Experimental Determination of Drug Diffusion ... - ACS Publications

Subscriber access provided by MT ROYAL COLLEGE

Article

Experimental determination of drug diffusion coefficients in unstirred aqueous environments by temporally resolved concentration measurements Massimiliano Pio di Cagno, Fabrizio Clarelli, Jon Våbenø, Christina Lesley, Sokar Darsim Rahman, Jennifer Cauzzo, Erica Franceschinis, Nicola Realdon, and Paul C. Stein Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b01053 • Publication Date (Web): 20 Feb 2018 Downloaded from http://pubs.acs.org on February 28, 2018

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

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 25 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

1

Experimental determination of drug diffusion coefficients in unstirred aqueous

2

environments by temporally resolved concentration measurements

3 4

Massimiliano Pio di Cagnoa,*, Fabrizio Clarellib, Jon Våbenøc, Christina Lesleyd, Sokar

5

Darsim Rahmand, Jennifer Cauzzoa,e, Erica Franceschinise, Nicola Realdone and Paul C.

6

Steind

7 8

a

9

Arctic University of Norway, Tromsø, Norway

Drug Transport and Delivery Research Group, Department of Pharmacy, University of Tromsø, The

10

b

11

University of Norway, Tromsø, Norway

12

c

13

d

14

Denmark

15

e

Computational Pharmacology Group, Department of Pharmacy, University of Tromsø, The Arctic

Helgeland Hospital Trust, Sandnessjøen, Norway Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Odense,

Department of Pharmaceutical Sciences, University of Padova, Padova, Italy

16 17 18

*Corresponding author: Massimiliano Pio di Cagno; Tel: +47 776 45301; e-mail:

19

[email protected]

20

1 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

21

Table of contents graphics

22


Page 2 of 25

Page 3 of 25 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


23

Abstract

24

The diffusion coefficient (also known as diffusivity) of an active pharmaceutical ingredient

25

(API) is a fundamental physicochemical parameter that affects passive diffusion through

26

biological barriers and, as a consequence, bioavailability and biodistribution. However, this

27

parameter is often neglected, and it is quite difficult to find diffusion coefficients of small

28

molecules of pharmaceutical relevance in the literature. The available methods to measure

29

diffusion coefficients of drugs all suffer from limitations that range from poor sensitivity to

30

high selectivity of the measurements or the need for dedicated instrumentation. In this work, a

31

simple but reliable method based on time resolved concentration measurements by UV-visible

32

spectroscopy in an unstirred aqueous environment was developed. This method is based on

33

spectroscopic measurement of the variation of the local concentration of a substance during

34

spontaneous migration of molecules, followed by standard mathematical treatment of the data

35

in order to solve Fick’s law of diffusion. This method is extremely sensitive and results in

36

highly reproducible data. The technique was also employed to verify the influence of the

37

environmental characteristics (i.e. ionic strength and presence of complexing agents) on the

38

diffusivity. The method can be employed in any research laboratory equipped with a standard

39

UV-visible spectrophotometer, and could become a useful and straightforward tool in order to

40

characterize diffusion coefficients in physiological conditions and help to better understand

41

the drug permeability process.

42

43

Keywords: Diffusion coefficients, UV-visible spectroscopy, Fick’s laws of diffusion, data

44

fitting, molecular modelling, ion strength

45



Page 4 of 25

46

1. Introduction

47

Diffusion is the spontaneous migration of molecules of a substance from regions of higher

48

chemical activity (high concentration) to regions of lower chemical activity (low

49

concentration) through one or several isotropic environments. This process is of capital

50

importance in pharmaceutics since passive transport of active pharmaceutical ingredients

51

(APIs) remains a significant (and often primary) mechanism of biological membrane

52

permeation (1) and therefore the diffusion process is crucial for reaching an appropriate

53

bioavailability. According to Stokes-Einstein equation (Eq. 1), the diffusivity D (cm2/sec) is

54

directly proportional to the absolute temperature T (K) and Boltzmann constant Kb

55

(cm2*g/s2*K) but inversely proportional to the viscosity of the media η (g/cm*sec) and the

56

hydrodynamic radius r (cm) of the molecules.

=

6

Equation 1

57

The relation between diffusion and flux of a drug substance in aqueous environment can be

58

described by Fick’s first law as:

= −

Equation 2

59

Where j represents the net flux (mol/cm2*sec) of drug substance through the media, dc/dx

60

represents the concentration gradient and D is the diffusion coefficient. D is an essential

61

parameter to understand pharmaceutically relevant processes such as mass transport through

62

barriers (i.e. absorption) and bioavailability of APIs. However, pharmaceutical research has

63

for years overlooked the importance of this parameter in favor of the measurement of the

64

apparent permeability coefficient (Papp, 2-4). Since water is the primary component of the

65

human body, a proper quantification of D values for APIs should be of crucial importance for

66

a proper evaluation of the permeability process. Reliable D values, together with a proper

67

evaluation of partition coefficients (LogP), could help to elucidate the role of unstirred water 4 ACS Paragon Plus Environment

Page 5 of 25 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


68

layers (UWL) (5) and interface transitions in passive transport through barriers (6). The

69

scientific literature is significantly lacking in diffusivity values for APIs in aqueous

70

environments, and methods to measure them in a simple but reliable manner. Currently,

71

several techniques for measuring diffusion coefficients are available, including holographic

72

laser interferometry (HLI (7)), electron speckle pattern interferometry (ESPI (8)), mass

73

transfer through barrier studies (9), Taylor dispersion analysis (TDA (10,11)), nuclear

74

magnetic resonance (NMR (12)), dual-focus fluorescence correlation spectroscopy (FCA

75

(13)) and fluorescence recovery after photobleaching (FRAP (14)) just to mention the most

76

employed ones. Just a few of these techniques (TDA, NMR and mass transfer studies) have

77

been applied to estimate diffusivity of APIs, the reason being that all the techniques listed

78

above show limitations when facing pharmaceutically relevant problems such as poor

79

sensitivity (e.g. NMR), high specificity (e.g. FRAP) and the need for dedicated

80

instrumentation (e.g. HLI, ESPI, FCA, TDA and NMR which are not standard

81

instrumentations in most pharmaceutical research laboratories).

82

Another approach to estimate diffusivities of small molecules in aqueous environment is to

83

predict them from molecular properties such as molecular volume, molar volume or by

84

molecular simulation (15). However, all these in silico approaches are prone to uncertainty

85

and do not provide a complete interpretation of diffusion in the case of environmental changes

86

such as ion strength or presence of complexing agents. In this work, a simple, versatile and

87

easily implemented spectroscopic method (16, 17) for diffusion coefficient measurement is

88

introduced. This versatile method is based on the measurement of variation of local

89

concentration of a compound in aqueous media and subsequent temporal resolution of

90

diffusion equation by means of analytical as well as numerical mathematical methods. By

91

comparing different approaches to the solution of Fick’s diffusion equation we show that

92

simple non-linear least-squares fitting is sufficient to extract the diffusion coefficient. The



93

experiments are performed employing a regular quartz cuvette and a standard UV-visible

94

spectrometer. This method is simple, reliable, reproducible and extremely sensitive.

95

Moreover, it has the significant advantage that it could be employed in any research

96

laboratory equipped with a standard UV-visible spectrophotometer and very small changes

97

related to environmental alterations can be measured.

98

2. Experimental section

99

2.1 Materials

100

Sodium-di-hydrogen-phosphate monohydrate (NaH2PO4·H2O), di-sodium hydrogen-

101

phosphate dodeca-hydrate (Na2HPO4·12H2O), sodium chloride (NaCl), sodium hydroxide

102

(NaOH), hydrochloric acid (HCl, ≥ 37%), caffeine, calcein, chloramphenicol, ketoprofen,

103

nitrofurantoin, tetracycline hydrochloride, trimethoprim, vancomycin, and β-cyclodextrin

104

(βCD) were all purchased from Sigma Aldrich Chemie GmbH (Steinheim, Germany).

105

Paracetamol was purchased from Norsk Medisinaldepot AS (Oslo, Norway). Penicillin G

106

employed was USP Reference Standard (Timbrook, USA). All chemicals were of analytical

107

grade.

108

2.2. UV–visible localized spectroscopy

109

2.2.1 Preparation of the solutions

110

Phosphate buffer saline (PBS) 73 mM was prepared mixing a solution of NaH2PO4·H2O

111

(2.2% W/V) with a solution of Na2HPO4·12H2O (1.8 % W/V) in a ratio 1 to 5. The pH of the

112

PBS was adjusted to 7.3-7.4 (pH meter Lab 744, Metrohm AG, Herisau, Switzerland) by the

113

addition of NaOH solid pellets. Tonicity of the PBS solution was adjusted to 280-290 mOsm

114

(Semi-Micro Osmometer K-7400, Knauer, Berlin, Germany) by the addition of NaCl. Each of

115

the compounds was than adequately weighted and dissolved in the PBS in order to achieve a

116

concentration of approximately 0.5-1.5 mM. The molecular weights of the compounds and the

117

characteristics of the donor solutions (compound concentrations, buffer concentrations,


Page 6 of 25

Page 7 of 25 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


118

solution densities, wavelengths of maximal absorption (λMAX) and specific absorptivities (ε))

119

are reported in Table 1.

120

Table 1. General characteristics and composition of the aqueous solutions used as donor

121

environments. The diffusional medium was composed of plain distilled water (DW, density of

122

1.0015 g/mL). Each of the compounds was detected at its maximum wavelength of absorption

123

(λMAX) and the local concentration was calculated using its specific absorptivity (ε). Nominal conc. (mM)

Buffer conc. (mM)

Density*

λMAX

ε

name

Molecular Weight (g/mol)

(g/mL)

(nm)

(cm2/µmol)

DW

Caffeine

194.2

1.23

72.8

1.0086

269

11.3

DW

Calcein

622.6

1.23

72.8

1.0083

294

11.3

DW

Chloramphenicol

323.2

1.20

72.8

1.0095

277

10.0

DW

Ketoprofen

254.3

1.17

72.8

1.0064

257

18.4

DW

Nitrofurantoin

238.2

0.77

72.8

1.0113

274

13.7

DW

Paracetamol

151.2

1.42

72.8

1.0079

236

13.2

DW

Penicillin G

334.4

0.60

72.8

1.0071

207

19.9

DW

Tetracycline

444.4

0.83

72.8

1.0197

361

14.8

DW

Trimethoprim

290.3

1.19

72.8

1.0085

278

5.6

DW Vancomycin 1449.3 1.20 72.8 *Data are reported as mean of three measurements. SD below 0.5%

1.0011

280

6.7

124 125

Solutions of βCD (used in the diffusion experiments to investigate complexation effects) were

126

prepared by adding solid powder of the dextrin derivative (3.4-4 mg) directly into 2 mL of

127

ketoprofen neutral and acidic solution (pH 1.0, obtained by mixing 1.9 mL of neutral

128

ketoprofen solution plus 0.1 mL of HCl) in order to achieve a molar ratio drug:βCD of

129

approximately 1:1.

130

2.2.2 Experimental procedure

131

For the spectrophotometric measurements, a double array VWR (VWR International, Radnor,

132

USA) UV-visible spectrophotometer (model UV-6300 PC) equipped with a Hellma®

133

Suprasil® (Sigma-Aldrich) quartz absorption cuvettes (chamber volume of 700 µL and path

134

length of 10 mm) was employed. The experimental setup is schematized in Fig. 1.

Diffusional medium

Compound



135 136

Figure 1. Schematic representation of localized UV-visible spectroscopy experimental setup.

137

Both reference and sample cuvette were filled with 675 µl of distilled water and placed in the

138

respective compartment of the spectrophotometer. At time (t) = 0 sec (starting of the

139

experiment), 25 µL of drug solution was gently injected in the bottom of the sample cuvette

140

by a needle syringe. The reference and sample cuvette were both sealed with parafilm in order

141

to avoid evaporation of the solvent. Absorbance readings were recorded at fixed wavelength

142

(Table 1) at regular time intervals (120 sec) for 24 hours at room temperature (23-24°C).

143

Absorbance was recorded at a specific position (x1) corresponding to 0.51 cm from the bottom

144

of the cuvette (Figure 2). This was obtained by keeping the sample cuvette lifted 0.60 cm

145

(Hard Head calliper, Jula Norge AS, Lørenskog, Norway) from the bottom of the cuvette

146

holder.


Page 8 of 25

Page 9 of 25 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


147 148

Figure 2. Adaptation of the mono-dimensional diffusional model used to the cuvette

149

geometry and related relevant spatial points (initial point x0, measurement points x1 and x2

150

and total length of the diffusion chamber L).

151

For each of the API studied, two independent experiments were conducted (n=2) and the final

152

diffusional profile (employed for the mathematical data treatment and diffusivity calculations)

153

was obtained by averaging the respective absorbances at each time point. The average

154

variance (in percentage, %) between the absorbance registered at each time point between the

155

two parallel experiments was used to estimate the experimental error (≤ 5 %).

156

2.2.3 Mathematical model and UV-visible spectroscopy data interpretation

157

The mathematical approach used in this work to fit the experimental data obtained by UV-

158

visible spectroscopy to calculate the diffusion coefficients was based on the standard diffusion

159

equation (Eq. 3) and well-known classical physical models (9, 18). Due to the geometry of the

160

experimental setup (Figure 2), the diffusion process was approximated with a one-

161

dimensional model. The diffusion equation (Eq. 3) describes the concentration of a solute

162

over time and space in a homogenous medium (assuming constant diffusivity) as:



, , =

Page 10 of 25

Equation 3

163

Where c represents the concentration of the substance, t the time, x the position and D the

164

diffusivity. In theory, any x between 0.51 cm and 1.11 cm can be selected (Fig. 2) by lifting

165

the cuvette from its standard recording position. In this work, the recording point was

166

primarily set to 0.51 cm (see experimental procedure section).

167

The first mathematical approach used to find numerical solution of D was analytical. This

168

approach is based on the assumption that for times t, t

Experimental Determination of Drug Diffusion ... - ACS Publications

Recommend Documents