Mechanism of enhanced nifedipine dissolution by ... - ACS Publications

Subscriber access provided by University of Sussex Library

Article

Mechanism of enhanced nifedipine dissolution by polymerblended solid dispersion through molecular-level characterization Keisuke Ueda, Chisato Yamazoe, Yuki Yasuda, Kenjirou Higashi, Kohsaku Kawakami, and Kunikazu Moribe Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00523 • Publication Date (Web): 31 Jul 2018 Downloaded from http://pubs.acs.org on August 9, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Mechanism of enhanced nifedipine dissolution by polymer-blended solid dispersion through molecularlevel characterization Keisuke Ueda,∗,† Chisato Yamazoe,† Yuki Yasuda,† Kenjirou Higashi,† Kohsaku Kawakami,‡ and Kunikazu Moribe† †

Graduate School of Pharmaceutical Sciences, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 2608675, Japan ‡

International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan

∗

Corresponding Author: Graduate School of Pharmaceutical Sciences, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8675, Japan; Tel.: +81-43-226-2866; Fax: +81-43-226-2867; E-mail: [email protected]

ACS Paragon Plus Environment

1

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 34

Table of Contents/Abstract Graphic


2

Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ABSTRACT We investigated the effect of polymer composition on nifedipine (NIF) dissolution through molecularlevel characterization of NIF/hypromellose (HPMC)/Eudragit® S (EUD-S) ternary solid dispersions. The dissolution rates and molecular states of NIF and polymers were evaluated in NIF/HPMC/EUD-S spray-dried samples (SPDs) with different polymer compositions. Blending of HPMC and EUD-S improved the dissolution property of each polymer. Moreover, polymer blending enhanced NIF dissolution from the NIF/polymer SPD with EUD-S/polymer wt% of 50–75%. NIF dissolved simultaneously with polymers from the NIF/polymer SPDs with high EUD-S/polymer wt%. In contrast, NIF and polymers separately dissolved from the NIF/polymer SPDs with EUD-S/polymer wt% of 10– 25%, exhibiting a significantly reduced NIF dissolution rate. Fourier transform-infrared and solid-state NMR measurements revealed that HPMC and EUD-S formed molecular interactions with NIF via different interaction modes. Comprehensive analysis by spectroscopic measurements and modulated differential scanning calorimetry showed that the molecular interaction between NIF and EUD-S was stronger than that between NIF and HPMC. Furthermore, the

13

C-spin-lattice relaxation time

measurements revealed that EUD-S effectively restricted the molecular mobility of NIF compared with HPMC. The molecular interaction between NIF and EUD-S led to the simultaneous and fast dissolution of NIF with EUD-S from the NIF/polymer SPD with high EUD-S loading. Thus, enhanced NIF dissolution was ascribed to the fast dissolution properties of the blended polymer and to polymercontrolled NIF dissolution through the strong molecular interaction between NIF and EUD-S. To achieve efficient optimization of the formulation of polymer-blended solid dispersion with desired drug dissolution, it is necessary to consider both polymer–polymer and drug–polymer intermolecular interactions.

KEYWORDS


3


Page 4 of 34

solid dispersion, polymer blend, solid-state NMR, dissolution, intermolecular interaction

INTRODUCTION Amorphization is an efficient method to improve the solubility and dissolution properties of drugs without making any chemical modification to the drug molecules. Amorphized drugs possess higher energies than crystalline drugs; therefore, amorphized drugs exhibit improved dissolution.1,2 However, amorphous drugs can easily recrystallize upon exposure to external stresses, including temperature and humidity. Hence, solid dispersions, in which the drug is dispersed into the polymer as an amorphous state, have been widely studied as a promising method for stabilizing amorphous drug formulations. Various polymers, such as derivatives of vinyl,3 cellulose,4,5 polyethylene glycol,6 and methacrylate,7,8 have been used for preparing solid dispersions as they strongly inhibit the recrystallization of amorphous drugs. Some solid dispersion formulations are already available commercially.9 Solid dispersions offer rapid drug dissolution compared with drug crystals and forms drug supersaturated solutions in aqueous dispersions. Drug supersaturation directly improves drug absorption,5,10 and enhancement of drug supersaturation level using solid dispersion techniques effectively improves drug absorption. The level of drug supersaturation achieved by dissolving a solid dispersion is strongly dependent on the inhibition of drug crystallization by the polymer used as the solid dispersion carrier. Polymers that strongly inhibit the crystallization of a drug can maintain the drug in a highly supersaturated state.11 The dissolution rate of solid dispersions is also important for the effective improvement in the drug supersaturation level. The drug/polymer ratio in a solid dispersion strongly affects the drug dissolution rate from the solid dispersion,12 and sometimes alters the dissolution mechanism of the drug from the solid dispersion.13 When the drug contents are relatively low, the drug dissolves simultaneously with the polymer from the solid dispersion, and drug dissolution can be controlled by a carrier polymer. For a solid dispersion with carrier-controlled dissolution, the use of a fast-dissolving polymer as the solid dispersion carrier is an ACS Paragon Plus Environment

4

Page 5 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


effective way to enhance drug dissolution.14-16 Furthermore, the polymer dissolution rate itself can be enhanced by polymer blending, which modifies the molecular interaction between polymers. The molecular interaction between the drug and each polymer in the ternary solid dispersion is an important factor determining the drug dissolution properties.17 Our previous study indicated that the dissolution of hypromellose (HPMC)/Eudragit® S (EUD-S)-blended solid dispersion was rapid when compared to that of the solid dispersion of each polymer.18 The molecular interaction between HPMC and EUD-S enhanced the dissolution rate of each polymer, resulting in enhanced drug dissolution from the drug/HPMC/EUD-S ternary solid dispersion. However, enhanced drug dissolution was confirmed only for the ternary solid dispersion containing HPMC and EUD-S in equivalent amounts. The polymer composition in the ternary solid dispersion requires optimization to maximize drug dissolution based on the mechanism of enhanced drug dissolution from the polymer-blended solid dispersion. In contrast to a solid dispersion with relatively low drug content, an amorphous drug in a solid dispersion can be dissolved separately with a polymer especially when drug loading in the solid dispersion is relatively high.19 The drug-rich phase can be formed on the dissolution surface of solid dispersions owing to the faster dissolution of the polymer than that of the amorphized drug.13 Moreover, the remaining drug in the solid dispersion dissolves slowly depending on the drug dissolution property. In addition to the amount of drug loaded in the solid dispersion, the miscibility and molecular interaction between the drug and the polymer can affect the drug dissolution mechanism. Previous studies have indicated that insufficient miscibility and molecular interactions between the drug and polymers in solid dispersions result in a drug-rich phase remaining on the dissolution surface, thereby preventing effective dissolution enhancement of the drug from the solid dispersion.20-22 In the present study, we evaluated the effect of polymer composition on the dissolution properties of the poorly water-soluble drug nifedipine (NIF) from solid dispersions. HPMC and EUD-S were used as the carrier polymer. The molecular interaction between NIF and each polymer component in the NIF/HPMC/EUD-S solid dispersion prepared by spray-drying was characterized by Fourier transforminfrared (FT-IR) spectroscopy, solid-state NMR spectroscopy, and modulated differential scanning ACS Paragon Plus Environment

5


Page 6 of 34

calorimetry (MDSC). Finally, the effect of polymer composition on NIF dissolution from the solid dispersion is discussed through molecular-level analysis of NIF and the polymer in the solid dispersion.

EXPERIMENTAL SECTION MATERIALS NIF was purchased from Wako Chemicals Co. (Tokyo, Japan). HPMC (substitution type 2910, type TC-5E, MW: ~12,600) was a kind gift from the Shin-Etsu Chemical Co., Ltd. (Tokyo, Japan). EUD-S (methacrylic acid copolymer, type B, MW: ~135,000) was a gift from Evonik Japan Co., Ltd. (Tokyo, Japan). All other materials and solvents were of reagent grade. The chemical structures and peak assignments for the

13

C NMR spectra of NIF, HPMC, and EUD-S are shown in Figure 1. EUD-S was

composed of methacrylic acid and methyl methacrylate at a molar ratio of 1:2.

Figure 1. Chemical structures and peak assignments for

13

C NMR spectra of (a) NIF, (b) HPMC, and

(c) EUD-S


6

Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


METHODS Preparation of spray-dried sample (SPD) NIF and polymers at weight ratios of 25:75, 40:60, and 90:10 were dissolved in a dichloromethane/methanol (1:1, v/v) solution at a concentration of 2.5% (w/v), and fed into a spray dryer (ADL311S, Yamato Scientific Co., Ltd., Tokyo, Japan) at 5 g/min to prepare SPDs. HPMC, EUDS, or mixtures of HPMC and EUD-S at 90:10, 75:25, 50:50, or 25:75 (weight ratio) were used as polymers. The solution was spray-dried at an inlet temperature of 90 °C and an atomizing pressure of 0.05 MPa. The weight ratio of NIF, HPMC, and EUD-S in each SPD is henceforth presented in parentheses. Polymer SPDs without NIF were also prepared using the same method. Prepared SPDs were placed in vials with caps, and the vials were placed in a desiccator with silica gel and stored in a refrigerator. The desiccator was removed from the refrigerator approximately 1 h before analysis. All experiments were performed in the dark to prevent photodegradation of NIF.

Preparation of melt-quenched NIF Amorphous NIF was prepared by the melt-quench method. NIF was placed in a stainless-steel dish and heated to 185 °C, which is higher than the melting point of NIF (173 °C) for 10 min. The melted NIF was quenched by immersion in liquid nitrogen. The melt-quenched NIF was ground using a mortar and pestle.

SPD dissolution tests The dissolution of the polymer SPD and NIF/polymer SPD was tested using the rotating disk method. Briefly, 100 mg of the SPD was placed into a punch and die (diameter, 10 mm), and then compressed by a hydraulic pump (pressure, 10 MPa) for 2 min. The thickness of the compressed disk was 1 mm, and the disk was attached to the bottom of the shaft used for the rotary basket method by double-sided tape. The dissolution tests were performed using an NTR-VS6P system (Toyama Sangyo Co., Ltd., Osaka, Japan) with a shaft rotation speed of 150 rpm at 37 °C. For the dissolution medium, 40 mL of 100 mM ACS Paragon Plus Environment

7


Page 8 of 34

sodium phosphate buffer (pH 7.4) was used. The solution was sampled at predetermined time intervals, and NIF, HPMC, and EUD-S concentrations were determined using HPLC analysis.

HPLC analysis The sample solutions were diluted with acetonitrile before quantification, and the HPLC analysis was performed using an LC-2000 system (JASCO Co., Ltd., Tokyo, Japan). NIF was analyzed by injecting 5 µL samples onto a Sunfire® C18 (4.6 × 150 mm) column at 40 °C. The mobile phase was composed of acetonitrile/10 mM ammonium acetate buffer (pH 7.4, 1:1 v/v), and the flow rate was 1 mL/min. Samples were detected at 235 nm using a UV detector. The polymers were analyzed by injecting 20 µL samples onto a ShodexTM SB-804 HQ (8.0 × 300 mm) column at 40 °C. HPMC was detected using a refractive index detector. The mobile phase was composed of methanol/10 mM ammonium acetate buffer (pH 4.5, 1:1 v/v), and the flow rare was 1 mL/min. HPMC peak was detected at the retention time of 7.5 min, and the concentration was determined from the area of the HPMC peaks using the calibration curve (r2 > 0.996). EUD-S was detected using a UV detector at a wavelength of 210 nm. The mobile phase was composed of acetonitrile/30 mM sodium phosphate buffer (pH 8.3, 1:3 v/v), and the flow rare was 1 mL/min.

MDSC measurement MDSC measurements were performed using a Q2000 system (TA Instruments, New Castle, DE), which was periodically calibrated using indium and sapphire. For the measurement, approximately 3 mg of the powder sample was placed in a crimped aluminum pan. Inert dry nitrogen gas was passed at a flow rate of 50 mL/min, and the measurements were performed from 25 to 185 °C at a heating rate of 2 °C/min with a modulation of ± 0.5 °C every 60 s.

Solid-state NMR measurement


8

Page 9 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The solid-state NMR measurements were conducted using an ECX-400 NMR system (9.4 T, JEOL Resonance Inc., Tokyo, Japan). The 13C NMR spectra were acquired using cross polarization (CP) with magic-angle spinning (MAS) at 15 kHz and high-power 1H decoupling at 20 °C. Acquisition parameters included relaxation delays of 4–60 s, a CP contact time of 2 ms, and 1H 90° pulse of 2.95 µs. All spectra were externally referenced by setting the methane peak of hexamethylbenzene to 17.3 ppm. The

13

C-spin-lattice relaxation time (T1) was measured by the method previously reported by

Torchia.23 13C-T1 measurements were conducted at the spinning rate of 15 kHz and a contact time of 2 ms at 20 ºC. The

13

C-T1 values were calculated using the JEOL Delta software ver. 5.04 (JEOL

Resonance Inc.).

Powder X-ray diffraction (PXRD) measurements PXRD measurements were conducted using a MiniFlex II (Rigaku Co., Ltd., Tokyo, Japan) under the following conditions: target, Cu; voltage, 30 kV; and current, 15 mA. The scans were performed over an angular range of 2θ = 5–40° and a scanning speed of 4°/min.

FT-IR measurements FT-IR measurements were performed using an FT-IR spectrometer (Bruker Optik GmbH, Ettlingen, Germany) using the KBr tablet method under the following conditions: resolution, 4 cm-1; number of scans, 64; measurement range, 400–4000 cm-1.

RESULTS AND DISCUSSION Dissolution of polymer SPDs First, the dissolution property of the carrier polymer used for the solid dispersion was evaluated. The dissolution of the polymer SPD without NIF was tested using the rotating disk method. The dissolution profiles of the polymers from the polymer SPDs showed good linearity (r2 > 0.95). The dissolution rate


9


(A, µg/[min·cm2]) of each polymer from the disk surface of polymer SPDs was calculated using equation 1: =

× ×

(1)

where, k is the slope of the regression line (µg/[mL·min]), V is the volume of the dissolution medium (40 mL), S is the surface area of the disk (0.79 cm2), and x is the percentage of each component. The dissolution rate of HPMC and EUD-S from each polymer SPD was plotted against the EUD-S/polymer wt% in each polymer SPD (Figure 2). The dissolution rate of HPMC SPD was greater than that of EUDS SPD. However, in polymer SPDs containing both HPMC and EUD-S, the dissolution rates of HPMC and EUD-S were similar regardless of the weight ratio of the two polymers. HPMC and EUD-S simultaneously dissolved from the polymer SPD. Furthermore, the dissolution of HPMC and EUD-S from the polymer SPD was greater than that from each single polymer SPD. The dissolution rate of HPMC and EUD-S was enhanced when HPMC was blended with EUD-S. The dissolution rate of HPMC and EUD-S from the polymer SPD reached a maximum at an EUD-S/polymer wt% of 25%. Dissolution rate (µg/min·cm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 34

2000

HPMC EUD-S

1500

1000

500

0 0

25

50

75

100

EUD-S/polymer (wt%)

Figure 2. Dissolution rate of HPMC and EUD-S from polymer SPDs plotted against EUD-S loading (n = 3, mean ± S.D.).

Characterization of polymer SPDs


10

Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


MDSC measurements were performed for polymer SPDs to characterize the blended polymer. All polymer SPDs showed a single glass transition in the MDSC measurements (Figure S1). When the binary amorphous mixture contained an immiscible domain larger than 10–50 nm, more than two glass transitions can be separately observed.24 Therefore, the single glass transition of all polymer SPDs indicated that HPMC and EUD-S were miscible with a domain size of several tens of nanometers.25,26 The high miscibility of HPMC and EUD-S in polymer SPDs may contribute to the simultaneous dissolution of HPMC and EUD-S from polymer SPDs (Figure 2). The glass transition temperature (Tg) of polymer SPDs as a function of the EUD-S/polymer ratio is shown in Figure 3. The experimentally determined value of Tg is presented together with the predicted value of Tg. The predicted value of Tg was calculated using the Gordon–Taylor equation27 (Equation 2), as follows: =

∙ ∙ ∙

∙

(2)

where, Tg mix, Tg1, and Tg2 are the Tg of the mixture of component 1 and component 2, the Tg of component 1, and the Tg of component 2, respectively, and w1 and w2 are the mass fractions of each component. The K in the Gordon-Taylor equation can be calculated using Equation 3, as follows: =

∙ ∙

(3)

where, ρ1 and ρ2 represent the true densities of each component. The predicted value for the Tg of polymer SPDs was calculated using the experimentally determined Tg and the true density of HPMC and EUD-S SPDs. The true density was measured using AccuPyc II 1340 (Micromeritics Instrument Co., Norcross, GA). The experimentally determined Tg of polymer SPDs was almost consistent with the predicted value, regardless of the polymer composition. In the Gordon–Taylor equation, the predicted Tg was determined assuming that the components were well mixed and that there were no specific molecular interactions between the components.24 The predicted value of Tg was different from the experimentally determined value when the molecular interactions formed between the same component in each intrinsic material differed in number and/or strength compared with those formed between ACS Paragon Plus Environment

11


different components in the mixture.28-30 In contrast, the experimentally determined and predicted Tg coincided when the molecular interactions between the same components were similar in number and/or strength to those between different components in the mixture.31 The molecular interactions among polymers in HPMC/EUD-S SPDs were analyzed in detail by FT-IR and solid-state NMR spectroscopy.

●

175

Experimentally determined Predicted by the Gordon-Taylor equation

170 165

Tg (ºC)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 34

160 155 150 145 140 0

25

50

75

100

EUD-S/Polymer (wt%)

Figure 3. Tg of polymer SPDs. The broken line represents the Tg predicted by the Gordon–Taylor equation. The experimentally determined Tg represents the average value of duplicates.

In the comparison of FT-IR spectra of polymer SPDs, the C=O stretching peaks of EUD-S changed depending on the polymer composition in HPMC/EUD-S SPDs; however, other changes in HPMC and EUD-S peaks could not be clearly observed (Figure S2). EUD-S showed C=O stretching peaks at 1704 cm-1 of methacrylic acid and at 1732 cm-1 of methyl methacrylate.32 The shoulder peak of C=O stretching belonging to the methacrylic acid of EUD-S was reduced or shifted when HPMC was added to the polymer SPDs, indicating that HPMC formed molecular interactions with the methacrylic acid of EUD-S. The

13

C NMR spectra of polymer SPDs are shown in Figure 4. The

13

C peaks of HPMC and

EUD-S were assigned based on previous reports.18,33 The change in the chemical shift of

13

C peaks of

HPMC was small among polymer SPDs, while the peak shapes of Cβ, Cγ, Cδ, and Cε of HPMC were slightly changed by EUD-S blending; the peak width of HPMC at approximately 70–90 ppm was sharper for HPMC/EUD-S SPD than for HPMC SPD. This result is in agreement with a previous report ACS Paragon Plus Environment

12

Page 13 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


that indicated that the hydrogen bonds among the hydroxyl groups of HPMC were disrupted by the molecular interaction between HPMC and EUD-S, leading to a change in the peak shape of HPMC.18 The chemical shift and peak shape of EUD-S at 35–70 ppm did not significantly change by the HPMC blending, whereas the peak shape of Ca of EUD-S at 170–195 ppm clearly changed. The enlarged spectra of the Ca peak, belonging to the carbonyl group of EUD-S, with normalization by the maximum peak intensity of Ca are presented in Figure 5. The left shoulder of the peak area of Ca decreased as the HPMC content increased in polymer SPDs. The carbonyl peak of methacrylic acid was separately assigned to three peaks,33 which denote the cyclic-, open-, and non-dimer structures from the low magnetic field. The carbonyl group of the methyl methacrylate of EUD-S is likely involved in the nondimer structure.33 In our previous study, the relative carbonyl peak areas of the cyclic- and open-dimer structures at 186.3 ppm and 180.4 ppm in EUD-S were reduced in comparison with that of the nondimer structure at 177.7 ppm by HPMC blending with EUD-S.18 The polymer blending of HPMC with EUD-S led to the formation of hydrogen bonds between HPMC and EUD-S, which resulted in the cleavage of the dimeric structure of methacrylic acid in EUD-S. In the present study, it was demonstrated that the dimer structure in EUD-S was cleaved depending on the amount of HPMC in HPMC/EUD-S SPD. The cellulose polymer contains an internal network of hydrogen bonds,34,35 and EUD-S forms a hydrogen bond between methacrylic acid. Spray-drying of HPMC and EUD-S cleaved the molecular interaction in each polymer. The similarity between the experimentally determined and predicted Tg in all HPMC/EUD-S SPDs confirmed that HPMC and EUD-S formed intermolecular interaction in HPMC/EUD-S SPDs followed by disruption of the molecular interactions of HPMC-HPMC and EUDS-EUD-S in each single polymer. The molecular interactions between HPMC and EUD-S led to the high miscibility of HPMC and EUD-S in polymer SPDs, resulting in the simultaneous dissolution of these polymers. Furthermore, the replacement of the molecular interaction improved the dissolution rate of HPMC and EUD-S in aqueous medium.18


13


Figure 4.

13

Page 14 of 34

C CP/MAS NMR spectra of (a) HPMC SPD, (b) HPMC/EUD-S SPD (90/10), (c)

HPMC/EUD-S SPD (75/25), (d) HPMC/EUD-S SPD (50/50), (e) HPMC/EUD-S SPD (25/75), and (f) EUD-S SPD.


14

Page 15 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. 13C CP/MAS NMR spectra of the HPMC/EUD-S SPD with different EUD-S/polymer weight ratios. The NMR spectra were normalized by the maximum peak intensity of Ca of EUD-S.

Dissolution test of NIF/polymer SPDs The dissolution properties of NIF and each polymer from the NIF/polymer SPDs were evaluated by the same dissolution test used for testing the dissolution of the polymer SPDs. Figure S3 represents the PXRD patterns of NIF/polymer SPDs with NIF/polymer weight ratio of 25:75. The characteristic peaks of the NIF crystal disappeared in all NIF/polymer SPDs. The NIF was amorphized by spray-drying with HPMC and EUD-S. The dissolution profiles of the NIF and polymers from NIF/polymer SPDs showed good linearity (r2 > 0.95) in the dissolution test using the rotating disk method. The dissolution rate was calculated using Equation 1. The dissolution rate of NIF, HPMC, and EUD-S from each NIF/polymer SPD was plotted against the EUD-S/polymer wt% in each NIF/polymer SPD (Figure 6). The rates of NIF and EUD-S dissolution from the NIF/EUD-S SPD were similar, indicating the simultaneous dissolution of NIF and EUD-S. In contrast, dissolution of HPMC from the NIF/HPMC SPD was faster than that of NIF. NIF and HPMC dissolved separately from the NIF/HPMC SPD. Moreover, the dissolution rate of NIF was faster in the NIF/EUD-S SPD than in the NIF/HPMC SPD, although the polymer dissolution rate was faster in the HPMC SPD than in the EUD-S SPD (Figure 2). Therefore, the relatively slow dissolution of NIF from NIF/HPMC SPD was not due to the dissolution property of HPMC. It has been reported that the amorphous drug embedded in the solid dispersion could crystallize during the dissolution process, leading to a significant reduction in drug dissolution from the solid dispersion.21 To evaluate NIF crystallization in the dissolution process of the NIF/polymer SPD, PXRD measurements were conducted for the NIF/polymer SPD disks before and after the dissolution test (Figure S4). Characteristic peaks of the NIF crystal were not observed in the PXRD spectra of NIF/HPMC SPD disks after the dissolution test. Hence, the NIF remained in an amorphous state during the dissolution test of the NIF/HPMC SPD, although HPMC could not effectively control the NIF ACS Paragon Plus Environment

15


Page 16 of 34

dissolution in NIF/HPMC SPD. In contrast, the simultaneous dissolution of NIF and EUD-S from NIF/EUD-S SPD indicated that NIF dissolution from NIF/EUD-S was controlled by EUD-S, which effectively enhanced the dissolution of NIF owing to the rapid aqueous dissolution of the carrier polymer. In the dissolution test of NIF/polymer SPDs with EUD-S/polymer wt% of 50% and 75%, NIF dissolved simultaneously with HPMC and EUD-S. Among the NIF/EUD-S SPD and NIF/polymer SPD with EUD-S/polymer wt% of 50% and 75%, the NIF dissolution rates increased with a decrease in EUD-S/polymer wt%. The ranking of NIF dissolution rate was consistent with that of the polymer dissolution rate from each polymer SPD (Figure 2). Thus, NIF dissolution could be improved owing to the enhanced dissolution of the carrier polymer used for the NIF/polymer SPD with a high EUDS/polymer ratio. In contrast, NIF and each polymer were separately dissolved from the NIF/polymer SPD with EUD-S/polymer wt% of 10% and 25%. Consequently, the NIF dissolution rate was significantly reduced in NIF/polymer SPDs with a relatively high amount of HPMC. Among the NIF/HPMC SPDs and NIF/polymer SPDs with EUD-S/polymer wt% of 10% and 25%, the dissolution rate of NIF was almost constant and independent of the polymer dissolution rate. Therefore, NIF dissolution was not controlled by the carrier polymer, but by the NIF dissolution property itself in the NIF/polymer SPDs with high HPMC loading. Although the dissolution rate of polymers from each polymer SPD was the highest at the EUDS/polymer wt% of 25% (Figure 2), NIF dissolution from the NIF/polymer SPD was the fastest at the EUD-S/polymer wt% of 50%. An increase in the dissolution of NIF from the NIF/polymer SPD required both fast dissolution of the carrier polymer and polymer-control of NIF dissolution from the NIF/polymer SPD. The difference in the molecular state of NIF and polymer in each NIF/polymer SPD may lead to changes in the NIF dissolution mechanism, depending on the HPMC and EUD-S composition.


16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Dissolution rate (µg/min·cm2)

Page 17 of 34

1200

NIF HPMC EUD-S

800

400

0 0

25

50

75

100

EUD-S/polymer (wt%)

Figure 6. Dissolution rate of NIF, HPMC, and EUD-S from NIF/polymer SPDs (n = 3, mean ± S.D.). Each NIF/polymer SPD contained NIF at NIF/polymer weight ratio of 25:75.

Characterization of molecular interactions in the NIF/HPMC and NIF/EUD-S system First, the molecular interaction of NIF with HPMC or EUD-S in a binary solid dispersion was investigated prior to characterization of the ternary NIF/HPMC/EUD-S solid dispersion. The FT-IR spectra of crystalline NIF, melt-quenched NIF, NIF/HPMC SPD, and NIF/EUD-S SPD are presented in Figure 7. In addition to the NIF/HPMC and NIF/EUD-S SPD with NIF/polymer weight ratio of 25:75, those with NIF/polymer weight ratio of 40:60 were prepared (Figure S5) and characterized to allow a detailed evaluation of the molecular interaction between NIF and HPMC or EUD-S. Crystalline NIF showed characteristic peaks at 1680 and 1689 cm-1 of C=O stretching and at 3332 cm-1 of N-H stretching. It has been reported that NIF molecules form hydrogen bonds between the NH group and one of the two carbonyl groups in the NIF crystal.36 The different C=O stretching peaks in the NIF crystal indicate the coexistence of hydrogen-bonded and non-hydrogen-bonded carbonyls of NIF.36 The C=O and N-H stretching peaks were broadened and shifted to 1685 and 1704 cm-1 (C=O) and 3338 cm-1 (NH) in melt-quenched NIF. Tang et al. previously reported that amorphization resulted in a similar shift in NIF IR peaks.36 The amorphization of NIF reduced the strength of the hydrogen bonds between NIF, although the weak hydrogen bond between NIF molecules was still present in amorphous NIF.36 SolidACS Paragon Plus Environment

17


Page 18 of 34

state NMR measurements were also performed for crystalline NIF, NIF/HPMC SPD, and NIF/EUD-S SPD (Figure 8). The NIF/HPMC SPD with NIF/polymer weight ratio of 90:10 was used as the reference for amorphous NIF (Figure S5) instead of the melt-quenched NIF, which was recrystallized during the time-consuming solid-state NMR measurement with high-speed MAS to obtain well-resolved spectra. Enlarged 13C NMR spectra of the carbonyl carbon of NIF are shown in Figure 9. The NIF peaks in all the NIF/polymer SPDs were strongly broadened compared with those in crystalline NIF. NIF amorphization resulted in line broadening of NIF peaks owing to the wider distribution of isotropic chemical shifts in the same carbons belonging to different molecules in the disordered packing.37,38 The chemical shift of amorphized NIF in the NIF/HPMC SPD with NIF/polymer weight ratio of 90:10 was also different from that of crystalline NIF. The change in the molecular packing of NIF, including the weakening of intermolecular hydrogen bonds between NIF molecules caused by NIF amorphization, led to changes in the chemical environment around the NIF molecules. In the FT-IR spectra of the NIF/HPMC SPD, the peak top of N-H stretching of NIF was not clearly observed, owing to the overlapping of the broad peak of O-H stretching of HPMC at wavenumbers over 3000 cm–1 (Figure 7). The shape of the C=O stretching peaks of NIF in the NIF/HPMC SPDs differed from the shape of those in the melt-quenched NIF, and the change in peak shape was dependent on the amount of HPMC in the NIF/HPMC SPDs. The C=O stretching peaks of the hydrogen-bonded and nonhydrogen-bonded carbonyl of amorphous NIF were observed at 1685 and 1704 cm-1, respectively.36 The relative peak intensity belonging to the hydrogen-bonded carbonyl of NIF was reduced depending on the amount of HPMC in the NIF/HPMC SPDs. Hydrogen bonding via the carbonyl of NIF was reduced by the blending of HPMC. In the 13C NMR spectra, the carbonyl peak of NIF was observed with a peak maximum of approximately 168 ppm, with the shoulder peak at approximately 170 ppm (Figure 9). Comparison of the 13C peaks of the carbonyl group of NIF between NIF/HPMC SPDs showed that the peak maximum of the NIF carbonyl was shifted slightly upfield, and the shoulder peak of the NIF carbonyl at approximately 170 ppm was reduced with an increase in HPMC content, whereas the chemical shift of the other NIF peaks did not change upon HPMC blending. It has been reported that the ACS Paragon Plus Environment

18

Page 19 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


carbonyl peak in the 13C NMR spectra was shifted downfield by the formation of hydrogen bonds.39 The hydrogen-bonded carbonyl carbon of NIF was observed at a lower magnetic field than the nonhydrogen-bonded carbonyl carbon. Changes in the peak shape of the NIF carbonyl carbon in the solidstate NMR spectra indicated that hydrogen bonding via the carbonyl group of NIF molecules was weakened by HPMC blending, which supported the FT-IR results. Furthermore, the peak shape of HPMC at 65–90 ppm was altered by blending with NIF (Figure S6). The hydroxy group of HPMC could interact with NIF. It has been reported that HPMC forms molecular interactions with the NH group of felodipine, an NIF analog, in a felodipine/HPMC solid dispersion.31 Similarly, the NH group of NIF might form a molecular interaction with HPMC in the NIF/HPMC SPD. In summary, the molecular interaction was newly formed between the NH group of NIF and the hydroxy group of HPMC in NIF/HPMC SPDs, leading to weakening of the intermolecular hydrogen bonding between the NH group and the carbonyl group of NIF. In the case of NIF/EUD-S SPDs, the N-H stretching peak of NIF was shifted to 3363 and 3360 cm-1 in the NIF/EUD-S SPD with NIF/EUD-S weight ratio of 25:75 and 40/60, respectively, whereas that of the melt-quenched NIF was observed at 3338 cm-1 (Figure 7). The shift in the N-H stretching peak of NIF depended on the amount of EUD-S in NIF/EUD-S SPDs. Intermolecular hydrogen bonding between the NH group and the carbonyl group of amorphous NIF was weakened by the formation of the molecular interaction between NIF and EUD-S. EUD-S shows C=O stretching peaks at 1704 cm-1 of methacrylic acid and 1732 cm-1 of methyl methacrylate.32 The peak maxima of the C=O stretching peaks of both NIF and EUD-S could not be accurately evaluated by the FT-IR spectra owing to the overlapping of those peaks. In contrast, the carbonyl peaks of NIF could be separately observed in the 13C NMR spectra of NIF/EUD-S SPDs (Figure 9). The shoulder peaks of the NIF carbonyl at approximately 170 ppm increased with an increase in EUD-S, while changes in the peak shape and chemical shift of other NIF peaks were not clearly observed. Hydrogen bonding between carbonyl and carboxylic acid groups has been reported to induce a downfield shift of the carbonyl peak in the

13

C NMR spectrum.40 Hydrogen

bonding between the carbonyl group of NIF and methacrylic acid of EUD-S should increase the ACS Paragon Plus Environment

19


shoulder peaks of the NIF carbonyl. In contrast with the change in the NIF carbonyl peak, there were only minimal changes in the chemical shift and peak shape of the carbonyl region of EUD-S at 170–195 ppm by NIF loading (Figure S7). This may indicate that there was little interaction between the carbonyl group of EUD-S and NIF. However, the FT-IR measurements clearly demonstrated the intermolecular interaction of the NH group of NIF with EUD-S. It is assumed that the carbonyl group of EUD-S is the preferable accepter of hydrogen bonding with the NH group of NIF. Hence, it was speculated that NIF blending with EUD-S might disrupt the intermolecular interaction between EUD-S molecules. Nevertheless, the strength of the disrupted EUD-S-EUD-S interaction via the carbonyl groups was comparable with that of the newly formed intermolecular interaction between the carbonyl group of EUD-S and the NH group of NIF. Notably, there were clear chemical shifts in the Ce and Cb peaks of EUD-S next to the carbonyl carbon in the EUD-S chemical structure, which were dependent on the amount of NIF in the NIF/EUD-S SPD (Figure S8).

(a) (b)

(a) (b)

(c) (d)

(c) (d)

Transmittance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 34

(e) (f) (g)

(e) (f)

(h)

(g) (h)

3400

3350 3300 Wave number (cm-1)

3250 1850

1750 1650 Wave number (cm-1)

1550


20

Page 21 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. FT-IR spectra of (a) crystalline NIF, (b) melt-quenched NIF, (c) HPMC SPD, (d) EUD-S SPD, (e) NIF/HPMC SPD (25/75), (f) NIF/HPMC SPD (40/60), (g) NIF/EUD-S SPD (25/75), and (h) NIF/EUD-S SPD (40/60). The dotted line represents the peak maximum of N-H stretching and C=O stretching of NIF in melt-quenched NIF.

Figure 8.

13

C CP/MAS NMR spectra of (a) crystalline NIF, (b) NIF/HPMC SPD (90/10), (c)

NIF/HPMC SPD (25/75), (d) NIF/HPMC SPD (40/60), (e) NIF/EUD-S SPD (25/75), and (f) NIF/EUDS SPD (40/60).


21


Figure 9.

13

Page 22 of 34

C CP/MAS NMR spectra of (a) crystalline NIF, (b) NIF/HPMC SPD (90/10), (c)

NIF/HPMC SPD, and (d) NIF/EUD-S SPD. (c, d) Each NIF/polymer SPD contained NIF at NIF/polymer weight ratios of 25:75 (red) and 40:60 (blue); the dotted line represents the NMR spectrum of the NIF/HPMC SPD (90/10). The NMR spectra in (c) and (d) were normalized by the maximum peak intensity of C14,16 of NIF.

Characterization of NIF/polymer SPDs MDSC measurements were performed for the NIF/polymer SPDs containing NIF at NIF/polymer weight ratio of 25:75. All NIF/polymer SPDs showed a single glass transition (Figure S9). Phase separation of NIF with HPMC and EUD-S in NIF/polymer SPDs was not observed by MDSC measurements, indicating that NIF and the polymer were well mixed with a domain size of several tens of nanometers.25,26 To clarify the difference in molecular interaction between NIF and polymers depending on the polymer composition among NIF/polymer SPDs, the predicted value of Tg was calculated by the Gordon–Taylor equation.27 As represented in Figure S1, every HPMC/EUD-S SPDs ACS Paragon Plus Environment

22

Page 23 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


showed single glass transition without phase separation of HPMC and EUD-S. Hence, the predicted value of Tg for NIF/HPMC/EUD-S SPD was calculated assuming that NIF/HPMC/EUD-S SPD was a binary mixture of amorphous NIF and HPMC/EUD-S SPD. Experimentally determined Tg and the true density of melt-quenched NIF (Tg = 45.4 °C) and each polymer SPD with different polymer compositions were incorporated in equation 2. The experimentally determined Tg and predicted Tg of NIF/polymer SPDs are shown in Table 1. ∆Tg was calculated by subtracting the predicted Tg from the experimentally determined Tg. The experimentally determined Tg was lower than the predicted Tg for all NIF/polymer SPDs. The lower value of experimentally determined Tg of solid dispersion than the predicted Tg indicated that the drug–drug, polymer–polymer, and drug–polymer molecular interactions in the solid dispersion were smaller in number and/or strength than drug–drug and polymer–polymer molecular interactions in each intact.31 NIF formed molecular interactions between NIF molecules in an amorphous structure.36 Furthermore, the HPMC, EUD-S, and HPMC/EDU-S SPDs exhibited polymer– polymer interactions, as discussed in the previous section. The negative value of ∆Tg for all NIF/polymer SPDs indicated that the molecular interactions of NIF–NIF, NIF–polymer, and polymer– polymer formed in the NIF/polymer SPDs were smaller in number and/or strength than those formed in the amorphized NIF and polymer SPDs.

Table 1. Experimentally determined and predicted Tg of NIF/polymer SPDs. The experimentally determined Tg is given as the average of duplicate values. The predicted Tg was calculated by the Gordon–Taylor equation. ∆Tg was calculated by subtracting the predicted Tg value from the experimentally determined Tg. Each NIF/polymer SPD contained NIF at NIF/polymer weight ratio of 25:75.

EUD-S/polymer (wt%) Experimentally determined Tg of NIF/polymer SPDs

0

10

25

50

75

100

95.0 °C

99.0 °C

106.0 °C

111.3 °C

121.7 °C

131.2 °C


23


Page 24 of 34

Predicted Tg of NIF/polymer SPDs

114.8 °C

115.6 °C

120.0 °C

123.9 °C

129.5 °C

133.6 °C

∆Tg

-19.8 °C

-16.6 °C

-14.0 °C

-12.6 °C

-7.8 °C

-2.4 °C

The absolute value of ∆Tg in NIF/polymer SPDs increased with a decrease in the weight ratio of EUD-S/polymer. The large ∆Tg value implied a weak NIF–polymer interaction in the solid dispersion.31 The molecular interaction between NIF and polymers should be weakened with a decrease in the amount of EUD-S in the NIF/polymer SPDs. As noted, HPMC and EUD-S formed molecular interactions with NIF via different modes. The FT-IR spectra of the NH-stretching region of NIF in the NIF/polymer SPDs are shown in Figure 10, and the 13C NMR spectra of the carbonyl region of NIF in the NIF/polymer SPDs are shown in Figure 11. The N-H stretching peaks of NIF were shifted to higher wavenumbers as the EUD-S/polymer weight ratio in the NIF/polymer SPDs increased, although those peaks were not clearly detected in the NIF/polymer SPDs with high HPMC loading. The molecular interaction between NIF and the polymer should be gradually altered by the polymer composition. Furthermore, the relative height of the shoulder peaks of the NIF carbonyl at approximately 170 ppm increased depending on the EUD-S/polymer weight ratio in NIF/polymer SPDs (Figure 11). Hydrogen bonding between the carbonyl group of NIF with the methacrylic acid of EUD-S was strengthened with increased EUD-S loading in NIF/polymer SPDs. These results demonstrated that the molecular state of NIF could be altered by the polymer composition of NIF/polymer SPDs.


24

Page 25 of 34

(a)

(b) (c) Transmittance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(d) (e) (f) (g)

3400

3350

3300

Wave number (cm-1)

3250

Figure 10. FT-IR spectra of (a) melt-quenched NIF, (b–g, broken line) polymer SPDs, and (b–g, solid line) NIF/polymer SPDs. EUD-S/polymer wt% of each SPD was (b) 0%, (c) 10%, (d) 25%, (e) 50%, (f) 75%, and (g) 100%. Each NIF/polymer SPD contained NIF at NIF/polymer weight ratio of 25:75. The dotted lines represent the peak maximum of N-H stretching of NIF in melt-quenched NIF and NIF/EUD-S SPD (25/75).


25


Figure 11.

13

Page 26 of 34

C CP/MAS NMR spectra of NIF/polymer SPDs with different HPMC/EUD-S weight

ratios. Each NIF/polymer SPD contained NIF at NIF/polymer weight ratio of 25:75. The NMR spectra were normalized by the maximum peak intensity of C14,16 of NIF.

For the detailed investigation of the molecular state of NIF in NIF/polymer SPDs, the 13C-T1 values of NIF/polymer SPDs with the NIF/polymer weight ratio of 25:75 were evaluated by solid-state NMR spectroscopy. Figure 12 represents ∆13C-T1 value of NIF calculated by subtracting the

13

C-T1 value of

NIF in NIF/HPMC SPD from that in each NIF/polymer SPD. The ∆13C-T1 value of NIF changed depending on the polymer composition in NIF/polymer SPDs. The 13C-T1 of NIF in the slow motional regime measured by solid-state NMR spectroscopy reflects the localized mobility of specific functional groups, and an increase in the 13C-T1 value represents the suppression of their mobility.41,42 The 13C-T1 value of NIF in the NIF/EUD-S SPD (EUD-S/polymer = 100%) was significantly higher than that in NIF/HPMC SPD (EUD-S/polymer = 0%), whereas the 13C-T1 value of the carbonyl carbon at 168 ppm was almost constant independent from the polymer composition. As mentioned above, NIF possessed the hydrogen-bonded and non-hydrogen-bonded carbonyl, whose peaks were observed at around 170 ppm and 168 ppm, respectively. The constant value of

13

C-T1 of the

13

C peak at 168 ppm could have

been due to the slight change in the molecular motion of the non-hydrogen-bonded carbonyl carbon of ACS Paragon Plus Environment

26

Page 27 of 34

NIF. On the contrary, EUD-S formed a hydrogen bond with the carbonyl group of NIF, indicated by the increase in the shoulder peaks of the carbonyl carbon of NIF at around 170 ppm. However, the 13C-T1 value of the carbonyl carbon of NIF at around 170 ppm could not be accurately evaluated for every NIF/polymer SPD due to the relatively small peak intensity and strong peak broadening. In contrast, the mobility of other functional groups of NIF was strongly suppressed in NIF/EUD-S SPD. The 13C-T1 value of NIF increased with an increase in the EUD-S loading for NIF/polymer SPDs with EUD-S/polymer wt% of more than 50%. The restriction of the molecular motion of NIF in NIF/polymer SPD was progressed with the increase in EUD-S amount followed by strengthening molecular interaction between NIF and polymer. However, the

13

C-T1 value of NIF was almost constant in the

NIF/polymer SPDs with EUD-S/polymer wt% of 10% and 25% compared with NIF/HPMC SPD. The similar molecular mobility of NIF among the NIF/polymer SPDs suggested that NIF molecules are dispersed mainly into the HPMC matrix and that the NIF mobility was not effectively restricted by EUD-S in NIF/polymer SPDs with EUD-S/polymer wt% of 10 and 25%. EUD-S effectively interacted with NIF and suppressed the NIF mobility in the NIF/polymer SPDs with EUD-S/polymer wt% of more than 50%.

20

C3 (35 ppm) C8 (124 ppm) C9 (127 ppm) C10,C11 (132 ppm) C1,C5,C6,C7 (148 ppm) C14,C16 (168 ppm)

15

∆13C-T1 (s)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10

5

0

-5 0

25

50

75

100

EUD-S/Polymer (wt%)

Figure 12. ∆13C-T1 values of NIF in NIF/polymer SPDs with different EUD-S/polymer weight ratios (n = 3, mean ± S.D.). Each NIF/polymer SPD contained NIF at NIF/polymer weight ratio of 25:75. ACS Paragon Plus Environment

27


Page 28 of 34

The ∆13C-T1 value was calculated by the subtracting the 13C-T1 value of NIF in NIF/HPMC SPD from that in each NIF/polymer SPD.

Effect of polymer composition on the molecular state of NIF and the dissolution from NIF/polymer SPDs The dissolution property of NIF from NIF/HPMC/EUD-S ternary solid dispersion was discussed based on the differences in the molecular state of NIF in NIF/polymer SPDs (Figure 13). The dissolution rate of the polymer used for the NIF/polymer SPD is important for the fast dissolution of NIF. The present study clearly demonstrated the advantage of blending of HPMC and EUD-S to enhance the dissolution of each polymer owing to the disruption of the hydrogen bonding in HPMC and the dimeric form of methacrylic acid of EUD-S through the molecular interaction between HPMC and EUD-S. In addition to the polymer dissolution property, polymer-controlled NIF dissolution from each NIF/polymer SPD was important to enhance the dissolution of NIF followed by fast polymer dissolution. In the NIF/polymer SPD, HPMC and EUD-S could form molecular interactions with NIF, although the mode of molecular interaction differed. In the NIF/polymer SPD with a relatively high EUD-S/polymer weight ratio, NIF dissolution was dominated by the polymer dissolution property. EUD-S could strongly interact with NIF and suppress the molecular motion of NIF in the solid dispersion. Thus, NIF and EUD-S were simultaneously dissolved without phase separation of NIF molecules during the dissolution of the NIF/EUD-S SPD. In the NIF/polymer SPD with EUDS/polymer wt% of 50% and 75%, EUD-S effectively interacted with NIF and restricted the molecular motion of NIF, leading to the polymer-controlled dissolution of NIF in a manner similar to that of NIF/EUD-S SPD. Therefore, the fast dissolution owing to polymer blending may effectively enhance NIF dissolution. In contrast, in the NIF/polymer SPD with a relatively high HPMC/polymer weight ratio, NIF dissolved separately from the polymers, indicating that NIF dissolution from the solid dispersion was not controlled by the polymer. The molecular interaction of NIF with HPMC was ACS Paragon Plus Environment

28

Page 29 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


relatively weak compared with that with EUD-S. NIF retained relatively high mobility in the NIF/polymer SPD with a high HPMC loading. Therefore, in the dissolution process, some NIF molecules probably remained on the disk surface of the NIF/polymer SPD after fast HPMC dissolution. The reduced ability to control NIF dissolution with an increase in HPMC loading resulted in the relatively slow dissolution of NIF from the NIF/polymer SPD.


29


Page 30 of 34

Figure 13. A schematic illustration of the molecular state and dissolution mechanism of NIF from the NIF/HPMC/EUD-S SPD with different polymer compositions.

CONCLUSIONS The mechanism of NIF dissolution differed between NIF/polymer SPDs composed mainly of HPMC and those composed mainly of EUD-S. NIF dissolution was significantly improved by the simultaneous dissolution of NIF and polymers from the NIF/polymer SPD composed mainly of EUD-S. In contrast, NIF dissolved separately from the polymers in the NIF/polymer SPD composed mainly of HPMC. The MDSC, FT-IR, and solid-state NMR measurements revealed that EUD-S interacted strongly with NIF, whereas NIF molecules that were dispersed into the HPMC matrix weakly interacted with HPMC. NIF mobility was effectively restricted in the NIF/polymer SPD composed mainly of EUD-S. The present study indicates that effective improvement in NIF dissolution by polymer-blended solid dispersions requires polymer-controlled NIF dissolution through strong molecular interactions between NIF and the polymer as well as an improved aqueous dissolution property of the blended polymer. Optimization of the polymer components in polymer-blended solid dispersions can significantly improve drug dissolution. Although the polymer dissolution was effectively improved owing to the intermolecular interaction between blended polymers, drug–polymer interaction in the polymer-blended solid dispersion is critical for rapid drug dissolution. Optimization of formulations that consider both polymer–polymer and drug–polymer intermolecular interactions will enable efficient preparation of polymer-blended solid dispersions with the desired drug dissolution.

SUPPORTING INFORMATION MDSC profiles, FT-IR spectra, PXRD patterns, 13C CP/MAS NMR spectra

ACKNOWLEDGMENTS ACS Paragon Plus Environment

30

Page 31 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


This study was partly supported by the Grant-in-Aid for Young Scientists (B) (JSPS, 16K18859) from the Japan Society for the Promotion of Sciences. We thank Shin-Etsu Chemical Co., Ltd., (Tokyo, Japan) for donating HPMC and Evonik Japan Co., Ltd., (Tokyo, Japan) for donating EUD-S.


31


Page 32 of 34

REFERENCES

(1) Baghel, S.; Cathcart, H.; O'Reilly, N. J. Polymeric amorphous solid dispersions: A review of amorphization, crystallization, stabilization, solid-state characterization, and aqueous solubilization of biopharmaceutical classification system class II drugs. J. Pharm. Sci. 2016, 105 (9), 2527-2544. (2) Yu, L. Amorphous pharmaceutical solids: preparation, characterization and stabilization. Adv. Drug Deliv. Rev. 2001, 48 (1), 27-42. (3) Mistry, P.; Mohapatra, S.; Gopinath, T.; Vogt, F. G.; Suryanarayanan, R. Role of the strength of drug–polymer interactions on the molecular mobility and crystallization inhibition in ketoconazole solid dispersions. Mol. Pharm. 2015, 12 (9), 3339-3350. (4) Yun, F.; Kang, A.; Shan, J.; Zhao, X.; Bi, X.; Li, J.; Di, L. Preparation of osthole-polymer solid dispersions by hot-melt extrusion for dissolution and bioavailability enhancement. Int. J. Pharm. 2014, 465 (1), 436-443. (5) Ueda, K.; Higashi, K.; Limwikrant, W.; Sekine, S.; Horie, T.; Yamamoto, K.; Moribe, K. Mechanistic differences in permeation behavior of supersaturated and solubilized solutions of carbamazepine revealed by nuclear magnetic resonance measurements. Mol. Pharm. 2012, 9 (11), 30233033. (6) Zhu, Q.; Toth, S. J.; Simpson, G. J.; Hsu, H.-Y.; Taylor, L. S.; Harris, M. T. Crystallization and dissolution behavior of naproxen/polyethylene glycol solid dispersions. J. Phys. Chem. B 2013, 117 (5), 1494-1500. (7) Kojima, T.; Higashi, K.; Suzuki, T.; Tomono, K.; Moribe, K.; Yamamoto, K. Stabilization of a supersaturated solution of mefenamic acid from a solid dispersion with EUDRAGIT® EPO. Pharm. Res. 2012, 29 (10), 2777-91. (8) Moustafine, R. I.; Bukhovets, A. V.; Sitenkov, A. Y.; Kemenova, V. A.; Rombaut, P.; Van den Mooter, G. Eudragit E PO as a complementary material for designing oral drug delivery systems with controlled release properties: Comparative evaluation of new interpolyelectrolyte complexes with countercharged eudragit L100 copolymers. Mol. Pharm. 2013, 10 (7), 2630-2641. (9) Brough, C.; Williams, R. O. Amorphous solid dispersions and nano-crystal technologies for poorly water-soluble drug delivery. Int. J. Pharm. 2013, 453 (1), 157-166. (10) Dahan, A.; Beig, A.; Lindley, D.; Miller, J. M. The solubility–permeability interplay and oral drug formulation design: Two heads are better than one. Adv. Drug Deliv. Rev. 2016, 101, 99-107. (11) Ueda, K.; Higashi, K.; Yamamoto, K.; Moribe, K. The effect of HPMCAS functional groups on drug crystallization from the supersaturated state and dissolution improvement. Int. J. Pharm. 2014, 464 (1–2), 205-213. (12) Van Duong, T.; Van den Mooter, G. The role of the carrier in the formulation of pharmaceutical solid dispersions. Part II: amorphous carriers. Expert Opin. Drug Deliv. 2016, 13 (12), 1681-1694. (13) Craig, D. Q. M. The mechanisms of drug release from solid dispersions in water-soluble polymers. Int. J. Pharm. 2002, 231 (2), 131-144. (14) Papageorgiou, G. Z.; Bikiaris, D.; Kanaze, F. I.; Karavas, E.; Stergiou, A.; Georgarakis, E. Tailoring the release rates of fluconazole using solid dispersions in polymer blends. Drug Dev. Ind. Pharm. 2008, 34 (3), 336-346. (15) Chen, Y.; Liu, C.; Chen, Z.; Su, C.; Hageman, M.; Hussain, M.; Haskell, R.; Stefanski, K.; Qian, F. Drug–polymer–water interaction and its implication for the dissolution performance of amorphous solid dispersions. Mol. Pharm. 2015, 12 (2), 576-589. (16) Onoue, S.; Sato, H.; Ogawa, K.; Kawabata, Y.; Mizumoto, T.; Yuminoki, K.; Hashimoto, N.; Yamada, S. Improved dissolution and pharmacokinetic behavior of cyclosporine A using high-energy amorphous solid dispersion approach. Int. J. Pharm. 2010, 399 (1), 94-101. ACS Paragon Plus Environment

32

Page 33 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(17) Prasad, D.; Chauhan, H.; Atef, E. Amorphous stabilization and dissolution enhancement of amorphous ternary solid dispersions: Combination of polymers showing drug–polymer interaction for synergistic effects. J. Pharm. Sci. 2014, 103 (11), 3511-3523. (18) Ohyagi, N.; Ueda, K.; Higashi, K.; Yamamoto, K.; Kawakami, K.; Moribe, K. Synergetic role of hypromellose and methacrylic acid copolymer in the dissolution improvement of amorphous solid dispersions. J. Pharm. Sci. 2017, 106 (4), 1042-1050. (19) Van Duong, T.; Van den Mooter, G. The role of the carrier in the formulation of pharmaceutical solid dispersions. Part I: crystalline and semi-crystalline carriers. Expert Opin. Drug Deliv. 2016, 13 (11), 1583-1594. (20) Chen, Y.; Wang, S.; Wang, S.; Liu, C.; Su, C.; Hageman, M.; Hussain, M.; Haskell, R.; Stefanski, K.; Qian, F. Initial drug dissolution from amorphous solid dispersions controlled by polymer dissolution and drug-polymer interaction. Pharm. Res. 2016, 33 (10), 2445-2458. (21) Higashi, K.; Hayashi, H.; Yamamoto, K.; Moribe, K. The effect of drug and EUDRAGIT® S 100 miscibility in solid dispersions on the drug and polymer dissolution rate. Int. J. Pharm. 2015, 494 (1), 9-16. (22) Meng, F.; Trivino, A.; Prasad, D.; Chauhan, H. Investigation and correlation of drug polymer miscibility and molecular interactions by various approaches for the preparation of amorphous solid dispersions. Eur. J. Pharm. Sci. 2015, 71, 12-24. (23) Torchia, D. A. The measurement of proton-enhanced carbon-13 T1 values by a method which suppresses artifacts. J. Magn. Reson. 1978, 30 (3), 613-616. (24) Baird, J. A.; Taylor, L. S. Evaluation of amorphous solid dispersion properties using thermal analysis techniques. Adv. Drug Deliv. Rev. 2012, 64 (5), 396-421. (25) Kawakami, K. Miscibility analysis of particulate solid dispersions prepared by electrospray deposition. Int. J. Pharm. 2012, 433 (1), 71-78. (26) Marks, J. A.; Wegiel, L. A.; Taylor, L. S.; Edgar, K. J. Pairwise polymer blends for oral drug delivery. J. Pharm. Sci. 2014, 103 (9), 2871-2883. (27) Gordon, M.; Taylor, J. S. Ideal copolymers and the second-order transitions of synthetic rubbers. i. non-crystalline copolymers. J. Appl. Chem. 1952, 2 (9), 493-500. (28) Brekner, M. J.; Schneider, H. A.; Cantow, H. J. Approach to the composition dependence of the glass transition temperature of compatible polymer blends: 1. Polymer 1988, 29 (1), 78-85. (29) Gashi, Z.; Censi, R.; Malaj, L.; Gobetto, R.; Mozzicafreddo, M.; Angeletti, M.; Masic, A.; Di Martino, P. Differences in the interaction between aryl propionic acid derivatives and poly(vinylpyrrolidone) K30: A multi-methodological approach. J. Pharm. Sci. 2009, 98 (11), 42164228. (30) Dengale, S. J.; Hussen, S. S.; Krishna, B. S. M.; Musmade, P. B.; Gautham Shenoy, G.; Bhat, K. Fabrication, solid state characterization and bioavailability assessment of stable binary amorphous phases of Ritonavir with Quercetin. Eur. J. Pharm. Biopharm. 2015, 89, 329-338. (31) Konno, H.; Taylor, L. S. Influence of different polymers on the crystallization tendency of molecularly dispersed amorphous felodipine. J. Pharm. Sci. 2006, 95 (12), 2692-2705. (32) Sancin, P.; Caputo, O.; Cavallari, C.; Passerini, N.; Rodriguez, L.; Cini, M.; Fini, A. Effects of ultrasound-assisted compaction on Ketoprofen/Eudragit® S100 Mixtures. Eur. J. Pharm. Sci. 1999, 7 (3), 207-213. (33) Fortier-McGill, B.; Toader, V.; Reven, L. 13C MAS NMR study of poly(methacrylic acid)– polyether complexes and multilayers. Macromolecules 2014, 47 (13), 4298-4307. (34) Jarvis, M. Cellulose stacks up. Nature 2003, 426, 611-612. (35) Nishiyama, Y.; Langan, P.; Chanzy, H. Crystal structure and hydrogen-bonding system in cellulose Iβ from synchrotron X-ray and neutron fiber diffraction. J. Am. Chem. Soc. 2002, 124 (31), 9074-9082. (36) Tang, X. C.; Pikal, M. J.; Taylor, L. S. A spectroscopic investigation of hydrogen bond patterns in crystalline and amorphous phases in dihydropyridine calcium channel blockers. Pharm. Res. 2002, 19 (4), 477-483. ACS Paragon Plus Environment

33


Page 34 of 34

(37) Lefort, R.; De Gusseme, A.; Willart, J. F.; Danede, F.; Descamps, M. Solid state NMR and DSC methods for quantifying the amorphous content in solid dosage forms: an application to ball-milling of trehalose. Int. J. Pharm. 2004, 280 (1-2), 209-219. (38) Geppi, M.; Guccione, S.; Mollica, G.; Pignatello, R.; Veracini, C. A. Molecular properties of ibuprofen and its solid dispersions with Eudragit RL100 studied by solid-state nuclear magnetic resonance. Pharm. Res. 2005, 22 (9), 1544-1555. (39) Tsuchiya, K.; Takahashi, A.; Takeda, N.; Asakawa, N.; Kuroki, S.; Ando, I.; Shoji, A.; Ozaki, T. Hydrogen-bonding effect on 13C NMR chemical shifts of amino acid residue carbonyl carbons of some peptides in the crystalline state. J. Mol. Struct. 1995, 350 (3), 233-240. (40) Gu, Z.; Zambrano, R.; McDermott, A. Hydrogen bonding of carboxyl groups in solid-state amino acids and peptides: Comparison of carbon chemical shielding, infrared frequencies, and structures. J. Am. Chem. Soc. 1994, 116 (14), 6368-6372. (41) Aso, Y.; Yoshioka, S. Molecular mobility of nifedipine–PVP and phenobarbital–PVP solid dispersions as measured by 13C-NMR spin-lattice relaxation time. J. Pharm. Sci. 2006, 95 (2), 318-325. (42) Oksanen, C. A.; Zografi, G. Molecular mobility in mixtures of absorbed water and solid poly(vinylpyrrolidone). Pharm. Res. 1993, 10 (6), 791-799.


34

Mechanism of enhanced nifedipine dissolution by ... - ACS Publications

Recommend Documents