Molecular-Level Control of Ciclopirox Olamine Release from Poly

Subscriber access provided by SUNY DOWNSTATE

Article

Molecular-level control of ciclopirox olamine release from poly(ethylene oxide)-based mucoadhesive buccal films: Exploration of structure-property relationships with solid-state NMR Martina Urbanova, Marketa Gajdosova, Milos Steinhart, David Vetchy, and Jiri Brus Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.6b00035 • Publication Date (Web): 28 Mar 2016 Downloaded from http://pubs.acs.org on April 3, 2016

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 free 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 accessible to all readers and 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.

Molecular Pharmaceutics 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 42

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

Molecular-level control of ciclopirox olamine release from poly(ethylene oxide)-based mucoadhesive buccal films: Exploration of structure-property relationships with solid-state NMR Martina Urbanova,1Marketa Gajdosova,2 Miloš Steinhart,1 David Vetchy,2* Jiri Brus1*

1)

Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky sq. 2, 162 06 Prague 6, Czech Republic 2)

Veterinary and Pharmaceutical University, Faculty of Pharmacy, Department of Pharmaceutics, Palacky street 1946/1, 612 42 Brno, Czech Republic

E-mail: [email protected], [email protected], [email protected], [email protected], [email protected]

KEYWORDS: Mucoadhesive buccal films; Ciclopirox olamine; PEO; Solid state NMR; Polymorphism; Polymer-drug interactions; 1 ACS Paragon Plus Environment


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

TOC (GRAPHICAL ABSTRACT)

Controlled CPX release

ABSTRACT: Mucoadhesive buccal films (MBFs) provide an innovative way to facilitate the efficient site-specific delivery of active compounds while simultaneously separating the lesions from the environment of the oral cavity. The structural diversity of these complex multicomponent and mostly multiphase systems as well as an experimental strategy for their structural characterization at molecular scale with atomic resolution were demonstrated using MBFs of ciclopirox olamine (CPX) in a poly(ethylene oxide) (PEO) matrix as a case study. A detailed description of each component of the CPX/PEO films was followed by an analysis of the relationships between each component and the physicochemical properties of the MBFs. Two distinct MBFs were identified by solid-state NMR spectroscopy: i) at low API (active pharmaceutical ingredient) loading, a nanoheterogeneous solid solution of CPX molecularly dispersed in an amorphous PEO matrix was created; and ii) at high API loading, a pseudo-cocrystalline system containing CPX:2-aminoethanol nanocrystals incorporated into the interlamellar space of a crystalline PEO matrix was revealed. These structural differences were found to be closely related to the mechanical and physicochemical properties of the prepared MBFs. At low API loading, the polymer chains of PEO provided sufficient quantities of binding sites to stabilize the CPX that was molecularly dispersed in the highly amorphous semiflexible polymer matrix. Consequently, the resulting MBFs were soft, with 2 ACS Paragon Plus Environment

Page 2 of 42

Page 3 of 42

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


low tensile strength, plasticity and swelling index, supporting rapid drug release. At high CPX content, however, the active compounds and the polymer chains simultaneously cocrystallized, leaving the CPX to form nanocrystals grown directly inside the spherulites of PEO. Interfacial polymer-drug interactions were thus responsible not only for the considerably enhanced plasticity of the system but also for the exclusive crystallization of CPX in the thermodynamically most stable polymorphic form, Form I, which exhibited reduced dissolution kinetics. The bioavailability of CPX olamine formulated as PEO-based MBFs can thus be effectively controlled by inducing the complete dispersion and/or microsegregation and nanocrystallization of CPX olamine in the polymer matrix. Solid-state NMR spectroscopy is an efficient tool for exploring structure-property relationships in these complex pharmaceutical solids.

INTRODUCTION Buccal mucosal diseases, which usually manifest as lesions (e.g., recurrent aphthous stomatitis, herpetic stomatitis, and stomatitis simplex), affect the majority of the population. Among them, oral candidiasis is one of the primary causes of local and systemic mycoses. The prevalence of these diseases grows with increased use of antibiotics, corticosteroids and immunosuppressives. Infections induced by Candida are often accompanied by leukaemia or AIDS and also commonly occur in chronic smokers.1,2 Consequently, great interest has been focused on the buccal mucosa as a site for the systemic delivery of various drugs. However, the buccal mucosa is one of the few surfaces of the human body that is permanently exposed to external factors related to food intake, breathing and speaking processes, which can lead to the onset of some disorders. The effectiveness of therapies for these diseases can thus be improved by the application of mucoadhesive buccal films (MBFs) used as dressings that separate the lesion from the environment of the oral cavity.3 Moreover, MBFs provide an innovative way to deliver drugs 3 ACS Paragon Plus Environment


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 4 of 42

and have many advantages: MBFs do not obstruct speech or food and beverage intake; their application does not generate the unpleasant sensation of a foreign body in the oral cavity; and they act directly at the affected area of the lesion.3 Ideal MBFs should be flexible, adaptable, comfortable, strong and sufficiently mechanically durable to withstand damage due to stress from oral activities. In addition, MBFs should extend the otherwise short residence time of APIs (active pharmaceutical ingredients) compared to semisolid or liquid drugs used for the oral mucosa .4,5 One antifungal agent with robust local effects on Candida is ciclopirox olamine (CPX), a broad-spectrum antifungal agent that acts against dermatophytes, yeasts, moulds, Candida species and some bacteria. Depending on the API concentration and its contact time with the fungus, ciclopirox can be either fungistatic or fungicidal. The advantages of ciclopirox are its nontoxicity and low potential for irritation. Ciclopirox also displays mild anti-inflammatory effects in biochemical and pharmacological models, excellent tolerability and the complete absence of serious adverse effects.6 Treatments for isolated oral candidiasis can be local or both local and systemic, which is perceived as less harmful for the organism.7 The construction of MBFs is based on the formulation of a two-component API-polymer system that is structurally similar to pharmaceutical solid dispersions.8 Polymers suitable for the preparation of MBFs should have the ability to bind to mucin, a glycoprotein responsible for

mucoadhesion.

Generally,

cellulose

derivatives,

chitosan,

polyacrylic

acid,

polyvinylpyrrolidone or poly(ethylene oxide) (PEO) are used as film-forming polymers. PEO is a semicrystalline nonionic polymer4 and is widely used due to its solubility in water, its high viscosity, its ability to form hydrogen bonds and its compatibility with other bioactive substances.9 The vast majority of scientific publications focus on the preparation, dissolution and stabilization of MBFs,10 and detailed structural research discussing morphology of the buccal films at micrometer scale is scarce.11,12 However, drug-polymer, drug-drug, and

4 ACS Paragon Plus Environment

Page 5 of 42

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


polymer-polymer interactions are the most critical factor in the design and performance of MBFs. Therefore, detailed structural characterization is a fundamental prerequisite not only for the rational design of these pharmaceutical solids due to hardly predictable polymorphism and related differences with APIs bioavailability13,14,15 but also for the protection of intellectual property. Moreover, detailed physicochemical and structural characterization is a mandatory requirement of (inter)national authorities, such as the FDA and the EMA. As mucoadhesive buccal films are complex multicomponent and mostly multi-phase liquisolid systems based on solid dispersions the structural diversity of which has been recently well documented,8,16-19 traditional experimental approaches involving structural analysis are often not as useful for studies of these systems. Therefore, one of the purposes of this study was to develop an easy-to-implement experimental strategy for comprehensive characterising the structures of these films allowing for probing the structure of the systems at nanometer scale and molecular level with atomic resolution. Among numerous analytical techniques that are applicable to the analysis of multi-phase, multi-component and largely semicrystalline pharmaceutical solids, solid-state nuclear magnetic resonance (NMR) spectroscopy is the most suitable technique. The potential applications and reliability of the proposed strategy, which is based on domain-selective solid-state NMR spectroscopy combined with two-dimensional 1H-13C FSLG HETCOR NMR experiments, were demonstrated using novel mucoadhesive buccal films consisting of ciclopirox olamine, glycerol and poly(ethylene oxide) designed for the local treatment of oral candidiasis. By using the abovementioned strategy, two distinct types of CPX/PEO mucoadhesive buccal films were identified: i) at low API loading, a nanoheterogeneous solid solution of CPX molecularly dispersed in an amorphous PEO matrix was created, whereas ii) at high API loading, a pseudo-co-crystalline system containing CPX:2-aminoethanol nanocrystals incorporated into the interlamellar space of a crystalline



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

PEO matrix was revealed. These structural differences were found to be closely related to the mechanical and physicochemical properties of the prepared films. Moreover, the molecular mechanisms controlling the structural and physicochemical properties of these mucoadhesive buccal films were revealed.

MATERIALS AND METHODS Chemicals: The following chemicals were used to prepare MBFs: ciclopirox olamine (CPX, 6-cyclohexyl-1-hydroxyl-4-methyl-2(1H)-pyridone and 2-aminoethanol) produced by Zentiva Group (Praha, Czech Republic); poly(ethylene glycol) (PEO, Polyox® WSR 1105; Mw=900,000) produced by Colorcon® Ltd. (Dartford, United Kingdom); glycerol (SigmaAldrich); and Eudragit® NM 30D (Poly(ethyl acrylate-co-methyl methacrylate) 2:1) purchased from Evonik Industries (Darmstadt, Germany). Mucin from porcine stomach (Type II), for the preparation of artificial mucus for adhesion testing in the form of a 5% dispersion (w/w) in phosphate buffer pH 6.8 [Ph. Eur. 7], was provided by Sigma-Aldrich® Chemie GmbH (Steinheim, Germany). For primary structural characterisation of individual components used for preparing MBFs the

13

C CP/MAS NMR spectra were measured (Figure 1). Signal assignment was

performed on the basis of our experience and 13C CPPI/MAS NMR experiments. As indicated by the relatively small line-widths of the signals of neat CPX (ca. 50-100 Hz in Figures 1a and b) the original as-received active substance is a highly crystalline system with molar ratio 1:1 between ciclopirox and 2-aminoethanol. This is typical characteristics for the polymorphic Form II of CPX (confirmed by x-ray data see below).15 In contrast to most organic solids, semicrystalline PEO shows unconventional behavior in

13

C CP/MAS NMR spectra. This unusual behavior has been investigated and thoroughly

analyzed by combining

13

C CP/MAS NMR with WAXS and DSC and follows from the

extensive segmental motion of PEO segments in the crystalline domains where crystalline 6 ACS Paragon Plus Environment

Page 6 of 42

Page 7 of 42

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


PEO chains adopt helical conformation and execute high-amplitude motions. Frequency of this motion is comparable to the strength of rf field used for 1H high-power decoupling (ca. 50-100 kHz expressed in frequency units). This very competition between frequency of segmental reorientations and frequency of 1H dipolar decoupling leads to destructive interference that causes ineffectiveness of the dipolar decoupling and from this following severe broadening of the detected signal. In contrast, amorphous PEO at room temperature is far above Tg (ca. -67°C). Consequently, rapid segmental motion of PEO chains in amorphous phase is nearly isotropic and leads to narrowing of corresponding NMR signal. Therefore at room temperature the

13

C MAS NMR spectrum of semicrystalline PEO consists of a broad

(linewidth ∼1 kHz) and a narrow (∼200 Hz) signal with similar chemical shifts which in contrast to most other semicrystalline polymers are assigned to the crystalline and amorphous phase, respectively.20,21,22 The above-mentioned behavior of PEO is clearly documented in Figure 1. The broad signal in the

13

C CP/MAS NMR spectrum of neat PEO measured with a short cross-

polarization contact time (CP=0.2 ms) reflects the crystalline part of the polymer (Figure 1c), whereas the narrow signal detected in the 13C CP/MAS NMR spectrum measured with a long contact time (CP=1.0 ms) demonstrates the amorphous fraction of PEO (Figure 1d). Quantitative data extracted from the single-pulse

13

C MAS NMR spectrum measured with a

long repetition delay (240 s, not shown here) indicate crystallinity of the system, ca. 85-90%, while the amount of liquid-like amorphous PEO phase is relatively low, ca. 10-15% (determined by spectral deconvolution and further confirmed by x-ray data). In contrast, EUD as a typical amorphous polymer with Tg close to the room temperature (Tg = 11°C) is reflected by broad 13C CP/MAS NMR signals (Figure 1e).



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

Ciclopirox olamine 6

H2N

14

Page 8 of 42

Eudragit

Poly(ethylene oxide)

CH3 b

H

OH

13

CH3

d

H

c

a 3 4

2

5 8 9

7

10

12

O

N

O

O

O

g

H

H

1

n

O

e

O

O

h,i

C 2H 5

f

CH3 n

OH

11

Scheme 1: Chemical structures of the applied active pharmaceutical ingredient (ciclopirox olamine) and the polymers PEO and Eudragit (EUD).

9,10,11 7

a) CPX

3

1

4 2

5

8,12

6

13 14

b) CPX CH2

PEO cryst

c) PEO CH2

d) PEO

PEO amorph

e) EUD

h

f

b,a,c

d,i

e,g

180

160

140

120

100

80

60

40

20

0

chemical shift, ppm

Figure 1. 13C CP/MAS NMR spectrum of neat CPX polymorphic Form II (a); 13C CPPI/MAS NMR spectrum of neat CPX polymorphic Form II (b); 13C CP/MAS NMR spectrum of neat PEO measured with a short cross-polarization contact time, CP=0.2 ms (c);

13

C CP/MAS

NMR spectrum of neat PEO measured with a long contact time, CP=1 ms (d), and

13

C

CP/MAS NMR spectrum of neat EUD (e)

Preparation of mucoadhesive buccal films: Mucoadhesive mono- and bilayer oral films were prepared using the solvent casting method.23 CPX was dissolved in distilled water. 8 ACS Paragon Plus Environment

Page 9 of 42

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


Glycerol and PEO were then added to a drug solution. This solution was mixed using a paddle stirrer (RZR 2021, Heidolph Instruments GmbH & Co.KG, Schwabach, Germany) at 500 rpm for 2 h. Depending on the prepared dispersion, Eudragit (EUD) was eventually added. The solution was stirred for an additional 2 h. The solutions were allowed to swell for another 72 h and were stirred using Ultra-turrax® (type T 25 basic, IKA® Werke GmbH & Co. KG, Staufen, Germany) before casting for 2 min (13,000 rpm). The compositions of the prepared solutions are shown in Table 1. Using an automatic pipette 18 ml of the prepared uniform solutions was cast into a round plastic mould (63 mm in diameter), and the solvents were allowed to evaporate at room temperature for 72 h. Next, 9 ml of the selected solution was added to the same moulds and dried at room temperature. The solutions used for the preparation of the final films are shown in Table 2. Labelling of the prepared MBFs (i.e. CPX0.2/PEO or CPX0.2/PEO/EUD) reflects the absence/presence of EUD in the prepared films, whereas the numbers in the codes CPX0.2 or CPX1.8 reflect the weight of CPX in the final muccoadhesive patches (e.g. 0.2 or 1.8 mg of CPX in the prepared products). These codes also indicate the therapeutic dose of the active compound loaded into the polymer matrix of the prepared patches. Solutions with high contents of CPX were homogenized before casting using a paddle stirrer for another 30 min.

Table 1. Compositions of solutions used for preparing CPX mucoadhesive oral films. Solution CPX PEO Glycerol EUD Water No. (mg) (mg) (mg) (mg) (mg) 1 0.2 2 3 94.8 2 1.8 2 3 93.2 3 0.2 2 3 6 88.8 4 1.8 2 3 6 87.2



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 42

Table 2. Solutions used for preparing CPX oral films using the solvent casting method. For the first and second castings, 18 or 9 ml of the solution was used, respectively. Code of final MBFs CPX0.2/PEO CPX0.2/PEO/EUD CPX1.8/PEO CPX1.8/PEO/EUD

Solution No. 1st casting 2nd casting 1 1 3 1 2 2 4 2

Solid-state NMR spectroscopy: Solid-state NMR spectra were measured at 11.7 T using a Bruker Avance III HD 500 WB NMR spectrometer with 4-mm ZrO2 rotors. The polarization (CP) magic-angle spinning (MAS) and T1-filtered

13

C cross-

13

C MAS NMR spectra were

measured at a spinning frequency of 11 kHz, a B1(13C) field nutation frequency of 62.5 kHz, a contact time of 0.2-3 ms, and a repetition delay of 5-40 s. A cross-polarization-polarizationinversion (CPPI) experiment24, with a polarization-inversion (PI) period of 40 µs, was used to discriminate between CH2, CH and quaternary carbons. The

13

C-detected T1(1H) relaxation

times were measured using a saturation-recovery experiment in which the initial train of 1H saturation pulses was followed by a variable delay (0.01–15 s). To measure the

13

C-detected

T1ρ(1H) relaxation times, an initial 1H(90°) pulse was followed by a variable 1H spin-lock pulse for 0.1–25 ms. The intensity of the 1H spin-locking field was 80 kHz. The 2D 1H-13C FSLG-HETCOR experiments25 were performed with FSLG (frequency-switched LeeGoldburg) decoupling during the t1 evolution period, consisting of 64 increments consisting of 64-128 scans with a dwell time of 100 µs. The 13C scale was calibrated with glycine as an external standard (176.03 ppm – low-field C=O signal). The experiments were performed at 283 K with active cooling to compensate for the frictional heating of the spinning samples.26 Wide-angle X-ray scattering (WAXS): WAXS experiments were performed using a pinhole camera (Molecular Metrology System) attached to a micro-focused X-ray beam generator (Osmic MicroMax 002) operating at 45 kV and 0.66 mA (30 W). The samples were measured 10 ACS Paragon Plus Environment

Page 11 of 42

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


in transmission mode using interchangeable 23 x 25 cm imaging plates (Fujifilm). The q range was 0.25–3.5 Ǻ-1 (q = (4π/λ)sinθ, λ=1.54 Ǻ, and 2θ was the scattering angle). Mechanical properties: A Texture Analyzer CT3 (Brookfield, Middleborough, USA) equipped with a 4.5 kg load cell and TexturePro CT software (Brookfield) was used to determine the tensile strength of the prepared films using a modification of Shidhaye’s method.27 Rectangular samples (10 x 40 mm) were held between two clamps of a TA-DGA probe positioned at a distance of 2 cm. The measurement was repeated three times for each film sample. The texture analyser with a TA39 probe (cylindrical probe 2 mm in diameter; probe motion speed 0.5 mm/s) was used for a penetration test. The force needed for the penetration of square samples (25 x 25 mm) fixed in the jig TA-CJ, the work performed during this process, and the deformation of the film at the moment of penetration were measured. The measurements were repeated three times for each film sample. Because the films had different thicknesses, the values measured with the texture analyser were recalculated for a film thickness of 100 µm.3 Optical analysis: Film thickness was measured using an optical microscope (SMZ 1500, Nikon, Tokyo, Japan), with a camera (DFK 72AUC02, Imaging Source, Maisach, Germany) and software (NIS-Elements AR 4.00.06, Nikon). Rectangular samples were vertically fixed in a holder. The microscope was focused on the edge of the film, and the sample thickness was measured at 5 locations. This procedure was repeated 3 times for each film sample. Swelling index: The swelling properties and erosion characteristics of the film were evaluated by determining the percentage of hydration (water uptake) and the matrix erosion or dissolution. For each formulation, a circular sample 15 mm in diameter was weighed (W1 – dry weight) and placed on a Petri dish filled with a sponge containing phosphate buffer at pH 6.8. MBFs were weighed (W2) at predetermined time points (10, 20, 30, 45, 60, 75, 90, 120,



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 42

150, 180, 210, 240, 270 and 300 min). The percentage swelling index SI (%) was calculated using the following equation.28 SI (%) = [(W2 - W1) / W1] * 100

(1)

The measurement was repeated 3 times for each film sample. The maximum swelling index and the time when the maximum swelling index was achieved t(SImax) were recorded. In vitro residence time: A modified disintegration apparatus29 was used to determine the in vitro residence time (RT). Buccal mucosa was simulated using a cellophane membrane covered by a 5% mucin dispersion (w/w) in phosphate buffer at pH 6.8 (10 µl per 1 cm2) glued to the surface of the plastic slab. The phosphate buffer (pH 6.8) was maintained at 37±0.5 °C and was used as the testing medium. The slab with the attached circular samples (15 mm in diameter) of MBFs was allowed to move up and down (50 drafts per minute), so that the samples were completely immersed in the buffer solution at the lowest point and were out of the solution at the highest point. The time necessary for the complete detachment or erosion of the film was measured. The measurement was repeated three times.3 In vitro release study: A paddle apparatus (SOTAX AT 7, SOTAX, Aesch, Switzerland) (50 rpm) and software for dissolution measurement (WinSOTAX plus, SOTAX) were used to evaluate the CPX dissolution rate. Phosphate buffer at pH 6.8 (500 ml) was used as the dissolution medium. Circular MBF samples (15 mm in diameter) were attached to circular glass slabs by 5% mucin dispersion (10 µl per 1 cm2). Samples of the dissolution media were withdrawn at 15 min intervals and were tested spectrophotometrically (Lambda 25 UV/VIS spectrophotometer, PerkinElmer®, Waltham, USA) at 310 nm. The duration of MBF attachment to the slab with mucin was determined visually. Based on the obtained results, the dissolution rate of CPX, the kinetic order of drug release and the release mechanism according to dissolution kinetic models (Korsmeyer-Peppas model; Higuchi model; zeroorder kinetics and first-order kinetics) were evaluated.30,31


Page 13 of 42

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


RESULTS Two types of MBF were prepared by the solvent casting method: i) low-API-loaded samples (CPX0.2/PEO and CPX0.2/PEO/EUD) with ca. 5-3 wt. % of CPX dispersed in the polymer matrix; and ii) high-API-loaded samples (CPX1.8/PEO and CPX1.8/PEO/EUD) with ca. 25 wt. % of CPX in the matrix. All of the prepared samples were milk-white, macroscopically homogeneous, and relatively soft films.

Structural analysis of CPX/PEO MBFs: solid-state NMR spectroscopy and WAXS Solid-state NMR spectroscopy has long been recognized as a powerful method for providing detailed structural information on intra- and intermolecular interactions, molecular microsegregation and the segmental dynamics of multicomponent pharmaceutical solids, particularly in systems where the active compound forms nanosized amorphous or semicrystalline domains. Therefore, primary information regarding the polymorphism of the API and the structure, composition and segmental mobility of the prepared MBFs was obtained from 1H MAS, T1-filtered 13C MAS and 13C CP/MAS NMR spectra. The T1-filtered 13

C MAS NMR experiment is based on selective detection of

13

C magnetization of rapidly

relaxing, usually highly mobile components. On the other hand, the 13C magnetization of rigid segments usually exhibits long relaxation times and thus requires long repetition delays to be effectively recovered into the thermal equilibrium. This way, signals from carbon species with a long relaxation time will be greatly diminished in the resulting T1-filtered spectra. In contrast,

13

13

C MAS NMR

C magnetization of these rigid segments is selectively excited and

relatively enhanced in the 13C CP/MAS NMR spectra measured with short cross-polarization contact times. The degree of homogeneity of the prepared systems at the nanometre scale was further examined using

13

C-detected T1(1H) and T1ρ(1H) spin-lattice relaxation experiments

and 1H-13C FSLG HETCOR measurements.



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 14 of 42

Structural characterization of low-CPX-loaded MBFs Considerable structural changes in CPX upon the formation of low-API-loaded MBFs were revealed by comparing the

13

C CP/MAS NMR spectra of neat substances and the prepared

products CPX0.2/PEO and CPX0.2/PEO/EUD (Figure 2). Whereas the spectrum of neat CPX (Figure 2a) indicated a highly crystalline polymorphic Form II of ciclopirox:olamine 1:1, the observed broadening of CPX signals clearly indicated its complete amorphization (Figure 2b). After the binding of a small amount of CPX into the PEO matrix the corresponding

13

C

CP/MAS NMR spectrum of CPX0.2/PEO systems showed clear increase in the intensity of the narrow signal at 70 ppm, which is attributed to the amorphous phase of PEO.20,21,22 This finding thus indicates a decrease in crystallinity of the PEO matrix. This fact was additionally supported by WAXS diffractograms (Figure 3) the deconvolution of which revealed decrease in crystallinity from ca. 90-85% (neat PEO) to ca. 50-40%. The structure of the residual PEO crystallites remained unmodified as indicated by the unchanged 2θ positions of the X-ray reflections. A comparison of the T1-filtered

13

C MAS and

13

C CP/MAS NMR spectra revealed

considerable concentration fluctuations, which resulted in a heterogeneous multiple-phase character of the CPX0.2/PEO system. Whereas the

13

C CP/MAS NMR spectrum reflected a

rigid (semiflexible) phase consisting of immobilized CPX molecules and a small fraction of glycerol both dispersed in the semicrystalline PEO matrix, the T1-filtered

13

C MAS NMR

spectrum showed a highly mobile liquid-like phase that was formed predominantly by glycerol, PEO, 2-aminoethanol and a very small fraction of dissolved CPX molecules (narrow signals in Figures 2c and 2f). As inferred from the deconvolution of 1H NMR spectra (Figure 4), the quantity of entrapped glycerol reflected by narrow signals was ca. 30%.


Page 15 of 42

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


x1/20

DMSO

9,10,11

7 8,12

a) CPX 3

1

4

5

2

13

6 14

PEO am

b) CPX0.2/PEO 4

9,10,11 7 8,12

2

6

3,1,5

14 glycerol 8,12

c) CPX0.2/PEO

PEO am

4 2

14 7

8,12

d) CPX

1 5

3

PEO am

4 2

13

7

9,10,11 6

9,10,11 6

d,i

e) CPX0.2/PEO/EUD

h,13 f

e,g 4

180

3,1,5

160

d,i PEO am

*SSB

140

6

glycerol

f) CPX0.2/PEO/EUD e,g

b,a,c 8,12

2

4 2

120

100

80

60

40

20

chemical shift, ppm

Figure 2.

13

C CP/MAS NMR spectrum of the neat polymorphic Form II of CPX (a);

13

C

CP/MAS NMR spectrum of the CPX0.2/PEO film (b); T1-filtered 13C MAS NMR spectrum of the CPX0.2/PEO film (c); solution state

13

C NMR spectrum of dissolved CPX (d),

CP/MAS NMR spectrum of the CPX0.2/PEO/EUD film (e); and T1-filtered

13

13

C

C MAS NMR

spectrum of the CPX0.2/PEO/EUD film (f). The spectral region from 55 to 90 ppm is multiplied by a factor 1/20 to reduce intensities of PEO and glycerol signals.

For further quantitation of the prepared low-CPX loaded films we recorded and analyzed the single-pulse

13

C MAS NMR spectra measured with a long repetition delay (240 s).

However, despite the extremely long measurement time (3 days) the resulting spectrum still exhibits relatively low signal-to noise ratio. Moreover, it is worthy to note that the total 15 ACS Paragon Plus Environment


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 16 of 42

amount of the active compound in these MBFs is very low with ca. 3-5 wt. %. Consequently, the obtained quantitative data are rather inaccurate. Nevertheless, it is still possible to estimate that the liquid-like phase represents ca. 40% of the system (in total), while the semi-solid part represents ca. 60%. Furthermore, we were also able to estimate the amount of CPX dissolved in liquid-like domains, where according to our data approximately 30-40% of CPX is dissolved in liquid-like domains, while 60-70% remain dispersed in the semisolid amorphous matrix. a) CPX

CPX1.8/PEO

Amorphous phase

CPX0.2/PEO

PEO

10

15

20

25

30

35

40

45

50

°2 theta

b) CPX

CPX1.8/PEO

CPX0.2/PEO

PEO

16

18

20

22

24

26

28

30

°2 theta

Figure 3. WAXS diffractograms of neat CPX polymorphic Form II; the CPX1.8/PEO film; the CPX0.2/PEO film; and neat PEO. The difractograms are displayed in the whole 2θ range to show a broad signal of the amorphous phase (a), and in the expanded region (b).


Page 17 of 42

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


Glycerol (CH2O) Glycerol (OH)

PEO

a) CPX0.2/PEO glycerol PEO

b) CPX1.8/PEO

5.5

5.0

4.5

4.0

3.5

3.0

ppm

c) CPX0.2/PEO

d) CPX1.8/PEO

e) CPX

f) PEO 40

20

0

-20

-40

ppm

Figure 4. Expanded region of 1H MAS NMR spectra of CPX0.2/PEO film (a); and the CPX1.8/PEO film (b); and the full range 1H MAS NMR spectra of CPX0.2/PEO film (c); the CPX1.8/PEO film (d); neat CPX polymorphic Form II (e); the neat PEO (f).

An analysis of the

13

C CP/MAS NMR spectra of the samples with EUD

(CPX0.2/PEO/EUD, Figure 2e) revealed no differences in the structure or dynamics of the PEO matrix in the form of dispersed CPX molecules. In summary, the low-CPX-loaded films CPX0.2/PEO and CPX0.2/PEO/EUD were multiple-phase liquisolid systems in which fluctuations in the glycerol concentration resulted in the formation of a semiflexible solid phase dominated by a highly amorphous PEO matrix in which amorphous CPX and confined



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

glycerol molecules were dispersed. The secondary liquid phase (i.e., domains with locally increased concentrations of glycerol) was formed by the mixture of 2-aminoethanol and glycerol in which CPX molecules were dissolved. The extent of mixing of CPX in the PEO semicrystalline matrix was investigated using

13

C-detected T1(1H) relaxation experiments,

which are discussed below.

Structural characterization of high-CPX loaded MBFs Contrary to the results obtained for the low-API-loaded MBFs, the narrow signals of CPX and the broad signal of PEO detected in the 13C CP/MAS NMR spectra of CPX1.8/PEO and CPX1.8/PEO/EUD films indicate the formation of a highly crystalline structure (Figure 5). The observed symmetric splitting and changes in the resonance frequencies of CPX carbon atoms reflect the formation of a polymorphic form of CPX olamine, the crystal structure which is characterized by two symmetry-independent molecules of ciclopirox and one molecule of 2-aminoethanol in the crystal unit. As confirmed by WAXS, the most stable polymorphic Form I of CPX:olamine 2:1 was created upon the crystallization of the PEO matrix.15 Similar to the low-CPX systems, approximately 30% of residual glycerol was found in the high-CPX MBFs (Figure 5). Interestingly, in contrast to the low-CPX MBFs, the glycerol signals were absent in the corresponding

13

C CP/MAS NMR spectra (Figures 5b and 5c).

However, these signals were clearly apparent in the T1-filtered 13C MAS NMR spectra (Figure 5d and 5f). These findings indicate that glycerol formed a minor secondary liquid phase that was not directly incorporated into the solid PEO matrix but instead filled voids between crystallites. Despite this finding, the T1-filtered

13

C MAS NMR spectrum showed very low

dissolution of CPX in liquid glycerol (Figures 5d and 5f, dissolved CPX is marked by asterisks). As follows from the single-pulse

13

C MAS NMR spectrum recorded with a long


Page 18 of 42

Page 19 of 42

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


repetition delay (240 s) we found approximately 5% of CPX dissolved in the glycerol-rich liquid-like domains, while ca. 95% remained dispersed as crystalline domains in the PEO matrix. At the same time, by combining WAXS experiments with the measurements of

13

C

CP/MAS NMR spectra employing long CP time (2 ms), the PEO matrix was found to be highly crystalline (ca. 90%) with a minimal amount of amorphous phase.

9,10,11 7 8,12 6

a) CPX

2

4

5

1

3

13 14

9,10,11 7

b) CPX1.8/PEO

13 4

1

3

14

PEO-cryst

2

8,12 6

5 9,10,11

7

6

c) CPX1.8/PEO 1

3

13 4

5

4

d) CPX1.8/PEO 8x 3

1

2

glycerol

2 *

8,12 14

PEO-am

*

PEO-am

14 7

13

5

6

7

e) CPX1.8/PEO/EUD

f,13

PEO-cryst

e,g

4 3

1

14 8,12

6

b,a,c h

2

5

9 10 11 d,i

glycerol

f) CPX1.8/PEO/EUD

4 *

10x e,g

180

1

3

160

2 * f

5

140

120

100

80

14 b,a,c

60

6

7

40

d,i

20

chemical shift, ppm

Figure 5.

13

C CP/MAS NMR spectrum of the neat polymorphic Form II of CPX (a);

13

C

CP/MAS NMR spectrum of CPX1.8/PEO measured with a short and a long contact time CP=0.1 and 2 ms, (b) and (c), respectively; T1-filtered

13

C MAS NMR spectrum of

CPX1.8/PEO (d); 13C CP/MAS NMR spectrum of CPX1.8/PEO/EUD measured with a short contact time CP=0.1 (e); and T1-filtered 13C MAS NMR spectrum of CPX1.8/PEO/EUD (f).



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

In addition to high crystallinity, the recorded WAXS data (Figure 3) showed a slight shift and broadening of the main PEO reflections. These findings reflect changes in the crystal lattice d-spacing of approximately 0.04 Å and a reduction in the size of the PEO crystallites. Based on published data32,33, these changes can be interpreted as the simultaneous cocrystallization of CPX in the interlamellar space of PEO spherulites. This phenomenon had previously been detected for small organic molecules, allowing strong interactions with PEG chains.34 We also hypothesize that the crystallites of CPX that were intimately mixed with polymer chains partially blocked the growth of large PEO crystals and induced the formation of small crystallites.

Distribution of the active compound in the PEO matrix To further explore the homogeneity of the prepared MBFs at the nanometre scale, the 13Cdetected 1H spin-lattice relaxation times, T1(1H) and T1ρ(1H), were measured. Using a wide range of two-component solids, differences in 1H relaxation times between individual components (e.g., the active compound and the polymer matrix) were previously demonstrated to indicate their microsegregation and formation of domains. This rule follows from the fact that 1H-1H spin diffusion, which is generally very rapid in organic solids, is not able to equilibrate the nuclear spin properties of 1H nuclei in all parts of the material. Typically, 1H magnetization is transferred over a distance of approximately 1.1-1.2 nm during 1 ms.35 In the measurements of T1(1H) spin-lattice relaxation times, the relevant times of 1H spin diffusion are in the range of several seconds. Consequently, 1H magnetization can be effectively transferred over several hundred nanometres. Therefore, if the T1(1H) spin-lattice relaxation times of the API and polymer matrix are different, micro-segregation occurred and the size of the domains would be larger than ca. 200-300 nm.36 A similar approach also applies to measurements of T1ρ(1H) spin-lattice relaxation times, although the 1H spin


Page 20 of 42

Page 21 of 42

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


diffusion times are in the range of milliseconds. Therefore, 1H magnetization can be effectively transferred over several tens of nanometres. Consequently, a multi-component system with a uniform T1ρ(1H) relaxation time can thus be considered to be homogeneous with a domain size less than several nanometres.37,38 To quantify the extent of mixing, the maximum diffusive path length (L) can be estimated using the following equation:39,40 L = (6DTi)1/2

(1)

where D is the spin-diffusion coefficient, typically 0.8 nm2 ms-1 for organic solids41, and Ti is either T1(1H) or T1ρ(1H).

Table 3. T1(1H) and T1ρ(1H) spin-lattice relaxation times obtained for neat CPX, EUD and PEO and the corresponding MBFs. The obtained relaxation parameters represent average values of three independent measurements with an experimental error ca. 0.8-0.5 s for T1(1H) relaxation times and 0.6-0.4 ms for T1ρ(1H) spin-lattice relaxation times.

MBFs Neat CPX Form I Neat CPX Form II Neat PEO crystalline Neat PEO amorphous Neat EUD CPX0.2/PEO CPX0.2/PEO/EUD CPX1.8/PEO CPX1.8/PEO/EUD

T1(s) CPX 3.1 58.1 2.1 2.4 2.4 2.3

T1(s) T1(s) T1ρ(ms) PEO EUD CPX 98.7 5.1 7.2 3.7 1.2 2.4 1.2 2.7 2.0 1.2 2.5 65.7 2.5 2.0 62.2

T1ρ(ms) PEO 0.4 10.1 1.6 1.4 0.3 0.3

T1ρ(ms) EUD 4.2 2.7 4.7

It is important to note that the T1 relaxation constants obtained for the neat components (CPX, EUD and PEO) exhibited significantly different values (Table 3), which is the key precondition for a successful analysis. To obtain a complete set of relaxation parameters, we selectively measured the relaxation times for crystalline and amorphous fractions of neat PEO. Moreover, we also obtained relaxation data for neat polymorphic Form I of CPX that



was found in the prepared high-CPX loaded films. CPX Form I was prepared by the thermal conversion of CPX Form II.15 In this regard, it is worthy to note that the T1(1H) relaxation time of CPX Form I (T1(1H)=3.1 s) spontaneously formed in high-CPX loaded MBFs is considerably shorter than that of the polymorphic Form II (T1(1H)=58.1 s) used as the precursor for preparing these films (Figure 6). This finding can be explained by the released dynamics of the solvent molecules (2-aminoethanol) that are incorporated in the crystal lattice of Form I. According to the literature data15, hydrogen bonding between aminoethanol molecules and CPX are considerably weaker in the polymorphic Form I in comparison to the strength of these hydrogen bonds in the polymorphic Form II. Consequently, the 1H spinlattice relaxation processes are more efficient. 1

relative intensities

T1( H) relaxation CPX1.8/PEO (cryst) EUD CPX-I

1 0,75

PEO (cryst) CPX-II

0,5

CPX1.8/PEO (cryst)

0,25 0 0

5

10

15

20

time, s

25

1

T1ρρ( H) relaxation relative intensities

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 22 of 42

1

CPX-I 0,75

CPX1.8/PEO (cryst)

CPX-II

0,5

EUD 0,25

CPX1.8/PEO (cryst)

PEO (cryst)

0 0

Figure 6.

13

5

10

15

20

25

time, ms

C-detected T1(1H) and T1ρ(1H) spin-lattice relaxation dependences recorded for

neat CPX Form I, CPX Form II, EUD, crystalline fraction of PEO and both components of MBF CPX1.8/PEO (the relaxation curve of PEO is marked as CPX1.8/PEO; whereas the relaxation curve of CPX is marked as CPX1.8/PEO). 22 ACS Paragon Plus Environment

Page 23 of 42

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


From the data summarized in Table 3, it is clear that the T1(1H) and T1ρ(1H) relaxation times of CPX, EUD and PEO components in the prepared low-CPX-loaded systems are completely equilibrated, indicating homogeneity of these systems at the scale of tens of nanometres. Distinctly different relaxation behaviour, however, was found for the high-CPX loaded systems. The obtained very similar relaxation times T1(1H)=2.0-2.5 s indicate extensive mixing and tight interactions between all components in the prepared high-CPX-loaded systems. Bearing in mind the T1(1H) relaxation times of CPX Form I (3.1 s), crystalline fraction of PEO (7.2 s), EUD (1.2) and amorphous fraction of PEO (3.7 s), the considerable shortening of the relaxation times observed for the high-CPX-loaded systems reflects multiple-component spin-diffusion in which not only crystalline domains of CPX Form I and PEO but also domains of solvated amorphous phase of PEO are involved. However, as a result of incomplete 1H polarization transfer, the T1ρ(1H) relaxation times of CPX and PEO remained different. The T1ρ(1H) of the crystalline PEO matrix in the high-CPX MBFs was short (ca. 0.3 ms) and nearly identical to that of crystalline fraction of neat PEO, whereas the T1ρ(1H) of CPX Form I was considerably longer (ca. 60-70 ms). These findings reflect the nanoheterogeneous character of the high-CPX-loaded MBFs. Calculating the maximum diffusive path length (L) revealed that the CPX domains were not larger than ca. 100-120 nm, indicating that CPX had been incorporated into the interlamellar space PEO matrix with spherulitic morphology.30,31

2D 1H-13C HETCOR experiments: drug-polymer interactions The small size of CPX crystallites found in the high-CPX-loaded MBFs indicates a relatively large interfacial area. This large area provides the opportunity to explore intercomponent CPX-PEO interactions using 1H-13C HETCOR spectroscopy. However, we are balancing on the physical limits of this experimental technique. Moreover, due to the fast



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 24 of 42

T1ρ(13C) spin relaxation, the detection of crystalline PEO magnetization in 1H-13C FSLG HETCOR spectra requires the application of short cross-polarization times no longer than ca. 200 µs. Consequently, 1H-13C polarization transfer can be probed at relatively short 1H-13C distances, typically less than 2.5 Å.42 With these specifications, the intra-domain 1H-1H spin diffusion is limited and molecules from the interfacial area are predominantly involved in the intercomponent polarization transfer.

a)

13 OL

PEO crystal

2

4

14 OL

7 6

ppm 0

CH2

5

HPEO-CPEO HCPX-CPEO

H4’-C4’

CH3 OCH2

H6’’-C6’’ H13-C13

H2’-C2’

Ar

H2-C13

H2’’-C2’’ 10

H6’-C6’

H12-C12

H6’-C2’

H4’’-C4’’

CPX1.8/PEO

tmix=150 µs

15 120 110 100

PEO amorph

b) ppm

CH G

CH2 G

90

80

70

13 OL

60

50

14 OL

40

30

7

20

10

ppm

6

0

CH2 CH3 OCH2

5

Ar 10

NOH

CPX1.8/PEO

tmix=2000 µs

15 75

70

65

60

55

50

45

40

35

30

25

20

15

ppm

Figure 7. 1H-13C FSLG HETCOR NMR spectra of the CPX1.8/PEO film recorded with crosspolarization mixing times, tmix=150 and 2000 µs; (a) and (b), respectively.

Such a situation was demonstrated in the 1H-13C FSLG HETCOR spectrum of the CPX1.8/PEO film recorded at 150 µs of cross-polarization mixing time (Figure 7a). In this spectrum, strong correlation signals, such as H4CPX-C4CPX, H6CPX-C6CPX or H13OL-C13OL, 24 ACS Paragon Plus Environment

Page 25 of 42

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


indicate one-bond polarization transfer, whereas weaker signals (e.g., H6CPX-C2CPX) reflect medium-range transfer over at least 2.5 Å. According to published data8,43,44,45, the predicted 1

H resonance frequency of crystalline PEO is ca. 3.2-3.8 ppm. Due to interference with

dipolar decoupling, this expected one-bond correlation signal of PEO (HPEO-CPEO) was broadened and weak but was still clearly detectable at 3.5 ppm. Therefore, the secondary PEO correlation signal found at 4.2 ppm (in the 1H resonance frequency) can potentially be interpreted as a medium-range correlation signal reflecting polarization transfer from protons of the CH2-N and CH2-O groups of the 2-aminoethanol molecules. The 2-aminoethanol molecules released due to the transformation of CPX:olamine Form II (1:1) to the polymorphic Form I (2:1) can act as surfactants15 covering the surface of crystallites. These surfaces offer numerous opportunities for hydrogen bonding and can mediate intercomponent interactions, thereby allowing for tight contact between PEO CH2-O segments and CPX crystallites. As cross-polarization mixing time increases, 1H magnetization is transferred over longer distances. However, as the

13

C magnetization of the crystalline PEO is rapidly destroyed

during long cross-polarization periods, the narrow one-bond and long-range correlation signals at δ(1H)=4.5 and 9.0 ppm detected in the 1H-13C FSLG HETCOR spectrum (Figure 7b) reflect dipolar interactions involving residual amorphous PEO. The same 1H resonance frequencies were also recorded for the one-bond and long-range correlations of glycerol. This magnetization equilibration indicates that the amorphous PEO and glycerol molecules were intimately mixed. As only rigid components can be involved in the cross-polarization process, we conclude that the detected signals must reflect the fraction of glycerol molecules that are considerably immobilized by incorporation into the polymer matrix. In contrast, no longrange correlation signals with CPX molecules were found. These findings suggest that



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 26 of 42

amorphous PEO and glycerol molecules formed a single liquid-like phase that preferentially filled the voids between PEO spherulites.

Mechanical properties of CPX/PEO MBFs Ideal buccal films should exhibit adequate flexibility, elasticity, softness and robust mucoadhesive strength and be able to withstand movement and activity in the buccal cavity. Moreover, an ideal buccal film should exhibit a thickness between 50 and 1000 µm to provide adequate bioadhesion. In our case, the prepared low-CPX-loaded MBF samples (CPX0.2/PEO) were thinner (ca. 340 µm) than those with a high CPX content (CPX0.2/PEO; ca. 410 µm; Table 4). The bilayer films with EUD were additionally thickened by approximately 100 µm. In all cases, however, the dimensions of the prepared MBFs were suitable for buccal mucosal treatment without generating feelings of interference.46

Table 4. Mechanical properties of prepared MBFs: thickness (l, µm); stress at break (σ, MPa); E modulus (E, MPa); elongation at break (ε, %); and penetration force (F, N). Work (W, mJ) and deformation (d, mm) were determined by the penetration test. Tensile test

Penetration test

l MBFs (µm)

σ

E

ε

(MPa) (MPa) (%)

F (N)

W

d

(mJ) (mm)

CPX0.2/PEO

339

0.45

10.6

38

0.61 5.98

2.7

CPX0.2/PEO/EUD

473

0.62

13.4

77

0.57 3.78

1.4

CPX1.8/PEO

408

0.42

4.9

129 0.53 4.54

1.9

CPX1.8/PEO/EUD

498

0.58

8.3

156 0.66 4.19

1.5


Page 27 of 42

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


Generally, a suitable buccal film should exhibit high tensile strength and elongation at break and low elastic modulus. Moreover, strain at break is considered to be an indicator of the overall mechanical quality of the film. In this respect, the penetration and tensile tests showed rather neutral effects of EUD on the quality of CPX/PEO films. As required, stress and elongation at break increased considerably in the presence of EUD. On the other hand, these positive effects were partly compensated by the decrease in deformation under penetration and the slight increase in E modulus. Surprisingly, the most significant positive effect was attributed to the action of CPX crystallites because the fully crystalline high-CPX-loaded films exhibited considerably enhanced plasticity regardless of the presence of EUD. The tensile strain of the CPX1.8/PEO system increased to nearly 330% of that of the original CPX0.2/PEO system. We suggest that, when directly grown into the interlamellar space of PEO spherulites, the nanosized crystallites of CPX possess a sufficiently large surface area for noncovalent interfacial interactions to bind both components together and enhance the resistance of the systems against external stress. The interfacial hydrogen bonding between PEO chains and 2-aminoethanol molecules then hinders the slipping of the CPX crystallites from the interlamellar space of PEO, thereby increasing the maximum deformation at break. A similar phenomenon was previously observed in PEO nanocomposites modified by layered nanoparticles.32

Swelling test and in vitro residence time Adequate swelling is an essential property for uniform and prolonged drug release and for effective film mucoadhesion. Shortly after the start of swelling, adhesion occurs, but the adhesive strength is not very high. Thus, from a practical perspective, solid bioadhesive formulations should begin swelling within 5-15 min after contact with the mucus, and swelling should continue over a prolonged period of time.47



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 28 of 42

Table 5. In vitro residence time (ti, min); maximum swelling index (SImax, %), and time when SImax was achieved (t(SImax), min). ti

SImax

(min)

(%)

t(SImax) (min)

CPX0.2/PEO

27 ±2

81 ± 9

20

CPX0.2/PEO/EUD

33 ± 2

148 ± 16

75

CPX1.8/PEO

132 ± 10

167 ± 5

60

CPX1.8/PEO/EUD

x*

205 ± 10

120

MBFs

*)

- Phosphate buffer was absorbed into the mucoadhesive and EUD layer for a prolonged time. The weight of the film increased significantly, and the film slid down from the cellophane membrane during the test. The prepared samples began to swell extensively within 10 min. As liquid penetrated into

the film, a gel formed, the diameter of the film increased, and a distinct sol-gel boundary developed. After hydration and swelling, the PEO matrix dissolved and eroded due to its hydrophilic properties. At low CPX loading, the MBFs CPX0.2/PEO and CPX0.2/PEO/EUD with an amorphous PEO matrix had lower and less prolonged swelling compared to highCPX films with highly crystalline PEO matrixes (Table 5). High degrees of PEO crystallinity thus strongly enhanced the required swelling properties of the films. At the same time, the hydrophobic EUD further increased the maximum swelling index (SImax) and doubled the t(SImax) time. Similarly, the co-crystalline nature of the high-CPX-loaded films (CPX1.8/PEO and CPX1.8/PEO/EUD) considerably increased the in vitro residence time (Table 5). The obtained residence time of the films with low CPX content (CPX0.2/PEO and


Page 29 of 42

CPX0.2/PEO/EUD) was ca. 30 min regardless of the presence of EUD, whereas the residence time recorded for the high-CPX systems (CPX1.8/PEO) was almost five times longer and thus had superior end-use properties.

In vitro release study In vitro drug release kinetics, as a predictor of in vivo behaviour, provides critical information about dosage form behaviour and is a key parameter used to assess product safety and efficacy. In our case, the low-CPX products exhibited significantly faster drug release compared to the high-CPX films (Figure 8). For the CPX0.2/PEO and CPX1.8/PEO films, the entire drug release occurred within 90 and 360 min, respectively. The rate of drug release was further decreased by adding EUD to the films. Consequently, complete drug release was achieved within 180 and 420 min for CPX0.2/PEO/EUD and CPX1.8/PEO/EUD, respectively. 100

80

Dissolved %

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


60

CPX1.8/PEO CPX1.8/PEO/EUD CPX0.2/PEO CPX0.2/PEO/EUD

40

20

0 0

100

200

300

400

500

600

700

Time (min)

Figure 8. CPX release from the prepared mucoadhesive buccal films CPX0.2/PEO, CPX1.8/PEO, CPX0.2/PEO/EUD and CPX1.8/PEO/EUD.

The release characteristics of all of the prepared films were examined further using Korsmeyer-Peppas, Higuchi, zero-order kinetics and first-order dissolution kinetics models. 29 ACS Paragon Plus Environment


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 30 of 42

The consistency between the models and experimental data as inferred from R2 factors are summarized in Table 6. Overall, the dissolution data were more consistent with the zero-order kinetics than with the first-order kinetics model, indicating that the control mechanism of CPX release is erosion of the film rather than via Fick’s diffusion. To estimate the relative extent of the diffusion and film erosion processes during CPX release, the Korsmeyer-Peppas model was used. As reflected by the diffusion exponent n>1, erosion, which accelerated over time, is the most likely mechanism for the initial release of drugs from CPX/PEO films. In contrast, the parameter n, ranging between 0.5 and 1.0 (0.5

Molecular-Level Control of Ciclopirox Olamine Release from Poly

Recommend Documents