2024 年 72 巻 10 号 p. 894-898
N-[(4R)-5,7-Difluoro-2′-(5-methylpyridin-2-yl)-4′-oxo-2,2′,3,4′,5′,7′-hexahydrospiro[1-benzopyran-4,6′-pyrazolo[4,3-c]pyridin]-3′-yl]-2-(methanesulfonyl)acetamide (S-309309) is an anti-obesity drug developed by Shionogi & Co., Ltd. that has a monoacylglycerol acyltransferase 2 inhibitory effect. S-309309 has poor wettability, and the amount of the degradation product (4R)-3′-amino-5,7-difluoro-2′-(5-methylpyridin-2-yl)-2,2′,3,7′-tetrahydrospiro[[1]benzopyran-4,6′-pyrazolo[4,3-c]pyridin]-4′(5′H)-one (compound 8) increases over time under acidic conditions. In this study, we have tried to improve S-309309 wettability and suppress the amount of degradation product increased under acidic conditions. As a result of the study, we found that by mixing with a water-soluble polymer, the wettability of S-309309 and its dissolved concentration in fluid were increased suggesting that its dissolution behavior should be enhanced. In addition, by encapsulating S-309309, the increase of degradation product amount was suppressed under acidic conditions, suggesting that the suppression of degradation product formation would be expected in the stomach after oral dosing. Overall, these results suggest that the drug property issues of S-309309 can be overcome by mixing S-309309 with a water-soluble polymer and encapsulation.
N-[(4R)-5,7-Difluoro-2′-(5-methylpyridin-2-yl)-4′-oxo-2,2′,3,4′,5′,7′-hexahydrospiro[1-benzopyran-4,6′-pyrazolo[4,3-c]pyridin]-3′-yl]-2-(methanesulfonyl)acetamide (S-309309) is an anti-obesity drug that was developed by Shionogi & Co., Ltd. S-309309 has an inhibitory effect on monoacylglycerol acyltransferase 2 (MGAT2), which makes it useful for treating diseases such as obesity.1,2) Three isoforms of MGAT have been identified: MGAT1, MGAT2, and MGAT3. Of these, MGAT2 and MGAT3 are highly expressed in the small intestine, where they are thought to be involved in fat absorption. It has been reported that high-fat diet loading in wild-type mice increased MGAT2 expression in the small intestine and increased MGAT activity.3) Furthermore, MGAT2 knockout mice showed suppressed weight gain, insulin resistance, blood cholesterol elevation, fatty liver formation, and energy consumption even under high-fat diet loading.4)
In general, drug absorption in the gastrointestinal tract is often rate-limited by the dissolution of the drug from the dosage form.5) The wettability of a drug is an important property that affects its dissolution6); wettability can be quantitatively evaluated by contact angle measurements.7) S-309309 has poor wettability, low solubility, and low stability under acidic condition, which are challenges for formulation development. Poor-wettability and low-solubility drugs may not be absorbed sufficiently in the body, resulting in an ineffective blood concentration. In addition, excessive production of related substances in the body not only impairs the stability and efficacy of the drug, but also increases the risk to patient health.
In this study, we have tried to improve dissolution of S-309309 and suppress the amount of the degradation product (4R)-3′-amino-5,7-difluoro-2′-(5-methylpyridin-2-yl)-2,2′,3,7′-tetrahydrospiro[[1]benzopyran-4,6′-pyrazolo[4,3-c]pyridine]-4′(5′H)-one (compound 8) that forms under acidic conditions by simple methods.
To evaluate the wettability of S-309309, S-309309 alone or mixed with a water-soluble polymer was compressed to form tablets and then the water contact angle of each tablet was evaluated. The results are listed in Table 1. The contact angle of all tablets containing S-309309 and a water-soluble polymer was lower compared with that of the tablet of S-309309 alone. Also, the higher the water-soluble polymer ratio to S-309309, the lower the contact angle of tablets, indicating that the wettability of the tablets was improved by including a water-soluble polymer. Talukder et al.8) reported the contact angles of the poorly soluble drugs ibuprofen, nifedipine, and carbamazepine with 0.5% Hydroxypropyl cellulose (HPC) solution and water. It was found that the contact angles of 0.5% HPC solution on the drug surfaces were lower than the corresponding contact angles for water. They concluded that the relatively non-ionic polymer HPC markedly lowered the surface tension of water, resulting in improved drug surface wetting. In this study, the water contact angle of tablets prepared by mixing S-309309 and a water-soluble polymer was evaluated. The results indicated that the contact between the water-soluble polymer on the tablet surface and the media lowered the surface tension of the water droplet on the tablet surface, leading to improved wettability of the S-309309-based formulations. Based on the suggestion by Lippold et al.9) that improved drug wettability leads to an increase of the effective surface area, which in turn results in an accelerated dissolution rate, it is expected that improving the wettability of S-309309 by adding the water-soluble polymer to the formulation would improve its dissolution in aqueous solution.
| Sample | Ratio of API to polymer | Contact angle (°) Mean ± S.D. | Polymer type |
|---|---|---|---|
| Control (S-309309 only) | — | 91.1 ± 1.5 | — |
| S-309309 + HPC | 10 : 1 | 77.3 ± 6.0** | Cellulose |
| 7 : 1 | 76.7 ± 4.4** | ||
| 5 : 1 | 79.2 ± 2.5** | ||
| 2 : 1 | 67.9 ± 1.8** | ||
| S-309309 + HPMC | 10 : 1 | 87.2 ± 1.8 | Cellulose |
| 7 : 1 | 85.9 ± 2.5 | ||
| 5 : 1 | 84.8 ± 3.5 | ||
| 2 : 1 | 84.0 ± 3.2* | ||
| S-309309 + PVP | 5 : 1 | 78.9 ± 5.6** | Vinyl |
Contact angle is expressed as mean ± standard deviation (S.D.) for four or five experiments. * p < 0.05, ** p < 0.01 compared with the control.
Next, the dissolved concentration of S-309309 alone and S-309309 mixed with different water-soluble polymers was evaluated in phosphate buffer (pH 6.8) at 2 and 24 h (Table 2). The dissolved concentration of S-309309 in phosphate buffer (pH 6.8) after 2 and 24 h was improved by mixing a water-soluble polymer, regardless of the polymer type, compared with that of S-309309 alone. Loftsson et al.10) reported the addition of a small amount of the polymers results in increasing the aqueous solubility of drugs and suggested that a large fraction of drug molecules in aqueous polymer solutions be bound to the polymers. The polymers can interact with drug substances via electrostatic bonds (i.e., ion-to-ion, ion-to-dipole, and dipole to-dipole bonds), van der Waals’ forces and hydrogen bridges, may work on the complex formation with drugs.11) Therefore, intermolecular interaction between S-309309 and the polymer could increase the dissolved concentration of S-309309, resulting in enhance its dissolution.
| Sample | Dissolved concentration of S-309309 (µg/mL) | Polymer type | |
|---|---|---|---|
| 2 h | 24 h | ||
| Control (S-309309 only) | 73.7 ± 0.8 | 73.1 ± 1.1 | — |
| S-309309 + HPC | 101.6 ± 0.3* | 90.8 ± 0.6* | Cellulose |
| S-309309 + HPMC | 149.1 ± 6.0* | 111.4 ± 0.5* | Cellulose |
| S-309309 + PVP | 92.7 ± 0.2* | 83.6 ± 2.3* | Vinyl |
The dissolved concentration of S-309309 is expressed as mean ± S.D. for three experiments. * p < 0.01 compared with the control.
S-309309 has poor stability under acidic conditions; it decomposes to form compound 8 (Fig. 1). Therefore, it is important to minimize contact between S-309309 and gastric fluid to suppress the amount of compound 8 formed in the body. We tried to encapsulate S-309309 and evaluated the effect of encapsulation on the increase of compound 8 amount in acidic solution. To prevent the decomposition of drugs in the stomach, enteric-coated capsules,12) tablets with enteric coating13) or enteric-coated particles by particulate coating14) are usually prepared. However, when enteric-coated formulations are used, a decrease in bioavailability is often observed.15,16) Moreover, enteric-coating techniques often require an organic solvent, and has difficulties to control the amount and density of a dense coating film that significantly affects dissolution and absorption in the gastrointestine. Therefore, in this study, we selected the simple encapsulation method and used a rapidly dissolving capsule considering the gastric emptying rate in the fasting state,17,18) which has a low pH. The time-dependent formation of the compound 8 amount from S-309309 capsule or unencapsulated S-309309 in hydrochloric acid solutions was evaluated (Fig. 1). When S-309309 was encapsulated, no compound 8 was observed for 20 min, and even after 30 min, the increase of compound 8 amount was significantly suppressed compared with the case for unencapsulated S-309309. It has been reported that the gastric emptying rate constant in humans under the fasted state is approximately 5 h−1,17) so most of the stomach contents would be emptied after 30 min. Therefore, the degradation product formation in the stomach by a simple encapsulation method would be expected to be suppressed, which would lead the higher safe and efficacious.
Results are expressed as the mean of three experiments with S.D. shown as an error bar. N.D. means not detected. White column, control (S-309309 only); black column, S-309309 with HPMC capsules, * p < 0.05, ** p < 0.01 compared with the control.
To estimate the dissolution of the designed S-309309-based formulations in the small intestine, formulations (Table 3) were prepared using HPC or hypromellose (HPMC), which increased the wettability and the dissolved concentration of S-309309, and encapsulated to suppress the increase of compound 8 amount under acidic conditions. The dissolution profile of each formulation in phosphate buffer (pH 6.8) was assessed, as shown in Figs. 2(A)–(C). In addition, the effect of HPC and HPMC on the dissolution of S-309309 was evaluated. Although a lag time of up to 5 min after the start of the test was observed regardless of the presence or absence of water-soluble polymer, all the capsules containing HPC or HPMC showed rapid dissolution behavior compared with the capsules without them regardless of S-309309 dose. These results demonstrated that the wettability improvement and solubilization induced by water-soluble polymer accelerated the dissolution rate of S-309309. Thus, the formulations containing water-soluble polymer can be expected to show better dissolution and subsequent absorption in the small intestine than the formulations without water-soluble polymer.
| Formulation | 3 mg formulation | 10 mg formulation | 30 mg formulation | ||||||
|---|---|---|---|---|---|---|---|---|---|
| With HPC | With HPMC | Without water- soluble polymer | With HPC | With HPMC | Without water- soluble polymer | With HPC | With HPMC | Without water-soluble polymer | |
| S-309309 (mg) | 3 | 3 | 3 | 10 | 10 | 10 | 30 | 30 | 30 |
| Mannitol (mg) | 41.5 | 41.5 | 43.0 | 34.5 | 34.5 | 36.0 | 103.5 | 103.5 | 108.0 |
| Hydroxypropyl cellulose (HPC) (mg) | 1.5 | — | — | 1.5 | — | — | 4.5 | — | — |
| Hypromellose (HPMC) (mg) | — | 1.5 | — | — | 1.5 | — | — | 4.5 | — |
| Low-substituted hydroxypropyl cellulose (mg) | 3.5 | 3.5 | 3.5 | 3.5 | 3.5 | 3.5 | 10.5 | 10.5 | 10.5 |
| Magnesium stearate (mg) | 0.5 | 0.5 | 0.5 | 0.5 | 0.5 | 0.5 | 1.5 | 1.5 | 1.5 |
| Total (mg) | 50 | 50 | 50 | 50 | 50 | 50 | 150 | 150 | 150 |
Results are expressed as the mean of three or six experiments with S.D. shown as an error bar. ●, with HPC; ■, with HPMC;▲, without water-soluble polymer.
We tried to develop S-309309 formulation by simple manufacturing methods with the goals of achieving rapid dissolution of S-309309 in the small intestine and suppression of compound 8 formation in the stomach. The results suggested that the mixing S-309309 with a water-soluble polymer improved the wettability of S-309309 and solubilized, thereby enhanced its dissolution rate. We also found that encapsulating S-309309 should suppress the increase of compound 8 amount in the stomach. These findings provide useful information for designing and developing a S-309309 formulation that is both safe and efficacious.
S-309309 was designed by Shionogi & Co., Ltd. (Osaka, Japan). The chemical structure of S-309309 is shown in Fig. 3. The molecular formula and the molecular weight of S-309309 are C23H21F2N5O5S and 517.51, respectively. S-309309 was synthesized as a crystal, and the synthesis is described in Supplementary Materials. The reversible crystal transformation of S-309309 between the anhydrate and the hydrate depending on the humidity were reported.19) Therefore, the amount of S-309309 used in each experiment was calculated based on the actual water content measured before this study. The particle sizes of S-309309 hydrate used in this study were D50 = 2.74 µm, D90 = 8.34 µm for one sample (Lot No. B1) and D50 = 3.29 µm, D90 = 10.94 µm for the other sample (Lot No. A1); the purity (quantitative value) of these samples was 99.4 and 99.6%, respectively. D50 or D90 is defined as the size value corresponding to cumulative size distribution at 50 or 90%, which represents the size of particles below which 50 or 90% of the sample lies. HPC, hypromellose (HPMC), and povidone (PVP), which are water-soluble polymers, were purchased from Nippon Soda Co., Ltd. (Tokyo, Japan), Shin-Etsu Chemical Co., Ltd. (Tokyo, Japan), and BASF SE (Ludwigshafen, Germany), respectively. The viscosity of the HPC used in this study was 2.0–5.9 mPa·s (2% (w/v) aqueous solution, 20 °C), with a molecular weight of 40000–100000, and substitution type of the HPMC used in this study was 2910. The viscosity of the PVP was approximately 1.2 mPa·s (1% in water). Mannitol was purchased from Roquette Frères (Lestrem, France), and its particle sizes was D50 = 100 µm. Low-substituted hydroxypropyl cellulose was purchased from Shin-Etsu Chemical Co., Ltd., magnesium stearate was purchased from SpecGx LLC (St. Louis, MO, U.S.A.), and HPMC capsules were purchased from Lonza K.K. (Sagamihara, Kanagawa, Japan). All other chemicals and solvents were analytical reagent-grade commercial products.
The binary mixtures consisting of HPC, HPMC, or PVP with S-309309 hydrate were prepared. A ratio of S-309309 hydrate and polymers were 10 : 1, 7 : 1, 5 : 1, and 2 : 1 for HPC and HPMC, and 5 : 1 for PVP. and mixtures consisting of PVP with S-309309 hydrate were prepared in a ratio of 5 : 1. The flat tablets consisting of each mixture (total weight 60 mg) were manufactured using a single tablet press equipped with a 7.5 mm diameter at a compression force of 5 kN. A drop of water (1 µL) was dropped onto the flat surface of each tablet and then the water contact angle was measured using an automatic contact angle meter (DMo-601; Kyowa Interface Science Co., Ltd., Niiza, Saitama, Japan). The control group consisted of tablets containing only S-309309, without a water-soluble polymer. Measurements were performed in quadruplicate or quintuplicate and the data were presented as arithmetic means with standard deviations for quadruplicate or quintuplicate studies. The statistical significance of the difference between the means of experimental groups was determined using Dunnett’s test for multiple comparisons.
Evaluation of the Dissolved Concentration of S-309309Fifty point six mg of S-309309 hydrate (as 50 mg of S-309309) alone as a control or a mixture of 50.6 mg of S-309309 hydrate and 10 mg of a water-soluble polymer (HPC, HPMC, or PVP) were added to a test tube and then 30 mL of 200 mM phosphate buffer (pH 6.8) was added. Each mixture was shaken at 37 °C using a shaking water bath (Personal-11, Taitec Corporation, Koshigaya, Saitama, Japan) at 100 rpm for 2 and 24 h. Each sample was filtered through a 0.45-µm filter and then the concentration of S-309309 in the filtrate was measured by ultra-performance liquid chromatography (UPLC, ACQUITY UPLC H-Class; Waters Corporation, Milford, MA, U.S.A.) at a flow rate of 0.6 mL/min on a YMC–Triart C18 ExRS column (1.9 µm, 3.0 × 100 mm; YMC Co., LTD., Kyoto, Japan) under an isocratic condition of mobile phase (0.1% formic acid : acetonitrile, 3 : 2, v/v) for 5 min at 60 °C using a column heater. The detection wavelength was set at 267 nm and the injection volume was 10 µL. The statistical significance of the difference between the means of experimental groups was determined using Dunnett’s test for multiple comparisons.
Evaluation of the Amount of Degradation ProductsAn HPMC capsule was filled with 30.4 mg of S-309309 hydrate (as 30 mg of S-309309). The capsule with S-309309 was added to 50 mL of 0.1N hydrochloric acid in a small-volume vessel equipped with a small paddle stirring at 50 rpm. The mixture was heated to 37 °C and then 3 mL of aliquots were withdrawn after 5, 10, 15, 20, and 30 min. Each aliquot was filtered through a 0.45-µm filter and then diluted 2 times with 100 mM carbonate buffer (pH 9.7) to inject into the UPLC system. The amount of the degradation product compound 8 (Fig. 4) was measured by UPLC at a wavelength of 267 nm using a YMC–Triart C18 ExRS column (1.9 µm, 3.0 × 100 mm) at 60 °C using a column heater. The mobile phase consisted of 0.1% formic acid and acetonitrile. The separation was achieved in 35 min using the gradient program summarized in Table 4. The flow rate was 0.6 mL/min throughout the run. The injection volume was 10 µL. The amount of compound 8 was calculated as a percentage (%) of the total peak area of the chromatogram, which was set to 100%. The statistical significance of the difference between the means of experimental groups was determined using Student’s t-test for single comparison of experimental groups.
| Minutes | 0.1% formic acid (%) | Acetonitrile (%) |
|---|---|---|
| 0–0.5 | 90 | 10 |
| 0.5–19.0 | 90→40 | 10→60 |
| 19.0–21.0 | 40→10 | 60→90 |
| 21.0–25.0 | 10 | 90 |
| 25.0–26.0 | 10→90 | 90→10 |
| 26.0–35.0 | 90 | 10 |
S-309309 capsule formulations were composed of the common pharmaceutical excipients (mannitol, HPC, HPMC, low-substituted hydroxypropyl cellulose, magnesium stearate, and HPMC capsule) without any special excipients and were manufactured by simple manufacturing processes such as mixing, and encapsulation. The formulae used are listed in Table 3. Dissolution tests were conducted with a dissolution apparatus (NTR-6400AC; Toyama Sangyo Co., Ltd., Osaka, Japan) following the paddle method (the United States Pharmacopeia (USP) Apparatus II) at 50 rpm and 37 °C. S-309309 capsule formulations were tested in 900 mL of 200 mM phosphate buffer (pH 6.8). A sinker was used for the dissolution test. Five-mL aliquots of the test solution were collected at 5, 10, 15, 20, 30, 45, and 60 min (3 and 10 mg capsules) or at 5, 10, 15, 20, 30, 45, 60, 90, and 120 min (30 mg capsules). Each aliquot was filtered through a 0.45-µm filter. The filtrate was injected into UPLC system. UPLC analysis was performed at a wavelength of 267 nm with a YMC–Triart C18 ExRS column (1.9 µm, 3.0 × 100 mm) under an isocratic condition of mobile phase (0.1% formic acid : acetonitrile, 3 : 2, v/v) for 5 min at 60 °C using a column heater. The detection wavelength was set at 267 nm and the injection volume was 5 µL.
We thank Natasha Lundin, Ph.D., for editing a draft of this manuscript. We would also like to thank Ken-ichi Setsukinai, Ph.D., and Naomi Fujita for assisting with the publication of our manuscript.
The authors are employees of Shionogi & Co., Ltd.
This article contains supplementary materials.
