Effect of Solubility Improvement via Formation of an Amorphous Composite of Indomethacin and Sulindac on Membrane Permeability

Abstract

The importance of permeability as well as solubility of the drug has been recognized in improving the solubility of poorly water-soluble drugs. This study investigated the impact of amorphous composites of indomethacin (IMC) and sulindac (SLD) on the membrane permeability of drugs. The IMC/SLD (1/1) formulation prepared by dry grinding was amorphous with a single glass transition temperature. The Fourier transform IR spectra and Raman spectra revealed formation of hydrogen bonds between the OH group of IMC and the carbonyl group of SLD. These results suggest that an amorphous composite was formed between IMC and SLD through hydrogen bonding. The amount of dissolved IMC and SLD from the amorphous composite of IMC/SLD (1/1) was higher than that of the untreated IMC or SLD in the dissolution test. The permeated amounts and permeation rates of both drugs were enhanced by increasing the solubility of the amorphous composite. Conversely, the apparent membrane permeability coefficients (P_app) were almost same for untreated drugs and amorphous composites. In the case of hydroxypropyl-β-cyclodextrin and sodium dodecyl sulfate, P_app of the drugs decreased with the addition of these compounds, although the drug solubility was enhanced by the solubilization effect. This study revealed that an amorphous composite formed through hydrogen bonding is an attractive pharmaceutical way to enhance the permeated amount and permeation rate without changing the P_app of both the drugs.

Introduction

The permeability through the gastrointestinal membrane and the solubility of the drug in the gastrointestinal milieu are the fundamental parameters controlling oral absorption.¹⁾ Because solubility limitations in oral drug delivery are frequently encountered during the development of drugs, approaches such as inclusion complexes, micellar solubilization, and amorphization, have been established to address the issue.^1–3) The solubility-permeability interplay when using surfactants or cyclodextrins has the nature of a trade-off, in which any solubility gain is paid by permeability loss.⁴⁾ In general, the membrane permeability of a drug is expressed as the product of the diffusion coefficient of the drug in the membrane and membrane/aqueous partition coefficient divided by the thickness of the membrane.⁵⁾ The direct correlation between intestinal permeability and membrane/aqueous partitioning, which is dependent on the apparent solubility of the drug, suggests a close relation between the solubility and permeability.⁶⁾

Solubility-improving amorphous composites stabilize the amorphous state by decreasing the molecular mobility of the drug via intermolecular interactions between the two compounds in the solid state.^7–9) Drug–drug amorphous systems using drugs as coformers can be explored in combination therapy.^10,11) However, reports on the membrane permeation of drugs dissolved in amorphous composites are limited.^12,13) Although the effect of amorphous composite formation on the membrane permeability of the drug is unclear, considering these effects is important for designing a combination therapy.

The present study aimed to explore the effects of amorphous composites on the membrane permeability of drugs. Indomethacin (IMC) and sulindac (SLD), which are non-steroidal anti-inflammatory drugs applicable to drug–drug amorphous formulations, were selected as model drugs. Ground mixtures (GM) of the two drugs at different molar ratios were prepared, and the physicochemical properties of the amorphous particles were assessed. The membrane permeability of the drug in the amorphous composite formulation was investigated using a parallel artificial membrane permeability assay (PAMPA). Overall, this work provides the increased understanding of the membrane transport of the drugs in amorphous composite formulation. The clarification enables more efficient drug complexation strategies to facilitate oral absorption.

Results and Discussion

Physicochemical Properties Assessment

The physicochemical properties of the IMC/SLD are summarized in Fig. 1. The powder X-ray diffraction (PXRD) patterns indicated that dry grinding changed the IMC/SLD formulations to an amorphous state. Differential scanning calorimetry (DSC) measurements revealed that the GM of the IMC/SLD formulations showed a single glass transition temperature (T_g), indicating a single-phase amorphous state between the two compounds.¹⁴⁾ The T_g tends to increase with increasing SLD ratio, while the T_g of the formulation with IMC/SLD (1/2) was lower than that of (1/1). The positive deviation of the T_g (47.3 °C) of the IMC/SLD (1/1) from the T_g (43.5 °C) calculated from the Gordon–Taylor equation indicated the formation of amorphous composite via new interaction between IMC and SLD.¹⁵⁾ On the other hand, Jensen et al. reported that an excess amount of the molecule relative to the ratio forming the interaction have a possibility to decrease T_g.¹⁶⁾ The presence of excess SLD to IMC might reduce the T_g of IMC/SLD (1/2). The stability of the amorphous state was evaluated among the amorphous formulations after storage at 25 °C in 75% relative humidity (RH) (Supplementary Fig. S1). The formulation with a 1/1 molar ratio of IMC/SLD formed the most stable amorphous state without recrystallization peaks originating from the IMC and SLD crystals. In the Fourier transform (FT)-IR spectra of untreated IMC, the peak derived from the C=O symmetric stretching motion of the carbonyl group was observed at 1718 cm⁻¹, and the peak derived from the C=O asymmetric stretching motion was observed at 1692 cm⁻¹.¹⁷⁾ These peaks were observed at the same wavenumber in the GM of IMC alone. A characteristic peak at 1701 cm⁻¹ derived from C=O stretching of the carbonyl group, was observed in the untreated SLD. The peak corresponding to the carboxyl group of SLD was identified at a higher wavenumber of 1720 cm⁻¹ in the GM of SLD alone. GM of IMC/SLD (1/1), the peaks derived from the carbonyl group of IMC were shifted to lower wavenumbers at 1682 cm⁻¹. GM of IMC/SLD (1/1), the peaks derived from the carbonyl group of IMC were shifted to lower wavenumbers at 1682 cm⁻¹. The peak derived from the carboxylic group of SLD in GM of IMC/SLD (1/1) was observed at 1717 cm⁻¹, which is an intermediate wavenumber between the carboxyl group of untreated SLD and GM of SLD alone. The peak shift of carbonyl group of IMC suggested the possibility of hydrogen bond formation between IMC and SLD although further research is needed for the carboxylic group of SLD. This result suggests the formation of hydrogen bonds between the IMC and SLD. The Raman spectra of the untreated IMC, untreated SLD, GM of IMC alone, GM of SLD alone, and GM of IMC/SLD (1/1) are shown in Fig. 1 (D). The carbonyl C=O stretching peak was observed at approximately 1700 cm⁻¹ in the Raman spectra of untreated IMC and GM of IMC alone.¹⁸⁾ The peak shifted to a single broad peak around 1680 cm⁻¹ in the Raman spectrum of the GM of IMC/SLD (1/1). These results indicate a mobility change of the carbonyl group compared to the GM of IMC alone, which may support a hydrogen bond between the carbonyl group of IMC and the carboxylic group of SLD, as expected from the IR results.

Fig. 1. Physicochemical Properties of Indomethacin (IMC) and Sulindac (SLD) Composite: (A) Powder XRD Patterns, (B) DSC, (C) FT-IR Spectra, (D) Raman Spectra, and Chemical Structure of IMC and SLD

Dissolution Test and Solubility Test

Figures 2 (A) and (B) show the dissolution profiles of IMC and SLD under non-sink conditions. The concentrations of dissolved IMC and SLD from the untreated drugs were approximately 300 and 1200 µg/mL, respectively. The dissolved IMC and SLD from GM alone, as single amorphous formulations, were approximately 800 and 1800 µg/mL, respectively. The GM of IMC/SLD (1/1) showed an enhanced IMC solubility. Conversely, the dissolved SLD from the GM of IMC/SLD (1/1) decreased slightly compared to the GM of SLD alone. It has been reported that simultaneous dissolution could occur from drug–drug amorphous composite systems according to interaction between drugs.^19,20) SLD could simultaneously dissolve into the aqueous solution with IMC from the solid phase of amorphous composite. On the other hand, the solubility of IMC is lower compared to that of SLD. Therefore, the solubility of SLD decreased owing to the limited solubility of IMC from the solid phase of the amorphous composite although the solubility of IMC was enhanced by the amorphous composite formation with SLD. Therefore, the dissolved IMC and SLD from the GM of IMC/SLD (1/1) showed similar values.

Fig. 2. Dissolution Profiles of (A) IMC and (B) SLD in pH 6.5 Phosphate Buffer

Data are presented as the mean ± standard deviation (S.D.) (n = 3).

The relationship between the concentration of the additives and the amount of dissolved IMC and SLD is shown in Fig. 3. The solubilities of IMC and SLD increased linearly with the addition of hydroxypropyl-β-cyclodextrin (HP-β-CD) and sodium dodecyl sulfate (SDS), indicating that IMC and SLD formed an inclusion complex with the hydrophobic portion of HP-β-CD or were solubilized in the micellar structure of SDS.^21–24)

Fig. 3. The Dissolved (A) IMC and (B) SLD in pH 6.5 Phosphate Buffer with Different Concentrations of Hydroxypropyl-β-cyclodextrin (HP-β-CD) or SDS

Data are presented as the mean ± S.D. (n = 3).

Membrane Permeability Evaluation

The membrane permeability of IMC and SLD was evaluated using PAMPA. Completely dissolved IMC and SLD at various concentrations were added to the donor side in PAMPA. Figure S2 shows the time-membrane permeation profiles of IMC and SLD. The permeated amounts of IMC and SLD increased linearly with time for all formulations. Figure 4 shows the calculated values of Flux and P_app from the results shown in Supplementary Fig. S2. Figures 4 (A) and (B) show the relationship between flux and drug concentration. The Flux of IMC and SLD increased linearly with drug concentration in GM. The increase rates of the flux of the IMC and SLD were approximately 5-fold and 1.5-fold, respectively. The Flux of IMC and SLD with HP-β-CD or SDS increased slightly, except for the PM of SLD/HP-β-CD. Figures 4 (C) and (D) show the relationship between P_app and drug concentration. The P_app of IMC and SLD from the GM of IMC/SLD was almost the same as that of untreated IMC. The inclusion complex formation and solubilization into the micellar structure resulted in decreased P_app of both drugs, although drug solubilities were dramatically enhanced by the addition of their additives. Therefore, it is important to consider the correlation between solubility enhancement and membrane permeability. The most important factor for enhancing the permeated amount is increasing the amount of membrane-permeable drugs on the membrane surface.²⁵⁾ Drug–CD complexation and solubilization with a surfactant often reduce membrane permeability because drug release from the complex or micelle is the rate-limiting step.⁵⁾ The amorphous composite enhanced the permeated amounts and flux of both drugs without changing P_app, suggesting an increase in membrane-permeable drugs on the membrane surface. The formation of amorphous composites enhanced the solubility of both drugs compared to that of crystalline drugs. In the dissolved condition, the hydrogen bonds between IMC and SLD formed in the solid state might not be strong enough to affect the membrane permeability or be broken in aqueous solution, resulting in an increased free drug concentration. Enhanced solubility contributes to increased concentration of membrane-permeable drugs on the surface of the membrane. It is expected that increased passive diffusion and drug permeation can be achieved by the increased local drug concentration on the membrane surface. Therefore, amorphous composite is an attractive pharmaceutical technique to simultaneously enhance the permeated amount and flux of both drugs.

Fig. 4. Flux of (A) IMC and (B) SLD and Apparent Permeability (P_app) of (C) IMC and (D) SLD with Different Concentrations of Drugs

The dashed line represents P_app of IMC or SLD. Data are presented as the mean ± S.D. (n = 3).

This study demonstrates the influence of amorphous composites via hydrogen bonds on membrane permeability. In conclusion, the formation of an amorphous composite of IMC and SLD improved the solubility of both drugs without decreasing their permeability. The amorphous composite of drugs can be applied as a combination therapy because it can simultaneously enhance the permeated amount and flux of both the drugs. The amorphous composite formulation may contribute to the development of orally administrated formulations of drugs for multidrug combination therapy.

Experimental

Materials

IMC and SLD were purchased from the Tokyo Chemical Industry Co. (Tokyo, Japan). HP-β-CD and potassium bromide (KBr) were obtained from FUJIFILM Wako Pure Chemical Corporation (Tokyo, Japan). SDS was purchased from Nacalai Tesque Inc. (Kyoto, Japan).

Preparation of GM

IMC and SLD were weighed into 15-mL glass vials at molar ratios of 2/1, 1/1, and 1/2, and physically mixed using a vortex mixer. Physical mixtures (PM) were milled for 90 min using a vibrational ball mill MM400 (Retsch, Haan, Germany) to obtain GM.

Characterization of Amorphous Samples

The crystallinity of each sample before and after storage at 25 °C and 75% RH was determined using a Rigaku Mini-Flex powder X-ray diffractometer (Rigaku Corporation, Tokyo, Japan). DSC measurements were performed on a Hitachi DSC7000X instrument (Hitachi High-Tech Science Corporation, Tokyo, Japan) to determine the T_g of amorphous samples. For measurements, each formulation was placed in the aluminum pan after the drying process in the desiccator. The temperature was increased at a rate of 3 °C/min over 60 s and a temperature amplitude of 1 °C. T_g was the intersection of the tangent line at the baseline and the inflection point. Spectroscopic measurements were performed to confirm the intermolecular interaction between IMC and SLD. IR spectroscopy was performed using an IR Affinity-1S (Shimadzu Corporation, Kyoto, Japan) using the KBr method. Spectra were acquired using 64 scans at a resolution of 2 cm⁻¹. Raman spectra were recorded using a Raman Touch (Nanophoton Corp., Osaka, Japan). A 12-mW laser with a wavelength of 785 nm was used to obtain the spectra. A grating of 300 grating/mm was used for measurements.

Dissolution Test and Solubility Test

GM containing 30 mg equivalent of IMC was dispersed in 15 mL of pH 6.5 phosphate buffer. The suspension was filtered at the determined time after shaking and the solution was quantified by HPLC.

An excess amount of untreated IMC or SLD was added to 15 mL of pH 6.5 phosphate buffer containing different concentrations of HP-β-CD or SDS. After 24 h of incubation, the suspension was filtered, and the solubility of the drug was determined using HPLC.

In Vitro Permeability Test

Membrane permeation studies were performed using PermeaPad^® 96-well plates (InnoMe GmBH, Espelkamp, Germany).²⁶⁾ The filtered solution was used as the donor solution after 1 h of the dissolution test. The flux of the drug was calculated as follows (Eq. (1)):

  

(1)

where A is the membrane area (cm²). The apparent permeability coefficient (P_app) of the drug was calculated using the following equation.

  

(2)

where C_d is the concentration in the donor chamber (µg/cm³).²⁷⁾

Acknowledgments

This research was supported by grants from Hosokawa Powder Technology Foundation and JSPS KAKENHI (Grant No. 20K07191; Tokyo, Japan).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Higashi T., Chem. Pharm. Bull., 67, 289–298 (2019).
• 2) Tanida S., Yoshimoto A., Yoshida M., Uchiyama H., Kadota K., Tozuka Y., Chem. Pharm. Bull., 67, 921–928 (2019).
• 3) Tanida S., Kurokawa T., Sato H., Kadota K., Tozuka Y., Chem. Pharm. Bull., 64, 68–72 (2016).
• 4) Dahan A., Beig A., Lindley D., Miller J. M., Adv. Drug Deliv. Rev., 101, 99–107 (2016).
• 5) Beig A., Miller J. M., Dahan A., Eur. J. Pharm. Biopharm., 85 (3 Pt B), 1293–1299 (2013).
• 6) Miller J. M., Beig A., Krieg B. J., Carr R. A., Borchardt T. B., Amidon G. E., Amidon G. L., Dahan A., Mol. Pharm., 8, 1848–1856 (2011).
• 7) Hirakawa Y., Ueda H., Takata Y., Minamihata K., Wakabayashi R., Kamiya N., Goto M., Eur. J. Pharm. Sci., 158, 105667 (2021).
• 8) Yamamoto K., Kojima T., Karashima M., Ikeda Y., Chem. Pharm. Bull., 64, 1739–1746 (2016).
• 9) Hatanaka Y., Uchiyama H., Kadota K., Tozuka Y., J. Drug Deliv. Sci. Technol., 65, 102715 (2021).
• 10) Arnfast L., Kamruzzaman M., Löbmann K., Aho J., Baldursdottir S., Rades T., Rantanen J., Pharm. Res., 34, 2689–2697 (2017).
• 11) Han J., Wei Y., Lu Y., Wang R., Zhang J., Gao Y., Qian S., Expert Opin. Drug Deliv., 17, 1411–1435 (2020).
• 12) Ruponen M., Rusanen H., Laitinen R., J. Pharm. Sci., 109, 2252–2261 (2020).
• 13) Ruponen M., Visti M., Ojarinta R., Laitinen R., Eur. J. Pharm. Biopharm., 129, 247–256 (2018).
• 14) Newman A., Engers D., Bates S., Ivanisevic I., Kelly R. C., Zografi G., J. Pharm. Sci., 97, 4840–4856 (2008).
• 15) Liu J., Rades T., Grohganz H., Mol. Pharm., 17, 1335–1342 (2020).
• 16) Jensen K. T., Larsen F. H., Löbmann K., Rades T., Grohganz H., Eur. J. Pharm. Biopharm., 107, 32–39 (2016).
• 17) Taylor L. S., Zografi G., Pharm. Res., 14, 1691–1698 (1997).
• 18) Hédoux A., Guinet Y., Capet F., Paccou L., Descamps M., Phys. Rev. B, 77, 094205 (2008).
• 19) Allesø M., Chieng N., Rehder S., Rantanen J., Rades T., Aaltonen J., J. Control. Release, 136, 45–53 (2009).
• 20) Löbmann K., Laitinen R., Grohganz H., Gordon K. C., Strachan C., Rades T., Mol. Pharm., 8, 1919–1928 (2011).
• 21) Faiz U., Sami U., Muhammad F. A. K., Fahad K. T., Nazeer H. K., Biomedical Letters, 8, 170–176 (2022).
• 22) Bandi N., Wei W., Roberts C. B., Kotra L. P., Kompella U. B., Eur. J. Pharm. Sci., 23, 159–168 (2004).
• 23) Fillet M., Hubert P., Crommen J., Electrophoresis, 18, 1013–1018 (1997).
• 24) Maity B., Chatterjee A., Ahmed S. A., Seth D., J. Phys. Chem. B, 119, 3776–3785 (2015).
• 25) Aungst B. J., J. Pharm. Sci., 82, 979–987 (1993).
• 26) Eriksen J. B., Christensen S. B., Bauer-Brandl A., Brandl M., J. Pharm. Sci., 111, 1667–1673 (2022).
• 27) Sormunen H., Ruponen M., Laitinen R., Int. J. Pharm., 570, 118653 (2019).