Introduction

Tetracaine belongs to the aminoester family of local anesthetics (LA) an important class of nonciceptive agents whose action involves blockage of nervous impulse transmission. It is believed that both the cationic and the uncharged species of LA—that coexist at physiologic pH—bind to the Na+ channels of the nerve membranes, stabilizing its inactivated state and thus blocking the initiation and propagation of nervous impulses. However, LA show a relatively short duration of action (1–4 h) and may have adverse side effects such as cardiac and CNS toxicity, accompanied sometimes by allergic reactions [1].

The use of cyclodextrins (CD), calixarenes, liposomes and polymers as drug carrier systems has become an increasingly successful method to improve the potency of many therapeutic molecules whose bioavailability is threatened by problems such as limited (water/membrane) solubility, low chemical stability, fast serum clearance, etc. [2].

Among the available complexing agents, CD are the most widely used in drug formulations [3], while calixarenes belong to a new class of cyclooligomers, formed via phenol-formaldehyde reactions [4]. Calixarenes and CD share common features since they are macrocyclic molecules with a repeating unit, and both are cage-like molecules with a hydrophobic cavity. The natural cyclodextrins (α-, β- and γ-CD) have inner-cavity diameters of 5.7, 7.8, and 9.5 Å, respectively [2a]. In contrast, the inner cavity diameters of calix[4]arene, calix[6]arene, and calix[8]arene are 3.0, 7.6, and 11.7 Å, respectively [5]. Thus, the inner-cavity diameter of calix[6]arene is comparable to that of β-CD. Just like micelles, CD and sulphonate-calix[n]arenes can provide hydrophobic environments to guest molecules (through the glycoside and benzene rings, respectively), and hydrophilic outer surfaces (OH and SO −3 , respectively) [6]. However, there is a significant difference imposed by the oligosaccharide units of CD and the phenol units of calixarenes: whereas CD are quite rigid molecules, calixarenes are highly flexible molecules, possessing the ability to undergo complete ring inversion [7].

The characterization of the delivery systems formed by complexation of charged tetracaine 1 in β-cyclodextrins 2 or p-sulphonic acid calix[6]arene 3 (Scheme 1) will provide important information to optimize their future performance, which requires a better knowledge of their molecular properties [8]. Two pieces of information are crucial for the characterization of these complexes: the formation/dissociation constant and the relative positioning of the carrier/guest inclusion complex, both of which can be obtained from PGSE (pulsed field gradient spin-echo) [9] and NOE (nuclear Overhauser effect) [10] experiments, respectively. Such NMR techniques have been successfully applied to other supramolecular host-guest structures [11]. The aim of the present paper is to describe tetracaine 1 complexation with 2 and 3 (Scheme 1), by application of PGSE and NOE NMR methodologies.

scheme 1

Results and discussion

Sulphonate-calix[n]arenes have been used for the encapsulation of commercial steroids, furosemide, nifedipine and niclosamine [12] and the literature contains plenty of examples of CD based pharmaceutical formulations [2a, 13], most of them involving LA [14].

The association of 1 with 2 or 3 was first evaluated by complexation-induced hydrogen chemical shifts (Δδ) in the 1/2 and 1/3 complexes, relative to free 1. Interestingly, chemical shift differences in the 1H NMR spectra of 1—free or in 1/2 complex—were mainly observed for H-3 hydrogens (Table 1). Complexation between 1 and 3 induced large shielding effects in all the hydrogens of 1, mainly H-3′, H-4′ and H-6′ (Δδ 0.43, 0.72 and 0.63, respectively, Table 1), indicating interactions between the ammonium group of 1 with the SO −3 group of 3 [15].

Table 1 The values of the chemical shifts 1H NMR of pure 1 in D2O, complexed 1/2, 1/3 and chemical shift differences (Δδ = δ 1 fre e−δ 1 comple x)
Full size table

Diffusion-ordered spectroscopy (DOSY) NMR experiments were pivotal to demonstrate that 1 and 2 or 1 and 3 form stable complexes. It serves also to distinguish compounds or complexes by their differences in diffusion coefficients [9a]. The diffusion coefficients of pure 1, 2 and 3 (D 1 = 5.58 × 10−10, D 2 = 3.27 × 10−10 and D 3 = 3.05 × 10−10 m2 s−1, respectively) were first obtained (Table 2). In the presence of 2 or 3, compound 1 showed a significant reduction of the diffusion rate (D 1/2 = 4.24 × 10−10 and D 1/3 = 3.70 × 10−10 m2 s−1, respectively) (Table 2, representative Figs. 1 and 2), indicating that 1 formed host-guest complexes with either 2 or 3. Moreover, the diffusion rate values (Table 2) of 1 in the 1/3 complex are in agreement with the assumption of a strong association. Taking into account that we are dealing with a rapidly equilibrating system, both the chemical shifts and the diffusion coefficients are weight-averaged NMR values between free and bound 1 species. From these diffusion coefficients and applying a well established methodology [16], we have calculated the complexed population (%p bound) and apparent binding constants (K a) of the complexes. In D2O %p bound was found to be 55% and 70% for 1/2 and 1/3, respectively. The values of K a (1358 M−1 and 3889 M−1 for 1/2 and 1/3, respectively) confirmed that a stronger association took place between 1 and 3 than between 1 and 2, due to the negatively charged sulphonic groups of calix[6]arene.

Table 2 Diffusion coefficients of pure 1, 2, 3 and in the 1/2 and 1/3 complexes, in D2O (2 mmol l− 1 samples, 298 K)
Full size table
Fig. 1
figure 1

Representative 1H DOSY NMR experiment (499.885 MHz, D2O, 298 K, 2 mmol l−1) of pure 1

Full size image
Fig. 2
figure 2

1H DOSY NMR experiment (499.885 MHz, D2O, 298 K, 2 mmol l−1) of the 1/3 complex

Full size image

At pH 8.2 there is an equilibrium between the protonated and non protonated form of 1 [17]. Studies conducted at pH 10.0 revealed an increase in the association between uncharged tetracaine 1 and 2 (K a  = 6597 M−1). The association constant between uncharged 1 and 3 was not determined due to the alkaline hydrolysis of the ester bond of 1 (t 1/2 < 0.1 h). Fortunately alkaline hydrolysis was prevented (t1/2 > 16 h) by the strong association of 1 and 2 at pH 10.0, allowing the DOSY experiments to be carried out.

We further established the stoichiometry for the 1/2 and 1/3 complexes, using the Job plot method [18]. The plots obtained from the NMR analysis indicated the predominant formation of 1:1 complexes, both between 1 and 2 and with 1 and 3 (Figures not shown).

To gain more insight into the topological aspects of these two complexes (1/2 and 1/3) we have performed 1H-ROESY NMR experiments, which are usually suited to measure NOEs in complexes with ωτc close to 1 [19]. Specific ROE signals were observed between H-2 and H-3 of 1 with H-3 (enhancement of 0.46% and 0.45%, respectively) and H-5 (1.01% and 0.15% of signal enhancement respectively) of β-CD 2. We therefore suggest that the aromatic moiety of tetracaine 1 was inside the CD cavity. We have also addressed the association issue however no dipolar interaction between the H-1, H-2 and H-4 of the β-CD 2 was observed with the tetracaine 1, thus the 1/2 complex’ topology was proposed and is depicted in Fig. 3.

Fig. 3
figure 3

Proposed topology for the 1/2 complex, based on 1H NMR evidences

Full size image

The signal enhancements in 1/3 (ROE) between H-6′ of the tetracaine 1 and H-3 (1.27% signal enhancement) of the p-sulphonic acid calix[6]arene 3 indicate that the ammonium group of 1 can be included in the cavity of 3, as suggested in Fig. 4. The lack of more ROE signal is probably due to the conformational behavior of the p-sulphonic acid calix[6]arene 3 which will be better investigated in the near future.

Fig. 4
figure 4

Proposed topology for the 1/3 complex, in fast equilibrium, according to 1H NMR evidences

Full size image

Conclusions

The combined use of PGSE and ROE techniques has helped the determination of the host-guest structure, kinetic stability and degree of guest encapsulation in solution, for the 1/2 and 1/3 complexes.

The complexation of 1 with either 2 or 3 has been investigated in view of its potential use for the preparation of new therapeutic formulations, to increase the bioavailability and to decrease the systemic toxicity of tetracaine 1 in anesthesia procedures. In water, we determined a strong association between 1 and 2. The interaction between 1 and 3 was found to be even more stable, due to the binding of the ammonium group of 1 and the SO −3 group of 3. This was confirmed by experiments performed at pH 10.0, which revealed that the association between 1 and 3 decreased in relation to that in D2O, while that of 1 and 2 increased (K a  = 6597 M−1), preventing the alkaline hydrolysis of 1.

The proposed topologies of the 1/2 and 1/3 complexes were established using ROESY 1D and they are intrinsically distinct. In 1/2, 1 is almost totally inserted into the cyclodextrin 2 cavity while in 1/3 it is located at the sulphonic rim so that we can say that 1/2 complexation is governed by hydrophobic interactions while complexation of 1/3 is governed by ion pair interactions. A topic of major interest in the development of drug delivery system is to understand the specific features that determine the interaction between host-guest molecules and the study of the molecular aspects involved in it is continuing in our laboratory.

Experimental

Chemicals and reagents

Tetracaine 1 (99%), β-CD 2 (99%), and D2O (99.75%) were purchased from Aldrich, Acros Organics and Merck, respectively. All other reagents were of analytical grade. p-sulphonic acid calix[6]arene 3 was synthesized in our laboratory following literature procedures [20].

Preparation of solid inclusion complexes

Inclusion complexes (1/2 or 1/3) with 1:1 molar ratios were prepared by shaking appropriate amounts of 1 and 2 or 3, e.g. 2 mmol l−1, in deionized water at room temperature (298 ± 1 K) for 1 h. Kinetic experiments revealed that equilibrium was reached after 40 min incubation (data not shown).

After reaching equilibrium, the solution was freeze-dried in a Labconco Freeze-dry system (Freezone 4.5) and stored at 253 K until further use.

NMR spectroscopy

All experiments were performed at 298 K in D2O. For the experiments with uncharged 1, the pH value of the solutions was adjusted by addition of 0.02 mol l−1 carbonate buffer, prepared in D2O.

Routine 1D 1H experiments were acquired with an INOVA-500 Varian spectrometer operating at 499.885 MHz for 1H (64 k data points, 30° excitation pulse duration of 2.2 μs, spectral width of 6 kHz, acquisition time of 3.3 s and relaxation delay of 10 ms) in a 5 mm probe with inverse detection mode at room temperature unless stated otherwise.

NOE measurements The ROESY 1D experiments were obtained with a selective 180° and a non-selective 90° pulse, a mixing time of 0.5 s was used during the spin-lock. The selective pulses were generated by a waveform generator, which automatically attenuates the shape, power, and pulse duration to obtain the required selectivity. The subtraction of the on- and off-resonance acquisition furnished the ROESY 1D experiment. All spectra were acquired with a 5 mm inverse probe at 298 K in 5 mm tubes.

HR-DOSY experiments were carried out by carefully choosing the correct pulse sequence and gradients for the experiments. The measurements were made using: (a) 5 mm inverse probe with Z-gradient coil; (b) the GCSTESL (Gradient Compensated Stimulated Echo Spin Lock) HR-DOSY sequence; (c) amplitudes of the gradient pulses ranging from 0.000685 to 0.003427 T cm−1, where an approximately 90%–95% decrease in the resonance intensity was achieved at the largest gradient amplitudes. For all experiments, 25 different gradient amplitudes were used. The baselines of all arrayed spectra were corrected prior to processing the data. The processing program (the DOSY macro in a Varian instrument) involves the determination of the peak heights of all signals above a pre-established threshold and the fitting of the decay curve for each peak to an exponential decay. The DOSY macro was run with data transformed using fn = 64 K. Very crowded spectra were processed in sections due to the limitation of handling only 512 lines at a time. The results of the DOSY method of analysis are pseudo two-dimensional spectra with NMR chemical shifts along one axis and calculated diffusion coefficients (m2 s−1 × 10−10) along the other.

Determination the stoichiometry of complexation

Job plots have been prepared with 2 mmol l1− stock solutions of 1 and 2 or 1 and 3 [21].