Nuclear Magnetic Resonance Spectroscopic and Thermodynamic Characterization of Hydroxypropyl-cyclodextrin Inclusion Complexes with Bioactive Molecules from Plant Mother Tinctures

Abstract

This study investigates the formation of inclusion complexes using the cyclodextrins, specifically HP-β-CD and HP-γ-CD, with bioactive molecules that mask the unpleasant tastes found in herbal extracts, such as horehound, eucalyptus, plantain, and thyme. The selected bioactive molecules (BAMs), including thymol, carvacrol, and eucalyptol, were subjected to inclusion complexation with host hydroxypropyl-cyclodextrins (HP-CDs) and studied by ¹H nuclear magnetic resonance (NMR) spectroscopy. The stoichiometry of the complexes was determined to be 1:1 (HP-CD:BAM). The analysis of ¹H chemical shift changes in BAM NMR spectra allowed the calculation of binding constants, indicating better stability with HP-γ-CD due to its larger cavity size and enhanced flexibility, which favors stronger and more stable inclusion with bioactive molecules. Two-dimensional rotating-frame Overhauser effect spectroscopy experiments provided insight into the binding mode. A simple thermodynamic model highlighted the hydrogen bonding contribution to complexation. Furthermore, the inclusion complex of the carvacrol and eucalyptol mixture showed that carvacrol consistently maintains the most stable complex, even in the presence of both active compounds combined with HP-CD.

Highlights

• 1. ¹H NMR spectroscopy was used to investigate the formation of inclusion complexes of cyclodextrins and bioactive molecules from plant mother tinctures.
• 2.The stoichiometry of inclusion complexes was 1:1 for all the studied bioactive molecules.
• 3.The most stable inclusion complexes are formed with hydroxypropyl-γ-cyclodextrin.
• 4.Hydroxypropyl-γ-cyclodextrin is a promising candidate for masking the unpleasant taste of mother tinctures.

1. Introduction

The plant world is considered a true mosaic due to its diversity of natural resources, with an extremely important contribution of bioactive molecules used in numerous cosmetic, food, and pharmaceutical fields. Indeed, medicinal plants remain a reliable source of active molecules known for their therapeutic properties. These plants are traditionally used by oral or topical administration in the form of infusions, capsules, alcoholic extracts, tinctures, and extracts. Among these medicinal plants, those endemic to the Mediterranean region, such as eucalyptus, horehound, plantain, and thyme, are traditionally used in preparations to relieve coughs and treat respiratory infections. − The combination of an active pharmaceutical ingredient with a specific dosage form is intended to optimize its administration by enhancing absorption, safeguarding or stabilizing the compound to extend its presence in the body, and, in some cases, directing it to a particular site of action. When developing pharmaceutical products, taste frequently plays an important role, especially since many active substances have a bitter flavor that can be problematic for pediatric oral medications. To address this, the pharmaceutical industry employs a variety of taste-masking strategies. Commonly, sweetening agents such as sugars, sweet-tasting polyols (like glycerol, sorbitol, and xylitol), and artificial sweeteners are incorporated into formulations like syrups to improve palatability. Nevertheless, when the bitterness is particularly strong, these additives may be insufficient and must be combined with other agents such as anesthetics or cooling excipients. Despite their potential, taste blockers and flavor modifiers are not widely used in pharmaceutical development. One effective solution involves encapsulating bioactive compounds within cyclodextrins (CDs), which can significantly enhance the taste profile of medications by concealing their undesirable sensory attributes. Cyclodextrins are cyclic oligosaccharides produced through the enzymatic breakdown of starch. They feature a distinctive structure with a hydrophobic cavity surrounded by a hydrophilic exterior, , enabling them to form inclusion complexes with a wide range of organic and inorganic molecules. , CDs hide the unpleasant tastes by complexing with active chemicals and preventing bitter functional groups from engaging directly with the taste receptors on the tongue. Hydroxypropyl-substituted cyclodextrins (HP-CDs), such as HP-β-CD and HP-γ-CD, are particularly suitable for masking undesirable flavors in plant-based formulations due to their improved aqueous solubility and complexation efficiency, making them especially suitable for masking undesirable flavors in plant-based formulations. In this study, the unpleasant tastes of four mother tinctures (thyme, eucalyptus, plantain, and horehound) were masked by encapsulating their active ingredients with hydroxypropyl-cyclodextrins, thanks to the latter’s ability to form molecular inclusion complexes. Each mother tincture used in this study contains several bioactive molecules, many of which have an unpleasant taste. As indicated by the referenced studies, specific molecules responsible for these undesirable tastes were identified in each mother tincture (Table ). Thymol, carvacrol and eucalyptol have been identified as significant contributors to the unpalatable taste associated with thyme (TMTH), plantain (TMPL), eucalyptus (TMEU), and horehound (TMMA). − Thymol and carvacrol are both monoterpene phenols found in the extracts of aromatic plants like thyme and plantain. These compounds share a nearly identical chemical structure, consisting of an aromatic ring and a differently positioned hydroxyl group, which gives them phenolic properties. These active ingredients are renowned for their wide range of applications in both traditional medicine and phytotherapy. Thymol has strong antiseptic and antimicrobial properties, providing the distinctive, strong flavor of the culinary herb thyme. Carvacrol, on the other hand, has a thymol odor. They have a strong taste and smell that can be felt unpleasant. − Eucalyptol has a pungent, spicy odor typical of cloves and a slightly acidic taste. Belonging to the category of bicyclic monoterpenes, this molecule is valuable in a wide range of applications, especially in the pharmaceutical field due to its anti-inflammatory, antimicrobial, antiviral, and antioxidant activities. As part of this study, bioactive molecules such as thymol, carvacrol, and eucalyptol were selected as target compounds for complexation with hydroxypropyl-substituted cyclodextrins (HP-CDs) (Figure ). These molecules were chosen due to their contribution to the undesirable taste associated with each mother tincture. − , By encapsulating these bioactive molecules with HP-CDs, the inclusion complexation process not only helps to reduce their unpleasant sensory properties but also provides a valuable molecular model to investigate host–guest interactions. In this work, we focus on a detailed ¹H NMR study to explore the complexation behavior of selected bioactive compounds responsible for the unpleasant taste of plant mother tinctures. The stoichiometry and binding constants of the inclusion complexes were determined through chemical shift analysis, while two-dimensional rotating-frame Overhauser effect spectroscopy nuclear magnetic resonance (2D ROESY NMR) provided spatial insights into the host–guest architecture. Additionally, competitive binding experiments were performed to assess molecular selectivity and stability within binary mixtures. These NMR-based investigations provide a structural and thermodynamic understanding of the inclusion phenomena, which contributes to the rational design of taste-masked pharmaceutical formulations using cyclodextrins.

1. Abbreviations of Mother Tinctures Used in the Study and Their Corresponding Most Common Bioactive Molecule According to the Bibliography.

mother tinctures (MT)	abbreviation	most common bioactive molecule	refs.
eucalyptus	TMEU	eucalyptol	and
horehound	TMMA	carvacrol, thymol
plantain	TMPL	eucalyptol
thyme	TMTH	carvacrol, eucalyptol

1.

Structure of (A) hydroxypropyl-substituted cyclodextrins and (B) bioactive molecules used in this study.

2. Experimental Section

2.1. Materials and Methods

Table lists the cyclodextrins and active compounds used in this study, along with their suppliers, acronyms, and some of their physicochemical properties. Both hydroxypropyl-cyclodextrins were vacuum-dried overnight at 120 °C before use. Deuterium oxide (D₂O, 99.8%) was purchased from Eurisotop. We used a Bruker Avance III 500 Ultra-Shield Plus spectrometer to record ¹H NMR spectra using a solution of D₂O. 128 scans were used to record classical 1D ¹H NMR spectra. S stands for singlet, d for doublet, t for triplet, and m for multiplet. The ROESYGPPH and NOESYGPPH sequences from the Bruker library were used to perform 2D ROESY and 2D NOESY correlation spectra with the following acquisition parameters: mixing time = 800 ms, relaxation delay = 1 s, 90° pulse width = 9.8 μs, and spectral width = 4800 Hz (giving an acquisition time of 0.2 s).

2. Characteristics and Abbreviations of Cyclodextrins and Active Molecules Used in the Study .

cyclodextrin and active molecules	abbreviation	suppliers	molar mass (g mol–1)	aqueous solubility at 25 °C (g L–1)	ref.
hydroxypropyl-β-cyclodextrin	HP-β-CD	Acros Organics (DS = 0.65)	1400	>600
hydroxypropyl-γ-cyclodextrin	HP-γ-CD	Acros Organics (DS = 0.60)	1576	>500
thymol	THY	Thermo Scientific (purity ≥98.5%)	150.22	0.9
eucalyptol	EUC	Thermo Scientific (purity 99%)	154.249	3.5
carvacrol	CV	Acros Organics (purity 99%)	150.217	1.25

^aDS: degree of substitution.

2.2. Maceration Procedure of Mother Tinctures

The preparation of the mother tinctures involved macerating fresh plant material in 65° ethanol for three weeks. The ratio of plant matter to alcohol used in the maceration process was 1:10 (m/V). After the maceration period, the mixture was pressed and filtered to remove any inactive residue. The resulting mother tinctures contained concentrated active compounds from the plant, ensuring their therapeutic efficacy.

2.3. NMR Solvent Removal Technique

In order to reduce the intensity of the solvent peaks and to highlight the presence of compounds with low-intensity peaks, the “solvent removal” technique was used during the analysis of the plant mother tincture. When samples contain a substantial amount of water, this approach is especially useful. Indeed, the presence of water can mask signals from guest molecules, making detection difficult or even impossible when their peaks are overlapping solvent peaks. Water suppression was achieved using the one-dimensional nuclear Overhauser effect spectroscopy (1D NOESY) pulse sequence, which incorporated presaturation during the relaxation delay and mixing time (150 ms). The presaturation power was set to the minimum required to effectively suppress the water peak. The receiver gain setting for each pulse sequence was manually adjusted during preliminary experiments and kept consistent across all spectra.

2.4. Experimental Approach to Host–Guest Complexation Investigations

Thus, with the total concentrations of G and CD being denoted as [G]_t and [CD]_t, respectively, the mole fraction of G is x = [G]/([G] + [CD]). The sum of the CD and G concentrations was equal to 3.99 × 10^–3 M, which corresponds to 1.2 × 10^–3mmol for ([CV]_t + [CD]_t), for ([THY]_t + [CD]_t), and for ([EUC]_t + [CD]_t). First, the maximum of the bell-shaped Job plot of (δ_exp – δ₀)x against x gives the stoichiometry m:n of the inclusion complexes G _m :CD _n ; a maximum at x = 0.5 indicates a 1:1 stoichiometry. Second, modeling the Job plot provides the equilibrium constant K _nm by fitting the model’s prediction to the experimental chemical shifts as explained in Section . The algorithmic method involves collecting spectra of solutions with a constant concentration of active molecules (G) and varying concentrations of cyclodextrin, in order to determine the binding constant of cyclodextrins with guest molecules (G) by fitting the same model to the experimental data. However, this is not the original algorithmic method; because the concentration range of CD was not restricted to [CD]_t ≫ [G]_t, a wide range of [CD]_t was used and full calculations of the concentrations of CD, G, and CD _n :G _m allowed fitting the theoretical variation of chemical shifts to the experimental data. The concentration of the host [CD]_t was varied from 1.99 to 15.97 mM, which is equivalent to 0.01–0.08 mmol for a guest concentration of 3.99 mM based on these molar ratios (CD/G): 0.5, 1, 1.5, 2, 2.5, 3, 3.5, and 4. For all measurements, to ensure complete solubilization, the correct amounts of cyclodextrin and guest molecule were weighed out accurately and dissolved in an appropriate solvent. The samples were then kept under magnetic stirring for 24 h at room temperature until equilibrium was achieved. The stoichiometry and stability constants of the inclusion complexes were then determined by analyzing the resulting solutions.

2.5. Process of Establishing Equilibrium Binding Constants

Most methods for determining the association constant (K _mn) are graphical, allowing linear relationship to be established between the chemical shifts and the constant K _mn. The binding constant for the complex was calculated using a mathematical model based on the NMR studies used for the Job plot. A theoretical approach has therefore been developed to determine the pair of values of K ₁₁ and ${\delta }_{{\mathrm{CD}}_n:{\mathrm{G}}_m}$ for which the theoretical curve approaches the measured values. This method does not involve any approximation; it is based on the material conservation equations for the involved entities.

In general, the formation of the inclusion complex of G with CD is a reversible process, and it is described by the following equation:

$$m\mathrm{G}+n\mathrm{CD}\stackrel{K_{\mathrm{mn}}}{\rightleftharpoons }{\mathrm{CD}}_n:{\mathrm{G}}_m$$ 1

where m and n are the stoichiometric coefficients and K _mn is the equilibrium binding constant.

The affinity of the guest (G) for the CD cavity can be studied by measuring the association constant. A popular algorithmic method for determining the association constant K ₁₁ is a graphical method that relies on the linear relationship between the chemical shift variation and the total concentration of CD, when the latter is small enough with respect to that of the guest for the linear approximation is valid. Therefore, this method is restricted to a limited range of concentrations and only applies to a 1:1 stoichiometry. In this study, an exact calculation of the concentrations of all species was done according to an equilibrium model described by a mass action law. The theoretical chemical shift variation was then calculated for each mixing scheme.

At the equilibrium, the formation of the 1:1-type complex means that the active molecule and cyclodextrin are in two states: the free form (G) and the complex form (CD:G).

In the case of fast exchange between the free and complexed forms, the observed chemical shift of the guest molecule G is the average value of the chemical shifts of the free (δ_G) and complexed ( ${\delta }_{{\mathrm{CD}}_n:{\mathrm{G}}_m}$ ) forms weighted by their respective mole fractions.

$${\delta }_{\mathrm{G},\mathrm{obs}}=x_{\mathrm{G}}{\delta }_{\mathrm{G}}+m{x}_{{\mathrm{CD}}_n:{\mathrm{G}}_m}{\delta }_{{\mathrm{CD}}_n:{\mathrm{G}}_m}=\frac{[\mathrm{G}]}{{[\mathrm{G}]}_{\mathrm{t}}}{\delta }_{\mathrm{G}}+m\frac{[{\mathrm{CD}}_n:{\mathrm{G}}_m]}{{[\mathrm{G}]}_{\mathrm{t}}}{\delta }_{{\mathrm{CD}}_n:{\mathrm{G}}_m}$$ 2

In the case of a 1:1 stoichiometry, the complexation equilibrium is given by

$$\mathrm{G}+\mathrm{CD}\stackrel{K_{11}}{\rightleftharpoons }\mathrm{CD}:\mathrm{G}$$ 3

$$K_{11}=\frac{[\mathrm{CD}:\mathrm{G}]}{[\mathrm{G}][\mathrm{CD}]}$$ 4

The concentrations [G] and [CD:G] were determined from the total concentrations of the active molecule [G]_tot and the cyclodextrin [CD]_tot by solving the system of three equations:

$${[\mathrm{G}]}_{\mathrm{t}}=[\mathrm{G}]+[\mathrm{CD}:\mathrm{G}]$$ 5

$${[\mathrm{CD}]}_{\mathrm{t}}=[\mathrm{CD}]+[\mathrm{CD}:\mathrm{G}]$$ 6

The system of three equations with three variables is solved for [CD:G], yielding a second-degree equation:

$${[\mathrm{CD}:\mathrm{G}]}^2-({[\mathrm{G}]}_{\mathrm{t}}+{[\mathrm{CD}]}_{\mathrm{t}}+\frac{1}{K_{11}})[\mathrm{CD}:\mathrm{G}]+{[\mathrm{G}]}_{\mathrm{t}}{[\mathrm{CD}]}_{\mathrm{t}}=0$$ 7

where the solution is

$$[\mathrm{CD}:\mathrm{G}]=\frac{1}{2}\{({[\mathrm{G}]}_{\mathrm{t}}+{[\mathrm{CD}]}_{\mathrm{t}}+\frac{1}{K_{11}})-\sqrt{{({[\mathrm{G}]}_{\mathrm{t}}+{[\mathrm{CD}]}_{\mathrm{t}}+\frac{1}{K_{11}})}^2-4{[\mathrm{G}]}_{\mathrm{t}}{[\mathrm{CD}]}_{\mathrm{t}}}\}$$ 8

The chemical shift δ_G was determined from a solution of pure G, and the pair of variables (K ₁₁ and δ_CD:G) was adjusted to achieve the most accurate fit of the model to the observed chemical shift data for both the Job and algorithmic methods. The complexation constant and the chemical shifts of the complex, as derived from NMR spectra were adjusted to minimize the average relative error (ARE) for n data points and p adjustable parameters using the nonlinear GRG algorithm of the Excel Solver.

$$\mathrm{ARE}=\frac{1}{n-p}\sum _{i=1}^n\left|\frac{{\delta }_{\exp }-{\delta }_{\mathrm{calc}}}{{\delta }_{\exp }}\right|$$ 9

The number of data points is the product of the number of samples of different concentrations and the number of NMR lines for which the chemical shift was measured. A global fitting procedure was applied to each bioactive molecule, using a single binding constant (K ₁₁) to fit all the chemical shift variations simultaneously across multiple proton signals. This approach provides a more robust and consistent estimation of the binding constant.

On the same footing, the case of competitive complexation of a mixture of two guests by a CD model of 1:1 complexation was analyzed using the following equations:

$${\mathrm{G}}_1+\mathrm{CD}\stackrel{K_1}{\rightleftharpoons }\mathrm{CD}:{\mathrm{G}}_1{K}_1=\frac{[\mathrm{CD}:{\mathrm{G}}_1]}{[{\mathrm{G}}_1][\mathrm{CD}]}$$ 10

$${\mathrm{G}}_2+\mathrm{CD}\stackrel{K_2}{\rightleftharpoons }\mathrm{CD}:{\mathrm{G}}_2{K}_2=\frac{[\mathrm{CD}:{\mathrm{G}}_2]}{[{\mathrm{G}}_2][\mathrm{CD}]}$$ 11

$$\mathrm{concentration}\mathrm{balance}\mathrm{of}\mathrm{G}1:{[{\mathrm{G}}_1]}_{\mathrm{t}}=[{\mathrm{G}}_1]+[\mathrm{CD}:{\mathrm{G}}_1]$$ 12

$$\mathrm{concentration}\mathrm{balance}\mathrm{of}\mathrm{G}2:{[{\mathrm{G}}_2]}_{\mathrm{t}}=[{\mathrm{G}}_2]+[\mathrm{CD}:{\mathrm{G}}_2]$$ 13

$$\mathrm{concentration}\mathrm{balance}\mathrm{of}\mathrm{CD}:{[\mathrm{CD}]}_{\mathrm{t}}=[\mathrm{CD}]+[\mathrm{CD}:{\mathrm{G}}_1]+[\mathrm{CD}:{\mathrm{G}}_2]$$ 14

The system of five equations (eqs to ) with five variables ([G₁], [G₂], [CD], [CD:G₁], and [CD:G₂]) is solved for [CD], giving a third-degree equation of the variable [CD] as:

$${[\mathrm{CD}]}^3+{[\mathrm{CD}]}^2(-{[\mathrm{CD}]}_{\mathrm{t}}+\frac{K_1+K_2}{K_1{K}_2}+{[{\mathrm{G}}_1]}_{\mathrm{t}}+{[{\mathrm{G}}_2]}_{\mathrm{t}})+[\mathrm{CD}](-{[\mathrm{CD}]}_{\mathrm{t}}\frac{K_1+K_2}{K_1{K}_2}+\frac{1}{K_1{K}_2}+\frac{{[{\mathrm{G}}_1]}_{\mathrm{t}}}{K_2}+\frac{{[{\mathrm{G}}_2]}_{\mathrm{t}}}{K_1})-\frac{{[\mathrm{CD}]}_{\mathrm{t}}}{K_1{K}_2}=0$$ 15

Equation was solved numerically, and all the other concentrations were straightforwardly calculated from [CD]:

$$[{\mathrm{G}}_1]=\frac{{[{\mathrm{G}}_1]}_{\mathrm{t}}}{1+K_1[\mathrm{CD}]}[\mathrm{CD}:{\mathrm{G}}_1]=K_1[\mathrm{CD}][{\mathrm{G}}_1]$$ 16

$$[{\mathrm{G}}_2]=\frac{{[{\mathrm{G}}_2]}_{\mathrm{tot}}}{1+K_2[\mathrm{CD}]}[\mathrm{CD}:{\mathrm{G}}_2]=K_2[\mathrm{CD}][{\mathrm{G}}_2]$$ 17

3. Results and Discussion

3.1. Selection of Bioactive Molecules for Evaluation of Inclusion Complexes

The selection of target compounds for inclusion complex formation was informed by identifying key constituents that contribute to the unpleasant taste of the chosen mother tinctures. Thymol, carvacrol, and eucalyptol were retained based on both their sensory impact and their high abundance in the respective extracts. In addition to their well-documented therapeutic potential, these molecules were considered particularly relevant due to their prominent role in the organoleptic profile of the formulations. − , The study focused on inclusion complexes in an aqueous solution, which is important for liquid medicinal formulations such as syrups. Several studies have emphasized the importance of solution-state characterization for predicting the behavior of bioactive molecules in liquid dosage forms where taste masking is crucial for ensuring patient compliance. Therefore, the present study was designed to evaluate host–guest interactions in solution, with particular emphasis on their potential application in syrup formulations.

The ¹H NMR spectrum of each bioactive molecule was compared with that of its corresponding mother tincture. This method was chosen because it offers a direct, molecular-level view of host–guest interactions through chemical shift and cross-peak studies. This is possible regardless of solubility fluctuations caused by factors such as pH, temperature, or solvent composition. Moreover, as bioactive molecules (BAMs) are present in plant extracts at low concentrations, phase-solubility measurements would not accurately replicate the concentrated formulations predicted by this study.

The results relating to carvacrol are discussed in the main text, while the ¹H NMR spectra of the other bioactive molecules (thymol and eucalyptol) and their corresponding mother tinctures are presented in the Supporting Information (Figures S1–S4). The ¹H NMR spectrum of carvacrol (Figure ) shows two regions appearing at 1.20–2.85 ppm being assigned to the aliphatic protons H_a (1.20 ppm), H_b (2.17 ppm), and H_c (2.85 ppm) and 6.82–7.16 ppm corresponding to the protons of the aromatic ring H_d (6.82 ppm), H_e (6.85 ppm), and H_f (7.16 ppm). The coupling constants and multiplicities are presented as follows:

2.

¹H NMR spectrum of carvacrol in D₂O.

¹H NMR (500 MHz, D₂O): δ_Hf = 7.15 ppm (d, J = 8.0 Hz, 1H, Ar–H), δ_He = 6.85 ppm (dd, J = 8.0, 2.0 Hz, 1H, Ar–H), δ_Hd = 6.82 ppm (d, J = 2.0 Hz, 1H, Ar–H), δ_Hc = 2.85 ppm (s, 3H, Ar–CH₃), δ_Hb = 2.17 ppm (mult, J = 6.9 Hz, 1H, CH of i-Pr), δ_Ha = 1.20 ppm (d, J = 6.9 Hz, 6H, 2 × CH₃ of i-Pr).

To confirm the presence of carvacrol in the thyme extract, the ¹H NMR spectrum of the thyme mother tincture was recorded and the characteristic resonance signals of CV were searched for. Figure shows that the peaks in the NMR spectrum of the thyme mother tincture predominantly belong to carvacrol, amino acids, organic acids, and sugars. , The ¹H NMR spectrum clearly showed that the thyme mother tincture contained a significant amount of CV. The observed shift of CV resonances in the mother tincture with respect to the pure CV is ascribed to interactions of carvacrol with other compounds present in thyme.

3.

(A) Superimposition of ¹H NMR spectra in D₂O of (a) horehound mother tincture, (b) thyme mother tincture, and (c) CV. (B) Detection of the different proton peaks of CV in the mother tincture of thyme and horehound.

3.2. Stoichiometry and Binding Constants of Inclusion Complexes

Carvacrol as well as thymol and eucalyptol were selected for investigation into the formation of inclusion complexes with the different cyclodextrins (HP-β-CD and HP-γ-CD) in a D₂O solution through their analysis by ¹H NMR spectroscopy. The primary objectives of our study were 2-fold: first, to determine the stoichiometry and stability constants of these inclusion complexes and second, to gain insight into the architecture of the host–guest interactions. The ¹H NMR spectra exhibited single and narrow lines, indicating that the complexed and free forms of BAMs and HP-cyclodextrins were undergoing fast exchange due to dynamic equilibrium between the free and complexed states. Accordingly, the chemical shifts of the BAMs were averaged from the chemical shifts of the free and complexed forms, weighted by their mole fractions (eq ). The determination of the stoichiometry of the inclusion complexes was carried out using the Job continuous variation method, which is a well-established and straightforward approach for assessing the stoichiometry of associated species. ¹H NMR spectra were recorded for a series of mixtures containing different proportions of HP-CD, revealing changes in the ¹H chemical shifts of the bioactive molecule in the presence of HP-CD (Figures S5–S14 and Tables S1–S6). For the two HP-CDs, the maximum in the Job plots was found at a BAM mole fraction of x = 0.5 for all BAM proton peaks (Figure and the odd-numbered Figures S7, S9, S11, S13, and S15 in the SI file). This indicated a 1:1 stoichiometry for the BAM complexes with all of the studied HP-CDs.

4.

Job plot of the H_a to H_f lines of CV for the complexation of CV with HP-γ-CD in D₂O. The total concentration of the two species was kept constant (3.99 × 10^–3 M). The solid lines are the best fits of the theoretical model to experimental data.

The equilibrium binding constants K ₁₁ were determined by fitting the model presented in Section to the experimental data. The curvature at the top of the Job plot is the most sensitive part of the information providing an estimate of K ₁₁. The values of the binding constants of BAMs with all the studied HP-cyclodextrins are given in Table .

3. Stoichiometry and binding constants of complexes in D₂O inferred from experiments using the Job mixing scheme.

complexes	stoichiometry	K 11	ARE
HP-β-CD:CV	1:1	2609	0.0022
HP-γ-CD:CV	1:1	3507	0.0040
HP-β-CD:THY	1:1	2322	0.0047
HP-γ-CD:THY	1:1	3010	0.0036
HP-β-CD:EUC	1:1	1050	0.0227
HP-γ-CD:EUC	1:1	1275	0.0134

The binding constants (K _{1 1}) reflect the relative affinity of each bioactive molecule for the HP-cyclodextrin cavity and highlight the influence of both steric and electronic factors. Among the studied compounds, carvacrol exhibited the highest affinity, particularly with HP-γ-CD, suggesting an optimal fit of its hydrophobic moiety within the larger γ-cyclodextrin cavity. This favorable interaction may also be reinforced by additional van der Waals forces or weak hydrogen bonding. Thymol, which is structurally similar to carvacrol, showed slightly lower K _{1 1} values but followed a comparable trend, confirming the role of molecular shape and functional group positioning in complex stability. However, eucalyptol presented lower binding constants with both HP-β-CD and HP-γ-CD due to the reduced affinity attributed to its rigid, bicyclic structure and lack of polar functionalities, which limit its ability to establish stabilizing interactions with the HP-cyclodextrin interior or rim. Furthermore, for all bioactive molecules tested, HP-γ-CD consistently yielded higher K _{1 1} values than HP-β-CD, underscoring the beneficial effect of the larger cavity size and enhanced conformational flexibility provided by hydroxypropyl substitution.

The binding constant and stoichiometry were also determined from measurements using the algorithmic method for the complexes of BAMs with all the HP-CDs. According to the mixing scheme in which the guest concentration is held constant while the HP-CD host concentration is increased, the complex concentration increases continuously with respect to the HP-CD/BAM molar ratio. Figure shows the spectra of CV in the HP-γ-CD/CV mixed solutions, from which the chemical shifts variations against the concentration ratio [HP-γ-CD]/[CV] were obtained. Additional inclusion complex data, along with the algorithmic treatment and plots for the other BAMs, are provided in the SI file. By comparing the spectra of the cyclodextrin–carvacrol (CD:CV) complexes across a range of molar ratios of CD/CV (ranging from 0.5 to 4), with the corresponding spectra of the free native or modified cyclodextrins, along with carvacrol recorded under the same conditions, induced chemical shift variations can be discerned. These variations were particularly evident when observing the chemical shifts of carvacrol protons. The results are compiled in Table .

5.

(A) ¹H NMR spectra of CV at 3.99 mM in D₂O as a function of molar ratios of HP-γ-CD/CV:CV alone, (a) 0.5/1, (b) 1/1, (c) 1.5/1, (d) 2/1, (e) 2.5/1, (f) 3/1, (g) 3.5/1, and (h) 4/1. (B) Expansion of the isopropyl and aromatic region showing the shifted peaks for H_a, H_b, H_c, H_d, H_e, and H_f protons of carvacrol.

4. Chemical Shifts of Carvacrol Protons for the Free and Complexed Forms in D₂O (3.99 mM CV and 3.99 mM HP-γ-CD and HP-β-CD, with r _[CD]/[CV] = 1).

complexes	protons	δ (free)	δ (obs)	Δδ (obs‑free)	δ CD:CV
HP-γ-CD:CV	Ha	1.1999	1.2864	0.0865	1.3085
	Hb	2.1761	2.2067	0.0306	2.2147
	Hc	2.8502	2.8330	–0.0172	2.8313
	Hd	6.8221	6.6839	–0.1382	6.6582
	He	6.8515	6.7015	–0.1500	6.6735
	Hf	7.1599	7.0262	–0.1337	7.0016
HP-β-CD:CV	Ha	1.1999	1. 2810	0.0811	1.3105
	Hb	2.1761	2.2055	0.0294	2.2138
	Hc	2.8502	2.8350	–0.0152	2.8320
	Hd	6.8221	6.6951	–0.1270	6.6621
	He	6.8515	6.7130	–0.1385	6.6750
	Hf	7.1599	7.0365	–0.1234	7.0031

As shown Figure , Table , and Table S7, the CV protons H_a and H_b undergo a downfield shift and H_c, H_d, H_e, and H_f experience an upfield shift, clearly indicating the association between the carvacrol guest and the cyclodextrin hosts. See Figures S7 and S8 in the SI file for other CDs. The largest variations were observed for the aromatic protons H_d, H_e, and H_f. These shifts are undoubtedly caused by magnetic perturbations resulting from the aromatic part of the carvacrol molecule penetrating the HP-γ-CD cavity. The differences of observed line shifts depending on the type of proton suggested a more or less complete inclusion of the carvacrol molecule within the cyclodextrin cavity. Moreover, the H_b signal supports the idea that the carvacrol guest penetrates into the CD cavities through its isopropyl and aromatic sides as it went through a tiny chemical shift, while leaving the methyl group outside the cavity. In addition, H_a and H_b are less impacted, causing a little alteration in the chemical shift, due to their location at the aromatic ring’s hydroxyl-substituted carbon, where they are partially exposed outside the CD cavity. Moreover, the observed chemical shifts vary in the same way from those of the free BAM to those of the pure complex at high HP-CD concentrations. The progress of complexation can easily be visualized by plotting the chemical shift against the concentration of HP-CD. Conversely, maximum complex formation is reached at x = 0.5 in the Job mixing scheme. The chemical shift variations were quite large, ranging from 0.02 to 0.18 ppm depending on the observed NMR line. These variations were much larger than the spectral resolution of 10^–3 ppm.

The model presented in Section fitted quite well to the observed variations in the chemical shifts of all lines in the CV (Figure , Table , and even-numbered Figures S16 and S17 and Table S8), confirming a 1:1 host–guest stoichiometry once again. The binding constant values K _{1 1} providing the best fit, obtained by minimizing the average relative error (ARE), are summarized in Table . Additional fitting curves and data sets for thymol and eucalyptol with both HP-β-CD and HP-γ-CD are included in the Supporting Information (Figures S19–S25 and Tables S9–S12), which further validates the consistency of the model across all the studied inclusion complexes.

6.

Experimental variations Δδ (ppm) of carvacrol chemical shifts with respect to free carvacrol as a function of [HP-γ-CD]/[CV] and nonlinear fits of the model to them.

5. Stoichiometry and Binding Constants of Complexes in D₂O Inferred from Experiments Using the Algorithmic Treatment.

complexes	stoichiometry	K 11	ARE
HP-β-CD:CV	1:1	2900	0.0023
HP-γ-CD:CV	1:1	3791	0.0024
HP-β-CD:THY	1:1	2658	0.0020
HP-γ-CD:THY	1:1	3268	0.0022
HP-β-CD:EUC	1:1	1425	0.0072
HP-γ-CD:EUC	1:1	1895	0.0082

The binding constants (K _{1 1}), as determined from the Job plots and algorithmic analyses (Tables and ), show good agreement albeit with slight variations, which can be explained by the specific experimental conditions of each method. The Job plot method involves continuously varying the mole fractions of the host and guest while keeping their total concentration constant. This approach does not involve a significant excess of HP-cyclodextrin, and the maximum host-to-guest ratio usually remains close to 1:1. This reflects equilibrium conditions close to stoichiometric complexation, which is ideal for determining the complex stoichiometry. In contrast, the algorithmic method uses an excess of the host (HP-CD), with [HP-CD]/[BAM] ratios of up to 1:4. This excess favors the formation of the inclusion complex and leads to more pronounced chemical shift changes, thereby improving the sensitivity of the fitting procedure. However, this experimental condition can result in slightly higher apparent K _{1 1} values due to the enhanced complexation environment. Notably, for all the bioactive molecules (BAMs) studied, carvacrol, thymol, and eucalyptol, the binding constants derived from the algorithmic method were consistently higher than those obtained using the Job plot approach. This consistent trend further confirms the influence of experimental conditions on the apparent affinity measurements and highlights the importance of combining both methods to obtain a more comprehensive and accurate picture of the complexation behavior. Among all complexes, HP-γ-CD/ CV exhibited the highest binding affinity, a result that was consistent across both methods.

6. Standard Free Energy of Inclusion Complexes Formed between HP-CD and Bioactive Aroma Molecules (BAMs).

	Δcomp G 0 (kJ mol–1)
HP-CD	carvacrol	thymol	eucalyptol
HP-β-CD	–19.49	–19.20	–17.23
HP-γ-CD	–20.22	–19.84	–17.71

The inclusion constants of the complexes studied using HP-β-CD are consistent with previous reports in the literature. For instance, Kfoury et al. reported K _{1 1} values for thymol and carvacrol via UV–visible spectroscopy. Despite methodological differences, their results (e.g., K _{1 1} ≈ 2154 for carvacrol with HP-β-CD) closely match those obtained using NMR (K _{1 1} ≈ 2900). This confirms the robustness and reproducibility of the complexation process when using different analytical techniques. Furthermore, the study by Kfoury et al. confirms the consistency and reliability of the inclusion results obtained for thymol and carvacrol when combined with cyclodextrins. They found that carvacrol had higher stability constants than thymol when combined with various types of cyclodextrins, particularly β-CD and HP-β-CD. This trend is consistent with our own results and provides further support for the idea that carvacrol forms stronger inclusion complexes due to its favorable molecular geometry and higher hydrophobicity. As emphasized in the paper, the position of the hydroxyl group significantly influences binding affinity; the meta position in carvacrol allows it to fit better into the cyclodextrin cavity, whereas the ortho position in thymol causes steric hindrance, resulting in less stable complexation. This interpretation reinforces the idea that subtle structural differences between isomers can have a significant impact on their inclusion behavior.

Moreover, Ciobanu et al. found that among a range of volatile flavor compounds tested with various cyclodextrins, including HP-β-CD, eucalyptol exhibited one of the lowest binding affinities. Their study confirmed that eucalyptol generally forms weak complexes with cyclodextrins, likely due to its rigid bicyclic structure and limited ability to form hydrogen bonds.

In addition, the increase in stability constants for HP-γ-CD compared to HP-β-CD for all three compounds (CV, THY, and EUC) is supported by the findings of other studies. For example, D’Aria et al. investigated the complexation of onco-A using isothermal titration calorimetry (ITC) and found that the binding constants were higher for HP-γ-CD (K _{1 1} ≈ 3175) than for HP-β-CD (K _{1 1} ≈ 890). These confirm that the larger cavity of the γ-CD derivatives can accommodate bulky or flexible molecules more effectively.

The stability constants (K _{1 1}) determined from the NMR data are consistent with values reported in the literature for similar host–guest systems. Typically, K _{1 1} values in the range of 100–5000 indicate efficient yet reversible inclusion behavior (Yin et al.). The values obtained in this study fall within this range, confirming the reliability of our approach. Rodríguez-López et al. reported a binding constant of 2583 for the HP-β-CD:thymol complex, in excellent agreement with our results. Similarly, Rodríguez-López et al. found that the binding constant of HP-β-CD:carvacrol ranged between 198 and 5000 depending on the experimental conditions. Kfoury et al. estimated the stability constant for the HP-β-CD:eucalyptol complex to be between 1112 and 1200, which also aligns with our findings. In addition, Li et al. observed that γ-CD-based materials exhibited larger inclusion constants with thymol compared to α- and β-CDs, further supporting our results.

These comparisons validate the accuracy of our NMR-based determination and confirm that the host–guest interactions observed in this work are consistent with previously reported data from phase-solubility and other experimental techniques.

3.3. Structure of the Inclusion Complexes of Carvacrol with HP-γ-CD

The ¹H NMR spectrum of the inclusion complex is shown in Figure . Within the spectrum, both the aromatic protons and the aliphatic protons of carvacrol are noticeable. The aromatic protons H_d, H_e, and H_f are shielded by −0.1382, −0.1500, and −0.1337 ppm, while all other protons of CV which are H_a, H_b, and H_c were shielded by 0.0865, 0.0306, and −0.0172 ppm, respectively (Table ). In this case, the aromatic protons exhibited the most significant alterations in the chemical shift, which can be attributed to changes in their surroundings. This finding confirms the entry of the carvacrol molecule into the HP-γ-CD cavity, highlighting the significance of aromatic protons in host–guest interactions. These observations confirm the formation of an IC between CV and HP-γ-CD.

7.

¹H NMR spectrum of the HP-γ-CD:CV inclusion complex (r = 1).

More detailed structural insights into the inclusion complex were gained through 2D ROESY and NOESY NMR experiments. These experiments were conducted to provide complementary and confirmatory information about host–guest spatial interactions in our inclusion complexes. NOESY was primarily used to detect the spatial proximity of the guest to the inner cavity protons of cyclodextrin (notably H₃ and H₅), thereby highlighting the inclusion phenomenon. ROESY was used to confirm these interactions and to enhance the detection of cross-dipolar relaxation, particularly when NOE signals are weak or negative. This typically occurs in cyclodextrin complexes because of guest dynamics and indirect molecular exchange in aqueous environments. ROESY is therefore particularly advantageous under these conditions. In our case, the ROESY spectra clearly confirmed the NOESY observations, and each technique mutually reinforced the interpretation of the inclusion complex. ,

The presence of cross-peaks in the contour plots of these spectra (Figures and ) reveals dipolar interactions through space between protons located in close proximity (typically within <5 Å), which is indicative of spatial proximity and thus inclusion phenomena. In both the ROESY (blue/red contours) and NOESY (red/orange contours) spectra, distinct intermolecular cross-peaks are observed between the internal protons of HP-γ-CD (H-3 and H-5), which are located inside the cyclodextrin cavity, and the isopropyl H_a as well as the aromatic H_d, H_e, and H_f protons of carvacrol. Notably, additional cross-peaks involving the H-6 protons of HP-γ-CD, which are located at the narrow rim, near the primary hydroxyl groups, and in the vicinity of the CH and CH₂ groups of the 2-hydroxypropyl substituents, confirm the hypothesis that inclusion occurs through the primary face of the cyclodextrin. These NOE and ROE interactions confirm the insertion of the aromatic moiety of carvacrol into the hydrophobic cavity of HP-γ-CD, most likely via the narrow rim. The observed interactions with H-3 and H-5 indicate a deep aromatic ring insertion into the cavity, while correlations with H-6 suggest that the isopropyl group remains at or near the entrance. This orientation is consistent with previously reported inclusion complexes of phenolic compounds with cyclodextrins, supporting the formation of a stable, well-defined 1:1 host–guest complex. ,

8.

2D ROESY spectrum of HP-γ-CD/CV mixed solution at a 1:1 mol ratio in D₂O and illustrative model of the inclusion complex between CV and HP-γ-CD: (a) proposed inclusion model between carvacrol and HP-γ-CD; (b) 2D ROESY spectrum of HP-γ-CD/CV; (c) expanded region of the 2D ROESY NMR spectrum showing the correlation between H_a in carvacrol and H₆ in HP-γ-CD; (d) expanded region of the 2D ROESY NMR spectrum showing the correlations between the aromatic protons of carvacrol and H₃, H₅, and H₆ of HP-γ-CD.

9.

2D NOESY spectrum of HP-γ-CD/CV mixed solution at a 1:1 molar ratio in D₂O and illustrative model of the inclusion complex between CV and HP-γ-CD: (a) proposed inclusion model between carvacrol and HP-γ-CD; (b) 2D NOESY spectrum of HP-γ-CD/CV; (c) expanded region of the 2D NOESY NMR spectra showing the correlation between H_a in carvacrol and H₆ in HP-γ-CD; (d) expanded region of the 2D NOESY NMR spectrum showing the correlations between the aromatic protons of carvacrol and H₃, H₅, and H₆ of HP-γ-CD.

3.4. Thermodynamic Characterization and Contribution of Hydrogen Bonding to Inclusion Complex Stability

In this study, we focused on bioactive molecules that are commonly used in pharmaceutical formulations and are typically prepared and administered at room temperature. Therefore, our experiments were performed under these practical conditions. The formation of inclusion complexes with cyclodextrins in aqueous solutions is partly due to the hydrophobic nature of the CD cavity, which allows association with organic molecules via a hydrophobic effect. However, other interactions also play a role. In particular, hydrogen bonding between the hydroxyl groups of cyclodextrins and the active molecules plays a crucial role in enhancing the complexation process. This effect can be explained using a simple thermodynamic model that combines two types of binding free energy. The standard free energy of complexation (Table ) is related to the stability constant K ₁₁ as:

$${\mathrm{\Delta }}_{\mathrm{comp}}{G}^0=-RT\ln (K_{11})$$ 18

The total free energy of complexation can be split into two contributions, the hydrogen bonding contribution and the nonhydrogen bonding contribution:

$${\mathrm{\Delta }}_{\mathrm{comp}}{G}^0={\mathrm{\Delta }}_{\mathrm{comp}}{G}^0(\mathrm{HB})+{\mathrm{\Delta }}_{\mathrm{comp}}{G}^0(\mathrm{non}‐\mathrm{HB})$$ 19

The contribution of hydrogen bonding represents the standard free energy of complexation of the active molecule through an acidic or basic group involving a heteroatom. Conversely, the nonhydrogen bonding involves nonspecific interactions, such as the hydrophobic effect and dispersion interactions. According to this model, the standard free energy of complexation Δ_comp G ⁰ should vary depending on the presence and the nature of functional groups. The presence and type of hydrogen bonding between hydroxyl groups located on the wide and narrow rims of cyclodextrins and the functional groups of the active molecules that do not enter the cavity deeply significantly stabilize inclusion complexes. Hydroxyl groups located on the aromatic ring may bind to the oxygen atoms of the glycosidic linkages within the CD cavity. The results given in Table are consistent with this view. Carvacol and thymol are similar in this respect, as they both have one hydroxyl group on their aromatic ring and no polar groups apart from this ring. Finally, eucalyptol forms the weakest complex of the series because its bicyclic core contains a basic ether oxygen and lacks an acidic hydroxyl group that could form a hydrogen bond with the glycosidic linkages. In conclusion, the presence and strength of hydrogen bonding are responsible for the enhanced complexation of the active molecules in this series.

3.5. Competitive Complexation

The bioactive ingredients are often mixed together, resulting in a complex environment in which multiple molecules compete to bind to the HP-CDs. This competition may result in the release of the active molecules with the lowest affinity for the HP-CDs. Therefore, a simple approach was considering the inclusion complexation of an equimolar mixture of carvacrol and eucalyptol by HP-γ-CD, as this exhibited the best complexation performance for an NMR study using the algorithmic treatment (Figures S26–S31 and Tables S13–S16 in the SI file). Using the binding constants and chemical shifts of the previously obtained complex from individual measurements, a fair fit of the theoretical prediction to experimental chemical shifts was obtained for the model of two competitive equilibria (eqs – in Section ) with no adjustable parameter (Figure ). However, the distribution of species was not the sum of the distributions of species measured in separate experiments. Compared to individual experiments, CV has taken a part of HP-γ-CD molecules from the HP-γ-CD:EUC complex to form the more stable HP-γ-CD:CV complex. However, the stoichiometry of the complexes remained 1:1. These experiments validate the use of equilibrium constants measured individually for the active molecules can be used in complex mixtures.

10.

Algorithmic method plots for complexation of CV and EUC by HP-γ-CD in a D₂O solution of a 1/1 mixture of CV and EUC. (A) Algorithmic method plot for HP-γ-CD:CV. (B) Algorithmic method plot for HP-γ-CD:EUC.

The equilibrium concentrations of all species in an HP-γ-CD can be calculated from the known K ₁₁ values and the concentrations of free and complexed CV and EUC, as well as the concentration of HP-γ-CD (Figure ). Due to the higher value of K ₁₁ for CV than for EUC, the HP-γ-CD:CV concentrations are higher than the HP-γ-CD:EUC concentrations, despite the total of CV and EUC concentrations being the same. This behavior highlights the stronger affinity of CV for the HP-γ-cyclodextrin cavity, shifting the equilibrium more strongly toward complex formation. In contrast, the lower stability constant of the HP-γ-CD:EUC complex results in a greater amount of free EUC remaining in solution. The curves also demonstrate that the formation of the CV complex forms more rapidly and reaches saturation earlier than the EUC complex at higher HP-γ-CD concentrations. This suggests that CV is encapsulated more efficiently than EUC due to more favorable interactions, such as hydrogen bonding and a better geometric fit, between CV and the cyclodextrin cavity.

11.

Distribution of the concentrations of the different species of carvacrol and eucalyptol in equilibrium with HP-γ-CD: [HP-γ-CD:CV], [CV], [HP-γ-CD: EUC], [EUC], and [HP-γ-CD].

In addition, the binary mixture of carvacrol CV and eucalyptol EUC shows no strong interactions in the absence of cyclodextrin due to a lack of hydrogen bonding or π–π stacking between the two molecules. This is contrary to the binary mixture of thymol and carvacrol. However, in the presence of cyclodextrins, these hydrophobic compounds preferentially escape the polar aqueous medium and are individually accommodated within the apolar cavity of the cyclodextrin. This competitive inclusion behavior significantly reduces or prevents direct guest–guest interactions in solution. This is supported by the chemical shift patterns observed in the ¹H NMR spectra, which suggest that each molecule is independently included.

3.6. Impact of HP-Cyclodextrins on Mother Tinctures

The formation of inclusion complexes of active compounds in mother tinctures using HP-β-CD and HP-γ-CD was investigated using the same methodology and ¹H NMR. Notably, significant shifts in the aromatic proton peaks of carvacrol and thymol, as well as the methylene group of the cyclohexyl moiety of eucalyptol, were observed when these cyclodextrins were added to thyme, horehound, plantain, and eucalyptus mother tinctures. Additionally, the ¹H NMR spectrum of TMTH revealed that the locations of the H_e and H_d aromatic proton peaks of carvacrol were inverted following its complexation by HP-β-CD and HP-γ-CD (Figure ). This change definitively confirms the inclusion of carvacrol in the cavity of the HP-cyclodextrins. These results are consistent with earlier findings on the complexation of pure bioactive compounds. Therefore, HP-CDs are a suitable option for masking the undesirable taste of pharmaceutical formulations containing plant extracts.

12.

(A) ¹H NMR spectra in D₂O of (a) CV alone, (b) TMTH, (c) HP-β-CD-TMTH, and (d) HP-γ-CD-TMTH. (B) Expansion of the aromatic region showing the shifted peaks for H_d, H_e, and H_f protons of carvacrol.

4. Conclusions

In this study, we investigated the formation of inclusion complexes between selected bioactive molecules found in plant mother tinctures, such as thymol, carvacrol, and eucalyptol, and two hydroxypropyl-cyclodextrins using ¹H NMR spectroscopy. These bioactive molecules were selected for their known contribution to the characteristic taste of the tinctures, with carvacrol chosen as a model compound to elucidate the complexation mechanism. The results demonstrated the formation of stable 1:1 inclusion complexes, as evidenced by systematic chemical shift variations and intermolecular cross-peaks in the 2D ROESY/NOESY spectra. HP-γ-CD exhibited the highest affinity toward carvacrol, reflecting favorable geometric and hydrophobic complementarity between the host and guest. The calculated binding constants and thermodynamic parameters further confirmed the high stability and spontaneous formation of these complexes. Overall, these findings provide molecular-level insight into the host–guest interactions governing complex stability and geometry. While the results confirm efficient encapsulation of bioactive molecules by hydroxypropyl-cyclodextrins, the actual taste-masking performance remains to be validated experimentally (Figure ). A forthcoming study will address this aspect through comprehensive sensory and electronic tongue (e-tongue) analyses to quantitatively assess the effect of cyclodextrin complexation on flavor perception.

13.

Illustrative scheme of taste masking by cyclodextrins.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c09162.

• Additional characterization and analysis results: (S.1) selection of bioactive molecules for evaluation of inclusion complexes, (S.2) stoichiometry and binding constants of inclusion complexes, and (S.3) competitive complexation (PDF)

Kallel Jihene: data curation, formal analysis, writing of the original draft, and editing; Najeh Jaoued: formal analysis, validation, and visualization; Elaiech Riahi: formal analysis; Claire Bordes: validation and visualization; Yves Chevalier: validation and visualization; Hbaieb Souhaira: supervision, project administration, conceptualization, investigation, methodologies, validation, and visualization.

The authors declare no competing financial interest.