Plant Alkylbenzenes and Terpenoids in the Form of Cyclodextrin Inclusion Complexes as Antibacterial Agents and Levofloxacin Synergists

Allylpolyalkoxybenzenes (APABs) and terpenoids from plant essential oils exhibit a range of remarkable biological effects, including analgesic, antibacterial, anti-inflammatory, antioxidant, and others. Synergistic activity with antibiotics of different classes has been reported, with inhibition of P-glycoprotein and impairment of bacterial cell membrane claimed as probable mechanisms. Clearly, a more detailed understanding of APABs’ biological activity could help in the development of improved therapeutic options for a range of diseases. However, APABs’ poor solubility in water solutions has been a limiting factor for such research. Here, we found that complex formation with β-cyclodextrins (CD) is an efficient way to transform the APABs into a water-soluble form. Using a combination of spectroscopic (FTIR, NMR, UV) methods, we have estimated the binding constants, loading capacity, and the functional groups of both APABs and monoterpenes involved in complex formation with CD: ethylene, aromatic, methoxy and hydroxy groups. In the presence of a molar excess of CD (up to 5 fold) it was possible to achieve the complete dissolution of APABs and terpenoids in an aqueous medium (at 90–98% encapsulation) higher by 10–1000 times. Further, we have demonstrated that CD-APABs, if used in combination with levofloxacin (Lev), can be antagonistic, indifferent, additive, or synergistic, mostly depending on the concentration ratio: at high Lev concentration with the addition of APAB is typically neutral or even antagonistic; while at a Lev concentration below MIC, the addition of CD-APAB is either additive or synergistic (according to FICI criteria). An over three-fold increase in Lev antibacterial activity was observed in combination with eugenol (EG), as per the growth inhibition diameter measurement in agar. Interestingly, a synergistic effect could be observed with both Gram-positive and Gram-negative bacteria. So, obviously, the APAB-CD and terpenoid-CD mechanism of action is not limited to their interaction with the bacterial membrane, which has been shown earlier for CDs. Further research may open new prospects for the development of adjuvants to improve the therapeutic regimens with existing, as well as with new anti-infective drugs.


Plant extracts from dill
The formation of inclusion complexes of terpenoid's and APAB's with MCD was investigated by UV spectroscopy (Figure 1). In the UV spectra of EG ( Figure 1A), safrole ( Figure 1B) and its MCD-complexes, three peaks were observed: (1) 279 and 287 nm, respectively, correspond to the absorption of the aromatic system; (2) 205 nm-absorption of the double bond of the allyl group; (3) 230 and 235 nm, respectively, correspond to the resonant structures of a double bond with a benzene ring. The region from 240-350 nm was analyzed for all compounds studied except for linalool absorbing due to two double bonds at 200-220 (ε 200 = 24,300). For pure APAB, the molar absorption coefficient was in the order of thousands ( Figure 1C), in the series of apiol's analogues increased from 1580 for myristicin to 4570 for allyltetramethoxybenzene symbiotically with an increase in the number of CH 3 O groups and their spatial convergence (dillapiol vs. apiol).
The formation of APAB inclusion complexes with MCD led to an increase in the UV absorbance of the samples at 278-288 nm (Figure 1), which corresponded to an increase in the concentration of these substances in aqueous fraction (APAB (aq.) alone and APAB-MCD (aq.) ). In Figure 1A,B, the brown spectra correspond to EG and safrole dissolved in water without MCD, while the cyan spectra represent a system APAB-MCD (molar ratio 1:5) with increased solubility of APAB by 5 and 25 times, respectively, for EG and safrole.
To optimize the system composition, similar experiments were carried out on the formation of EG and apiol inclusion complexes with HPCD ( Figure S1). In addition to changes in intensity, the following changes were observed in the UV spectra of apiol and EG: the peaks of apiol and eugenol shift of 205 nm into the long-wavelength region occurs due to a decrease in the energy of the π→π*, n→π* allyl group transitions into the hydrophobic cavity of the CD compared to aqueous solution. According to these changes, the dissociation constants of the EG-HPCD and apiol-HPCD complexes ( Figure S2) were calculated by Hill's equation (Table 1). Similar changes were observed in the complexation of EG and apiol with MCD. The physicochemical parameters of the interaction of APAB and terpenoids (X) with CD were studied by determining the concentration of substances in the aqueous fraction ([APAB] + [APAB-MCD]) according to the calibration dependence ( Figure 1C) for systems with a constant amount of the essential oil component X and the concentration gradient MCD (Table 1). Theta (θ) is a fraction of the bound substance X:θ = [X-MCD]/([X-MCD] + [X]) and was calculated based on the dependence of the absorption intensity X on the MCD concentration in the system in the form of a sigmoidal Hill curve with subsequent linearization (Sections 3, 3.3 and 3.4).
APAB and terpenoids form inclusion complexes with CD with an excess of tori per guest molecule from 1.1 to 1.7, which is presumably explained by the smaller volume of the included molecule in comparison with the CD cavity, as well as the possibility of interaction additionally with the outer hydrophilic shell of the CD. The distinctive properties were Pharmaceuticals 2022, 15, 861 4 of 28 demonstrated by myristicin, which forms inclusion complexes of molar ratio APAB:CD close to 2 to 1. Typical values of K d are of the order of magnitude 10 −3 -10 −4 M. In general, MCD forms inclusion complexes stronger by 0.5-1.5 orders of magnitude than HPCD due to the increased hydrophobicity of the inner cavity due to methyl groups [27]. The inclusion complex is stronger in the case of larger and more hydrophobic guest molecules. Safrole, myristicin and allyltetramethoxybenzene are the leaders among the APABs. The quantitative loading of drugs into the CD can be estimated using the parameters entrapment efficiency (EE) and loading capacity (LC). EE characterizes the proportion of molecules that form a complex with CD. In the presence of 0.01 M MCD, almost all of the considered compounds effectively form complexes with CD, EE > 40%, although this parameter is lower for apiol. At the same time, Lev, menthol, myristicin and safrole are more than 90% in a complex form at the studied conditions. LC characterizes the mass fraction of the "guest" in the complex formation. In the considered series, LC ≈ 10%, increases for compounds with a higher molecular weight and stronger binding to CD. APAB and terpenoids form inclusion complexes with CD with an excess of tori per guest molecule from 1.1 to 1.7, which is presumably explained by the smaller volume of the included molecule in comparison with the CD cavity, as well as the possibility of interaction additionally with the outer hydrophilic shell of the CD. The distinctive properties were demonstrated by myristicin, which forms inclusion complexes of molar ratio APAB:CD close to 2 to 1. Typical values of Kd are of the order of magnitude 10 −3 -10 −4 M. In general, MCD forms inclusion complexes stronger by 0.5-1.5 orders of magnitude than HPCD due to the increased hydrophobicity of the inner cavity due to methyl groups [27]. The inclusion complex is stronger in the case of larger and more hydrophobic guest molecules. Safrole, myristicin and allyltetramethoxybenzene are the leaders among the APABs. The quantitative loading of drugs into the CD can be estimated using the parameters entrapment efficiency (EE) and loading capacity (LC). EE characterizes the proportion of molecules that form a complex with CD. In the presence of 0.01 M MCD, almost all of the considered compounds effectively form complexes with CD, EE > 40%, although this parameter is lower for apiol. At the same time, Lev, menthol, myristicin and safrole are more than 90% in a complex form at the studied conditions. LC characterizes the mass fraction of the "guest" in the complex formation. In the considered series, LC ≈ 10%, increases for compounds with a higher molecular weight and stronger binding to CD.
The graphical dependence of EE (calculated based on the absorption at 280 nm and peak shift at 205 nm in the UV spectra) on the amount of HPCD is given in the Supplementary Materials ( Figure S3). The entropy factor is the driving force behind the formation of APAB-CD complexes [22]. The dissociation of EG-HPCD and apiol-HPCD complexes is kinetically impeded: the amount of the complex remains at a level above 95% of the initial at 100-fold dilution (from 10 mM to 0.1 mM-relevant concentrations during biological testing) at 22-37 °C. The formation of APAB and terpenoid inclusion complexes is important for a significant increase in the solubility of these compounds in water (Table  Table 1. Physicochemical parameters of formation of MCD and HPCD inclusion complexes with Lev, APAB and terpenoids. Dissociation constants X-MCD and X-HPCD, where X is a "guest" compound, for the equilibrium: X aq + n MCD aq X·nMCD aq . Entrapment efficiency (EE) and loading capacity (LC) are shown, C(MCD) = 0.01 M. Solubility of X in water C 0 = [X aq ] and solubility of X in water in the presence of 2 and 10 mM MCD − C max are given. The data are indicated for MCD unless otherwise agreed. A 5 mM sodium phosphate buffer was used (pH 6.2). T = 22 • C. 1). A sample of 2 and, more effectively, 10 mM CD enables an increase in the solubility of hydrophobic compounds by several times, which allows them to be used as the biologically active adjuvants of antibacterial drugs. In fact, solubility is limited by the concentration of CD, since the EE for many of the compounds under consideration is >90%, which means that almost any reasonable concentration of the complex form is achievable. 1). A sample of 2 and, more effectively, 10 mM CD enables an increase in the solubility of hydrophobic compounds by several times, which allows them to be used as the biologically active adjuvants of antibacterial drugs. In fact, solubility is limited by the concentration of CD, since the EE for many of the compounds under consideration is >90%, which means that almost any reasonable concentration of the complex form is achievable. 1). A sample of 2 and, more effectively, 10 mM CD enables an increase in the solubility of hydrophobic compounds by several times, which allows them to be used as the biologically active adjuvants of antibacterial drugs. In fact, solubility is limited by the concentration of CD, since the EE for many of the compounds under consideration is >90%, which means that almost any reasonable concentration of the complex form is achievable. 1). A sample of 2 and, more effectively, 10 mM CD enables an increase in the solubility of hydrophobic compounds by several times, which allows them to be used as the biologically active adjuvants of antibacterial drugs. In fact, solubility is limited by the concentration of CD, since the EE for many of the compounds under consideration is >90%, which means that almost any reasonable concentration of the complex form is achievable.   1). A sample of 2 and, more effectively, 10 mM CD enables an increase in the solubility of hydrophobic compounds by several times, which allows them to be used as the biologically active adjuvants of antibacterial drugs. In fact, solubility is limited by the concentration of CD, since the EE for many of the compounds under consideration is >90%, which means that almost any reasonable concentration of the complex form is achievable.

FTIR Spectroscopy Approach
Fourier transformed infrared spectroscopy in the mode of attenuated total reflection (ATR) is an advanced method for studying the interactions of substances at the atomicmolecular level. Changes in the microenvironment of chemical bonds, functional groups, protonation/deprotonation, or the formation of non-valent interactions between the receptor and the ligand (in this case, the formation of guest-host inclusion complexes of CD with EG and apiol molecules) leads to a change in the intensity or position of characteristic peaks in the IR spectra. Thus, when the apiol was included in HPCD or MCD (data for MCD are similar and not shown), a significant increase in the intensity of all characteristic peaks was observed ( Figure 2A [19]. This fact indicates the inclusion of apiol not only in the CD cavity, but also its interaction with the external hydrophilic shell of HPCD. The formation of inclusion complexes of APAB with a CD cavity was confirmed with a change in the intensity of peaks in the IR spectra, first of all corresponding to the fluctuations of aromatic C-C bonds (1505 cm −1 ) and ethylene C=C (1650 cm −1 ). These data allowed the estimation of the apiol-CD complex dissociation constant value as 4.0 ± 0.3 mM ( Figure S4). In this case, one CD molecule interacts with two apiol molecules, (by Hill linearization model). The above-indicated peaks in the IR spectra are composite which was visually confirmed by the changes in the shape of peaks and even the appearance of shoulders when the APAB was turned on in the CD cavity. To separate the complex peaks into components, increase sensitivity and obtain more detailed information, the differentiation method was used. The shifts of minima/maxima of the second derivative of spectra ( Figure 2B) in allyl ν(C=C) and aromatic ν(C-C) regions were the most prominent. At the same time, when the apiol was included in CD, similar, but less vivid shifts of peaks corresponding to the C-H, C-O bonds occurred, which indicated the inclusion of the entire APAB molecule into the CD cavity.
The interaction of EG with CD leads to opposite changes in the IR spectra: quenching of almost all characteristic peaks was observed ( Figure 2C). This can be explained by the

FTIR Spectroscopy Approach
Fourier transformed infrared spectroscopy in the mode of attenuated total reflection (ATR) is an advanced method for studying the interactions of substances at the atomicmolecular level. Changes in the microenvironment of chemical bonds, functional groups, protonation/deprotonation, or the formation of non-valent interactions between the receptor and the ligand (in this case, the formation of guest-host inclusion complexes of CD with EG and apiol molecules) leads to a change in the intensity or position of characteristic peaks in the IR spectra. Thus, when the apiol was included in HPCD or MCD (data for MCD are similar and not shown), a significant increase in the intensity of all characteristic peaks was observed ( [19]. This fact indicates the inclusion of apiol not only in the CD cavity, but also its interaction with the external hydrophilic shell of HPCD. The formation of inclusion complexes of APAB with a CD cavity was confirmed with a change in the intensity of peaks in the IR spectra, first of all corresponding to the fluctuations of aromatic C-C bonds (1505 cm −1 ) and ethylene C=C (1650 cm −1 ). These data allowed the estimation of the apiol-CD complex dissociation constant value as 4.0 ± 0.3 mM ( Figure S4). In this case, one CD molecule interacts with two apiol molecules, (by Hill linearization model). The above-indicated peaks in the IR spectra are composite which was visually confirmed by the changes in the shape of peaks and even the appearance of shoulders when the APAB was turned on in the CD cavity. To separate the complex peaks into components, increase sensitivity and obtain more detailed information, the differentiation method was used. The shifts of minima/maxima of the second derivative of spectra ( Figure 2B) in allyl ν(C=C) and aromatic ν(C-C) regions were the most prominent. At the same time, when the apiol was included in CD, similar, but less vivid shifts of peaks corresponding to the C-H, C-O bonds occurred, which indicated the inclusion of the entire APAB molecule into the CD cavity.
The interaction of EG with CD leads to opposite changes in the IR spectra: quenching of almost all characteristic peaks was observed ( Figure 2C). This can be explained by the

FTIR Spectroscopy Approach
Fourier transformed infrared spectroscopy in the mode of attenuated total reflection (ATR) is an advanced method for studying the interactions of substances at the atomicmolecular level. Changes in the microenvironment of chemical bonds, functional groups, protonation/deprotonation, or the formation of non-valent interactions between the receptor and the ligand (in this case, the formation of guest-host inclusion complexes of CD with EG and apiol molecules) leads to a change in the intensity or position of characteristic peaks in the IR spectra. Thus, when the apiol was included in HPCD or MCD (data for MCD are similar and not shown), a significant increase in the intensity of all characteristic peaks was observed ( [19]. This fact indicates the inclusion of apiol not only in the CD cavity, but also its interaction with the external hydrophilic shell of HPCD. The formation of inclusion complexes of APAB with a CD cavity was confirmed with a change in the intensity of peaks in the IR spectra, first of all corresponding to the fluctuations of aromatic C-C bonds (1505 cm −1 ) and ethylene C=C (1650 cm −1 ). These data allowed the estimation of the apiol-CD complex dissociation constant value as 4.0 ± 0.3 mM ( Figure S4). In this case, one CD molecule interacts with two apiol molecules, (by Hill linearization model). The above-indicated peaks in the IR spectra are composite which was visually confirmed by the changes in the shape of peaks and even the appearance of shoulders when the APAB was turned on in the CD cavity. To separate the complex peaks into components, increase sensitivity and obtain more detailed information, the differentiation method was used. The shifts of minima/maxima of the second derivative of spectra ( Figure 2B) in allyl ν(C=C) and aromatic ν(C-C) regions were the most prominent. At the same time, when the apiol was included in CD, similar, but less vivid shifts of peaks corresponding to the C-H, C-O bonds occurred, which indicated the inclusion of the entire APAB molecule into the CD cavity.
The interaction of EG with CD leads to opposite changes in the IR spectra: quenching of almost all characteristic peaks was observed ( Figure 2C). This can be explained by the The graphical dependence of EE (calculated based on the absorption at 280 nm and peak shift at 205 nm in the UV spectra) on the amount of HPCD is given in the Supplementary Materials ( Figure S3). The entropy factor is the driving force behind the formation of APAB-CD complexes [22]. The dissociation of EG-HPCD and apiol-HPCD complexes is kinetically impeded: the amount of the complex remains at a level above 95% of the initial at 100-fold dilution (from 10 mM to 0.1 mM-relevant concentrations during biological testing) at 22-37 • C. The formation of APAB and terpenoid inclusion complexes is important for a significant increase in the solubility of these compounds in water (Table 1). A sample of 2 and, more effectively, 10 mM CD enables an increase in the solubility of hydrophobic compounds by several times, which allows them to be used as the biologically active adjuvants of antibacterial drugs. In fact, solubility is limited by the concentration of CD, since the EE for many of the compounds under consideration is >90%, which means that almost any reasonable concentration of the complex form is achievable.

FTIR Spectroscopy Approach
Fourier transformed infrared spectroscopy in the mode of attenuated total reflection (ATR) is an advanced method for studying the interactions of substances at the atomicmolecular level. Changes in the microenvironment of chemical bonds, functional groups, protonation/deprotonation, or the formation of non-valent interactions between the receptor and the ligand (in this case, the formation of guest-host inclusion complexes of CD with EG and apiol molecules) leads to a change in the intensity or position of characteristic peaks in the IR spectra. Thus, when the apiol was included in HPCD or MCD (data for MCD are similar and not shown), a significant increase in the intensity of all characteristic peaks was observed (Figure 2A [19]. This fact indicates the inclusion of apiol not only in the CD cavity, but also its interaction with the external hydrophilic shell of HPCD. The formation of inclusion complexes of APAB with a CD cavity was confirmed with a change in the intensity of peaks in the IR spectra, first of all corresponding to the fluctuations of aromatic C-C bonds (1505 cm −1 ) and ethylene C=C (1650 cm −1 ). These data allowed the estimation of the apiol-CD complex dissociation constant value as 4.0 ± 0.3 mM ( Figure S4). In this case, one CD molecule interacts with two apiol molecules, (by Hill linearization model). The above-indicated peaks in the IR spectra are composite which was visually confirmed by the changes in the shape of peaks and even the appearance of shoulders when the APAB was turned on in the CD cavity. To separate the complex peaks into components, increase sensitivity and obtain more detailed information, the differentiation method was used. The shifts of minima/maxima of the second derivative of  Figure 2B) in allyl ν(C=C) and aromatic ν(C-C) regions were the most prominent. At the same time, when the apiol was included in CD, similar, but less vivid shifts of peaks corresponding to the C-H, C-O bonds occurred, which indicated the inclusion of the entire APAB molecule into the CD cavity.
parameters EE, LC) of the integral intensity of the C-Carom peak, which characterizes the allocation of eugenol in the sample. EG was effectively (EE > 70%) included in two-thirds of the sample by quantity, in the remaining part EE exceeded 25-35%.
Visually, the dissolution of APAB and the homogenization of the system (disappearance of oil droplets) when they were included into CD (both in the MCD and in the HPCD) were observed in a light microscope (Figures S7 and S8). The dependence of the dissolution degree on the content of CD in the system was determined. Visual observation indicated that for a noticeable conversion to the complex/soluble form of EG, an equimolar amount of CD is required, and almost complete dissolution is achieved with a 5-fold molar excess. For apiol, due to its lower affinity for CD, 2-and 10-15-fold excess of CD are required, respectively. The degree of EG inclusion practically does not increase with long incubation. Since soluble complex forms of EG, apiol and other individual components of essential oils have been obtained, a quantitative study of the adjuvant and synergistic activities of APAB is relevant but previously impossible because of the stratification of substances and the formation of oil fractions. The following are the results of experiments to determine the antibacterial and synergistic effects of adjuvant-CD complexes. . Spectra of HPCD in the corresponding concentration were subtracted as background.

NMR Spectroscopy
NMR spectroscopy is a useful tool, and complementary to FTIR, to provide the evidence for the inclusion of guest molecules in the CD cavity and interaction with the outer hydrophilic shell of the CD. The 1 H NMR spectra of MCD and HPCD alone, their complexes with EG and apiol (2:1) as well as double inclusion complexes MCD-Lev-EG and HPCD-Lev-EG in D2O are presented in Figures 3 and S9. The interactions of MCD and HPCD with EG, apiol and Lev and the structure of complexes can be characterized by induced chemical shifts Δδ, equal to the difference of chemical shifts in the complex and single substances ( Table 2). The inclusion of an apolar fragment of the guest molecule into the host hydrophobic cavity induced a shielding of the inner protons of the glucose units of MCD, namely, H3 and H5, whereas the protons on the exterior of the torus (H1, H2 and H4) were relatively unaffected ( Figure 3A-insert); this was previously shown in systems where CD formed inclusion complexes with Lev, atropine, nitrobenzene, nicardipine [35][36][37][38][39].
The effect of MCD on the chemical shift of EG protons ( Figure 3B-insert), Lev protons ( Figure 3C-insert) and apiol protons ( Figure S9) is presented in Table 2. This effect is split into two groups: first chemical shifts are shifted upfield and the other downfield. A downfield chemical shift of the drug protons indicates that they are close to an electronegative atom, oxygen [38,[40][41][42]. An upfield chemical shift is probably due to a change in the local polarity when the protons of the guest molecule are immersed in the MCD cavity (screening effect due to van der Waals forces between the drug and the carbohydrate chains of the CD). The interaction of EG with CD leads to opposite changes in the IR spectra: quenching of almost all characteristic peaks was observed ( Figure 2C). This can be explained by the possibility of apiol interacting with the outer hydrophilic shell of the CD (in addition to the CD cavity itself), which is less typical for EG, which forms an order of magnitude stronger guest-host inclusion complex. Moreover, this phenomenon depends on the amount of CD added, as we have previously demonstrated by the interaction of EG with mannosylated polymer and molecular containers [29]. Nuchuchua et al. [22] have also observed a decrease in the intensity of the EG peak in the solid-phase EG-CD spectra. In addition, there are examples of the "disappearance of guest molecules" in the CD cavity [30] or peak's quenching [31][32][33], especially for model system CD-fluoroquinolone [34]. However, at the same time, a bifurcated peak appears and increases in intensity (1738, 1730 cm −1 )-which corresponds to the overtone of deformation vibrations of C-H bonds of the terminal methylene group. Secondly, the aromatic peak of EG 1514 cm −1 shifts to 1515.5 cm −1 (Figures S4b,c and S5), which made it possible to calculate the dissociation constant of the EG-HPCD complex equal to (4.1 ± 0.3) mM. The graphs of the second derivative ( Figure S5) provided a visual representation of the change in the microenvironment of functional groups and the aromatic system of the molecule included in the CD cavity. These parameters were in a good agreement with the UV spectroscopy data as well as the shift of the peak at 205 nm to the long-wavelength region in the UV spectrum indicating that EG was included in HPCD cavity.
To level the overlap of bond fluctuations in water molecules and the effect of hydration, the spectra of the EG-HPCD complex in solid form were registered as complementary materials ( Figure S6). There are characteristic peaks of EG, which were quenched in the aqueous phase, C-C arom and C=C of the ethylene group. In the solid phase, we observed peaks corresponding to the stretching oscillations of CH 2 , CH 3 groups (2900-3000 cm −1 ) and O-H (3200-3500 cm −1 ) without solvent contribution. Figure S6 shows the distribution (significant for visualization and semi-quantitative determination of drug loading parameters EE, LC) of the integral intensity of the C-C arom peak, which characterizes the allocation of eugenol in the sample. EG was effectively (EE > 70%) included in two-thirds of the sample by quantity, in the remaining part EE exceeded 25-35%.
Visually, the dissolution of APAB and the homogenization of the system (disappearance of oil droplets) when they were included into CD (both in the MCD and in the HPCD) were observed in a light microscope (Figures S7 and S8). The dependence of the dissolution degree on the content of CD in the system was determined. Visual observation indicated that for a noticeable conversion to the complex/soluble form of EG, an equimolar amount of CD is required, and almost complete dissolution is achieved with a 5-fold molar excess. For apiol, due to its lower affinity for CD, 2-and 10-15-fold excess of CD are required, respectively. The degree of EG inclusion practically does not increase with long incubation. Since soluble complex forms of EG, apiol and other individual components of essential oils have been obtained, a quantitative study of the adjuvant and synergistic activities of APAB is relevant but previously impossible because of the stratification of substances and the formation of oil fractions. The following are the results of experiments to determine the antibacterial and synergistic effects of adjuvant-CD complexes.

NMR Spectroscopy
NMR spectroscopy is a useful tool, and complementary to FTIR, to provide the evidence for the inclusion of guest molecules in the CD cavity and interaction with the outer hydrophilic shell of the CD. The 1 H NMR spectra of MCD and HPCD alone, their complexes with EG and apiol (2:1) as well as double inclusion complexes MCD-Lev-EG and HPCD-Lev-EG in D 2 O are presented in Figure 3 and Figure S9. The interactions of MCD and HPCD with EG, apiol and Lev and the structure of complexes can be characterized by induced chemical shifts ∆δ, equal to the difference of chemical shifts in the complex and single substances ( Table 2). The inclusion of an apolar fragment of the guest molecule into the host hydrophobic cavity induced a shielding of the inner protons of the glucose units of MCD, namely, H3 and H5, whereas the protons on the exterior of the torus (H1, H2 and H4) were relatively unaffected ( Figure 3A-insert); this was previously shown in systems where CD formed inclusion complexes with Lev, atropine, nitrobenzene, nicardipine [35][36][37][38][39].
The effect of MCD on the chemical shift of EG protons ( Figure 3B-insert), Lev protons ( Figure 3C-insert) and apiol protons ( Figure S9) is presented in Table 2. This effect is split into two groups: first chemical shifts are shifted upfield and the other downfield. A downfield chemical shift of the drug protons indicates that they are close to an electronegative atom, oxygen [38,[40][41][42]. An upfield chemical shift is probably due to a change in the local polarity when the protons of the guest molecule are immersed in the MCD cavity (screening effect due to van der Waals forces between the drug and the carbohydrate chains of the CD).
It is noteworthy that inclusion in the CD cavity is more pronounced for MCD, and to a lesser extent for HPCD having hydrophilic substituents, which causes induced chemical shifts in a weak field (Table 2; Figures 3E,F and S9). Thus, an increase in the solubility of APAB and other terpenoids is achieved by including cyclodextrin in the cavity and interacting with the outer shell of the CD. The results obtained in the paper [27] show the role of methyl CD's substituents that increase the hydrophobicity of the internal cavity, which affects chemical shifts: a shift in a strong field is characteristic only for MCD-therefore, the complexes formed are stronger and the solubility of APAB is higher. According to computer simulation data, the sizes of the hydrophobic cavities CD and MCD are practically the same and approximately equal to 10.2-10.3 Å (O6-O3) ( Figure S9). At the same time, the HPCD cavity size is a bit smaller (8.9 Å) due to volumetric substituents. The height of the CD tori increases in the series CD < MCD < HPCD: 6.2, 6.6 and 8.6 Å ( Figure S9). The simulation data were consistent with the literature data calculated based on X-ray diffraction ( Figure S9h) [43]. These values confirm the preference for the inclusion of hydrophobic molecules of the APAB type and terpenoids inside the MCD rather than HPCD, given the dimensional aspect and increased hydrophilicity of HPCD, which destabilizes the complex.
The formation of complexes with CD is additionally confirmed by the chemical shifts of protons of the MCD due to theimpact of the guest molecule (Lev/EG/apiol). The upfield shifts observed for the H3 (δ EG > δ EG-Lev > δ apiol ), H5 and H6 (δ EG > δ apiol > δ EG-Lev ) protons of MCD related to water replacement by the hydrophobic aromatic benzene rings of the EG, apiol and Lev molecules inside the cavity. The greatest induced shift is characteristic of a system with EG, which indicates in favor of a greater thermodynamic benefit of complexation. The downfield shifts of protons H2, H4 MCD for EG and Lev indicate additional interactions of polar groups (-OH, -COO -, C=O) with the outer hydrophilic shell of the CD, which increases the strength of the guest-host complex. In contrast, such stabilization is practically not observed for apiol, which explains the small values of K d for apiol and higher values for EG and Lev determined by the UV and FTIR methods (Table 1).    It is noteworthy that inclusion in the CD cavity is more pronounced for MCD, and to a lesser extent for HPCD having hydrophilic substituents, which causes induced chemical      Figure 3E)  Figure 3B)

EG Proton Chemical Shifts (Insert of
EG proton δ (free EG) [41] δ (MCD-EG) δ (HPCD-EG) δ (MCD-Lev-EG) δ (HPCD-Lev-EG) Apiol Proton Chemical Shifts (Insert of Figure 3F) Apiol proton δ (free apiol) [12] δ (MCD-apiol) δ (HPCD-apiol) Induced shifts of EG protons confirm this mechanism ( Table 2). The downfield shifts of the Hd proton indicate the electrostatic interactions of the -OH group of the EG with the outer shell of the CD. The upfield shifts observed for the remaining protons confirms the inclusion of the aromatic system and the allyl group in the CD cavity. In the presence of Lev, there is competition from the guest molecules for CD, as a result of which the apparent constants of the formation of inclusion complexes decrease, which is reflected in a decrease in the magnitude of induced shifts. However, the δ(Hc) in the Lev-EG mixture is greater than for a simple EG, which is presumably due to additional π-π and π-p stacking between the allyl group of the EG and the aromatic Lev system directly inside the CD cavity. In other words, it can form a complex of inclusion of both molecules (Lev and EG) into one CD torus at once.
The interaction of Lev with MCD is characterized by the inclusion of a predominantly aromatic system inside the CD cavity, since the largest induced shifts (Table 2) are characteristic of the H5 and H8 quinolone structure. Moreover, the H14 signal is absent, which indicates that the carboxyl group is in a deprotonated state and its interaction with the outer hydrophilic shell. Induced proton shifts of the piperazine ring H2 and H6 are large, while H3 and H5 are hardly noticeable, which is explained by the location of half of the piperazine ring Lev outside the CD cavity (the molecule does not fit completely). Thus, the structure of the Lev-MCD complex is a complex of inclusion of the quinolone and half of the piperazine structure into the CD cavity and additional stabilization of the "emerging parts of the Lev" by the methyl and hydroxyl groups of the outer surface of the MCD.
The interaction of apiol with MCD leads to weak induced shifts due to low EE ( Table 2). The upfield shifts observed for protons of methyl and methylene groups of apiol confirm the inclusion of MCD in the hydrophobic cavity. In the case of the interaction of apiol with HPCD, very strong downfield shifts are observed (Table 2), indicating the interaction of apiol with the hydrophilic outer shell of HPCD and the retention of molecules by hydroxypropyl groups.
The NMR spectroscopy data were in good agreement with those obtained by FTIR. Namely, both methods confirmed the inclusion of aromatic systems of APAB and Lev in the hydrophobic cavity of the CD; the possibility of including the carboxyl group of Lev in the CD only in the protonated state or as an alternative, the deprotonated COOinteracts with the outer polar shell; the change in the microenvironment of the allyl group EG. Therefore, these methods are complementary and together allow us to study the mechanism of formation of the complexes.
Based on the above changes in the NMR spectra, we conclude that the most likely structure of the inclusion complexes EG, apiol and Lev is that shown in Figure 3D, which is in a good agreement with the data presented in the work [43].
Using spectroscopic methods, the formation of complexes of inclusion of aromatic and hydrophobic molecules with both MCD and HPCD was proved. The complex formation significantly increases solubility and protect functional groups responsible for antibacterial properties from destruction. Therefore, it is expected to increase the antibacterial efficiency of Lev, enhanced by APAB and terpenoids, in the form of complex formulations with CD.

Antibacterial and Synergistic Effects of APABs and Monoterpene Adjuvants
APAB adjuvants can be used to reduce the concentration of the potentially toxic Lev in the complex formulation due to the synergy effect, as well as to increase the therapeutic potential of the drug. Presumably, the use of complex forms of Lev in combination with APABs would increase the circulation time of Lev in the bloodstream and "save" a significant part of the drug from misuse (destruction, accelerated excretion). Indeed, EG has an antibacterial effect on some bacteria [8,9]; EG [10], apiol and its analogues [1,12] as well as monoterpenes, menthol and linalool [13], showed synergism with several antibiotics and antimicrobials (chloramphenicol, minocycline, amoxicillin and ciprofloxacin). In this paper, we studied the antibacterial effect of adjuvants (EG, apiol, dillapiol, myristicin, allyltetramethoxybenzene, linalool, menthol and safrole) in the form of inclusion complexes with MCD, as well as the effect of these formulations on the activity of Lev on Gram-positive and Gram-negative bacteria. The choice of Lev was based on its wide applicability in the treatment of a wide range of bacterial infections. In addition to APABs, two monoterpenes, linalool and menthol, were also studied as adjuvants since they were reported to exhibit antimicrobial effect [7,44,45]. A favorable adjuvant configuration was identified by comparison of compounds with different numbers and positions of substituents in the benzene ring (Table 3).

Antibacterial Effect of Lev, Lev-MCD and Adjuvants-MCD
(1) E. coli. According to the agar diffusion test, Lev in its free form inhibited growth of E. coli with MIC = 0.1 µg/mL, whereas the application of the Lev inclusion complex in MCD reduced the MIC value to 0.06 µg/mL (Table 3). A similar tendency was observed according to the broth micro-dilution test, where MCD improved the effect of Lev by 30%. This phenomenon can be explained by defects in cell membranes arising due to CD, and as a consequence, an increase in the penetration of the Lev through the bacterial plasma membrane (as shown by Tychinina et al. [18,46] on the model experiment with liposome). The presented values of MIC50 and MIC90 determined for APAB and terpenoids in the form of inclusion complexes with MCD by broth micro-dilution technique were less in terms of diffusion in agar due to differences in the experiment conditions due to a slow diffusion rate in agar and a large bacterial seeding area. The values obtained by the two methods correlated with each other in terms of relative characteristics. Comparison of adjuvant MIC values yielded EG as the optimum molecular scaffold, namely, allylbenzene with 3-methoxy-4-hydroxy substituents. Safrole with a methylenedioxy moiety was second in the activity range, whereas apiol and dillapiol featuring a dimethoxy-methylenedioxy fragment were less potent ( Table 3). The antimicrobial effect of methylenedioxy-containing APABs might be attributed to their facile hydrolysis in cells yielding molecules with a hydroxy-methoxysubstituted benzene ring [1]. CD complexed with myristicin, allyltetramethoxybenzene, and linalool were inactive on both bacteria species up to 3-5 mg/mL concentration. Thus, the ranked list of substances according to their activity against E. coli is the following: Lev-MCD > Lev >>> EG-MCD > dillapiol-MCD ≈ apiol-MCD > myristicin-MCD >> other (no significance).
(2) B. subtilis. According to the agar diffusion test, it was found that Lev and adjuvants inhibited growth of Gram-positive B. subtilis at higher concentrations than required for Gram-negative bacteria (Table 3, Figure S10). Menthol was the only exception. It failed to exhibit an effect on Gram-negative bacteria, while it suppressed the growth of Grampositive B. subtilis, although at a relatively high concentration (MIC = 2.6 mg/L). Lev in its free form inhibited growth of B. subtilis with MIC = 0.45 µg/mL, whereas application of the Lev inclusion complex in MCD reduced the MIC value to 0.25 µg/mL. According to the broth micro-dilution method, MCD reduced MIC 50 of Lev by half, and MIC 90 by more than a third. Among the APAB and terpenoids studied, only EG (MIC = 1 mg/mL), apiol (MIC = 5.3 mg/mL), safrole (MIC = 3.9 mg/mL) and menthol (MIC = 2.6 mg/mL) showed an inhibitory effect on the growth of Gram-positive bacteria in reasonable concentrations. According to the broth micro-dilution method, safrole turned out to be more active than menthol, which is probably due to the slow diffusion of the latter in the agarose gel. The structure-activity relationship for adjuvants on Gram-positive and Gram-negative bacteria (considered for E. coli) was similar (Table 3). Thus, a ranked list of substances according to their activity against B. subtilis is the following: Lev-MCD > Lev >>> EG-MCD > safrole-MCD ≈ menthol-MCD > myristicin-MCD >> other (no significance).

Synergy of Adjuvants with Lev
We obtained complex formulations containing adjuvants-MCD and Lev-MCD. These complexes were referred to as enhanced Lev. The synergy effect was studied at a constant concentration of Lev and a variable concentration of an enhancer. Table 4 shows the diameters (D Lev and D) of E. coli and B. subtilis growth inhibition zones surrounding Lev-MCD and Lev-MCD with adjuvant-MCD, respectively. Based on the D/D Lev ratio, the effectiveness of cooperative antibacterial action (φ) of the adjuvant and Lev was calculated. The fractional inhibitory concentration (FIC) parameter was introduced: at a fixed Lev concentration, the FIC of adjuvant corresponds to the minimum fractional adjuvant concentration at which the antibacterial effect was still manifest (MIC adjuvant in the presence of Lev) [47]. The fractional inhibitory concentration index (FICI) [27] characterizes the ratio of the active Lev and adjuvant concentrations to the MIC values of these substances individually: As discussed above [13], the addition of APAB-CD and terpenoid-CD complexes to an antibacterial drug reduced its MIC, resulting in the drug dosage decrease and, consequently, reducing the systemic toxicity and development of drug-resistant pathogens.
(1) E. coli. Significant enhancement of Lev inhibitory effect corresponded to a decrease of MIC Lev : up to twice in the presence of menthol ( Figure S11) and apiol ( Figure S12), up to 1.25-1.5 times when dillapiol, myristicin and allyltetramethoxybenzene were used as adjuvants. The most powerful effect was shown for safrole and EG (Figure 4a): MIC Lev can be reduced from 0.10 to 0.015-0.02 µg/mL.
The greatest increase in antibacterial power φ was demonstrated using the following combinations (Table 4) Table 3). The adjuvant effect correlates with the individual antibacterial activity of APABs, while the resulting enhancement of the Lev effect is not limited to a simple sum of activities, but is characterized by significant synergism.
FICI values characterize the type of Lev interaction with APABs or terpenoid. Almost all combined forms studied were characterized by an enhancing of the effect compared to the action of individual components. The additivity type already indicated a good potential of a pair of components, and the synergism type indicated a significant decrease in the fractional concentration of one of the components. The desired reduction of the MIC of potentially toxic Lev was demonstrated by EG, apiol, dillapiol, myristicin, menthol and safrole. The most powerful synergist, EG, exhibited maximal values of the two essential parameters, φ and FICI, at the relevant Lev concentrations of 0.5-2 MIC. However, the increase in Lev concentration up to >8-10 MIC resulted in indifference or even antagonism due to a saturation effect.
The graphical dependences of the E. coli growth inhibition zone diameters by Lev at varying adjuvant concentrations (from 60 µg/mL to 1-3.3 mg/mL) are shown in Figure 5. The greatest synergistic effect corresponded to safrole, apiol and menthol ( Figure 5, Table 4). In the first experiment (Figure 4a), the Lev concentrations were selected in such a way that Lev concentration without an adjuvant was close to the MIC. In the second experiment ( Figure 5b), inhibition by Lev itself was not achieved; however, the addition of adjuvants resulted in significant increase inhibition zone diameters from 10 to 13-14.5 mm. Table 4. Parameters of E. coli and B. subtilis growth inhibition by mixtures of Lev-MCD and adjuvant-MCD (1:2). D Lev and D-diameter of bacterial inhibition zone surrounding Lev-MCD and Lev-MCD with APAB-MCD, respectively. φ Lev -cooperative antibacterial action of APABs or terpenoids and Lev. The conditions and adjuvant concentrations are similar to those given in Table 3. Values are presented as the mean ± SD of three experiments.  ; *** NI-no inhibition.
The strongest synergist of levofloxacin among the APAB and terpenoids was eugenol (Table 4). Clearly and in detail this effect is presented in the form of 3D dependence of the inhibitory effect of the drug composition on the concentration of Lev and EG (Figure 6a): yellow area-weak inhibition, orange-medium, red-significant inhibition. MIC of Lev-MCD on E. coli was reduced from 0.06 to 0.015-0.02 µg/mL due to introduction of 1 mg/mL of the adjuvant EG-MCD. At minimal Lev concentrations the greatest adjuvant effect of EG (a 3-fold amplification effect) is achieved due to synergism.
(2) Bacillus subtilis. The synergistic effect of EG-MCD ( Figure S13), safrole-MCD ( Figure S14) and menthol-MCD (Figure 4b) with Lev-MCD was also confirmed on Grampositive bacteria (Table 4). All three adjuvants demonstrated an increase in antibacterial activity with an efficiency parameter φ from 1.1 to 1.5; the most powerful effect was observed for menthol: with φ value of 2.4. Moreover, the type of interaction of safrole and menthol with Lev was a «bright synergy» (FICI-0.08-0.11). However, in the case of Gram-positive cells, eugenol showed only additivity. Figure 6b shows the 3D dependence of the inhibitory effect of the drug composition on Lev and menthol concentrations: purple area-weak inhibition, blue and green-medium, and yellow-red-significant inhibition. Menthol-MCD at 1-3 mg/mL was identified as the strongest enhancer of Lev. MIC Lev reduced from 0.45 to 0.15 µg/mL. However, EG-MCD was the strongest B. subtilis growth inhibitor itself. Presumably, the antibacterial properties of eugenol are due to its ability to destroy cell membranes; in contrast, the adjuvant power of menthol is explained by the predominant intracellular action. Facilitating drug permeability through bacterial membrane is a crucial factor for an antibacterial effect on Gram-negative bacteria. As an example, EG demonstrated the greatest antibacterial activity along with pronounced membrane destruction, which resulted in a powerful synergy for a number of other antibiotics [10].  The graphical dependences of the E. coli growth inhibition zone diameters by Lev at varying adjuvant concentrations (from 60 μg/mL to 1-3.3 mg/mL) are shown in Figure 5. The greatest synergistic effect corresponded to safrole, apiol and menthol ( Figure 5, Table  4). In the first experiment (Figure 4a), the Lev concentrations were selected in such a way that Lev concentration without an adjuvant was close to the MIC. In the second experiment (Figure 5b), inhibition by Lev itself was not achieved; however, the addition of adjuvants resulted in significant increase inhibition zone diameters from 10 to 13-14.5 mm. The strongest synergist of levofloxacin among the APAB and terpenoids was eugenol (Table 4). Clearly and in detail this effect is presented in the form of 3D dependence of the inhibitory effect of the drug composition on the concentration of Lev and EG (Figure 6a): yellow area-weak inhibition, orange-medium, red-significant inhibition. MIC of Lev-MCD on E. coli was reduced from 0.06 to 0.015-0.02 μg/mL due to introduction of 1 mg/mL of the adjuvant EG-MCD. At minimal Lev concentrations the greatest adjuvant effect of EG (a 3-fold amplification effect) is achieved due to synergism. According to the literature data, the mechanism of action of essential oil components (such as terpenoids and APABs) is not the simple addition of the effects of Lev and the adjuvant, but the synergy effect that occurs [10,25,26,48]. The proposed mechanism of action of adjuvants is based on several aspects: (i) formation of bacterial membrane defects by an aromatic molecule due to its ability to interfere with the synthesis of the cell wall and cytoplasmic membrane, eventually leading to a leakage of intracellular material and, as a consequence, to the increased penetration of the antibiotics and antimicrobials (flu-conazole, amphotericin, azithromycin, ciprofloxacin chloramphenicol) into the cell [10,44]; (ii) inhibition of bacterial efflux pumps resulting in an increase in the intracellular drug concentration reported previously for doxorubicin [49]; (iii) inhibition of bacterial enzymes, including dehydrogenases; (iv) suppression of the production of bacterial virulence factors, such as violacein, elastase, and pyocyanin, that prevents biofilm formation [50]. Importantly, similarly to eukaryotic cells, bacteria have multidrug transporter proteins responsible for transmembrane drug efflux and multidrug resistance development [51]. In this respect, it should be noted that apiol was reported to exhibit synergistic cytotoxic effects with doxorubicin and vincristine associated with blocking the P-glycoprotein transmembrane efflux pump [52].
In summary, the experiments have demonstrated a significant potential of APAB-CD and terpenoid-CD complexes to enhance the antibacterial effect of Lev. The results can be considered promising for the further design of novel drug formulations with improved characteristics, such as low toxicity and increased therapeutic effect.

Preparation of β-Cyclodextrin Inclusion Complexes
(1) EG and apiol inclusion complexes with HPCD were obtained by mixing a drop of oil (1-2 mg, apiol and menthol were melted at 40 • C) with a solution of HPCD (0.23 mM) or MCD (50 and 100 mg/mL) to a total volume of 0.4 mL. The molar excess of CD varied from 0.04 to 10). Further, the complexes were incubated for 1 h at 37 • C and then centrifuged (rpm, 15 min, Eppendorf tubes) to separate the insoluble fractions. Complexes were studied by UV and FTIR spectroscopy. MCD was selected for experiments on bacterial cells due to higher LC and EE parameters.
(2) Lev, EG, apiol, dillapiol, myristicin, allyltetramethoxybenzene, linalool, menthol and safrole inclusion complexes with MCD (molar ratio 1:2) were prepared by mixing corresponding samples (approximately 2-5 mg per aromatic substance and 25-50 mg per MCD) followed by addition of 50 µL of acetonitrile. Then the mixture was intensively rubbed (kneading method), and if necessary, 25 µL of acetonitrile was added again and the procedure was repeated. Finally, incubation for 1 h at 37 • C was performed.

UV Spectroscopy
UV spectra of solutions (point 3.3.1) were recorded on the AmerSham Biosciences UltraSpec 2100 pro device (USA) three times in the range of 200-400 nm in a quartz cell Hellma 100-QS with an optical path of 1 cm. Background spectrum (5 mM sodium phosphate buffer (pH 6.2) with 50% EtOH) was subtracted as a blank. Calibration dependences are shown in Figure 1C.

FTIR Spectroscopy
ATR-FTIR spectra of samples' solutions (point 3.3.1) were recorded using a Bruker Tensor 27 spectrometer equipped with a liquid nitrogen cooled MCT (mercury cadmium telluride) detector. Samples were placed in a thermostatic cell BioATR-II with ZnSe ATR element (Bruker, Ettlingen, Germany). The IR spectrometer was merged with a constant flow of N 2 (NiGen HF-1, Clint, Italy). FTIR spectra were acquired from 900 to 3000 cm −1 with 1 cm −1 spectral resolution. For each spectrum, 50-70 scans were accumulated at 20 kHz scanning speed and averaged. ATR-FTIR spectra of solid samples placed on KBr glass were recorded using a Bruker Lumos II IR microscope in the region from 700 to 4000 cm −1 with 1 cm −1 spectral resolution with scanning in the area on average 1 × 1 microns. Spectral data were processed using the Bruker software system Opus 8.2.28 (Bruker, Germany), which includes linear blank subtraction, baseline correction, differentiation (second order, 17-21 smoothing points), min-max normalization and atmosphere compensation [29,56]. If necessary, 5-point Savitsky-Golay smoothing was used to remove noise. Peaks were identified by standard Bruker picking-peak procedure.
The study of the antibacterial effect of Lev and adjuvants was carried out by (a) agar diffusion test and (b) broth micro-dilution method. (a) A measured 500 µL of 100-fold culture dilution was evenly distributed over a solid nutrient medium (LB, pH 7.2) on a Petri dish. After 20 min, agar disks with a diameter of 9 mm were removed from the cups and 50 µL of the test samples were placed in the formed wells at the required concentration in a sterile PBS buffer (see Section 3.3 (2)). After 30 min, Petri dishes were placed in a 37 • C thermostat. After 24 h, the diameters of the growth inhibition zones were measured. Minimum inhibitory concentrations (MICs) are determined as the infimum of multiple concentrations of an antimicrobial that inhibits the visible growth of a microorganism after overnight incubation. (b) According to the standard broth micro-dilution method, 200 µL of the overnight culture (5 × 10 6 CFU) was placed into a 96 micro-titer plate wells. Then 50 µL of tested samples was added. The range of tested Lev concentration was 4-200 ng/mL, adjuvants concentration was in the range of 0.008-2 mg/mL. Optical density measurements at 600 nm were taken at various times (0, 3, and 18 h) to determine the number of cells. The dependence of CFU (determined by A 600 ) on the concentration of Lev, APAB or terpenoid (briefly X) was approximated according to the Boltzmann equation (Origin software): CFU = (control CFU)/(1 + exp(([X] − MIC 50 )/constant value)). MIC 50 and MIC 90 are defined as concentrations at which the inhibition of bacterial growth by 50 and 90% is noticeable, respectively. In other words, the concentration at which the ordinate value (CFU) is equal to 50 or 10% of the control sample minus the background.

Mathematical Calculations and Equations
(1) Calculation of the dissociation constants X-MCD and X-HPCD, where X is a "guest" compound, considering the equilibrium: X aq + n MCD aq ↔ X·nMCD aq , where K d = [MCD aq ] n · [X aq ]/[X·nMCD aq ].
Calculations for the following items are performed when [MCD] = 0.01 M.
(3) Solubility of X in water C 0 = [X aq ] = A/(ε · l) was determined by UV absorbance after centrifugation of samples (10 min, 10,000× g). l = 1 cm. ε-molar absorbance coefficient ( Figure 1C).  4). The essence of this expression is to calculate the average change in the effective concentration compared to MIC. Note that the most important parameter corresponds to a decrease in MIC of Lev. However, FICI is designed to determine the type of interaction Lev with adjuvants.
(8) Statistical analysis of obtained data was carried out using the Student's t-test Origin 2022 software (OriginLab Corporation). Values are presented as the mean ± SD of three experiments. (9) The diameters of the hydrophobic cavities CD, MD and HPCD, as well as the heights of the rounds were calculated using a computer simulation in the Avogadro program (https://avogadro.cc/ accessed on 1 July 2022) using the GAFF force field, the steepest descent (n = 2000).

Conclusions
The formation of non-covalent guest-host complexes with CDs was previously known as a way to increase the solubility of poorly soluble organic molecules in an aqueous medium. Here, we have successfully applied this approach to a promising class of organic molecules-potential antibiotic enhancers, namely, EG, apiol, dillapiol, myristicin, allyltetramethoxybenzene, linalool, menthol and safrole.
In the presence of a molar excess of CD (up to 5-10 fold), it was possible to achieve complete dissolution of APABs and terpenoids in an aqueous medium (at 80-98% encapsulation) higher by 10 1 -10 3 times. The molecular details of the observed interaction were established using FTIR, 1 H MNR and UV spectroscopy. It turned out that CD and APAB form supramolecular structures, where the most hydrophobic aromatic and allyl parts of the APAB molecule are located in the depth of the hydrophobic cavity of the CD, whereas the most hydrophilic OH group actively interacts with the external environment. Antimicrobial terpenoids and APABs were used as adjuvants capable of modifying antibiotic potential.
The structure-activity relationship study of adjuvants with different numbers and positions of benzene ring substituents identified the optimal structure of synergic (or enhancing) molecules, namely, allyl-3-methoxy-4-hydroxybenzene, EG, which decreases Lev MIC by 3-4 times. Safrole with methylenedioxy moiety showed a somewhat weaker effect.
A synergistic antibacterial effect of APABs with Lev was detected when the Lev concentration was equal to MIC or even less. This effect was demonstrated on both Grampositive and Gram-negative bacteria, suggesting a complex mechanism of the observed synergy that requires further elucidation.
The observed synergy between Lev and the adjuvants is the main outcome of the study. Specifically, the two-three-fold increase in Lev efficacy caused by EG and menthol in inclusion complexes with MCD seems to be very promising in terms of reducing the dosage of toxic drugs, not only to increase the antimicrobial therapeutic effect but also for convenience in drug formulation handling. A deeper understanding of the mechanisms of these synergistic effects could facilitate further design of more effective and safe adjuvants for both existing and developed drugs.  Figure S3. Entrapment efficiency (EE) of apiol (a) and eugenol (b) into (2-hydroxypropyl)-β-cyclodextrin (HPCD) in aqueous solutions at 22 • C and A 280 . Hill plots. The EE plateau is reached at a level lower than 100% due to the partial presence of EG and apiol in dissolved form in water, not in the form of a complex with CD. However, solubility efficiency is aiming for 100%. C0-theoretical concentration under the assumption that 100% of the substance has dissolved, Cmax-the maximum practical achievable concentration; Figure S4.