Subscriber access provided by University of Florida | Smathers Libraries
Article
Computational and experimental models of the gastrointestinal environment 2. Phase behavior and drug solubilisation capacity of a Type I lipid-based drug formulation after digestion Woldeamanuel A. Birru, Dallas B. Warren, Sifei Han, Hassan Benameur, Christopher J.H. Porter, Colin W. Pouton, and David K Chalmers Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.6b00887 • Publication Date (Web): 12 Dec 2016 Downloaded from http://pubs.acs.org on December 16, 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 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
Molecular Pharmaceutics
Computational models of the gastrointestinal environment 2. Phase behavior and drug solubilization capacity of a Type I lipid-based drug formulation after digestion Woldeamanuel A. Birru,2 Dallas B. Warren,2 Sifei Han,2 Hassan Benameur,4 Christopher J. H. Porter,23 1 Colin W. Pouton23* and David K. Chalmers * 1
Medicinal Chemistry; 2Drug Delivery, Disposition and Dynamics; and 3ARC Centre of Excellence in Convergent Bio-Nano Science and Technology, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Pde Parkville, Victoria 3052, Australia. 4 Capsugel Research and Development, Strasbourg, France
ABSTRACT: Lipid based drug formulations can greatly enhance the bioavailability of poorly water soluble drugs. Following the oral administration of formulations containing tri- or di-glycerides, the digestive processes occurring within the gastrointestinal (GI) tract hydrolyze the glycerides to mixtures of free fatty acids and monoglycerides that are, in turn, solubilized by bile. The behavior of drugs within the resulting colloidal mixtures is currently not well characterized. This work presents matched in vitro experimental and molecular dynamics (MD) theoretical models of the GI microenvironment containing a digested triglyceride-based (Type I) drug formulation. Both the experimental and theoretical models consist of molecular species representing bile (glycodeoxycholic acid), digested triglyceride (1:2 glyceryl-1-monooleate and oleic acid) and water. We have characterized the phase behavior of the physical system using nephelometry, dynamic light scattering and polarizing light microscopy and compared these measurements to phase behavior observed in multiple molecular dynamics (MD) simulations. Using this model microenvironment, we have investigated the dissolution of the poorly water-soluble drug danazol; experimentally, using LC-MS, and theoretically by MD simulation. The results show how the formulation lipids alter the environment of the GI tract and improve the solubility of danazol. The MD simulations successfully reproduce the experimental results showing the utility of MD in modeling the fate of drugs after digestion of lipid-based formulations within the intestinal lumen.
KEYWORDS: lipid-based formulation, digestion, poorly water soluble drug, gastrointestinal tract, phase behavior, molecular dynamics, bile, nephelometry, dynamic light scattering, polarizing light microscopy
Introduction Although there is now a widespread appreciation of the importance of good biopharmaceutical properties in drug development,1, 2 compounds having characteristics that adversely impact solubility are still highly represented among development candidates3, 4 and marketed medicines.5-7 The administration of poorly water soluble drugs (PWSDs) in a formulation containing lipids can greatly improve their bioavailability. Following the oral administration of formulations containing lipids such as tri- or di-glycerides, the process of digestion converts the glycerides to mixtures of free fatty acids and monoglycerides that are, in turn, solubilized by biliary components, leading to the formation of a range of intestinal colloidal phases including vesicles and mixed micelles.8 These digested and biliary species increase the solubilization capacity of the gastrointestinal (GI) fluids for co-administered PWSDs, creating a high surface area for drug equilibration with the aqueous phase. In this work, we extend our previous model studies of the GI environment8 and investigate the behavior of a digested lipid formulation of the poorly water soluble drug (PWSD) danazol upon dispersion within the GI medium using (MD) simulations and a complementary in vitro model system. MD is a particularly attractive approach to the study of complex
systems of this type because it provides a detailed, atomic scale view of systems that enhances understanding of experimental observations. Accordingly, MD simulations are increasingly being used to assist our understanding of the diverse aspects of physical chemistry which influence drug delivery, including studies of drugs in solid form9, 10 and as liquid formulations,11, 12 subsequent dissolution,13, 14 and membrane permeation processes.15-17 Among a variety of approaches that are available for the formulation of PWSDs,18 lipid based drug delivery systems are an increasingly utilized approach to the problem of formulating water soluble drugs. Lipid formulations are classified into five groups (Type I, II, IIIA, IIIB and IV) on the basis of the formulation composition and properties.19, 20 Type I formulations, such as the system investigated in this work, are essentially oils, consisting primarily of triglycerides or mixed glycerides19 and are the most lipophilic of the five types. These formulations form coarse emulsions on dispersion in aqueous media and, when they enter the small intestine, the ester groups of the triglyceride are quickly hydrolyzed by pancreatic lipases (Figure 1a). The phospholipid component of bile which is secreted into the intestine in response to the presence of lipids in the gut is also hydrolyzed (Figure 1b). These digestive processes promote the dispersion of the lipid phase into
ACS Paragon Plus Environment
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
vesicles and micelles.8 In contrast, Type II, IIIA/B and Type IV formulations are self-emulsifying systems that produce nanosized emulsions and micelles within the GI tract. For these systems, digestion is not essential for dispersion20, although they are likely to be affected by digestion if they contain a significant mass of tri- and di-glyceride oils, e.g. Types II, IIA. The lipidic microenvironment of the GI tract can be studied using a variety of experimental techniques that provide information about the nature of the colloidal particles present; however, detailed understanding is hindered by the large number of molecular species present, the complexity of the phase behavior and the limited resolution of the experimental methods. We have previously established that MD simulations are a powerful method for obtaining atomic level information about the phase behavior of colloidal species11, 12, 21, 22 and, when used in conjunction with appropriate model experimental systems, MD can greatly improve our understanding of the consequences of the mixing of lipids and amphiphilic molecules within the GI tract.12, 22 MD can also be utilized to study the partitioning of PWSDs within the lipidic microenvironment, the fate of the drugs on dispersion of a formulation in the GI tract12 and can reveal any changes in formulation microstructure caused by the presence of the drug. This work aims to model the behavior of glyceride-based lipid formulations within the upper GI tract using simplified chemical systems made of pure molecular components. Specifically we aim to model the fasted state, which lacks addi-
a
Page 2 of 16
tional lipids and other materials which can be expected to modify the colloidal environment of the GI tract. These model systems are designed to allow the direct comparison of in silico models obtained from MD simulations, which are limited in size and complexity, with in vitro experimental models. In the first component of this work, we use nephelometry, dynamic light scattering and molecular simulation to investigate the effect of adding a digested, glyceride-based Type I formulation into the upper GI tract which contains digested bile. The triglyceride Type I formulation is assumed to be converted by digestion into a 1:2 molar mixture of glyceryl-1-monooleate (GMO) and oleic acid (OA). Bile is a complex mixture of bile salts, phospholipid and other components which is also assumed to be digested and is modelled as a 4:1 molar ratio mixture of sodium glycodeoxycholate (GDX, Figure 2) and digested POPC (lysophosphocholine, LPC and OA, Figure 1). GDX was chosen as a representative bile salt because pure GDX has been shown to behave similarly to a physiological bile salt mixture23. In the second section of this study, we introduce a model non-electrolyte PWSD, danazol (Figure 2), to the system to observe the distribution of the drug within the lipid phase. Danazol has a logP of 3.924 and aqueous solubility of 0.58 µg/ml25 and has been extensively studied as a representative PWSD.26-34 This is the first study of quaternary phase behavior using a model system of digested lipid, digested triglyceride, bile salt and water using both experimental and MD techniques.
b
dTGL
N+
N+ O P O O
O O
O O
O
O P O O
O
HO
O
HO
O
OH
O O
O
O
OH
HO
O
O O
O O
O +
+ Enzymatic hydrolysis
Enzymatic hydrolysis
TGL
GMO
2 x OA
POPC
LPC
OA
Figure 1. (a) Lipolysis of the simple Type I formulation trioleylglyceride (TGL) produces digested triglyceride (dTGL); a 1:2 mixture of glyceryl monooleate (GMO) and oleic acid (OA). (b) The action of lipase on the phospholipid present in bile (represented here by POPC) produces lysophosphocholine (LPC) and OA.
ACS Paragon Plus Environment
Page 3 of 16
Molecular Pharmaceutics O
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
N H
OH
O
HO
H
ONa +
GDX OH
N O
Danazol Figure 2. Structures of glycodeoxycholic acid, sodium salt (GDX), used in this study as a representative bile salt, and danazol, a model non-electrolyte poorly water-soluble drug.
Materials and Methods Materials 1-Palmitoyl-2-hydroxy-sn-glycerol-3-phosphocoline (LPC) was obtained from Avanti Polar Lipids, Inc. in powder form. Glycodeoxycholic acid, sodium salt (GDX) was obtained from Calbiochem. Oleic acid (> 99% pure) and glycerol 1-monooleate were obtained from Sigma-Aldrich. Danazol (17αpregna-2,4-dien-20-yno[2,3-d]isoxazole-17-ol) was obtained from Sterling Pharmaceuticals Pty Ltd. (Sydney, Australia). Sodium hydroxide (pellets), sodium phosphate monohydrate and sodium chloride were analytical grade. All water used was obtained from a Milli-Q water purification system (Millipore). Methanol and chloroform used in this work were HPLC grade from Merck (Melbourne, Australia). Fasted Simulated Intestinal Fluids Buffer Fasted state simulated intestinal fluid buffer (FaSSIF buffer) was based on the published composition of complete FaSSIF35, without the phospholipid and bile salt components and thus was composed of; 0.174 g of NaOH, and 1.977 g of NaH2PO4.H2O and 3.093 g of NaCl in a 500 ml of purified water. The pH was adjusted to 6.50 ± 0.02 using 1 M NaOH and 1 M HCl as required. Preparation of Lipid Stock Solutions Lipid solutions were prepared using the evaporated film method; 0.200 g of lipid (LPC+OA) was dissolved in 10 ml of methanol and the methanol evaporated within a round bottom flask using a rotary evaporator. The resulting lipid film was then dispersed in 7.00 g of the aqueous phase blank buffer, generating a 2.5% w/w stock solution. This stock solution was diluted and vortex-mixed for 5 minutes to prepare 0.47 % w/w solutions as required. Preparation of Mixtures of Bile Salt and Digested Triglyceride The aqueous stock solution of 2.5% w/w GDX was prepared by dissolution of 0.200 g of GDX into 7.800 g of blank buffer. In this stock solution 0.150 g of GMO and 0.238 g of OA were dissolved to give 6.0 % w/w mixture of GMO+OA+GDX solution. This stock was again diluted to give 4.54, 3.65, 3.04, 2.42, 1.81, 1.81, 1.53, 1.30, 1.06, 0.76, 0.48, 0.42, 0.37, 0.24,
0.12, and 0.02 % w/w digested triglyceride (dTGL) solutions as required.
Turbidity The required volume of the digested lipid together with mixtures of digested triglyceride and bile salt stock solutions were pipetted and mixed in situ within the individual wells of a 96 microwell plate and the plate was then introduced into the nephelometer. The delay between mixing of the solutions and the first turbidity measurement was recorded (approximately 9 to 10 minutes). Turbidity measurements for each plate were repeated every 10 minutes until the signal reached a stable value for three measurements. All measurements were performed at 37o C and the average of three data sets was taken for each solution. The turbidity of the mixtures, measured in nephelometery turbidity units (NTUs), was monitored using a NEPHLOstar Galaxy (BMG Labtechnologies, Germany) microplate nephelometer, which measured the turbidity as a function of forward scattered light, (not light absorption). The nephelometer program settings used were: gain = 70, cycle time = 30 s, measurement time per well = 0.30 s, positioning delay = 0.5 s. The forward scattered laser light (λ = 635 nm) was monitored at an angle of 80o. 96-microwell polystyrene plates with flatbottomed wells (NUNC, Thermo Scientific, Australia) were used. The turbidity curve for the quaternary mixture was plotted as signal versus mass fraction of digested triglyceride (WdTGL). The boundary between the micelle and vesicles phases was subsequently located as the point of intersection of lines fitted to the two adjoining regions of the turbidity curve. Dynamic Light Scattering A Malvern Zetasizer Nano ZS ZEN3600 (Worcestershire, UK) was used to measure the hydrodynamic diameter of the particles within the quaternary mixtures. Measurements were conducted at 37o C using low-volume disposable sizing cuvettes (cell type ZEN0112, Sarstedt, Germany). The scattered laser light (λ = 633 nm) was monitored at an angle of 173o. The viscosity of the dispersant (water) was used as sample viscosity. The equipment was calibrated using 60 nm ± 2.7 nm and 220 ± 6 nm diameter nanosphere size standards of polystyrene polymer latex (supplied by Duke Scientific Corpora-
ACS Paragon Plus Environment
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
tion, USA) in water. The polydispersity index (PDI) for the standards was < 0.2. The required volume of the digested lipid, bile salt and digested triglyceride stock solutions were pipetted, mixed in situ within the individual cuvette and then introduced into the Zetasizer. Solutions were prepared within the region of the phase boundary previously identified using turbidity measurements. Measurement was carried out one day after sample preparation, with the average of 6 data sets taken for each solution. Dynamic light scattering (DLS) determines particle size by measuring the random changes in the intensity of light scattered from a solution of particles36, 37. Using the refractive index of the lipid, the volume of the particles and number distribution of particle sizes can be calculated from the intensity distribution, although deviation from ideality (i.e. the lipid aggregates are not composed of homogenous spheres with a well-defined refractive index) may introduce some error into the particle size estimates. In this study, since we wished to determine the concentration where the first vesicles are formed, we used the intensity distribution to analyze our data. The intensity distribution is preferred over volume and number distributions, as it is sensitive to the appearance of larger particles. When only a small number of larger particles are present, the volume and number distribution will not detect their presence.
LC-MS assays The solubility of danazol in model formulation/bile systems was measured using a LCMS 2010 system (Shimadzu, Japan) comprising an LC-20AD binary pump, a SiL-20AC refrigerated autosampler, a mobile phase vacuum degassing unit (DGU20A5) and a temperature-controlled column compartment (CTO-20A), coupled with a single quadrupole mass spectrometer (Shimadzu LCMS 2010) with an electrospray ionization source. The autosampler was maintained at 15 °C and the column at 40 °C. A Phenomenex Gemini C6-phenyl column (50 × 2.0 mm, 3 µm) was used to allow separation. Samples were eluted via gradient elution at a flow rate of 300 µl/min. The mobile phases consisted of a mixture of solvent A (95% v/v Milli-Q water:5% v/v MeOH) and solvent B (5% v/v Milli-Q water:95% v/v MeOH) both containing 1 mM ammonium formate and 0.1% formic acid. The mobile phase gradient sequence (v/v) was initiated with 60 % mobile phase B, then linearly increased to 100% from 0 to 3 min, prior to holding at 100% mobile phase B from 3 to 7 min, returning to 60% mobile phase B from 7-8 min until the end of the 13 min run time. The retention times for danazol (m/z +338.10) and progesterone (internal standard, m/z +315.10) were 4.1 min and 6.0 min, respectively. The MS conditions were as follows: drying gas flow, 10 l/min; nebulizing gas flow, 1.5 l/min; CDL 250°C, heat block 200°C; interface voltage, 4.5 kV; and detector voltage, 1.5 kV. The chromatographic data were acquired and analyzed using the LabSolution software package, 5.31.277 (Shimadzu). The sample injection volume was 10 µl. A standard solution of danazol (1.0–40.0 ng/ml) was prepared by dilution of a concentrated 1 mg/ml stock solution with acetonitrile. Linearity across the working concentrations of the drug was confirmed during each LC-MS assay using standard measures of regression. The LC-MS assay for danazol was validated by replicate (n = 5) analyses of quality control samples at low (1 ng/ml), medium (10.0 ng/ml) and
Page 4 of 16
high (40.0 ng/ml) concentrations and was found to be accurate to within ±10% of target and precise to within 10% CV.
Cross Polarized Microscopy Micrographs were recorded using a Zeiss Axiolab E microscope equipped with HFS 91 hot stage with TP 93 temperature programmer (Linkam, Surrey, UK) and Zeiss MC-80 35 mm camera (Zeiss, Oberkochen, Germany). Molecular Dynamics Simulations MD simulations were performed using GROMACS38 version 4.6. Calculations were performed using the Victorian Life Sciences Computation Initiative (VLSCI) Barcoo supercomputer system, comprising 1120 Intel Sandy Bridge compute cores running at 2.7 GHz, running the RHEL 6 operating system. Parallel scaling of sample simulations (20,000 steps) were performed to determine the optimum conditions to operate the simulations under to achieve an efficient use of CPU time; for the 15 nm box length simulations the optimum utilization was determined to be 32 CPUs. The GROMOS 53a6 united atom forcefield39 was used to represent GDX, LPC, GMO and OA. This force field is parameterized to reproduce free energies of solvation in water and cyclohexane and has been used extensively to model proteins, micelles and membranes. The cis double bond in OA was modeled using dihedral parameters developed by Barchar et al40, 41. Heavy hydrogen atoms (4 amu) were used to enable an increased time step the additional mass of the hydrogen was offset by reducing the mass of the attached heavy atom.42 Water was modeled using rigid SPC water and constrained with SETTLE43. The remaining solute bonds were constrained by the LINCS algorithm44. Isotropic periodic boundary conditions were also employed. A cut-off distance of 0.9 nm for both electrostatic and van der Waals interactions was used, and the particle-mesh Ewald (PME) method45 was used for long range electrostatic interactions. Temperature coupling used the velocity rescale algorithm46 with a reference temperature of 310 K. The production run used Parrinello-Rahman47 pressure coupling algorithms with a reference pressure of 1 bar and a compressibility of 4.5 x 10-5 bar-1 The starting models were built using the Silico script48 random_box. The required numbers of LPC+OA, dTGL (GMO+2OA), GDX and water molecules were randomly positioned within the simulation cell, resulting in systems containing 200,000 to 300,000 atoms with approximate dimensions 15 × 15 × 15 nm. The following procedure was used to set up the simulations for each system: (1) A steepest descent minimization of 500 steps. (2) A constant volume simulation of 5,000 steps with a time step of 2 fs. (3) A constant pressure simulation of 10,000 steps using Berendsen isotropic pressure coupling47 (which is more robust for systems that are far from the equilibrium state), with a coupling time constant of 0.1 and 2 fs time step. (4) A pre-production simulation of 50,000 steps and a time step 2 fs using Parrinello-Rahman pressure coupling with a pressure coupling time constant of 0.1 ps and vrescale temperature coupling with a 0.1 ps coupling constant. A production simulation of 200 ns with time step of 5 fs. The pressure reference was 1 bar for Parrinello-Rahman pressure coupling with pressure coupling time constant of 2 ps. Temperature using the v-rescale thermostat with coupling time constant of 0.1 ps and a reference temperature of 310 K corresponding to the experimental conditions used in this study.
ACS Paragon Plus Environment
Page 5 of 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
Molecular Pharmaceutics
Molecular aggregation was analyzed using the Silico script48 find_aggregate, which combines molecules into aggregates by comparing distances between carbon atoms. Two molecules are considered to be part of the same aggregate if carbon atoms are separated by a distance of less than 0.4 nm for lipid aggregates or 0.5 nm for danazol aggregates. Solvent-exposed surface areas were calculated with the GROMACS analysis tool gmx sasa. Visualization of the simulation trajectories was performed using VMD49 and images for publication were produced using PyMOL50.
Results Experimental Phase Behavior of a Model System of Digested Bile and Digested Triglyceride Formulation The in vitro experimental system used to study the addition of triglyceride formulation to the upper GI tract is composed of model digested bile, lysophospholipid (LPC), oleic acid (OA), bile salt (GDX) and digested triglyceride, a 1:2 molar ratio of GMO and OA (dTGL). The amount of digested formulation present in the system is characterized as a mass fraction of the system (WdTGL). The formation of micelle and vesicular phases was initially measured in 96-well plates using nephelometry, a high-throughput turbidity measurement method51 that readily distinguishes small and large colloidal particles. The nephelometry results were cross-validated using dynamic light scattering (DLS) measurements. We also used cross-polarized light microscopy to observe the formation of liquid crystalline phases at high concentration of digested triglyceride. In the second component of the study, we measured the solubility of danazol in the micellar and vesicular phases using liquid chromatography-mass spectrometry (LC-MS). Figure 3a shows the changes in turbidity as digested triglyceride (dTGL) is added to a physiological mixture of bile salt and digested phospholipid. The bile salt/digested phospholipid only mixture (WdTGL = 0) forms micelles and only creates a relatively small amount of forward scattering. With the increasing addition of dTGL, the forward scattering increases gradually until a point is reached near WdTGL = 0.40-0.45, where vesicles begin to form and the turbidity then increases rapidly with further dTGL addition. This point represents the phase boundary between the micellar and vesiclular/lamellar phase. The nephelometry data were corroborated using dynamic light scattering, shown in Figure 3b. The micelles present in the range of WdTGL = 0.00-0.40 were 7 to 80 nm in diameter. The vesicles that formed at high lipid content (WdTGL = 0.55-0.70) had diameters in the range 80-600 nm. The average polydispersity index (PDI) of the measurements was in the range of 0.13 – 0.39. To avoid measurement artifacts due to the presence of two or more particle populations, all results with a polydispersity index greater than 0.5 were discarded and the experiment was repeated until a moderate PDI value was achieved. However, as the samples have a population of more than one particle size, a low polydispersity index was not attainable. The large variation in the particle sizes of vesicles is a consequence of the mixtures being formed by simple agitation. Vigorous agitation or sonication would result in a more reproducible particle size, but this treatment would bear little relation to the mechanisms of mixing within the GI tract.
Figure 3. The effects of adding digested triglyceride (dTGL) to a mixture of bile salt and digested phospholipid. (a) Turbidity measurements (in nephelometry turbidity units) showing a discontinuity that delineates the boundary between micellar and vesicular/lamellar phases. (b) Particle size measurement of the system by DLS. Where more than one population of particles was evident (WdTGL ≥ 0.6) the particle sizes of the two or three populations are plotted at a single WdTGL.
Samples of LPC+OA/GDX/dTGL/H2O mixtures were assessed using a cross polarized light microscope to determine the presence of liquid crystalline phases. The presence of a lamellar liquid crystalline phase in the vesicular region was indicated by the presence of birefringence, as shown in Figures 4a and 4b. No birefringence was observed within the micellar phase region (Figure 4c).
ACS Paragon Plus Environment
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 6 of 16
Figure 5. Experimental solubility of danazol in bile salt/digested triglyceride mixtures plotted as a function of WdTGL.
Figure 4. Representative cross-polarized light micrographs (magnification 20 X) of the brightly birefringent (pink and blue colors) lamellar liquid crystal phases present at (a) WdTGL= 0.70 and (b) 0.63. No birefringence was observed in the micellar region (c) WdTGL= 0.33.
Experimental Measurement of Danazol Equilibrium Solubility The equilibrium solubility of danazol within the digested phospholipid/bile salt/digested triglyceride system was measured using LC-MS. Danazol solubility in the digested phospholipid/bile salt mixture (WdTGL = 0) and in the presence of increasing WdTGL is presented in Figure 5. In the absence of dTGL, the danazol solubility was measured to be 0.069 mg/ml. The addition of a small amount of dTGL (WdTGL = 0.0027) almost doubles the solubility to 0.113 mg/ml. The solubility then continues to increase rapidly until WdTGL = 0.0365 after which, it remains constant until the phase boundary is reached near WdTGL = 0.04, beyond which it increases more gradually to a value of 0.20 mg/ml at WdTGL = 0.71.
MD Simulation of Danazol Dispersion within a Model Digested Bile/Digested Triglyceride System Model systems containing digested bile and digested triglyceride were investigated through a set of 42 individual MD simulations of digested bile, triglyceride formulation and the PWSD danazol in various ratios. All simulations contained digested lipid and bile salt at a total concentration of approximately 5 % w/w, which is greater than found under physiological conditions, to reduce the amount of water (and total number of atoms) in the system and correspondingly the computational cost of the model. Widely used in vitro digestion models52 have a total lipid/bile salt content ranging between 0.28 (early digestion) and 0.76 % w/w (late digestion) in the fed state and 0.18 % w/w in the fasted state and, accordingly, the total bile salt/lipid content of the MD models is between 6.6 and 18 times the concentration in the fed state and 28 times that present the fasted state. Although this total lipid concentration is greater than is found under physiological conditions, the phase behavior of these colloidal systems depends principally on the molecular composition of the system and we expect the behavior at high total lipid concentration to reflect the behavior at physiological concentrations. The interaction of digested triglyceride formulation with bile was modeled using a set of simulations that varied from 0 to 40 % w/w triglyceride content (WdTGL from 0 to 0.40). We estimate that the concentration of lipid formulation within the GI tract is likely to be less than 1 % w/w; assuming that a 1 g lipid formulation capsule is dispersed into 250 cm3 of GI contents12. In the first instance, we performed a set of simulations that contained digested bile (LPC+OA and GDX) in the presence of increasing amounts of digested triglyceride (LPC+OA) (Simulation Group A, Table 1). Each molecular dynamics simulation was started from a dispersed, random arrangement of molecular constituents and was run for a simulation time of 200 ns. As each simulation progressed, the hydrophobic components rapidly aggregated to form micelles or larger structures, with the final states described in Table 1 and shown in Figure 6 and the Supporting Information. Figure 6 shows that, as the amount of dTGL is increased, the lipidic structures formed progress from simple mixed micelles to secondary
ACS Paragon Plus Environment
Page 7 of 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
Molecular Pharmaceutics
micelles, cylindrical micelles, and then finally make a transition to phase separated systems which span the simulation cell. In future studies it would be insightful to compare the MD simulations with electron microscopy images of the colloidal structures formed. The sizes of molecular aggregates formed in each simulation were determined by grouping molecules based on hydrophobic contacts. It is important to note that these simulations are limited in size and that, in many of the simulations, the sizes of the molecular aggregates are limited by the number of lipid molecules present within the system and therefore, the simulation results should be interpreted with care when only a small number of aggregates are present at the end of the simulation. In subsequent simulations we investigated the dispersion of danazol within the colloidal environment. Danazol concentrations of 2.5, 5.0, 7.5, 10.0, 15.0 and 20.0 mg/ml were added to mixtures corresponding to simulations 1, 2, 5, 8 and 10, producing simulation Groups B-F (Table 2). These danazol concentrations are greater than would be expected to occur within the GI tract but, as discussed above, constraints on the size of the model system mean that the bile salt/lipid concentration are also greater than physiological levels and we have correspondingly increased the danazol concentrations to cover danazol:bile mass ratios of 0.05 to 0.57. Figures 7a and b show the distributions of the danazol molecules on completion
of twelve selected simulations. The final frames of the complete set of thirty simulations are reported in the Supporting Information (Figures S1 to S5 and coordinate files). In all systems, danazol was found to be preferentially buried within the hydrophobic core of the colloidal structures or, if exposed to the surface, to make contact with the aqueous environment through the more polar oxazole ring or hydroxyl group. In the simulations of digested bile alone (Group B), which form small micelles, the danazol is, to a large extent, forced to reside near the colloid surface. In simulations containing large proportions of dTGL the lipid component separates to form a separate oil phase with the long alkyl chains aligning to form leaflets. In these cases, most of the danazol aligns with the alkyl chains and is buried deep within the lipid region, with only a small fraction of danazol molecules contacting the surface through the oxazole or hydroxy groups. To investigate the aggregation of the danazol molecules into clusters that might promote precipitation, we calculated the clustering of danazol molecules alone using a distance of 0.5 nm between carbon atoms as the definition of a contact (Table 2). These calculations show that, in all cases, the large majority of danazol molecules are not clustered together but instead exist as isolated molecules, although the small number of aggregates that are present increase in size with danazol concentration.
Table 1. Compositions and final simulation structures of MD simulations of model digested bile with digested triglyceride (Simulation group A). NO. of Molecules
Composition (% w/w)
GMO LPC GDX OAa Water
GMO LPC GDX OA
SIM. No. WdTGL 1
0
-
37
149
-
2
0.38
66
36
144
168 103858 1.2
107070 -
Num. agg.b
Median agg. sizeb
0.9 3.6
0.5 10
24
0.9 3.5
2.3 7
36
Structure in final state Micelles and isolated molecules Spherical and oblate micelles
c
Spherical and lamellar micelles
3
0.51
110
35
141
255 101717 1.9
0.9 3.4
3.6 4
138
4
0.63
176
34
137
386 98505
0.8 3.3
5.4 4
182c
Spherical and lamellar micelles
6.6 3
164
c
Spherical and lamellar micelles
c
Lamellar micelles
5
0.69
220
33
133
473 96363
3.1 3.9
0.8 2.2
6
0.74
286
32
129
605 93151
5.0
0.8 3.1
8.4 2
526
7
0.78
330
32
126
692 91010
5.8
0.8 3.0
9.8 2
590c
Lamellar micelles c
8
0.81
396
30
122
823 87797
7.0
0.7 2.9
11.5 1
1370
9
0.81
440
30
119
910 85656
7.7
0.7 2.9
12.7 1
1499c c
Phase separated (rod) Phase separated (rod)
10
0.87
550
28
112
1129 80303
9.7
0.7 2.7
15.7 1
1819
11
0.90
660
26
104
1347 74949
11.6 0.6 2.5
18.8 1
2137c
Phase separated
24.8 1
c
Phase separated
12
0.93
881
22
89
1563 64242
15.5 0.5 2.1
2774
a
Phase separated (plane)
Oleic acid (OA) molecules are derived from digestion of TGL (1 GMO + 2 OA) and POPC (1 LPC + 1 OA). bAggregate numbers and median values exclude any isolated molecules. cA small number of aggregates are observed and the median value will not be a reliable estimate of true aggregate size.
ACS Paragon Plus Environment
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 8 of 16
Figure 6. Final frames taken from simulations of the LPC+OA/GDX/dTGL/H2O system showing a progression from simple micelles (1-2) through spherical and lamellar micelles (3-5), lamellar micelles (6-7) and a phase separated system at (8-12) at high dTGL concentration. Atom coloring; GDX is grey, LPC is orange, OA is pink, GMO is green and oxygen atoms are red. The box indicates the periodic boundary and the scale bar length is 3.0 nm. Water atoms have been omitted.
ACS Paragon Plus Environment
Page 9 of 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
Molecular Pharmaceutics
Table 2. Compositions and danazol aggregation data for MD simulations of model digested bile, digested triglyceride and danazol (Simulation groups B-F). Parent Sim. No.
Max Danazol Agg. Size
Sim Group
13
B
1
0
15
0.05
2
2
14
B
1
0
30
0.11
1
2
15
B
1
0
45
0.16
5
4
16
B
1
0
60
0.21
5
3
17
B
1
0
90
0.32
15
6
18
B
1
0
120
0.43
17
7
19
C
2
0.38
15
0.06
1
2
20
C
2
0.38
30
0.11
2
2
21
C
2
0.38
45
0.17
5
3
22
C
2
0.38
60
0.22
8
3
23
C
2
0.38
90
0.33
11
9
24
C
2
0.38
120
0.44
21
4
25
D
5
0.69
15
0.06
1
2
26
D
5
0.69
30
0.12
2
2
27
D
5
0.69
45
0.18
5
3
28
D
5
0.69
60
0.24
1
2
29
D
5
0.69
90
0.36
6
3
30
D
5
0.69
120
0.48
16
5
31
E
8
0.81
15
0.07
1
2
32
E
8
0.81
30
0.13
0
0
33
E
8
0.81
45
0.20
3
2
34
E
8
0.81
60
0.26
4
2
34
E
8
0.81
90
0.39
5
4
36
E
8
0.81
120
0.52
19
3
37
F
10
0.87
15
0.07
0
-
38
F
10
0.87
30
0.14
0
-
39
F
10
0.87
45
0.21
1
2
40
F
10
0.87
60
0.28
7
2
41
F
10
0.87
90
0.43
8
3
42
F
10
0.87
120
0.57
10
3
WdTGL
No. Danazol Molecules
Mass ratio Dan: BS/Lipid
Sim. No.
a
Aggregate numbers exclude isolated molecules.
ACS Paragon Plus Environment
No. Danazol Agg.a
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 10 of 16
Figure 7a. Localization of danazol within the final frames of selected simulations, with changing danazol concentration. Danazol is colored yellow and the remaining molecule coloring is as described in Figure 6. The simulation numbering is described in Table 2.
ACS Paragon Plus Environment
Page 11 of 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
Molecular Pharmaceutics
Figure 7b. Localization of danazol within the final frames of selected simulations, with changing danazol concentration. Danazol is colored yellow and the remaining molecule coloring is as described in Figure 6. The simulation numbering is described in Table 2. Note that the last cell is rotated by 90° relative to the box above.
ACS Paragon Plus Environment
Molecular Pharmaceutics
Surface area (nm2)
a
GMO GDX OA
1500
LPC DAN Total
A 1000
C B
E
F
D
500
0 0
10
20
30
40
Simulation Number
b Surface area per molecule (nm2)
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 16
GMO GDX OA
4.0
B A
3.0
C
LPC DAN
D E F
2.0
1.0
0.0 0
10
20
30
40
Simulation Number
Figure 8. (a) Total and (b) per-molecule solvent exposed surface areas of simulation components for the complete set of 42 simulations. The molecular constitutions of simulations in group A are described in Table 1 and groups B-F are described in Table 2.
Interaction of the lipid components with the aqueous medium provides much of the driving force for the formation of colloidal phases. To evaluate the changes in lipid/solvent interaction, the solvent exposed surface areas of the molecular components were determined for each simulation and are shown in as totals for each molecular component (Figure 8a) and as average exposed surface area per molecule (Figure 8b). The Group A simulations contain model digested bile consisting of bile salt (GDX) and digested phospholipid (LPC+OA) with increasing amounts of digested triglyceride (GMO+2OA). Simulation 1 contains only digested bile, which forms dynamic mixed micelles. The colloid surface in this system is dominated by exposed GDX molecules, with the LPC and OA molecules sequestered in the micelle interior. On addition of dTGL to the bile solution (simulations 2-12), the OA and GMO components become progressively more surface exposed. Figure 8b shows a steady reduction in the exposed surface areas of all components as the system progresses from small mixed micelles to larger phase-separated aggregates. The simulations in groups B-F investigate the progressive addition of danazol to systems containing digested bile alone (B) and with progressive amounts of digested triglycer-
ide (C-F). Increasing the danazol concentration within each series does not greatly perturb the structure of the colloidal systems. The simulation series B-F does show that increasing amounts of dTGL reduces the exposure of the danazol molecules to the aqueous environment with an average of 1.4 nm2 exposure in digested bile (B) dropping to an average of 0.6 nm2 in simulation series F.
Discussion It is known that the bioavailability of PWSDs can be improved by coadministration with lipids, which increase drug absorption by enhancing solubilization and dissolution.53 However, the understanding of the interaction of lipid-based formulations with GI fluids and endogenous biliary lipids has been hindered by a lack of predictive formulation assessment studies. The phases that are formed between lipid-based formulations, GI fluids and endogenous biliary lipids have been described as a simple range of colloidal intestinal phases.54, 55 Although the nature of these phases have not been understood at the molecular level. It is desirable to understand the issues associated with performance of lipid-based formulations in terms of their impact on the phase behavior of the GI fluids. In
ACS Paragon Plus Environment
Page 13 of 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
Molecular Pharmaceutics
particular we seek to understand the PWSD solubilization capacity of different phases formed on digestion, the impact that this has on maintenance of the drug within solution, and whether the drug might precipitate during the dispersion or digestion processes. The ability of a lipid-based formulation to maintain drug in a solubilized state and inhibit drug precipitation is highly dependent on the nature of the formulation excipients included, and this is further complicated by the realization that the properties of the excipients can change significantly during dispersion and digestion in the GI tract.31, 50, 55-59 This work is the first study to investigate the association structures formed after digested bile makes contact with a digested LBDDS. As the concentration of the digested triglyceride is increased, the phase behavior changes from a mixed micellar structure to a more complicated mixture of vesicular/lamellar and hexagonal structures (Figure 6). In parallel, the solubilization behavior of lipophilic drugs in the presence of digested simple triglyceride lipid formulations is also a function of the concentration of the digested triglyceride. The nature of the colloidal phases produced upon digestion of the formulation lipids is therefore linked to their ability to solubilize the drug (Figure 5). As the mass fraction of digested triglyceride within the model GI lumen system increases, a phase transition from smaller, predominately spherical, micelles to larger vesicular/multilamellar structures occurs as shown by the turbidity and particle size measurements (Figure 3) and the cross polarized light microscopy results (Figure 4) The MD results (Figure 6) correlate well with these experimental studies, showing the systems spontaneously organizing into micellar structures at low digested triglyceride concentrations. At intermediate concentrations, hexagonal phase type structures form. Then, at higher digested triglyceride concentrations, phase separation occurs (WdTGL = 0.87 to 0.90 - systems 10 and 11), as shown in Figure 6. The investigation found that danzol solubility increases with the concentration of digested triglyceride (Figure 5) which is in agreement with published reports that found that solubility of the related compound, hydrocortisone, increased with increasing concentrations of caprylic, lauric and oleic acids.60, 61 This increase in solublization capacity seen here in vitro was supported by the MD calculations (Figure 7). This indicates that more aggregation of danazol occurs with lower dTGL concentrations (which may represent crystallization). This was supported by the solubility measurements of danazol, which showed that when more digested triglyceride was present, a larger mass of danazol was solubilized. Note that some studies of the fate of PWSDs after digestion of a lipid formulation have suggested that digestion may lead to a loss of drug solubilization capacity, and a subsequent decrease in bioavailability.50, 58, 60, 62, 63 This effect occurs when the solvent capacity of the formulation outweighs that of the aqueous digestion products. The MD models used in this study provide information about the localization of the lipophilic drugs within the mixture of the formulation-lipid and the GI fluids, and is indicative of the dynamic processes, such as aggregation, that will result in poor solubilization properties; shown in Figures 8a and b. Images detailing the localization of danazol within the final frames of all simulation are available in the Supporting Information. This study provides a foundation for the in silico design of lipid formulations, to further the understanding of
the behavior of a wider range of drugs in different lipid formulations and their fate after they are released into the GI lumen.
Conclusion This study explores in detail the impact of digested triglyceride on the phase behavior of the GI tract (modeled using a mixture of digested phospholipid and bile salt) using both experimental and MD simulation techniques. Additionally, it investigates the impact of digested triglyceride on the solubilization of the poorly water-soluble drug danazol. The results suggest that knowledge of the impact of formulation excipients on the phase behavior of the GI tract, in particular the digested state, and drug solubility may give us a better understanding of the performance of lipid-based formulations and help in optimization of the concentration of different excipients in lipid-formulations. The positive correlation between in vitro and in silico results provides confidence in the use MD simulations to model more complex events that occur with dispersion and digestion of the formulation in the GI tract and explores phenomena in atomistic detail. Moreover, this study suggests that MD can be used as a prediction tool to model the fate of poorly water-soluble drugs in the GI lumen after they are released from the oral dose form within the stomach.
Associated content Supporting Information The supporting information contains additional figures showing localization of drugs within the final frames of all simulation and mean square displacement of LPC, OA, GDX, GMO, danazol and water. PDB files containing the final frames of the MD simulations are also included. This material is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION Corresponding Authors * E-mail:
[email protected],
[email protected] Acknowledgment WAB acknowledges the support of a PhD scholarship funded by ARC Linkage Grant LP120100600 awarded to Monash University in collaboration with Capsugel. We would also like to acknowledge the CPU time and technical support provided by the Victorian Life Sciences Computation Initiative (VLSCI) through grant VR0004, MASSIVE, and the Australian National Computational Infrastructure (NCI) through grant y96.
Abbreviations dTGL, digested triglyceride; DLS, Dynamic light scattering; FaSSIF, fasted state simulated intestinal fluid; GDX, glycodeoxycholic acid, sodium salt; GI, gastrointestinal; GMO, glycerol 1mono-oleate; HLB, hydrophilic-lipophilic balance; LBDDS, lipid based drug delivery system; LC-MS, liquid chromatography and mass spectrometry; LFCS, Lipid Formulation Classification System; LPC, 1-palmitoyl-2-hydroxy-sn-glycerol-3-phosphocoline; MD, molecular dynamics; NTU, nephelometry turbidity unit; OA, oleic acid; PWSD, poorly-water soluble drug.
References 1. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev 1997, 23, (1-3), 3-25.
ACS Paragon Plus Environment
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
2. Lipinski, C. A. Drug-like properties and the causes of poor solubility and poor permeability. J Pharmacol Toxicol 2000, 44, (1), 235-249. 3. Broccatelli, F.; Cruciani, G.; Benet, L. Z.; Oprea, T. I. BDDCS Class Prediction for New Molecular Entities. Mol Pharm 2012, 9, (3), 570-580. 4. Leeson, P. D.; Young, R. J. Molecular Property Design: Does Everyone Get It? Acs Med Chem Lett 2015, 6, (7), 722-725. 5. Wenlock, M. C.; Austin, R. P.; Barton, P.; Davis, A. M.; Leeson, P. D. A Comparison of Physiochemical Property Profiles of Development and Marketed Oral Drugs. . J Med Chem 2003, 46, 250–1256. 6. Takagi, T.; Ramachandran, C.; Bermejo, M.; Yamashita; Yu, S.; X., L.; Amidon, G. L. A. Provisional Biopharmaceutical Classification of the Top 200 Oral Drug Products in the United States, Great Britain, Spain, and Japan. . Mol Pharm 2006, (3), 631–643. 7. Benet, L. Z.; Broccatelli, F.; Oprea, T. I. BDDCS Applied to Over 900 Drugs. Aaps J 2011, 13, (4), 519-547. 8. Birru, W. A.; Warren, D. B.; Headey, S. J.; Benameur, H.; Porter, C. J. H.; Pouton, C. W.; Chalmers, D. K. Computational models of the intestinal environment 1. The effect of digestion and pH changes on the phase behaviour of intestinal fluids; molecular dynamics simulations and NMR studies. Submitted. 9. Xiang, T. X.; Anderson, B. D. Molecular Dynamics Simulation of Amorphous Indomethacin. Mol Pharm 2013, 10, (1), 102-114. 10. Gao, Y.; Olsen, K. W. Drug-Polymer Interactions at Water-Crystal Interfaces and Implications for Crystallization Inhibition: Molecular Dynamics Simulations of Amphiphilic Block Copolymer Interactions with Tolazamide Crystals. J Pharm Sci 2015, 104, (7), 2132-2141. 11. Warren, D. B.; Chalmers, D. K.; Pouton, C. W. Structure and dynamics of glyceride lipid formulations, with propylene glycol and water. Mol Pharm 2009, 6, (2), 604-14. 12. Warren, D. B.; King, D.; Benameur, H.; Pouton, C. W.; Chalmers, D. K. Glyceride Lipid Formulations: Molecular Dynamics Modeling of Phase Behavior During Dispersion and Molecular Interactions Between Drugs and Excipients. Pharm Res 2013. 13. Gao, Y.; Olsen, K. W. Molecular Dynamics of Drug Crystal Dissolution: Simulation of Acetaminophen Form I in Water. Mol Pharm 2013, 10, (3), 905-917. 14. Ru, Q.; Fadda, H. M.; Li, C.; Paul, D.; Khaw, P. T.; Brocchini, S.; Zloh, M. Molecular Dynamic Simulations of Ocular Tablet Dissolution. J Chem Inf Model 2013, 53, (11), 3000-3008. 15. Tejwani, R. W.; Davis, M. E.; Anderson, B. D.; Stouch, T. R. An Atomic and Molecular View of the Depth Dependence of the Free Energies of Solute Transfer from Water into Lipid Bilayers. Mol Pharm 2011, 8, (6), 2204-2215. 16. Awoonor-Williams, E.; Rowley, C. N. Molecular Simulation of Nonfacilitated Membrane Permeation. . Biochim Biophys Acta 2016, 1672–1687. 17. Milanetti, E.; Raimondo, D.; Tramontano, A. Prediction of the Permeability of Neutral Drugs Inferred From Their Solvation Properties. Bioinformatics 2016, 32, 1163–1169. 18. Williams, H. D.; Trevaskis, N. L.; Charman, S. A.; Shanker, R. M.; Charman, W. N.; Pouton, C. W.; Porter, C. J. H. Strategies to Address Low Drug Solubility in Discovery and Development. Pharmacol Rev 2013, 65, (1), 315-499. 19. Pouton, C. W. Lipid formulations for oral administration of drugs: non-emulsifying, self-emulsifying and 'selfmicroemulsifying' drug delivery systems. Eur J Pharm Sci 2000, 11, S93-S98. 20. Pouton, C. W. Formulation of poorly water-soluble drugs for oral administration: Physicochemical and physiological issues and the lipid formulation classification system. Eur J Pharm Sci 2006, 29, (3-4), 278-287. 21. Warren, D. B.; Chalmers, D. K.; Hutchison, K.; Dang, W.; Pouton, C. W. Molecular dynamics simulations of spontaneous aqueous phase bile salt aggregation. Colloid Surf A 2006, 280, 182193.
Page 14 of 16
22. King, D. T.; Warren, D. B.; Pouton, C. W.; Chalmers, D. K. Using Molecular Dynamics to Study Liquid Phase Behavior: Simulations of the Ternary Sodium Laurate/Sodium Oleate/Water System. Langmuir 2011, 27, (18), 11381-11393. 23. Birru, W. A.; Warren, D. B.; Ibrahim, A.; Williams, H. D.; Benameur, H.; Porter, C. J. H.; Chalmers, D. K.; Pouton, C. W. Digestion of Phospholipids after Secretion of Bile into the Duodenum Changes the Phase Behavior of Bile Components. Mol Pharm 2014. 24. Erlich, L.; Yu, D.; Pallister, D. A.; Levinson, R. S.; Gole, D. G.; Wilkinson, P. A.; Erlich, R. E.; Reeve, L. E.; Viegas, T. X. Relative bioavailability of danazol in dogs from liquid-filled hard gelatin capsules. Int J Pharm 1999, 179, (1), 49-53. 25. Alsenz, J.; Meister, E.; Haenel, E. Development of a partially automated solubility screening (PASS) assay for early drug development. J Pharm Sci 2007, 96, (7), 1748-1762. 26. Larsen, A.; Holm, R.; Pedersen, M. L.; Mullertz, A. Lipidbased Formulations for Danazol Containing a Digestible Surfactant, Labrafil M2125CS: In Vivo Bioavailability and Dynamic In Vitro Lipolysis. Pharm Res 2008, 25, (12), 2769-2777. 27. Clarysse, S.; Psachoulias, D.; Brouwers, J.; Tack, J.; Annaert, P.; Duchateau, G.; Reppas, C.; Augustijns, P. Postprandial Changes in Solubilizing Capacity of Human Intestinal Fluids for BCS Class II Drugs. Pharm Res 2009, 26, (6), 1456-1466. 28. Söderlind, E.; Karlsson, E.; Carlsson, A.; Kong, R.; Lenz, A.; Lindborg, S.; Sheng, J. J. Simulating fasted human intestinal fluids: understanding the roles of lecithin and bile acids. Mol Pharm 2010, 7, (5), 1498-507. 29. Bevernage, J.; Forier, T.; Brouwers, J.; Tack, J.; Annaert, P.; Augustijns, P. Excipient-Mediated Supersaturation Stabilization in Human Intestinal Fluids. Mol Pharm 2011, 8, (2), 564-570. 30. Clarysse, S.; Brouwers, J.; Tack, J.; Annaert, P.; Augustijns, P. Intestinal drug solubility estimation based on simulated intestinal fluids: Comparison with solubility in human intestinal fluids. Eur J Pharm Sci 2011, 43, (4), 260-269. 31. Anby, M. U.; Williams, H. D.; McIntosh, M.; Benameur, H.; Edwards, G. A.; Pouton, C. W.; Porter, C. J. H. Lipid Digestion as a Trigger for Supersaturation: Evaluation of the Impact of Supersaturation Stabilization on the in Vitro and in Vivo Performance of Self-Emulsifying Drug Delivery Systems. Mol Pharm 2012, 9, (7), 2063-2079. 32. Devraj, R.; Williams, H. D.; Warren, D. B.; Mohsin, K.; Porter, C. J.; Pouton, C. W. In vitro assessment of drug-free and fenofibrate-containing lipid formulations using dispersion and digestion testing gives detailed insights into the likely fate of formulations in the intestine. Eur J Pharm Sci 2013, 49, (4), 748-60. 33. Devalapally, H.; Silchenko, S.; Zhou, F.; McDade, J.; Goloverda, G.; Owen, A.; Hidalgo, I. J. Evaluation of a Nanoemulsion Formulation Strategy for Oral Bioavailability Enhancement of Danazol in Rats and Dogs. J Pharm Sci 2013, 102, (10), 3808-3815. 34. Devalapally, H.; Zhou, F.; McDade, J.; Goloverda, G.; Owen, A.; Hidalgo, I. J.; Silchenko, S. Optimization of PEGylated nanoemulsions for improved pharmacokinetics of BCS class II compounds. Drug Deliv 2015, 22, (4), 467-474. 35. Marques, M. Dissolution Media Simulating Fasted and Fed States. Dissolut Technol 2004, 11, (2), 16. 36. Chu, B.; Liu, T. B. Characterization of nanoparticles by scattering techniques. J Nanopart Res 2000, 2, (1), 29-41. 37. Pecora, R. Dynamic light scattering measurement of nanometer particles in liquids. J Nanopart Res 2000, 2, (2), 123-131. 38. Pronk, S.; Páll, S.; Schulz, R.; Larsson, P.; Bjelkmar, P.; Apostolov, R.; Shirts, M. R.; Smith, J. C.; Kasson, P. M.; van der Spoel, D.; Hess, B.; Lindahl, E. GROMACS 4.5: a high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics 2013, 29, (7), 845-854. 39. Oostenbrink, C.; Villa, A.; Mark, A. E.; Van Gunsteren, W. F. A biomolecular force field based on the free enthalpy of hydration and solvation: The GROMOS force-field parameter sets 53A5 and 53A6. J Comput Chem 2004, 25, (13), 1656-1676. 40. Martinez-Seara, H.; Rog, T.; Karttunen, M.; Reigada, R.; Vattulainen, I. Influence of cis double-bond parametrization on lipid
ACS Paragon Plus Environment
Page 15 of 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
Molecular Pharmaceutics
membrane properties: how seemingly insignificant details in forcefield change even qualitative trends. J Chem Phys 2008, 129, (10), 105103. 41. Bachar, M.; Brunelle, P.; Tieleman, D. P.; Rauk, A. Molecular dynamics simulation of a polyunsaturated lipid bilayer susceptible to lipid peroxidation. J Phys Chem B 2004, 108, (22), 7170-7179. 42. Feenstra, K. A.; Hess, B.; Berendsen, H. J. C. Improving efficiency of large time-scale molecular dynamics simulations of hydrogen-rich systems. J Comput Chem 1999, 20, (8), 786-798. 43. Miyamoto, S.; Kollman, P. A. Settle - an Analytical Version of the Shake and Rattle Algorithm for Rigid Water Models. J Comput Chem 1992, 13, (8), 952-962. 44. Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M. LINCS: A linear constraint solver for molecular simulations. J Comput Chem 1997, 18, (12), 1463-1472. 45. Darden, T.; York, D.; Pedersen, L. Particle Mesh Ewald an N.Log(N) Method for Ewald Sums in Large Systems. J Chem Phys 1993, 98, (12), 10089-10092. 46. Bussi, G.; Donadio, D.; Parrinello, M. Canonical sampling through velocity rescaling. J Chem Phys 2007, 126, (1). 47. Berendsen, H. J. C. P., J. P. M.; Vangunsteren, W. F.; Dinola, A.; Haak, J. R. Molecular Dynamics with Coupling to an External Bath. J Chem Phys 1984, 81, (8), 3684-3690. 48. Chalmers, D. K.; Roberts, B. P. Silico: A Perl Molecular Toolkit (http://silico.sourceforge.net ), 2010. 49. Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual molecular dynamics. J Mol Graphics 1996, 14, (1), 33-38. 50. Kaukonen, A. M.; Boyd, B. J.; Porter, C. J. H.; Charman, W. N. Drug solubilization behavior during in vitro digestion of simple triglyceride lipid solution formulations. Pharm Res 2004, 21, (2), 245-253. 51. Hoelke, B.; Gieringer, S.; Arlt, M.; Saal, C. Comparison of nephelometric, UV-spectroscopic, and HPLC methods for highthroughput determination of aqueous drug solubility in microtiter plates. Anal Chem 2009, 81, (8), 3165-72. 52. Marques, M. R. C.; Loebenberg, R.; Almukainzi, M. Simulated Biological Fluids with Possible Application in Dissolution Testing. Dissolut Technol 2011, 18, (3), 15-28. 53. Humberstone, A. J.; Charman, W. N. Lipid-based vehicles for the oral delivery of poorly water soluble drugs. Adv Drug Deliv Rev 1997, 25, (1), 103-128. 54. Staggers, J. E.; Hernell, O.; Stafford, R. J.; Carey, M. C. Physical-chemical behavior of dietary and biliary lipids during intestinal digestion and absorption. 1. Phase behavior and aggregation states of model lipid systems patterned after aqueous duodenal contents of healthy adult human beings. Biochemistry 1990, 29, (8), 2028-40.
55. Armand, M.; Borel, P.; Dubois, C.; Senft, M.; Peyrot, J.; Salducci, J.; Lafont, H.; Lairon, D. Physicochemical Properties of Emulsions and Lipolysis of Dietary-Fat in the Human Stomach. Gastroenterology 1993, 104, (4), A608-A608. 56. Hernell, O.; Staggers, J. E.; Carey, M. C. Physicalchemical behavior of dietary and biliary lipids during intestinal digestion and absorption. 2. Phase analysis and aggregation states of luminal lipids during duodenal fat digestion in healthy adult human beings. Biochemistry 1990, 29, (8), 2041-56. 57. Kaukonen, A. M.; Boyd, B. J.; Charman, W. N.; Porter, C. J. H. Drug solubilization behavior during in vitro digestion of suspension formulations of poorly water-soluble drugs in triglyceride lipids. Pharm Res 2004, 21, (2), 254-260. 58. Porter, C. J. H.; Kaukonen, A. M.; Boyd, B. J.; Edwards, G. A.; Charman, W. N. Susceptibility to lipase-mediated digestion reduces the oral bioavailability of danazol after administration as a medium-chain lipid-based microemulsion formulation. Pharm Res 2004, 21, (8), 1405-1412. 59. Porter, C. J. H.; Kaukonen, A. M.; Taillardat-Bertschinger, A.; Boyd, B. J.; O'Connor, J. M.; Edwards, G. A.; Charman, W. N. Use of in vitro lipid digestion data to explain the in vivo performance of triglyceride-based oral lipid formulations of poorly water-soluble drugs: Studies with halofantrine. J Pharm Sci 2004, 93, (5), 11101121. 60. Kossena, G. A.; Charman, W. N.; Boyd, B. J.; Dunstan, D. E.; Porter, C. J. Probing drug solubilization patterns in the gastrointestinal tract after administration of lipid-based delivery systems: a phase diagram approach. J Pharm Sci 2004, 93, (2), 33248. 61. Kossena, G. A.; Boyd, B. J.; Porter, C. J.; Charman, W. N. Separation and characterization of the colloidal phases produced on digestion of common formulation lipids and assessment of their impact on the apparent solubility of selected poorly water-soluble drugs. J Pharm Sci 2003, 92, (3), 634-48. 62. Cuine, J. F.; McEvoy, C. L.; Charman, W. N.; Pouton, C. W.; Edwards, G. A.; Benameur, H.; Porter, C. J. H. Evaluation of the impact of surfactant digestion on the bioavailability of danazol after oral administration of lipidic self-emulsifying formulations to dogs. J Pharm Sci 2008, 97, (2), 995-1012. 63. Cuine, J. F.; Charman, W. N.; Pouton, C. W.; Edwards, G. A.; Porter, C. J. H. Increasing the proportional content of surfactant (Cremophor EL) relative to lipid in self-emulsifying lipid-based formulations of danazol reduces oral bioavailability in beagle dogs. Pharm Res 2007, 24, (4), 748-757.
ACS Paragon Plus Environment
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 16 of 16
Table of Contents Artwork
16
ACS Paragon Plus Environment