Formulation and Evaluation of Thienorphine Hydrochloride

December 2012 Regular Article Chem. Pharm. Bull. 60(12) 1479–1486 (2012) 1479

Formulation and Evaluation of Thienorphine Hydrochloride Sublingual Delivery System Fei Liu,a,b Yumei Zhao, a Jianxu Sun,a Yongliang Gao,*,a and Zhenqing Zhang*,a a Beijing Institute of Pharmacology and Toxicology; No. 27 Taiping Road, Beijing 100850, P. R. China: and b Department of Pharmacy, The First Afﬁliated Hospital of PLA; No. 51 Fucheng Road, Beijing 10048, P. R. China. Received June 7, 2012; accepted September 4, 2012

Thienorphine hydrochloride (ThH) is a highly insoluble and readily metabolized partial-opioid agonist. It is used for the treatment of pain and heroin addiction. This study aimed to formulate and evaluate sublingual delivery systems containing ThH. Dimethyl-β-cyclodextrin (DM-β-CD) can enhance the solubility and permeability of hydrophobic drugs. In this paper, ThH cyclodextrin inclusion complexes were prepared and administrated sublingually with the objective of improving the drug’s aqueous solubility, in vitro permeation rate, and in vivo absorption rate. The formulation was prepared with DM-β-CD using the freeze-dried method and characterized using phase solubility, differential scanning calorimetry (DSC), X-ray and NMR analy- ses. The results of each test indicated the formation of dynamic inclusion complexes between ThH and DM- β-CD. The inclusion complexes also showed signiﬁcant increases in in vitro aqueous solubility and mucosal

permeability. According to the pharmacokinetic study of the complex in rats, the AUC and Cmax values of the sublingual delivery group were 40 and 46 times higher than those of the gastrointestinal group, whereas tmax was shorter, which proved that in vivo absorption and metabolism had been improved. It can therefore be concluded that the inclusion technology and sublingual delivery system were suitable for ThH development. Key words bioavailability; sublingual drug delivery; absorption; solubility; cyclodextrin; thienorphine

Thienorphine (21-cyclopropyl-7-[1-(R)-1-hydroxy-1-methyl- network and finally the oral membrane. In addition, perme- 3-(thien-2-yl) propyl]-6,14-endoethano-6,7,8,14-tetra hydro- ation across the sublingual mucosa can be easily achieved oripavine, Fig. 1, is a new type of partial-opioid agonist, was when the concentration of active ingredient is relatively high. synthesized by the chemists in our institute.1) Meanwhile, Therefore, to develop a preparation of ThH suitable for sublin- ThH, a hydrochloride salt with a crystal water of thienor- gual delivery, increasing solubility is the primary concern.6,7) phine, has been developed as drug candidate. It has fared Cyclodextrins (CDs), cyclic oligosaccharides consisting comparatively well with respect to aqueous solubility and of 6–8 glucopyanose units, can trap the lipophilic drug in a pharmacodynamic action compared to its parent compound cage-like meshwork.8) So far, CDs have been reported to in- and its other salts.2) The drugs currently in clinical use for crease the solubility, stability, and bioavailability of drugs in a the treatment of opioid dependence are either full-opioid few different delivery systems.9,10) Many studies have reported agonists or antagonists. Partial-opioid agonists could be more that β-cyclodextrin (β-CD, Fig. 2) is a good natural cyclodex- widely used because they provide certain advantages. They trin because of its efficient drug complexation and availability −2 have unique pharmacological properties that allow them to in pure form. However, its low water solubility (1.5×10 M) prevent the addiction that can occur with full-opioid agonists and toxicity limit their application in pharmaceutical formula- and there has been a lack of subjective side effects.3) Although tions.11,12) Recently, various kinds of chemically modified CD ThH tablets have already passed through clinical testing in derivatives have been prepared to extend the physicochemical China, further development of ThH has been restricted by its properties and inclusion capacity of parent CDs. One of the very low oral bioavailability. This has two causes. First, ThH derivatives is dimethyl-β-cyclodextrin (DM-β-CD, Fig. 2), is poorly soluble, which causes a great deal of dissolution in which has more pronounced solubilization (0.2 M) and com- the gastrointestinal (GI) tract and significant enteric excretion. plexation than its parent compound. It also has some features Second, it is extensively metabolized by the liver, as previ- that make it safer for use in clinical settings.13) The DM-β-CD ously indicated.4) in particular has been shown to be excellent absorption en- Oral mucosal drug delivery is an alternative dosage form hancers in mucosal drug delivery.14) These all indicate that for thienorphine. It has been suggested that the sublingual they are well suited to sublingual administration. route has the following advantages: sufficient blood supply, This study aimed to formulate and evaluate a sublingual avoidance of the liver and GI, rapid absorption, and high bioavailability. Buccal and sublingual sectors are commonly used routes for drug delivery. The sublingualis more permeable and thinner than the buccal mucosa, with higher blood flow.5) This suggests that one way of improving ThH bioavailability may be the development of a sublingual delivery system. How- ever, sublingual delivery has strict requirements. The compound must first dissolve and disperse into the saliva. Then it must cross the unstirred water layer consisting of the mucin

The authors declare no conﬂict of interest. Fig. 1. Chemical Structure of Thienorphine

prepared with water saturated with oil phase. One milliliter of each solution was then transferred to 10 mL centrifuge tubes containing 1 mL of oil phase saturated with water. The tubes were vortex-mixed for 2 h at 37°C and centrifuged at 3000×g for 5 min. After centrifugation, 100 µL was withdrawn from

the water phase and assayed by HPLC at time zero (C0) and after shaking to ensure partition (Cw). The partition coefﬁcient was Koil/water=(C0−Cw)/Cw. The experiments were performed in triplicate. HPLC analysis of thienorphine was performed on a Shimadzu 10A HPLC system (Shimadzu, Japan). The detection wavelength was set at 220 nm. Chromatographic

separation was carried out with a Varian ODS-C18 column (4.6 mm×250 mm, 5 µm; Varian, U.S.A.) with mobile phase consisted of methanol and phosphoric acid solution (pH= 3) in Fig. 2. Chemical Structure of β-Cyclodextrin (R=H) and Dimethyl- a ratio of 40 : 60 (v/v), and the column temperature was 35°C. β-CD (R=CH ) 3 The flow rate was 1.0 mL·min−1. Phase Solubility Test The phase solubility studies were delivery system for ThH-DM-β-CD inclusion complexes. performed according to the method of Higuchi and Connors.15) Based on the in vitro results, a suitable formulation was con- A schematic explanation of the principle of this technique is structed and administered to test rats by a sublingual route. given below, via Eq. 1: Pharmacokinetic parameters were compared to those of rats K given the drug via the gastrointestinal tract (GIT). The results ThH+ DM-ββ -CDr ThH:DM- -CD (1) K of each test indicate that the inclusion complexes prepared by d the co-lyophilization method were satisfactory with respect to Here Kr is the combination rate constant for the complex complexing and increasing solubility, permeability, and bio- formation and Kd is the dissociation rate constant of the com- availability. plex. Excess ThH was added into aqueous solutions of DM-β-CD Experimental at a serial concentrations in capped tubes and shaken at 25°C Materials ThH (>99%) was synthesized in the Institute for 48 h. After reaching equilibrium, the suspensions were fil- of Pharmacology and Toxicology (Beijing, China). Buprenor- tered through a 0.45 µm Millipore filter followed by the quan- phine standard (>99%) n-octanol (>99%) was purchased tification of the drug by HPLC. Phase solubility curves were from the Sigma-Aldrich Chemical Co. (U.S.A.). DM-β-CD represented as the total dissolved drug concentration relative

(>98%) was purchased from the Li Quan Co., Ltd. (Shanxi, to the concentration of DM-β-CD. The stable constant (Ks) China). Methanol, acetonitrile and other materials were of was calculated from the initial straight portion of the phase analytical grade. solubility diagram using Eq. 215): Preparation of ThH-DM-β-CD Complexes The bulk slope ThH-DM-β-CD complexes were carried out according to the Ks = (2) S (1 slope) freeze-dried method as follows: ThH and the corresponding 0 amount of DM-β-CD (molar ratio 1 : 2) were dissolved in etha- Here S0 is the solubility of ThH in the absence of DM-β-CD nol and water, respectively. The resulting aqueous mixtures (equal to the intercept of the diagram), slope is the slope of were stirred for 2 h at constant temperature of 60°C. The the experimental phase solubility diagram for ThH-DM-β-CD, above preparation was then frozen and freeze-dried for 24 h, and Ks is the stability constant (Kr/Kd). using a lyophilizer (MDF-382E, Sihuan, China) at −80°C for Differential Scanning Calorimetry (DSC) Test Thermal 24 h. properties of the powder samples were investigated with a Aqueous Solubility The aqueous solubility of the ThH, differential scanning calorimeter (CDR-4P, Shanghai, China). ThH/DM-β-CD physical mixture and its DM-β-CD complexes Approximately 10 mg of sample was analyzed in an open were compared using a constant-temperature shaker (THX-82, aluminium pan, and heated at scanning rate of 10°C·min−1 Shanghai, China). The suspensions with excess amounts of from room temperature to 500°C. The results indicated the ThH, physical mixture, and inclusion complex were placed differences in the profiles, which were attributed to ThH, DM- in capped tubes. The tubes were placed in a water bath at a β-CD, the physical mixture, and the ThH-DM-β-CD complex. constant temperature (37°C), and were shaken in horizontal X-Ray Diffractometry (XRD) Test Diffraction patterns movements (100 min−1 for 48 h to allow equilibrium). After were obtained using a high resolution X-ray diffractometer equilibrium was attained, an aliquot was filtered through a (D/max r-B, Rigaku, Japan), using Ni-filtered CuKα (1.542 Å) 0.45 µm Millipore filter, and the thienorphine content was radiation (40 kV, 40 mA). Powder samples were mounted on measured by HPLC. The experiments were performed in trip- a sample holder and scanned from 5° to 50° in 2θ at a speed licate. of 0.02°·min−1. The samples analyzed were the same as those Partition Coefficient Water and oil solutions (n-octanol) used in the DSC experiments. were saturated with each other before the experiment. Solu- Proton Nuclear Magnetic Resonance (NMR) Spectra tions of ThH and ThH/DM-β-CD physical mixture (molar Test All proton detection experiments including one-dimen- ratio 1 : 2) and ThH-DM-β-CD complexes (100 µg·mL−1) were sional 1H-NMR (400 MHz) and the nuclear Overhauser effect December 2012 1481 spectroscopy (NOESY) of the sample in deuterium oxide

(D2O) were conducted using a JNM-ECA-400C spectrometer operated at 25°C. To eliminate the interference of moisture, all the samples and NMR tubes were vacuum dried at 80°C over 6 h before analysis. The chemical shifts are expressed using tetramethylsilane as an internal standard. In Vitro Mucosa Permeation Test The permeability study was carried out with Franz diffusion cells. The fresh porcine sublingual mucosa was used as a model in this test, owing to the similarity of composition, structure and permeability measurements between pigs and humans.16,17) Sublin- Fig. 3. Profiles of ThH Saturation Solubility at 37°C for Bulk Powder, gual mucosal specimens were obtained from tissue removed Physical Mixture of ThH and DM-β-CD (ThH : DM-β-CD=1 : 4, Mole from freshly slaughtered pigs and equilibrated in phosphate Ratio) and Lyophilized ThH-DM-β-CD Complex buffered saline (PBS) for 1 h before starting experiments. The * versus bulk ThH, p<0.01; # versus physical mixture, p<0.01. Franz diffusion cells had a receiver volume of 6.5 mL and diffusional area of 0.7 cm2. The receiver compartments were filled with degassed dissolution medium, which was a mixture ThH-DM-β-CD complex solution GI delivery; II: ThH-DM- ethanol and PBS at pH 6.8 (2 : 3, v/v), to promote drug solu- β-CD complex solution sublingual delivery). When adminis- bilization in the receptor. This medium did not alter barrier tered sublingually, rats were anesthetized with halothane (5% permeability.18,19) Micromagnetic stirrer bars were added to induction, 0.7% maintenance dose), and a suitable volume of the receptor compartments and the complete cells were placed liquid formulation was administered underneath the tongue in a water bath at 37°C. The sample (drug or drug-CD system using a pipette. The lower jaws of the rats were supported equivalent to 0.8 mg of ThH) was applied over the mucosa in the horizontal position for 20 min to minimize the chance surface. The donor compartment was closed with the screw of swallowing. Serial blood samples were collected at prede- cap to preserve the humid environment of the mucosa. At termined points in time (at 5, 10, 20, 30, 40 min and 1, 2, 3, predetermined times, aliquots of the receptor medium were re- 4, 7, 12, 24 h after GI administration; at 2, 5, 10, 20, 40 min moved and replaced with fresh dissolution medium. The sam- and 1, 2, 3, 4, 7, 12, 24 h after sublingual administration). ples withdrawn from this system were diluted and analyzed by Then 0.3 mL blood samples were collected, and the sample HPLC. The HPLC conditions were the same as those used in was centrifuged (3000×g, 15 min) and plasma was collected the solubility test. Each test was performed eight times. and stored at −20°C until analysis. The methods of plasma −2 −1 The flux values (Js, µg·cm ·min ) across the mucosa were sample preparation and detection were identical to those used calculated at the steady state per unit area by linear regression by Kong et al.4) The concentration of ThH was quantitated in analysis of permeation data, as shown in Eq. 3: this test. Statistics Statistical differences were estimated using stu- Qr Js = (3) dent’s t-test. Pharmacokinetic parameters were obtained using A t the practical pharmacokinetic program version 87 (Committee

Here Qr is the quantity of ThH which passes through the of Mathematic Pharmacology of the Chinese Society of Phar- tissue into the receptor compartment (µg), A is the oriﬁce area macology, China). for diffusion (cm2), and t is the duration of exposure (min). −1 The permeability coefﬁcient (Kp, cm·min ) was then calcu- Results and Discussion lated via Eq. 4: Solubility Test The solubility of bulk ThH was only −1 J 479.23 µg·mL at 37°C. As shown in Fig. 3. The physical K = s (4) p C mixture (ThH : DM-β-CD=1 : 4, mole ratio) showed a slightly d increase in the solubility of ThH (903.77 µg·mL−1). The in-

Here Js is the ﬂux calculated at the steady state clusion complexes displayed much more aqueous solubility, −2 −1 −1 (µg·cm ·min ), and Cd is the drug concentration in the which was 10469.08 µg·mL . donor compartment (400 µg·cm−3).20) Cyclodextrins are useful functional excipients that have In Vivo Absorption Test. Animals Wistar rats (200± enjoyed widespread attention and use. The reason for this 20 g) supplied by the Academy of Military Medical Sciences popularity is the ability of these materials to interact with Animals Center (Beijing, China) were used for pharmacoki- poorly water-soluble drugs and drug candidates, increasing netic studies. The animal were acclimatized at a temperature their apparent aqueous solubility. ThH is a highly insoluble, of 25±2°C and a relative humidity of 70±5% under natural highly metabolized drug candidate. It was solubilized by DM- light/dark conditions for 1 week and were given food and β-CD via formation of a soluble complex. The characterization water ad libitum. Prior to the experiment, the animals were results of the complex indicate that ThH is amorphous and kept under fasting conditions overnight. All experimental pro- consistent with inclusion of the ThH in the DM-β-CD cavity. cedures abided by the ethics and regulation of animal experi- The fact that both ThH and DM-β-CD are both amorphous ments of the Academy of Military Medical Sciences, China. lead to an enhanced solubility, and DM-β-CD stabilizes the Pharmacokinetic Test The dose of ThH-DM-β-CD com- ThH thereby preventing crystallization. The signiﬁcant differ- plex was 3 mg·kg−1. This dosage was selected on the basis of ence between the saturation solubility of the physical mixture a report on ThH given to rats (4). Twelve rats were equally and that of the lyophilized complex indicated that the prepara- divided into two groups by different administrations. (I: tion process is vital for this combination. 1482 Vol. 60, No. 12

Partition Coefficient Several solvent systems have been used to relate partition coefficients to mucosal absorption. N-Octanol is a suitable system because its polar and non- polar nature mimics the complex nature of the mucosa.21) In this study, the partition coefficient of DM-β-CD complexes

(Koil–water≈13) was found to be approximately half that of the pure drug (Koil/water≈26). This showed that the lipophilic characteristics of ThH decreased when it was complexed with CDs. This result was not same as the experiment by Frijlink, which showed that the addition of β-CD was not due to the obvious decrease in partition coefficients. This result was consistent with another experiment, which showed that the partition coefficients of pure drugs decreased when those drugs were complexed with various CDs.22) Partition coefficients of papaverine alone or in the presence of DM-β-CD at pH 7.4 Fig. 4. Phase Solubility Profile of ThH with DM-β-CD in Distilled was tested, and the value was obviously decreases in the com- Water at 25°C (n=3, Mean±S.D.) plexes.23) We insist that the difference between those results was caused by the absence of drug-cyclodextrin interactions. In the solubility test, the physical mixture showed a slight in- cavity can cause its melting, boiling, and sublimation points to crease in the solubility of ThH, while the inclusion complexes shift or disappear.29) Figure 5 shows DSC thermograms of the displayed much higher aqueous solubility. pure components, the physical mixture, and lyophilized pow- Lipophilicity is one of the determinants of drug permeabil- der of ThH-DM-β-CD complexes. The curve of ThH (Fig. 5A) ity through membranes, and the partition coefficient is closely displayed two endothermic peaks: a broad endotherm between related to the lipophilicity.24) However, lower lipophilicity 80°C and 120°C corresponds to the loss of bounded water. does not equal to poor ability to cross the sublingual mucosa. The other is a sharp endothermic peak at 258.16°C, indicating The drug transport mechanism through the mucosa involves the melting point. The thermogram of pure DM-β-CD (Fig. two major routes: the transcellular (across the cells) and para- 5B) powder exhibited a broad endothermic peak near 100°C. cellular (between the cells) pathways. The presence of tight This is because β-CD did not produce any peaks of interest junctions between intestinal epithelial cells is the primary when examined by DSC.30) This result was consistent with barrier to paracellular drug transport through the intestine those of other reports.31–33) The physical mixture of ThH and and nasal mucosa. However, tight junctions are rare in the DM-β-CD (Fig. 5C) apparently contains only the free species, oral mucosa. Therefore, most compounds actually traverse the but they are present at a much lower level of intensity due to buccal mucosa via the paracellular route.25) The increase in the dilution by DM-β-CD. This indicated the absence of drug- aqueous solubility and decrease in lipophilicity of ThH with cyclodextrin interactions. A different pattern was observed in inclusion may not have led to poor ability to cross the sublin- the thermogram of the association complexes (Fig. 5D). No gual mucosa. appreciable endothermic event associated with the melting of Phase Solubility Test Phase solubility testing of the ef- the ThH was observed, suggesting the likelihood of formation fect of complexing agents on the compound being solubilized of true inclusion complexes between ThH molecules and DM- is a standard approach to determining not only the value of β-CD. This in turn led to an almost complete loss of crystal- the stability constant but also to finding insight into the stoi- linity in this binary system. chiometry of the equilibrium.26) The 1 : 1 drug–cyclodextrin XRD Test Figure 6 shows the X-ray diffraction patterns complex is the most common type of association. In this asso- of ThH, DM-β-CD, the physical mixture, and the ThH-DM- ciation, a single drug molecule is included in the cavity of one β-CD complex. The diffraction pattern of ThH (Fig. 6A) cyclodextrin molecule, and there is a stability constant for the displayed sharp peaks, which is characteristic of crystalline equilibrium between the free and associated species.27) The compounds. In contrast, DM-β-CD showed a broad, diffuse phase solubility profiles of the ThH-DM-β-CD systems are pattern (Fig. 6B), suggesting that it was essentially amor- presented in Fig. 4 (y=0.4915x+0.8632). They show that drug phous. In the physical mixture shown in Fig. 6C, drug crystal- solubility increased linearly with increasing concentration of linity peaks were still detectable. The profile consisted mostly various CDs. The solubility curve with correlation coefficient of the ThH characteristics, while some of the DM-β-CD char- values (r=0.998) can be classified as a straight line (AL type) acteristics remained, indicating that no new physical state had to the model proposed by Higuchi and Connors.15) It can be formed. Compared to the diffraction patterns of pure ThH and related to the formation of a soluble inclusion complex. Ac- DM-β-CD, the profile of the complex (Fig. 6D) was similar −1 cording to Eq. 2, the Ks was calculated to be 1160 M in our to that of the amorphous DM-β-CD shown in Fig. 5B. These study. Furthermore, the favorable Ks values of general drug- results suggest that ThH is no longer present as a crystalline −1 28) CD inclusions are reported in the range 50–2000 M . These material and its DM-β-CD solid complexes exhibit complete may indicate that it is feasible to prepare a ThH-DM-β-CD amorphization. This result was consistent with previous re- inclusion complex to improve the solubility of drug. ports.34) Differential Scanning Calorimetry Test DSC is used in 1H-NMR Test 1 H-NMR is a another powerful tool to the pharmaceutical field to establish the identity and purity of investigate the formation of the inclusion of a guest molecule solid-state systems and to detect interactions between their in to the CD cavity. AS shown in Fig. 7, 1H-NMR spectros- components. Incorporation of a guest molecule into the CD copy of ThH displayed four different types of proton signals, December 2012 1483

Fig. 5. Differential Scanning Calorimetry (DSC) Thermograms of Fig. 6. X-Ray Diffraction Patterns of (A) ThH, (B) DM-β-CD, (C) (A) ThH, (B) DM-β-CD, (C) Physical Mixture of ThH and DM-β-CD Physical Mixture of ThH and DM-β-CD (ThH : DM-β-CD=1 : 1, Mole (ThH : DM-β-CD=1 : 1, Mole Ratio), (D) Lyophilized ThH-DM-β-CD Ratio), (D) Lyophilized ThH-DM-β-CD Complex Complex

Fig. 7. 1H-NMR Spectra of (A) ThH, (B) DM-β-CD, (C) Lyophilized ThH-DM-β-CD Complex namely a-H, b-H, c-H (thiophene protons) as well as d-H is deeply included inside the CD cavity. In addition, it can be (oripavine protons). The proton signals of ThH are affected by seen after inclusion, there were slightly downfield chemical the inclusion process and the change of chemical shifts of var- shift of the protons (3-H, 5-H) in the CD cavity. The results ious relevant protons in the ThH is listed in Table 1. It is seen indicated that the environment of these protons was changed that the chemical shift changes (Δδ) of the c-H are larger com- after the inclusion of ThH with the DM-β-CD. pared with d-H. On the whole, the change of chemical shifts 2D-NMR test For a better understanding of the inter- showed the following order: Δδ c-H>Δδ b-H>Δδ a-H>Δδ action model of the inclusion complex of ThH-DM-β-CD, d-H. It seems resonable to postulate that the thiophene ring 2D-NMR spectroscopy has also been used to study the 1484 Vol. 60, No. 12 inclusion complexation of DM-β-CD with ThH. Figure 8 Table 1. 1H Chemical Shifts Corresponding to ThH and ThH-DM-β-CD shows a spatial contour plot of the NOESY spectrum for the Complex Solution ThH/DM-β-CD system. Several intermolecular cross-peaks ThH-DM-β-CD Proton ThH (δ /ppm) Δδ(δ−δ ) between a-H, b-H, c-H of ThH and 3-H, 5-H of DM-β-CD 0 complex (δ/ppm) 0 were observed, indicating that ThH inserted into the hydrophobic cavity of cyclodextrin and the thiophene ring of ThH a-H 7.19 7.16 −0.03 was deeply included in the CD cavity. b-H 6.93 6.85 −0.08 In Vitro Mucosa Penetration Test After applying the c-H 6.87 6.74 −0.13 d-H 6.76 6.74 −0.02 drug to the mucosa, it must remain physicochemically and 3-H 3.83 3.87 0.04 enzymatically stable for a certain period of time. All experi- 5-H 3.69 3.76 0.07 ments were performed for 2.5 h because it was found that pro- longing the experiments over this time would cause the cells could die and the data to become inaccurate. The permeation and 6.109·10−3 cm·min−1. profiles of ThH and the complexes are illustrated in Fig. 9, The complex crosses the mucosa more easily than its coun- expressed as the cumulative amount of ThH permeated per terpart lacking inclusion. This corresponds to the conclusion unit surface area (cm2) over time. This showed a linear trend reached in the partition coefficient test, in which increases (ThH: y=0.9872x−16.6, R2=0.9777, lyophilized complex: in aqueous solubility and decreases in lipophilicity caused 2 y=2.4436x−35.768, R =0.9896). The flux values (Js) of ThH better paracelluar transfer of ThH on through the sublingual and complex equal to the slope of the linear portion of the plot, mucosa. Beyond that, CD is an effective absorption enhancer which were 0.9872 µg·cm−2·min−1, and 2.4436 µg·cm−2·min−1, in mucosal drug delivery.14,35) Babu and Pandit have reported respectively. The Kp values were calculated via the Eq. 4: the that methylated β-CD can extract all the major lipid classes mean values of the drug and the complexes were 2.468·10−3 and proteins and reduce the barrier function of the skin.36)

Fig. 8. Partial Contour Plot of a NOESY Spectrum of ThH-DM-β-CD Complex December 2012 1485

Fig. 9. Permeation of Different ThH Solutions through the Porcine Sublingual Mucosa (■) ThH, (□) lyophilized ThH-DM-β-CD complex. n=8, means±S.D.

Fig. 10. Mean Plasma Concentration–Time Profile of Thienorphine Even though the specific mechanism of the action of CDs in after GI (◆) or Sublingual (□) Administration of ThH-DM-β-CD Com- increasing the transcellular route has not yet been clarified, plex Solution −1 it seems likely that the regular arrangement of the lipid mol- 3 mg·kg , n=6, mean±S.D. ecules that constitute the cell membrane is perturbed by the interaction of membrane lipids and CDs.37) Table 2. Mean Pharmacokinetic Parameters of Thienorphine after GI or In Vivo Absorption Test The mean plasma concentra- Sublingual Administration of ThH-DM-β-CD Complex Solution tion–time curves of both groups are shown in Fig. 10. The Administration tracts mean estimated pharmacokinetic parameters are listed in Pharmacokinetic Table 2. Group II showed higher thienorphine plasma con- parameters GI Sublingual centration than the other group. The AUC and C of 0–t max Tmax (h) 0.25±0.00 0.17±0.00* group II were 40 and 46 times higher than those of group I, −1 Cmax (ng·mL ) 15.69±10.72 714.80±163.61* respectively, whereas tmax was shorter. This proves that the T1/2 (h) 2.41±1.21 4.53±3.14* administration of DM-β-CD complex solution by sublingual Clz/F (L·h−1·g−1) 171.35±57.05 4.13±1.10* −1 delivery significantly increased the rate and extent of absorp- AUC0–t (µg·h·L ) 19.18±6.69 770.10±209.33* tion of ThH (p<0.01). Meanwhile the t1/2 was longer, indicat- 3 mg·kg−1, n=6, means±S.D. * p<0.01, statistically significant difference from ing that sublingual delivery decreased the rate and extent of GI group. metabolism (p<0.01). This was also proved by the obvious decrease in Clz/F. Drugs are rapidly and quantitatively released from their was also found to be an effective promoter of permeation complexes, and the CDs do not alter the intrinsic pharmaco- through the sublingual mucosa, as indicated by an in vitro kinetics of drugs.38) No changes in metabolism could have mucosa permeation test. Pharmacokinetical profiles and pa- been induced by the inclusion, but may have been caused by rameters in rats indicated that the inclusion complex was the avoidance of fist-pass effect, which was precisely discov- absorbed faster and to a greater overall degree. It was me- ered before.4) In addition, interactions between DM-β-CD and tabolized more slowly when delivered by the sublingual route biomembranes cannot be excluded; CDs have been shown to than through the GIT. The present study demonstrates that the interact with membrane lipids and improve permeation and ThH-DM-β-CD method of preparing the inclusion complex absorption.24) However, the absorption of group I was very and our sublingual delivery system selection are suitable for poor. This may have been caused by the high pH in the intes- the pharmaceutical development of ThH. tinal tract and the poor dissolution of the DM-β-CD complex under basic conditions. This suggests that the good absorption Acknowledgment The work is now supported by National observed after sublingual administration was not caused by Natural Science Foundation of China (30472059). swallowing. References Conclusion 1) Yu G., Yue Y. J., Cui M. X., Gong Z. H., J. Pharmacol. Exp. Ther., Formulation with DM-β-CD markedly improved the aque- 318, 282–287 (2006). ous solubility of hydrophobic compound ThH and decreased 2) Li J. X., Becker G. L., Traynor J. R., Gong Z. H., France C. P., J. its lipophilicity. The interactions between the drug and excipi- Pharmacol. Exp. Ther., 321, 227–236 (2007). ent were characterized by studies including phase solubility, 3) Johnson R. E., Strain E. C., Amass L., Drug Alcohol Depend., 70 (Suppl.), S59–S77 (2003). DSC, X-ray diffraction and NMR investigation, in addition, it 4) Kong Q., Qiao J., Wang X., Yuan S., Zhang Z., Gong Z., Ruan J., J. seems resonable to postulate that the thiophene ring is deeply Chromatogr. B, 859, 52–61 (2007). included inside the CD cavity. All studies suggested that the 5) Madhav N. V., Shakya A. K., Shakya P., Singh K., J. Controlled preparation of inclusion complex was successful. DM-β-CD 1486 Vol. 60, No. 12

Release, 140, 2–11 (2009). 21) Frijlink H. W., Schoonen A. J. M., Lerk C. F., Int. J. Pharm., 49, 6) Veuillez F., Kalia Y. N., Jacques Y., Deshusses J., Buri P., Eur. J. 91–102 (1989). Pharm. Biopharm., 51, 93–109 (2001). 22) Vianna R., Bentley M., Ribeiro G., Carvalho F., Neto A., Oliveira 7) Loftsson T., Vogensen S. B., Brewster M. E., Konrádsdóttir F., J. D., Collett J., Int. J. Pharm., 167, 205–213 (1998). Pharm. Sci., 96, 2532–2546 (2007). 23) Ventura C. A., Puglisi G., Zappalà M., Mazzone G., Int. J. Pharm., 8) Bellia F., La Mendola D., Pedone C., Rizzarelli E., Saviano M., 160, 163–172 (1998). Vecchio G., Chem. Soc. Rev., 38, 2756–2781 (2009). 24) Cho E., Gwak H., Chun I., Int. J. Pharm., 349, 101–107 (2008). 9) Matsuda H., Arima H., Adv. Drug Deliv. Rev., 36, 81–99 (1999). 25) Nicolazzo J. A., Reed B. L., Finnin B. C., J. Controlled Release, 10) Yavuz B., Bilensoy E., Vural I., Sumnu M., Int. J. Pharm., 398, 105, 1–15 (2005). 137–145 (2010). 26) Brewster M. E., Loftsson T., Adv. Drug Deliv. Rev., 59, 645–666 11) García-Río L., Hall R. W., Mejuto J. C., Rodriguez-Dafonte P., (2007). Tetra hedron, 63, 2208–2214 (2007). 27) Davis M. E., Brewster M. E., Nat. Rev. Drug Discov., 3, 1023–1035 12) Ohtani Y., Irie T., Uekama K., Fukunaga K., Pitha J., Eur. J. Bio- (2004). chem., 186, 17–22 (1989). 28) Loftsson T., Hreinsdóttir D., Másson M., Int. J. Pharm., 302, 18–28 13) Hussain A., Yang T., Zaghloul A. A., Ahsan F., Pharm. Res., 20, (2005). 1551–1557 (2003). 29) Kamphorst A. O., Mendes de Sá I., Faria A. M., Sinisterra R. D., 14) Merkus F. W., Verhoef J. C., Marttin E., Romeijn S. G., van der Kuy Eur. J. Pharm. Biopharm., 57, 199–205 (2004). P. H., Hermens W. A., Schipper N. G., Adv. Drug Deliv. Rev., 36, 30) Liu J., Qiu L., Gao J., Jin Y., Int. J. Pharm., 312, 137–143 (2006). 41–57 (1999). 31) Vlachou M., Papaïoannou G., J. Biomater. Appl., 17, 197–206 15) Higuchi T., Connors K. A., Advances in Analytical Chemistry and (2003). Instrumentation, 4, 117–212 (1965). 32) Rajewski R. A., Stella V. J., J. Pharm. Sci., 85, 1142–1169 (1996). 16) Heaney T. G., Jones R. S., Arch. Oral Biol., 23, 713–717 (1978). 33) Gao L., Liu G., Wang X., Liu F., Xu Y., Ma J., Int. J. Pharm., 404, 17) Squier C. A., Cox P., Wertz P. W., J. Invest. Dermatol., 96, 123–126 231–237 (2011). (1991). 34) Liu J., Qiu L., Gao J., Jin Y., Int. J. Pharm., 312, 137–143 (2006). 18) Cappello B., De Rosa G., Giannini L., La Rotonda M. I., Mensit- 35) Sinswat P., Tengamnuay P., Int. J. Pharm., 257, 15–22 (2003). ieri G., Miro A., Quaglia F., Russo R., Int. J. Pharm., 319, 63–70 36) Babu R. J., Pandit J. K., Int. J. Pharm., 271, 155–165 (2004). (2006). 37) Nakanishi K., Nadai T., Masada M., Miyajima K., Chem. Pharm. 19) Veuillez F., Rieg F. F., Guy R. H., Deshusses J., Buri P., Int. J. Bull., 40, 1252–1256 (1992). Pharm., 231, 1–9 (2002). 38) Stella V. J., Rao V. M., Zannou E. A., Zia V. V., Adv. Drug Deliv. 20) De Caro V., Giandalia G., Siragusa M. G., Paderni C., Campisi G., Rev., 36, 3–16 (1999). Giannola L. I., Eur. J. Pharm. Biopharm., 70, 869–873 (2008).