Introduction

The influenza A (H1N1) virus causes respiratory infections ranging from acute stage to fatal disease [1]. The neuraminidase (NA) enzyme of the H1N1 virus is a potential target to develop antiviral drugs. NA is an enzyme; it cleaves sialic acid groups from glycoproteins, which is required for flu viral infection replication [2,3,4]. So far, three neuraminidase inhibitors have been approved for the treatment of flu infection, namely, zanamivir, oseltamivir and peramivir [5]. Most researchers are developing plant-derived drugs for treating the influenza virus, as they have potential action against NA [6]. Liquiritigenin (LGN) is a flavanone derivative drug extracted from the Glycyrrhiza uralensis plant [7, 8]. In vitro studies have shown that LGN exhibits anti-H1N1 viral properties by inhibiting NA [9].

The bioactive nature of a drug molecule can be determined from the structural characterization and reactivity in the biological environment. In recent years, vibrational spectroscopic techniques such as FT-IR and FT-Raman have been used to probe the structure, bonding nature and biological activity of drug molecules [10]. Quantum chemical calculation based on density functional theory (DFT) is a significant research tool in modeling of structures, geometries and spectroscopic (vibrational and electronic) properties [11]. These studies are very useful for recognizing the active regions of the drug in the binding sites. The absorption, distribution, metabolism, excretion and toxicity (ADMET) properties of drug-like compounds combined with the pharmacology is very essential to drug development and drug safety [12, 13]. In modern days, the theoretical calculations for ADMET predictions are widely used for the initial study of drugs, which saves the time and investment [14]. The physicochemical, pharmacokinetic and toxicological properties have been theoretically computed using the PreADMET tool available online and has also proved much efficient in recent years (https://preadmet.bmdrc.kr/).

In the present study, the experimental and theoretical vibrational spectral analyses of LGN are compared. The structural parameters of the LGN molecule such as bond length and bond angle are analyzed. The HOMO–LUMO energy gap has been calculated from UV–Vis spectra using TD-DFT [15]. DFT-derived methods such as molecular electrostatic potential (MEP), global reactivity descriptors, Fukui functions and natural bond orbitals (NBO) are used to determine the reactivity of the LGN molecule [16]. The drug likeness and ADMET profiles of LGN are determined to find its drug potential and toxicity. Molecular docking was used to investigate the strength of the LGN molecular binding and intermolecular interactions with the H1N1 NA enzyme. This study will provide some new structural and electronic information for future drug design approaches against the H1N1 virus.

Materials and methods

Experimental details

Analytical reagent-grade LGN (stated purity of ≥ 97%) was purchased from Sigma-Aldrich Company and used without further purification for recording the experimental spectra. The FT-IR spectrum of the title compound was recorded with a Bruker Tensor 27 FT-IR spectrometer using a potassium bromide (KBr) pellet technique in the range of 4000–400 cm−1 with 4 cm−1 resolution in transmission mode. The FT-Raman spectrum of the sample was recorded using a Bruker RFS 27 FT-Raman spectrometer with a neodymium-doped aluminium garnet (Nd:YAG) laser source of 1064 nm in the region 4000–100 cm−1 with ± 2 cm−1 resolution. The LGN molecule is dissolved in methanol to examine the UV–visible (UV–Vis) absorption spectrum, and it was recorded in the range of 190–900 nm using a Perkin Elmer Lambda 25 UV–Vis spectrophotometer.

Computational details

The DFT calculations of the LGN molecule were carried out using the Gaussian 09W software package [17] with 6-31G, 6-31G(d,p) and 6-311G(d,p) basis sets. The GaussView 5.0 visualization program [18] was used to view the molecular geometry and plot the molecular electrostatic potential map (MEP). The potential energy distribution (PED) percentage of the vibrational assignments was computed using the VEDA 4xx program [19]. The electronic properties [UV absorption, highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO)] of the title molecule were computed through TD-DFT. The second-order Fock matrix was utilized to assess the donor–acceptor interactions in the NBO analysis [20]. Utilizing the GaussSum 2.2 software program [21], the molecular orbital’s contribution of the functional groups was analyzed with total density of states (TDOS or DOS), partial density of states (PDOS) and overlap population density of states (OPDOS) spectra. A molecular docking calculation was performed with the AutodockTools 1.5.6 software package [22]. The protein–ligand complex and intermolecular interactions between the NA enzyme and LGN ligand molecule were viewed with the PyMOL [23], Ligplot [24], Chimera [25] and Discovery Studio [26] software programs.

Results and discussion

Optimized geometry

The LGN molecular structure was drawn using ChemDraw 8.0 software. To obtain the exact optimized geometry and electronic parameters of the LGN molecule, DFT [27, 28] with the basis sets 6-31G, 6-31G(d,p) and 6-311G(d,p) is utilized to perform the minimum energy structure optimization. The optimized minimum energy value converged at threshold limits of 0.000450 and 0.001800 a.u. for the maximum force and displacement, respectively. Figure 1 displays the optimized geometry of the title molecule (scaled ball-and-stick model) with atom numbering scheme. The predicted geometrical data for the LGN molecule has not been published so far. The selected optimized geometrical parameters (bond length and bond angle) of LGN are computed for 6-31G, 6-31G(d,p) and 6-311G(d,p) basis sets and compared with the accessible X-ray diffraction (XRD) data related to the structure [29,30,31] and are listed in Table 1. The geometrical parameters of experimental values are slightly smaller than the calculated values, which is due to the fact that experimental data are observed in the solid phase and theoretical values in the gas phase. From the table it is found that comparing the other two basis sets, the 6-311G(d,p) basis set provides the values of bond lengths and bond angles nearer to the experimentally reported values in most cases, where these values have close reliability and good agreement with each other. This molecule has 16 C–C bond lengths, two O–H bond lengths, five C–O bond lengths and ten C–H bond lengths. The observed bond length values for C–C single bonds range from 1.473 to 1.531 Å. For C–C double bonds, the bond length values range from 1.372 to 1.404 Å. The bond lengths of C–H bonds present in the benzene ring range from ~ 1.372 to 1.404 Å and ~ 1.086 to 0.930 Å, respectively. Recently, Shahidha et al. reported that the bond lengths of C–C single bonds are high and for O–H are low (0.963 Å) in the benzene ring and are distorted, yielding variation in bond angles at the point of substitution [32]. The bond angle for C–C–C is found to be 113.7° in bridge bonds. In the bridge bond for the C–C–O angle, the observed values (108.1°) are found to be less than the experimental values (126.7°).  This variation in bond length and bond angles signifies the bonding nature of atoms. 

Fig. 1
figure 1

The ball-and-stick model of optimized structure of LGN with atom numbering scheme

Full size image
Table 1 Selected bond lengths and bond angles of LGN molecule compared with XRD data
Full size table

Vibrational frequency analysis

The theoretical vibrational frequencies were computed using DFT/B3LYP method (6-31G, 6-31G(d,p) and 6-311G(d,p) basis sets) and are scaled down by using scaling factors 0.962, 0.961 and 0.967 [33]. The combined experimental and theoretical FT-IR and FT-Raman spectra for the LGN molecule are represented in Figs. 2 and 3, respectively. The observed wavenumbers in the FT-IR and FT-Raman spectra and their scaled frequency, IR intensities, Raman intensities and vibrational assignments along with their PED% are presented in Table 2. The title compound belongs to the C1 symmetry group and it posses 31 atoms with 87 normal modes of vibration reported. The comparison of the frequency calculated at 6-31G, 6-31G(d,p) and 6-311G(d,p) basis sets using DFT shows that some of the frequencies are underestimated and many of the frequencies are overestimated for the basis sets 6-31G and 6-31G(d,p) compared with 6-311G(d,p). This difference may be due to the allocation of two different sizes of Gaussian type orbitals (GTOs) for extended valence orbitals used in 6-311G(d,p) basis sets for perfectly defining the orbitals of the molecules which consists of C, O and H atoms. For this reason, the FT-IR and FT-Raman frequencies obtained from experiment are compared with frequencies calculated from 6-311G(d,p) basis sets.

Fig. 2
figure 2

Experimental (blue) and theoretical (pink-B3LYP/6-311G(d,p), brown-B3LYP/6-31G(d,p), green-B3LYP/6-31G) FT-IR spectra of LGN. (Color figure online)

Full size image
Fig. 3
figure 3

Experimental (blue) and theoretical (pink-B3LYP/6-311G(d,p), brown-B3LYP/6-31G(d,p), green-B3LYP/6-31G) FT-Raman spectra of LGN. (Color figure online)

Full size image
Table 2 The observed (FT-IR and FT-Raman) and calculated vibrational frequencies using DFT/B3LYP methods for LGN
Full size table

C–C vibrations

Since stronger bonds are usually stiffer and absorb at higher frequencies than weaker bonds, C–C vibrations occur over the frequency range of 1600–700 cm−1 [34, 35]. In the LGN molecule, C–C stretching vibrations are reported in 1640, 1631 cm−1 in FT-IR and 1653, 1605 cm−1 in FT-Raman spectra. The computed bending vibrations have been found to be reliable with the experimental spectral values. It is supported with an average PED contribution of 59%.

O–H vibrations

Because of inter- or intramolecular hydrogen bonding in the molecules, the hydroxyl group vibrations show a large deviation in wavenumber, intensity and bandwidth. The O–H stretching vibrations are sensitive to hydrogen bonding [36]. Basically, the hydroxyl group vibrations are reported over the range of 3600–3200 cm−1 (when H-bonded) and 3700–3500 cm−1 (stretch free) [37]. For the title compound O–H stretching vibrations occurs at 3530 and 3528 cm−1 in B3LYP/6-31G method, 3673 and 3670 cm−1 in B3LYP/6-31G(d,p), 3705 and 3699 cm−1 in B3LYP/6-311G(d,p) method with a PED contribution of 100% which is in good agreement with the experimental data (step rise in FT-IR).

C–H vibrations

The C–H stretching frequencies generally lie in the region 3100–3000 cm−1 and are typically exhibited as weak to moderate bands [38, 39]. In the present study C–H stretching vibrations are reported in 2929, 2923 cm−1 in FT-IR and 3056, 2978, 2906 cm−1 in FT-Raman spectra. In contrast to the recorded values, the scaled frequencies over the range of 3099–2902 cm−1 in B3LYP/6-311G(d,p) method, 3106–2892 cm−1 in B3LYP/6-31G(d,p) method and 3138–2915 cm−1 in B3LYP/6-31G method are identified as C–H stretching vibrations with PED contributions not less than 93%.

C–O vibrations

The C–O bond is strongly polarized towards oxygen atom. According to the literature survey, the C–O vibration is pushed to the lower region by the influence of other vibrations because of the proximity. In the title compound, C–O out-of-plane bending vibrations are reported at 616 and 503 cm−1 in B3LYP/6-311G(d,p) method, 611 and 500 cm−1 in B3LYP/6-31G(d,p) method, 615 and 505 cm−1 in B3LYP/6-31G method, which are fitted well with the experimental values and literature values [40]. A medium band of C–O out of plane bending is also found at 284 cm−1 in B3LYP/6-311G(d,p) method, 283 cm−1 in B3LYP/6-31G(d,p) method, 286 cm−1 in B3LYP/6-31G method. The experimental data of 286 cm−1 in FT-Raman spectra fits well with the calculated value in B3LYP/6-311G(d,p) method.

Molecular electrostatic potential

The molecular electrostatic potential (MEP) map for the LGN molecule is shown in Fig. 4a, and the atoms are colored on the basis of Mulliken atomic charges in order to locate the suitable idea of the individual atoms, and the Mulliken graph is shown parallel to the MEP. The MEP of the title molecule is helpful for studying the important properties of the drug molecule and provides information about the reactivity, nucleophilic and electrophilic sites of the molecule and also shows the intermolecular interaction with the neuraminidase enzyme in the active site cavity. The nucleophilic regions are attracted to the positive potential and electrophilic regions are attracted to the negative potential [41,42,43]. This model is used to predict the strength of intermolecular interactions. These regions are represented by different colors to show the electrophilic and nucleophilic sites of the molecule in the order of red < orange < yellow < green < blue. Here, the red region (electrophilic) means a negative charge and the blue region (nucleophilic) indicates a positive charge. The oxygen atoms possess a high negative charge due to hydrogen bonding interactions with amino acids present in the active site of the enzyme. Similarly, the Mulliken charges of oxygen atoms also tend to have a negative charge owing to intermolecular interactions between the protein and ligand. The iso surface representation of the molecular electrostatic potential of the LGN molecule is shown in Fig. 4b. The positive (blue) and negative (red) regions of the MEP were plotted for the iso surface value of +0.7 eÅ−1 and −0.07 eÅ−1. A large electronegative region was found in the vicinity of O(10), O(11), O(18) and O(19) atoms. This shows the possibility for the occurrence of intermolecular interactions in these regions when the molecule is present in the biological environment.

Fig. 4
figure 4

a Molecular electrostatic potential map of the LGN molecule with Mulliken charges and b molecular electrostatic potential map of LGN molecule showing the electropositive (blue) and electronegative (red) regions of the molecule. The iso surface values are +0.7 eÅ−1for positive potential, −0.07 eÅ−1 for negative potential. (Color figure online)

Full size image

UV–Vis spectral analysis

The UV–Vis spectral analysis is used to investigate the nature of electronic transition properties in the LGN molecule. The theoretical UV–Vis spectra are computed using TD-DFT with basis set 6-311G(d,p) in the Gaussian 09W program. The UV radiation was absorbed by bonding \((\pi ,\sigma )\), anti-bonding (π*, σ*) and non-bonding (n) of the molecules containing transitions from HOMO to LUMO, i.e., highest filled level to unoccupied level. The experimental and theoretical UV–Vis absorption spectra of the title molecule are shown in Fig. 5. The absorption wavelength (λ), excitation energy (E) and oscillating strength (f) are computed for title molecule and are listed in Table 3. The absorption bands of \(\pi \to \pi ^{*}\) transition are strong and \(n \to \pi ^{*}\) transition are weak as reported to characterize the UV–Vis absorption spectra [44, 45]. The \(\pi \to \pi ^{*}\) transitions are observed in the maximum wavelength range of 391 nm for experiment and 320 nm for theory. The experimental \(\sigma \to \pi ^{*}\) electronic transitions are 221 nm, but theoretically the transition wavelengths are 241 nm. The major contribution of first transition is 51% for the LGN molecule from HOMO to LUMO levels. This wavelength range shows that the LGN molecule is biologically active.

Fig. 5
figure 5

Experimental and theoretical absorption spectra of LGN in methanol solution

Full size image
Table 3 Electronic properties of LGN calculated using TD-DFT (B3LYP)/6-311G(d,p)
Full size table

Global reactivity descriptors

The HOMO–LUMO energies of the molecule play an important role in finding the numerous chemical reactivity descriptors to understand several aspects of the LGN molecule, including pharmacological properties and drug design. The HOMO–LUMO for the title molecule is plotted by using GaussView 5.0 to visualize the charge localization and delocalization as shown in Fig. 6. The molecular descriptors of the title molecule, such as ionization potential (I = −EHOMO), Electron affinity (A = −ELUMO), global hardness (η = (I − A)/2), chemical softness (s = 1/2η), electronegativity (χ = (I + A)/2), chemical potential (μ = −χ) and electrophilicity (ω = μ2/2η) [31, 46,47,48], were estimated, and their values are shown in Table 4. In HOMO–LUMO calculations, a large energy gap means it is a hard molecule, and a small energy gap means it is a soft molecule [49]. The computed energy gap for the title molecule is 4.815 eV, which implies that LGN is a soft molecule. The Koopmans theorem for closed-shell molecules states the ionization potential (it approximates the negative of HOMO) and electron affinity (it approximates the negative of LUMO) of the molecule. The ionization potential refers to the change of energy of a system when an electron is removed, and electron affinity refers to the variation of energy of a system when an electron is added. A higher level of HOMO energy indicates the high reactivity nature of the molecule, expected to react with electrophiles, while a lower level of LUMO energy is essential for molecular reaction with nucleophiles. The electronegativity of the title molecule is related using HOMO and LUMO energies and is formulated using Mulliken [50] to illustrate the electron density of the molecule. The calculated values of electron affinity and electron negativity are very low compared to ionization potential. From this it is confirmed that LGN has less susceptibility to accept electrons, so this resembles a nucleophile. The global hardness of the molecule is defined by Parr and Pearson [31] and provides information about the absolute hardness, kinetic stability and reactivity of the molecule. The determined value for LGN is 2.408 eV, which implies that the title molecule has more kinetic stability and reactivity. The electrophilicity provides information about reactivity and toxicity of the molecule. The electrophilicity index of LGN is found to be very low, thus indicating low toxicity. The obtained results suggest that the LGN molecule possesses good biological activity in pharmacological properties and drug design.

Fig. 6
figure 6

The HOMO and LUMO orbitals of LGN

Full size image
Table 4 Calculated global reactivity descriptors for LGN
Full size table

Density of states (DOS)

As a reasonable description of the frontier orbitals cannot be obtained by simply analyzing HOMO–LUMO, TDOS, OPDOS and PDOS spectra are employed and plotted with the energies on the X-axis and the calculated respective density values on the Y-axis [51,52,53] using GaussSum 2.2 software. The occupied and virtual levels are clearly indicated in the TDOS plot, which provides the molecular orbital composition with respect to their chemical bonding. The OPDOS plot determines the donor–acceptor properties of the ligands and bonding and non-bonding interactions. The overlap of hydrogen with oxygen exhibits a positive value only, which clearly indicates the bonding interaction of hydrogen and oxygen atoms. The bonding nature of carbon atoms with hydrogen atoms is much higher (because of positive overlap population). Since it is very difficult to compare groups in terms of their bonding and anti-bonding properties of PDOS spectra, as plotted in Fig. 7, the PDOS plot mainly presents the composition of the fragment orbitals contributing to the molecular orbitals. The HOMO levels are localized on the ring, and their contributions are about 68%, and the LUMO levels are localized on the ring, and their contributions are 92% of the title compound.

Fig. 7
figure 7

a Total, b partial and c overlap population density of states of LGN molecule

Full size image

Local reactivity descriptors

The Fukui function is determined as local reactivity descriptors, and it plays an major role in studying chemical reactivity and selectivity. showing a greater tendency towards accepting or donating electrons that are more prone to undertake a nucleophilic or an electrophilic attack, respectively [28]. The Fukui functions are calculated on the basis of Mulliken atomic charges of neutral (N), cation (N + 1) and anion (N − 1) states of LGN using the following equations.

$$f_{k}^{ + } { = }[q_{(N+1)} - q_{(N)} ]\;{\text{for}}\;{\text{nucleophillic}}\;{\text{attack}}$$
(1)
$${\text{and}}\;f_{k}^{ - } { = }[q_{(N)} - q_{(N - 1)} ]\;{\text{for}}\;{\text{electrophilic}}\;{\text{attack}}$$
(2)

The Fukui functions \(\left( {f_{k}^{ + } ,f_{k}^{ - } } \right)\), local softness \(\left( {(sf)_{k}^{ + } ,(sf)_{k}^{ - } } \right)\) and local electrophilicity index \(\left( {(\omega f)_{k}^{ + } ,(\omega f)_{k}^{ - } } \right)\) [54, 55] for selected atomic sites in LGN are shown in Table 5. Local softness and electrophilicity index are computed using Eqs. (3) and (4). The local softness is derived as the product of the Fukui value and the global softness, while the local electrophilicity index is derived as the product of the Fukui value and global electrophilicity index.

Table 5 Fukui values fk and descriptors (sf)k and (ωf)K for LGN with Mulliken atomic charges
Full size table
$$s_{k}^{ + } = \, (sf)_{k}^{ + }\;{\text{and}}\;s_{k}^{-} = \, (sf)_{k}^{-}\;{\text{for}}\;{\text{local}}\;{\text{softness}}$$
(3)
$$\omega_{k}^{ + } = (\omega f)_{k}^{ + }\;{\text{and}}\;\omega_{k}^{-} = (\omega f)_{k}^{-}\;{\text{for}}\;{\text{local}}\;{\text{electrophility}}\;{\text{index}}$$
(4)

The \(f_{k}^{+}\) value tends to nucleophilic attack and the \(f_{k}^{-}\) value tends to electrophilic attack. From the table it is clear that the \(f_{k}^{+}\) value says that the possible sites for nucleophilic attack are O11, C3, C5, O19 and C7 atoms, respectively, and the \(f_{k}^{-}\) value says the possible site for electrophilic attack is the C1 atom. The result shows that the LGN molecule has biological properties.

Natural bond orbitals

The NBO were analyzed to find the electron density delocalized from donor (occupied Lewis-type NBOs) to acceptor (unoccupied non-Lewis-type NBOs) in the LGN molecule. The NBO analysis is performed to determine the donor–acceptor interaction energy, and it was evaluated by second-order perturbation theory using a Fock matrix with hyperconjugative interactions [56]. E(2) is given by the following equation

$$E(2) = \Delta E_{ij} = \frac{{q_{i} \left( {F_{ij} } \right)^{2} }}{{\varepsilon_{j} - \varepsilon_{i} }}$$
(5)

where \(q_{i}\) is the donor orbital occupancy, \(F_{ij}\) is the Fock matrix element between i and j NBO orbitals, \(\varepsilon_{i}\) and \(\varepsilon_{j}\) are the orbital energies of diagonal element.

The intramolecular hyperconjugative interactions of \(\pi \to \pi ^{*}\), \({\text{LP}} \to \pi ^{*}\) and \({\text{LP}} \to \sigma ^{*}\) are the possible interactions for LGN. The \(\pi \to \pi ^{*}\) interactions of \(\pi \left( {{\text{C}}7{-}{\text{C}}8} \right) \to \pi ^{*}\left( {{\text{C}}4{-}{\text{C}}9} \right)\;{\text{and}}\;\pi \left( {{\text{C}}5{-}{\text{C}}6} \right) \to \pi ^{*}\left( {{\text{C}}7{-}{\text{C}}8} \right)\) give rise to high stabilization energy, and the values are 26.09 and 25.32 kJ mol−1. The lone pair LP(2) of oxygen atom interacts with σ* and π* are LP(2)O11 to σ*(C2–C3), LP(2)O11 to σ*(C3–C4),  LP(2)O10 to π*(C4–C9), LP(2)O18 to π*(C15–C16) and LP(2)O19 to π*(C7–C8), and the values are 20.81, 19.05, 28.08, 28.83 and 31.51 kJ mol−1, respectively. This results shows that LGN molecule is highly stabilized. The natural bond energy values of LGN are presented in Table 6.

Table 6 The natural bond energy values of LGN at 6-311G(d,p) level of theory
Full size table

Drug likeness properties

Lipinski’s rule of five [57] was used to estimate the bioavailability of bulk materials to determine drug likeness properties; this rule plays a major role in drug discovery. In the present work, the physicochemical properties of bulk molecules were carried out with the Molinspiration Cheminformatics program available online (https://www.molinspiration.com). The drug likeness of the LGN molecule was calculated and presented in Table 7. The good membrane permeability of molecules should obey the following properties: H-bond donors (HBD) were found to be 2 (≤ 5), H-bond acceptors (HBA) were found to be 4 (≤ 10), partition coefficient (Mi logP) of LGN was observed to be 2.2 (≤ 5), molecular weight (MW) of the LGN is 256.26 g mol−1 (≤ 500 g mol−1), van der Waals topological polar surface area (TPSA) value for LGN is 66.76 Ǻ2 (< 120 Ǻ2) [27]. The LGN molecule obeys Lipinski’s rule of five for all cases.

Table 7 Drug likeness descriptor of LGN predicted from Molinspiration
Full size table

The bioactivity score of LGN calculated for GPCR ligand, ion channel modulator, kinase inhibitor, nuclear receptor ligand, protease inhibitor and enzyme inhibitor were 0.03, − 0.32, −0.39, 0.32, − 0.21 and 0.17, respectively, and presented in Table 8. A bioactivity score of more than 0 means that it is active, if − 0.5 to 0 then it is moderately active. This clearly states that LGN acts as a GPCR ligand, good nuclear receptor ligand and enzyme inhibitor. The result indicates that the LGN molecule can be treated as a potential candidate for bioactive application.

Table 8 Bioactivity score of LGN predicted from Molinspiration
Full size table

ADMET predictions

The pharmacokinetic properties such as absorption, distribution, metabolism, excretion and toxicity (ADMET) were analyzed for the LGN molecule as tabulated in Table 9. Under the ADMET analysis, various properties were analyzed, including blood–brain barrier (BBB) penetration, human intestinal absorption (HIA), Caco-2 cell permeability, water solubility in buffer, pure water solubility, skin permeability, MDCK cell permeability, P-glycoprotein inhibition, plasma protein binding, CYP (2C19, 2C9, 2D6, 3A4) inhibition, CYP (2D6, 3A4) substrate, Ames test for mutagenicity, carcinogenicity (mouse, rat), human ether-a-go-go-related gene (hERG) inhibition, Lipinski’s rule, WDI-like rule, lead-like rule, CMC-like rule and MDDR-like rule [58]. Through the identified results it is found that LGN is similar to the recommended range of 96% noted from the presently known drugs. The result of penetration potential in the BBB shows that it has less potential to cross the BBB in the central nervous system [59]. The obtained value for the LGN molecule is 0.6436, which indicates that the molecule causes fewer side effects in the central nervous system [58]. In human skin, the permeability value for LGN is − 3.4987 cm h−1 , which shows that the molecule cannot absorb through human skin. HIA is one of most important parameters for a drug molecule [60]. An HIA value between 70 and 100% shows good human intestinal absorption, and for LGN it is 92.3564% [61, 62]. Caco-2 and MDCK cell permeability are two predicted parameters for oral absorption of drugs [63]. The computed Caco-2 cell permeability value is 17.646 nm s−1 and MDCK cell permeability value is 41.7998 nm s−1 , respectively. The inhibited P-glycoprotein value for the LGN molecule shows that it is a non-inhibitor, and the protein is responsible for absorption, distribution, metabolism and excretion of LGN drug [64]. Through the analysis of water and pure water LGN attained higher values, which are 96.3932 mg L−1 (water) and 164.186 mg L−1 (pure water). The cytochrome P450 (CYP) protein indicates that the LGN molecule was an inhibitor of 2C19, 2C9 and 3A4, which reduces the capability of the drug and develops the pharmacological effects of the drug. LGN was found to be a non-inhibitor of CYP 2D6. The mutagenicity data by Ames test for the LGN molecule is positive for Ames TA100 (+S9 and –S9) isomers and negative for Ames TA 1535 (+S9 and −S9) isomers, respectively. LGN is not carcinogenic in mice, but is carcinogenic in rats [65]. The LGN molecule obeys Lipinski’s rule of five and World Drug Index (WDI) as probability of permeability and solubility is greater than 90% [66]. The lead-like rule has higher binding affinity greater than 0.1 μM. The LGN molecule is in the qualifying range for the CMC-like rule [67]. In this case, the MDDR-like rule shows mid-structure, which means it is intermediate between a potential and non-potential drug [68]. The ADMET profile for the LGN molecule shows a starting point for a lead optimization program.

Table 9 Prediction of ADMET profiles for LGN
Full size table

Molecular docking analysis

Molecular docking investigations were performed by Autodock software. The complex form of the NA enzyme was obtained from Brookhaven Protein Data Bank (PDB Code: 4B7N) with a resolution of 2.84 Å. The protein was separated by removing the ions, water and ligand molecules, and it can be used as a target for docking analysis. The optimized structure of LGN with DFT/6-311G(d,p) method is selected as a ligand. The docked ligand molecules were lifted from the active site of the NA enzyme. Autodock generated ten different conformers, and their corresponding docked energy values are listed in Table 10. Among ten possible conformational energy values, the fourth conformer exhibits lowest docked energy with a docking value of −7.05 kcal mol−1 with inhibition constant of 6.81 μM. The intermolecular interactions of LGN with NA enzyme is shown in Fig. 8. The surface view of LGN encapsulated in the active site of the NA enzyme is shown in Fig. 9. The ligplot displays the important hydrophobic and hydrogen bonding interactions as represented in Fig. 10. The LGN molecule was slightly twisted in the active site cavity, and it is fixed very well when compared to the gas phase, which is because the title molecule interacts with the nearest amino acids in the protein. The active site residues are Arg118, Ile149, Arg368, Ser400, Ile427, Pro431 and Lys432. The amino acids Ser400 and Arg368 form two conventional H bonds with oxygen atoms with distances of 2.82 and 1.88 Å, respectively. The H(31) atom also forms strong conventional hydrogen bonding interactions with Ile149 at a distance of 2.12 Å. The Pi-Alkyl interaction forms between the center of the ring and the residues center. The Pi-Alkyl interactions form between the amino acids Lys432, Ile427 and Arg368 and aromatic benzene ring of LGN. The residues Pro431 and Ile149 form Pi-Alkyl interactions with a hydroxychroman ring. These interactions enhance the binding affinity of LGN towards NA. From the docking analysis, it is clear that in active sites, the molecule adopts a new conformation, which is due to the hydrophobic and hydrogen bonding interactions between LGN and the NA enzyme.

Table 10 The docking score value of 10 possible conformers of LGN in the active site of NA enzyme
Full size table
Fig. 8
figure 8

Intermolecular interactions of LGN with the NA enzyme

Full size image
Fig. 9
figure 9

Surface view of LGN encapsulated in the active site of the NA enzyme

Full size image
Fig. 10
figure 10

The ligplot showing intermolecular interactions of LGN in the active site of the NA enzyme

Full size image

Conclusion

The molecular structure of LGN was optimized to its minimum energy level, and the optimized geometry was compared with available XRD data related to the structure of LGN. The experimental and observed FT-IR and FT-Raman spectral data for LGN shows good agreement, and UV spectral data is computed through TD-DFT approach. The absorbed wavelength of the LGN shows that the molecule is biologically active. The MEP provides the information about the drug-receptor interactions in drug discovery and permits to predict the nucleophilic and electrophilic sites of the molecule, in LGN it revealed that the negative region of the molecule was exposed to the electrophilic attack. The electrophilicity index (ω) values of LGN are found to be smaller; this means that the molecule has less toxicity. NBO and Fukui functions of the LGN molecule are computed and posses good chemical selectivity and high stability of the molecule. The LGN molecule obeys the properties of Lipinski’s rule of five. Hence, from the Molinspiration predictions, it could be stated that LGN is a good candidate for bioactive applications. The docking score for the lowest energy conformation of LGN in the active site is −7.05 kcal mol−1 with an inhibition constant of 6.81 μM. The docking analysis and quantum chemical calculation shows the exact conformational and molecular orientations of LGN in the active site of the NA enzyme. These crucial vibrational, structural and docking analysis details of LGN can provide useful insights for future anti-influenza A (H1N1) drug design.