¹H-NMR Spectroscopy Study of the Formation of Inclusion Complexes between Cucurbit[7]uril and Ciprofloxacin, Levofloxacin, and Lomefloxacin

Abstract

Owing to the recent detection of pharmaceutical residues in aquatic environments, the development of methods for their removal has attracted increasing research attention. Considering the rich host–guest chemistry of cucurbit[7]uril (CB[7]), which can form stable inclusion complexes with various compounds, we envisioned that CB[7] could be used for capturing pharmaceutical residues in aquatic environments. In this study, using ¹H-NMR spectroscopy, we examined the formation of inclusion complexes between CB[7] and new quinolone antibiotics that have been linked to the emergence of resistant bacteria, that is, ciprofloxacin hydrochloride monohydrate (CPFX), levofloxacin hydrochloride (LVFX), lomefloxacin hydrochloride (LFLX), and pazufloxacin mesylate (PZFX). The results showed that CPFX, LVFX, and LFLX formed inclusion complexes with CB[7] at a molar ratio of 1 : 1, with complex formation constants (K) of 0.529, 0.877, and 3.65 (×10⁴ M⁻¹), respectively, whereas PZFX did not. This difference was attributed to the presence or absence of a piperazine ring, indicating that it is a critical feature for the formation of inclusion complexes with CB[7]. In addition, the thermodynamic parameters calculated using van’t Hoff plots revealed that LVFX and LFLX with a methyl group on the piperazine ring expel high-energy water from the cavity of CB[7] more efficiently, resulting in larger K values. Because the piperazine ring structure is commonly found in many drugs, CB[7] can be expected to capture other drugs apart from those evaluated in this study. Therefore, CB[7] is a promising candidate as a host molecule for use in drug removal in aquatic environments through host–guest chemistry.

Introduction

Despite being indispensable for maintaining human and animal health, pharmaceuticals pose a serious risk when discharged into aquatic environments through human and animal waste or untreated domestic wastewater. Owing to their bioactivity, even trace amounts of pharmaceutical residues can exert a negative impact on ecosystems.^1–5) In fact, antibiotics, antipyretic analgesics, and antipsychotics have been detected in aquatic environments around the world.^6–8) Various methods, such as adsorption, decomposition, and membrane filtration, have been explored for water remediation in terms of the removal of pharmaceutical residues.^9–11) However, a definitive solution has not yet been found. Among pharmaceuticals, antibiotic residues have a particularly significant impact. The inappropriate use of antibiotics has already been highlighted as a contributing factor to the emergence of resistant bacteria. What is more, the presence of trace amounts of antibiotics in aquatic environments has been linked to the rise of resistant bacteria.^12,13) Antimicrobial resistance is a concern for WHO because it affects not only the treatment of common infections but also the prevention of infections in surgical procedures.¹⁴⁾ Furthermore, the spread of resistant bacteria could make the treatment of infections with antibiotics difficult, leading to an increase in mortality rates and a reduction in the healthy labor force, which could have a considerable economic impact.¹⁵⁾ According to Wilkinson et al., antibiotic residues have been detected in environments worldwide at concentrations exceeding the levels that raise concerns about the emergence of resistant bacteria.¹⁶⁾

Cucurbit[n]urils (CB[n]s) are macrocyclic compounds consisting of n glycolyl units bridged by methylene groups, and have recently attracted attention as host molecules capable of forming highly stable inclusion complexes by encapsulating various compounds into their cavities.¹⁷⁾ In particular, CB[7] (Fig. 1) is highly water-soluble and low in toxicity,^18,19) making it the subject of numerous reports in the pharmaceutical field. For example, Li et al. reported that the inclusion of clofazimine within the cavity of CB[7] led to enhanced solubility and reduced cardiotoxicity.²⁰⁾ Similarly, Ma et al. demonstrated that fluorofenidone formed an inclusion complex with CB[7], resulting in increased solubility and bioavailability.²¹⁾ While CB[7] is well known as a compound with encapsulation ability, recent findings have also revealed that CB[7] forms outer-surface adducts with electron-rich compounds via the outer-surface interaction and chaotropic effect due to its positive electrostatic potential.^22,23) As one example, we have found that CB[7], when added to solutions containing gallated catechins and non-gallated catechins, which are a type of polyphenol with various physiological activities and rich in electrons, forms outer-surface adducts preferentially with gallated catechins.²⁴⁾ This phenomenon leads to selective separation through precipitation. Building upon previous studies on inclusion complexes and outer-surface adducts, we hypothesized that CB[7] could not only capture various pharmaceutical residues present in aquatic environments as inclusion complexes but also remove them from aquatic environments by adding compounds that form outer-surface adducts with CB[7], thereby precipitating the pharmaceutical residues.

Fig. 1. Chemical Structure of CB[7]

In this study, new quinolone antibiotics, that is, ciprofloxacin hydrochloride monohydrate (CPFX), levofloxacin hydrochloride (LVFX), lomefloxacin hydrochloride (LFLX), and pazufloxacin mesylate (PZFX), were selected as the target drugs (Fig. 2). The ability of CB[7] to encapsulate these drugs was evaluated using ¹H-NMR spectroscopy. Based on the results, we assessed the potential of CB[7] as a promising host molecule for pharmaceutical removal from aquatic environments through host–guest chemistry.

Fig. 2. Chemical Structures of CPFX, LVFX, LFLX, and PZFX

Results and Discussion

Confirmation of the Formation of Inclusion Complexes between CB[7] and Drugs Using ¹H-NMR Spectroscopy

The formation of inclusion complexes between CB[7] and each drug was confirmed using ¹H-NMR spectroscopy. When CB[7] forms an inclusion complex, the signals of protons located inside the cavity exhibit an upfield shift, whereas those of the protons near the carbonyl groups at the rim undergo a downfield shift.^{25) 1}H-NMR measurements were conducted using 0.5 mM drug solutions in deuterium oxide (D₂O), as well as mixed solutions containing 0.5 mM drug and either 0.5 or 1 mM CB[7] (one or two equivalents relative to the drug) in D₂O. The proton signals of the drug in the drug solutions and the mixed solutions were compared to confirm the changes in the chemical shifts. When the chemical shift change values (Δδ: the difference in chemical shift values of the drugs observed in the CB[7]-containing solution relative to those in D₂O) were calculated between the drug solution and the mixed solution containing 1 mM CB[7], the H³ and H⁴ proton signals of the piperazine ring in CPFX exhibited upfield shifts of 0.605 and 0.401 ppm, respectively (Fig. 3a). Furthermore, in the cases of LVFX and LFLX, the methyl group proton (H¹) attached to the piperazine ring showed upfield shifts of 0.487 and 0.444 ppm, respectively (Figs. 3b, 3c). In the spectra of the solutions of CPFX, LVFX, and LFLX with CB[7], the protons of the piperazine ring showed a clear upfield shift, suggesting the formation of an inclusion complex (drug–CB[7]) with the piperazine ring inside the cavity of CB[7]. In all three drugs, the addition of CB[7] resulted in the observation of a single averaged proton signal, rather than separate signals for the free and complexed forms. As the amount of CB[7] increased, progressive changes in the chemical shifts were observed. In addition, peak broadening was also noted. These characteristic phenomena are commonly observed during the formation of inclusion complexes and indicate intermediate-to-fast chemical exchange on the NMR timescale between the free state and the drug–CB[7] state.²⁶⁾

Fig. 3. ¹H-NMR Spectra of the Pristine Drugs and Mixed Solutions of Each Drug and CB[7]: (a) CPFX, (b) LVFX, (c) LFLX, and (d) PZFX

The red arrows, blue arrows, and black dashed lines represent high-field shifts, low-field shifts, and no shift change, respectively.

Conversely, virtually no changes were observed in the spectrum of the mixture of CB[7] and PZFX, which lacks a piperazine ring, relative to that of the pristine drug (Fig. 3d), which allowed ruling out the formation of an inclusion complex. Accordingly, the formation of inclusion complexes between CPFX, LVFX, and LFLX with CB[7] can be attributed to the presence of the piperazine ring. This is in accord with previous studies describing that CB[7] forms stable inclusion complexes with amines,¹⁷⁾ suggesting that the piperazine ring in drug molecules is an important structure for the formation of inclusion complexes with CB[7].

Determination of the Stoichiometry and K Values in the Formation of the Inclusion Complexes

Having confirmed the formation of inclusion complexes between CB[7] and the CPFX, LVFX, and LFLX drugs, the K values were calculated to gain more insight into the ability of CB[7] to form these inclusion complexes. To select the equation for the regression curve for the calculation of the K values, the stoichiometry was first determined using Job’s plot (Fig. 4). For all drugs, the Job’s plot shows a maximum at a molar fraction of 0.5, indicating that the drugs form inclusion complexes with CB[7] in a 1 : 1 molar ratio.

Fig. 4. Job’s Plots for a Total Concentration of the Drugs and CB[7] of 2 mM at 298 K

The X value indicates the mole fraction of the drugs and ∆δ is (δ_mix–δ_D), where δ_mix and δ_D indicate the chemical shift value of the drug in the mixed solution with CB[7] and the chemical shift value of the drug in D₂O, respectively. The analysis of CPFX, LVFX, and LFLX was conducted using the chemical shift values of the H³, H¹, and H¹ protons, respectively.

Then, the K values were determined via NMR titration using mixed solutions with a drug concentration of 0.5 mM and varying concentrations of CB[7] from 0 to 5 mM. On the basis of the chemical shift changes in the resonances of the drugs, the K values for CPFX, LVFX, and LFLX with CB[7] were calculated to be 0.529, 0.877, and 3.65 (×10⁴ M⁻¹), respectively, at 298 K using the method reported by Thordarson.²⁷⁾ These results suggested that the new quinolone antibiotics containing a piperazine ring with a methyl group tended to have a higher K value. To confirm this tendency, the mechanism of inclusion complex formation between CB[7] and these drugs was investigated.

Calculation of Thermodynamic Parameters for the Inclusion Complex Formation

CB[7] is known to form stable inclusion complexes not only through hydrophobic effects and ion–dipole interactions but also through the expulsion of high-energy water from its cavity by the guest molecule.²⁸⁾ To investigate the mechanism of inclusion complex formation between CB[7] and CPFX, LVFX, and LFLX, the Gibbs free energy change (ΔG), enthalpy change (ΔH), and entropy change (ΔS) were determined from van’t Hoff plots constructed using NMR spectroscopy (Eq. (1)). To generate the van’t Hoff plots, since the equation used to calculate the K value depends on the molar ratio, the molar ratio of CB[7] and each drug in the corresponding complex was first determined at temperatures between 303 and 333 K using the same procedure as that used at 298 K. The corresponding Job’s plots confirmed that the drugs showed a maximum at a mole fraction of 0.5, indicating the formation of 1 : 1 complexes in all the temperature range evaluated (Supplementary Fig. S1). Similarly, the K values at each temperature were calculated by conducting NMR titrations between 303 and 333 K (Supplementary Figs. S2–S4). The thermodynamic parameters were calculated using the obtained K values and the van’t Hoff plots (Fig. 5). The results are shown in Table 1. The ΔH values of LVFX and LFLX were larger than that of CPFX, which indicates that molecules with a methyl group on the piperazine ring expel high-energy water from the CB[7] cavity more effectively, resulting in higher K values. Furthermore, when encapsulated by CB[7], the morpholine ring in LVFX causes steric hindrance, resulting in a smaller K value compared with that of LFLX:

  

$$\text{ln}K=-\frac{\Delta H}{RT}+\frac{\Delta S}{R}$$(1)

Fig. 5. van’t Hoff Plots for the Inclusion Complex Formation Reaction of (a) CPFX, (b) LVFX, and (c) LFLX with CB[7] from 298 to 333 K

Table 1.  Thermodynamic Parameters of CPFX, LVFX, and LFLX at 298 K

	∆G (kJ/mol)	∆H (kJ/mol)	–T∆S (kJ/mol)
CPFX	−21.3	−11.5	−9.77
LVFX	−22.6	−20.8	−1.80
LFLX	−26.0	−18.0	−8.09

Conclusion

Using ¹H-NMR spectroscopy, CB[7] was confirmed to form inclusion complexes with the new quinolone antibiotics CPFX, LVFX, and LFLX in a 1 : 1 molar ratio. The changes in the ¹H-NMR chemical shifts suggested that the piperazine ring of the drugs is located within the cavity of CB[7] in the inclusion complexes. This, together with the fact that PZFX lacking a piperazine ring did not form an inclusion complex, indicates that the piperazine ring plays a crucial role in the formation of these inclusion complexes. The K values for the formation of the inclusion complexes between CPFX, LVFX, and LFLX and CB[7] were 0.529, 0.877, and 3.65 (×10⁴ M⁻¹), respectively. Furthermore, the thermodynamic parameters revealed that molecules with a methyl group on the piperazine ring expel high-energy water from the cavity of CB[7] more efficiently, resulting in larger K values.

Although the inclusion complex formation between piperazine derivatives and CB[6] or CB[8] has been reported,^29,30) examples involving inclusion complex formation with CB[7] are relatively limited. In this study, we showed that CB[7] forms the 1 : 1 inclusion complexes with piperazine-containing molecules, contributing to a better understanding of host–guest interactions involving CB[7] and piperazine rings. Considering that the piperazine ring structure is commonly found in many drugs such as aripiprazole, cetirizine, and eszopiclone, CB[7] can be expected to capture other drugs. These results demonstrate the potential of CB[7] as a host molecule for use in drug removal from aquatic environments through host–guest chemistry.

Experimental

Materials

LVFX and LFLX were purchased from LKT Laboratories, Inc. (St. Paul, MN, U.S.A.) and BLD Pharmatech Ltd. (Shanghai, China), respectively. CPFX, PZFX, glycoluryl, and paraformaldehyde were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). D₂O (99.8% D, for NMR spectroscopy) was purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). Hydrochloric acid (Guaranteed Reagent) was purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). CB[7] was synthesized according to the method reported by Gomes et al.³¹⁾ All reagents were used as received without purification.

¹H-NMR Experiments

All NMR experiments were conducted on a JNM-ECZ600R spectrometer (600 MHz, JEOL, Tokyo, Japan) using D₂O as the solvent. The NMR spectra were recorded over a spectral width of 9000 Hz through 16 scans. Measurements were carried out at the temperatures specified in each section; when not specified, measurements were performed at room temperature. The chemical shifts were determined using the signal of sodium 3-(trimethylsilyl) propionate-2,2,3,3-d₄ as an external reference. The spectral analysis was performed using the Delta NMR data processing software version 6.3.0 (JEOL USA, Inc., Peabody, MA, U.S.A.).

Confirmation of the Formation of Inclusion Complexes

To confirm the formation of an inclusion complex between CB[7] and each drug, solutions containing a drug concentration of 0.5 mM in D₂O and a separate solution of CB[7] in D₂O were prepared. Then, mixed solutions with final concentrations of 0.5 mM for the drug and 0.5 or 1 mM for CB[7] were prepared in D₂O to measure the ¹H-NMR spectra. The proton signals of the drugs in the pristine drug solutions and the mixed solutions were compared to confirm the changes in the chemical shifts. Δδ was calculated as (δ_mix – δ_D), where δ_mix and δ_D represent the chemical shift values of the drug in the mixed solution with CB[7] and in D₂O, respectively.

Determination of the Stoichiometry (Job’s Plot)

Solutions containing each drug and CB[7] at concentrations of 2 mM were prepared in D₂O. Mixed solutions were then prepared by varying the ratios from 0 to 1. The ¹H-NMR spectra of these solutions were measured at temperatures ranging from 298 to 333 K. The Job’s plots were generated using the changes in the chemical shifts of the drugs, specifically, the H³, H¹, and H¹ protons of CPFX, LVFX, and LFLX, respectively.

Calculation of the K Values

To calculate the K values, mixed solutions of each drug and CB[7] were prepared in D₂O with final concentrations of 0.5 mM for the drug and 0–5 mM for CB[7]. The ¹H-NMR spectra of the prepared solutions were measured from 298 to 333 K.

After confirming the formation of inclusion complexes in a 1 : 1 molar ratio, the K values were calculated. Considering that the inclusion complex formation between the drug and CB[7] proceeds according to Eq. (2), the K values were calculated using Eq. (3),²⁷⁾ which is only valid when the molar ratio is 1 : 1. In Eq. (3), δ_D–CB[7] is the chemical shift of D–CB[7], and [D₀] and [CB[7]₀] are the mean initial concentration of D and CB[7], respectively. The chemical shift changes were determined using the H³, H¹, and H¹ signals of CPFX, LVFX, and LFLX, respectively:

  

$$\text{D}+\text{CB[7]}\rightleftarrows \text{D}-\text{CB[7]}$$(2)

  

$$\begin{array}{l}\Delta \delta =\left\{\frac{\text{1}}{\text{2}}\left(\left[{\text{D}}_0\right]+\left[\text{CB}{\left[\text{7}\right]}_0\right]+\frac{\text{1}}{K}\right)-{\left(\begin{array}{l}{\left(\left[{\text{D}}_0\right]+\left[\text{CB}{\left[\text{7}\right]}_0\right]+\frac{\text{1}}{K}\right)}^{\text{2}}\hfill \\ -\text{4}\left[{\text{D}}_0\right]\left[\text{CB}{\left[\text{7}\right]}_0\right]\hfill \end{array}\right)}^{\frac{\text{1}}{\text{2}}}\right\}\hfill \\ \times \left({\delta }_{\text{D}-\text{CB[7]}}-{\delta }_{\text{D}}\right)\times \frac{1}{\left[{\text{D}}_0\right]}\hfill \end{array}$$(3)

The K values were determined by plotting Δδ against [CB[7]₀]/[D₀]. The analysis was performed using KaleidaGraph version 3.5 J (Synergy Software, Reading, PA, U.S.A.).

Acknowledgments

This study was supported by funding from Fukuoka University (Grant No. GW2215).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1)  Robinson A. A., Belden B. J., Lydy J. M., Environ. Toxicol. Chem., 24, 423–430 (2005).
• 2)  Valenti T. W. Jr., Gould G. G., Berninger P. J., Connors A. K., Keele B. N., Prosser N. K., Brooks W. B., Environ. Sci. Technol., 46, 2427–2435 (2012).
• 3)  Péry A. R. R., Gust M., Vollat B., Mons R., Ramil M., Fink G., Ternes T., Garric J., Chemosphere, 73, 300–304 (2008).
• 4)  Han S., Choi K., Kim J., Ji K., Kim S., Ahn B., Yun J., Choi K., Khim S. J., Zhang X., Giesy P. J., Aquat. Toxicol., 98, 256–264 (2010).
• 5)  Weinberger J. II, Klaper R., Aquat. Toxicol., 151, 77–83 (2014).
• 6)  Pang L., He K., Zhang Y., Li P., Lin Y., Yue J., Sci. Total Environ., 916, 170204 (2024).
• 7)  Banjac Z., Ginebreda A., Kuzmanovic M., Marcé R., Nadal M., Riera M. J., Barceló D., Sci. Total Environ., 520, 241–252 (2015).
• 8)  Bouzas-Monroy A., Wilkinson L. J., Melling M., Boxall A. B. A., Environ. Toxicol. Chem., 41, 2008–2020 (2022).
• 9)  Thamer A. A., Mustafa A., Bashar Q. H., Van B., Le P., Jakab M., Rashed R. T., Kułacz K., Hathal M., Somogyi V., Nguyen D. D., J. Environ. Manage., 359, 121058 (2024).
• 10)  Cibati A., Gonzalez-Olmos R., Rodriguez-Mozaz S., Buttiglieri G., Chemosphere, 290, 133315 (2022).
• 11)  Alfonso-Muniozguren P., Serna-Galvis A. E., Bussemaker M., Torres-Palma A. R., Lee J., Ultrason. Sonochem., 76, 105656 (2021).
• 12)  Barathe P., Kaur K., Reddy S., Shriram V., Kumar V.,J. Hazard. Mater. Lett., 5, 100105 (2024).
• 13)  Rodriguez-Mozaz S., Chamorro S., Marti E., Huerta B., Gros M., Sanchez-Melsio A., Borrego M. C., Barcelo D., Balcazar L. J., Water Res., 69, 234–242 (2015).
• 14)  World Health Organization, “Antimicrobial resistance.”: <https://www.who.int/news-room/fact-sheets/detail/antimicrobial-resistance>, cited 2 October, 2024.
• 15)  Dadgostar P., Infect. Drug Resist., 12, 3903–3910 (2019).
• 16)  Wilkinson J. L., Boxall A.B. A., Kolpin D. W., et al., Proc. Natl. Acad. Sci. U.S.A, 119, e2113947119 (2022).
• 17)  Barrow S. J., Kasera S., Rowland M. J., del Barrio J., Scherman O. A., Chem. Rev., 115, 12320–12406 (2015).
• 18)  Lagona J., Mukhopadhyay P., Chakrabarti S., Isaacs L., Angew. Chem. Int. Ed., 44, 4844–4870 (2005).
• 19)  Uzunova V. D., Cullinane C., Brix K., Nau W. M., Day A. I., Org. Biomol. Chem., 8, 2037–2042 (2010).
• 20)  Li S., Chan J. Y. W., Li Y., Bardelang D., Zheng J., Yew W. W., Chan D.P. C., Lee S.M. Y., Wang R., Org. Biomol. Chem., 14, 7563–7569 (2016).
• 21)  Ma W. J., Chen H. Y., Huang Y. L., Chen J. M., Lu T. B., Mol. Pharm., 20, 4517–4527 (2023).
• 22)  Tian F., Cheng R., Chen K., Zhao J., Small Methods, 6, 202402186–202402232 (2025).
• 23)  Wang W., Wang X., Cao J., Liu J., Qi B., Zhou X., Zhang S., Gabel D., Nau W. M., Assaf K. I., Zhang H., Chem. Commun., 54, 2098–2101 (2018).
• 24)  Tsutsumi H., Miura A., Hanada A., Sonoda R., Nakashima R., Ohata T., Ikeda H., Chem. Commun., 60, 14093–14096 (2024).
• 25)  Assaf K. I., Faraj A. N., Abu-Nameh E. S. M., Alnajjar M. A., RSC Adv., 14, 13286–13290 (2024).
• 26)  Wang R., Bardelang D., Waite M., Udachin K. M., Leek D. M., Yu K., Ratcliffe C. I., Ripmeester J. A., Org. Biomol. Chem., 7, 2435–2439 (2009).
• 27)  Thordarson P., Chem. Soc. Rev., 40, 1305–1323 (2011).
• 28)  Biedermann F., Uzunova V. D., Scherman O. A., Nau W. M., De Simone A., J. Am. Chem. Soc., 134, 15318–15323 (2012).
• 29)  Burkhanbayeva T., Guslyakov A. N., Tarikhov F. F., Pavlov D. I., Erkasov R. S., Bakibaev A. A., Potapov A. S., Yanovsky V. A., Malkov V. S., J. Saudi Chem. Soc., 28, 101819 (2024).
• 30)  Wu X., Meng X., Cheng G., J. Incl. Phenom. Macrocycl. Chem., 64, 325–329 (2009).
• 31)  Gomes A. C., Magalhães C. I. R., Oliveira T. S. M., Lopes A. D., Gonçalves I. S., Pillinger M., Dalton Trans., 45, 17042–17052 (2016).