys 145 and His 41, showing their strong antiviral effects on the catalytic function of this enzyme ( Figure 5C,D). Taken together, these results show that the binding of the four natural inhibitors (extracted from A. millefolium) to the active site of 3CL pro may disrupt substrate binding and active site formation, thereby impeding catalytic functions. Cys 145 (located in domain II and acting as a nucleophile in the initial step of the hydrolysis reaction in the catalytic region of the enzyme with the assistance of His 41 as a base catalyst) [65]; therefore, the interaction with this residue may have remarkably caused a malfunction in its enzymatic activities. Moreover, the hydrophobic interactions between kessanyl acetate and the important residues located in domains I and II [65], including His 41, Met 49, Leu 141, Asn 142, Gly 143, and Gln 189, can ratify the potent inhibitory effect of this natural product on the active site of the enzyme.

Docking Studies of SARS-CoV-1 3CL pro
To investigate the potency of the extracted natural products against other major proteases, we selected SARS-CoV-1 3CL pro by using the same docking procedure. Based on the docking data, it could be asserted that four of the tested ligands showed higher binding energies (as they did against SARS-CoV-2 3CL pro ). The 3CL proteases of SARS-CoV-1 and SARS-CoV-2 share approximately 96% sequence identity, indicating that they are similar to each other [66], and SARS-CoV-1 3CL pro also possesses the three main domains, I, II, and III, and its catalytic dyad is composed of His 41 and Cys 145, in which its substrate-binding pocket is located in the cleft between I and II [67]. Interestingly, kessanyl acetate binds to this viral receptor through three H-bonds with Gly 143, Ser144, and Cys 145, which, in turn, may hinder the catalytic site and cause strong disorganization in the enzymatic function of the protease. In addition, hydrophobic interactions with other residues located in the active site (Leu 141 and Asn 142) can prove its high preventive effects on the catalytic site ( Figure 5E).
As shown in Figure 5F, all interacting residues were located at the active site. Moreover, Chavibetol (m-eugenol) formed five strong hydrogen bonds with amino acid residues located in the catalytic region of domain II (Gly 143, Gly 143, Ser144, and His 163), and the active site entry may have been impaired [67], and the catalytic activities of such enzymes can be remarkably diminished. With regard to the interaction of SARS-CoV-2 3CL pro with Farnesol and 7-epi-β-Eudesmol, it can be mentioned that both have formed H-bonds with Leu 141 and Ser 144, along with hydrophobic interactions with the significant catalytic dyad residues His 41 and Cys 145, indicating that these two natural inhibitors may possess highly potent activity for hindering the catalytic region of the protease, and they can potentially disrupt its enzymatic function ( Figure 5G,H). Thus, these four natural inhibitors may have strong inhibitory effects on the active site of SARS-CoV-1 3CL pro and could have a preventive and momentous disruptive impact on its catalytic function.
The four selected natural products were pharmacokinetically investigated using Swis-sADME. Interestingly, according to Tables 5 and 6, all four natural inhibitors satisfied the drug-likeliness properties, indicating a high bioavailability score. Remarkably, none of them served as P-glycoprotein (P-gp) substrates, showing excellent oral availability of these ligands and indicating that they have no inhibitory or inductive effect on metabolizing enzymes [68,69]. More importantly, all showed high gastrointestinal absorption and could cross the blood-brain barrier, indicating that passing the cell membrane is vital for the components to enter the cells infected with SARS-CoV-2 and obtain successful access to the viral 3CL pro of SARS-CoV-2. Consistent with our results, it should be noted that kessanyl acetate and farnesol possessed better pharmacokinetic characteristics, such as pharmacological and drug-likeliness properties. Although Kessanyl acetate did not interact with other medications, Farnesol and Chavibetol (m-eugenol) showed inhibitory effects on CYP1A2, which can be anticipated to interact with the metabolism of some typical anxiety or dispersive disorder drugs. Considering all of these properties, it can be postulated that these four naturally occurring compounds may be effectively absorbed, distributed, and diffused within the body. Finally, the toxicities of the four compounds were predicted using the ProTox-II online tool. As shown in Table 6, all compounds exhibited modest or light toxicity, and Farnesol and Kessanyl acetate showed higher LD50 values. Based on the results obtained in this study, it can be asserted that all four natural compounds (particularly kessanyl acetate and farnesol) may be the most desirable candidates for further in vitro and in vivo investigations of SARS-CoV-2.  ), farnesol, and 7-epi-β-Eudesmol) showed the highest binding energies and the maximum number in the clusters, revealing a high potential to demolish the active site of the proteases and ruin their catalytic functions by forming H-bonds with two important amino acid residues of CYS 145 and HIS 41. Furthermore, the ADME properties of the final compounds were determined to evaluate their pharmacokinetic characteristics. Our results revealed that kessanyl acetate and farnesol are the most suitable natural compounds with regard to drug pharmacokinetics. Based on the findings of the present study, it should be mentioned that kessanyl acetate and farnesol are more potent inhibitors of 3CL-proteases than the other studied natural compounds, and they may be applied in future experimental studies to identify effective anti-SARS-CoV-2 compounds.
Supplementary Materials: The following supporting information can be downloaded from: https: //www.mdpi.com/article/10.3390/life13020378/s1, Figure S1, Chromatogram of the GC-MS analysis of A. millefolium essential oil from the Zanjan region; Figure S2, Chromatogram of the GC-MS analysis of the essential oil of A. wilhelmsii C. Koch from the Zanjan region; Figure S3, Chromatogram of the GC-MS analysis of essential oil of the vegetative body of A. tenuifolia Lam from the Zanjan region; Figure S4, A 2D illustration of the interactions between the candidate compounds and viral polymerases (I: SARS-CoV-2 3CL pro : 6NUR, and II: SARS-CoV-2 3CL pro : 2H2Z) ((a) kessanyl acetate,