Regional Intestinal Permeability in Rats: A Comparison of Methods

Subscriber access provided by FLORIDA INTL UNIV

Article

Regional intestinal permeability in rats: a comparison of methods Carl Roos, David Dahlgren, E. Sjögren, Christer Tannergren, Bertil Abrahamsson, and H. Lennernas Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b00279 • Publication Date (Web): 18 Sep 2017 Downloaded from http://pubs.acs.org on September 19, 2017

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 35

Molecular Pharmaceutics Roos, C.

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

1

Regional intestinal permeability in rats: a comparison between methods.

1 (29)

Abstract

2 3

Currently, the screening of new drug candidates for intestinal permeation is typically based on in vitro

4

models which give no information regarding regional differences along the gut. When evaluation of

5

intestinal permeability by region is undertaken, two preclinical rat models are commonly used, the

6

Ussing chamber method and single-pass intestinal perfusion (SPIP). To investigate the robustness of in

7

vivo predictions of human intestinal permeability, a set of four model compounds was systematically

8

investigated in both these models, using tissue specimens and segments from the jejunum, ileum, and

9

colon of rats from the same genetic strain. The influence of luminal pH was also determined at two pH

10

levels. Ketoprofen had high and enalaprilat had low effective (Peff) and apparent (Papp) permeability in

11

all three regions and at both pH levels. Metoprolol had high Peff in all regions and at both pHs and high

12

Papp at both pHs and in all regions except the jejunum, where Papp was low. Atenolol had low Peff in all

13

regions and at both pHs, but had high Papp at pH 6.5 and low Papp at pH 7.4. There were good

14

correlations between these rat in situ Peff (SPIP) and human in vivo Peff determined previously for the

15

same compounds by both intestinal perfusion of the jejunum and regional intestinal dosing. The results

16

of this study indicate that both investigated models are suitable for determining the regional

17

permeability of the intestine; however, the SPIP model seems to be the more robust and accurate

18

regional permeability model.

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 2 of 35

2 (29)

19

Introduction

20

It is common to administer pharmaceutical products orally once daily as it is safe, convenient, and

21

associated with improved patient compliance 1, 2. To accommodate this, many active pharmaceutical

22

ingredients (APIs) are formulated as oral modified-release (MR) formulations. Currently, the majority

23

of candidate drugs permeate well through the intestinal membranes but are poorly water soluble (class

24

II according to the biopharmaceutical classification system; BCS) 3-5. Several of these APIs have been

25

formulated by various approaches into innovative pharmaceutical products and are consequently

26

protected by intellectual proprietary rights (IPR). Drugs designed to be MR orally administered

27

products need biopharmaceutical properties that will result in sufficiently high absorption in both the

28

small and large intestines to cause the desired therapeutic result 6, 7. In general, more innovative

29

formulation work and longer periods of development are required to turn BCS II drugs into successful

30

commercial MR pharmaceutical products with adequate bioavailability 4, 8, 9. In addition, a better

31

understanding of the dynamic and variable physiology of the gastrointestinal (GI) tract, especially the

32

permeability to APIs at different intestinal sites, is needed to define the rate-controlling GI absorption

33

processes and to predict the performance of novel MR oral formulations 10. Drug permeation is

34

currently commonly evaluated in in vitro screening models (e.g. Caco-2 cell studies and artificial

35

membrane assays such as PAMPA), which give no information regarding regional intestinal

36

absorption. In order to predict the regional intestinal absorption of candidate drugs, site-specific

37

permeability measurements performed in pre-clinical animal models are needed. Furthermore, at an

38

industrial level, there is a need to switch from an empirical development approach to more rational

39

strategies 7. In particular, it is important to explore the in vivo assessment of regional intestinal

40

permeability in humans compared to common in vitro and in vivo pre-clinical models, which will

41

subsequently form a basis for robust, reliable translation approaches and predictive models 6, 11-14.

42

Identifying the rate-limiting step for absorption along the intestine may also be crucial from an IPR

43

perspective as it may justify the patent claims for a specialized, patent-protected release mechanism

44

that governs the absorption rate of an API.

45


Page 3 of 35


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


3 (29)

46

Two commonly used rat models for evaluating the regional intestinal permeability to drugs, nutrients,

47

disease markers, and formulations are the Ussing chamber system and the in situ single-pass intestinal

48

perfusion (SPIP) model. The Ussing chamber method has been used to assess regional differences in

49

transport rates in several species 15-20. This in vitro transport model has been described in detail

50

elsewhere but, briefly, it is based on a vessel with two chambers filled with media (e.g. buffer with and

51

without the API), which are separated by a membrane (e.g. excised intestine or a synthetic membrane)

52

15

53

investigation is determined over a specified time and an apparent intestinal permeability (Papp) value is

54

calculated for the API from its appearance rate in the receiver chamber.

. The absorptive and/or efflux transport from the donor chamber across the barrier under

55 56

The in situ SPIP method is more complex and physiologically relevant than the Ussing chamber

57

method, because intact blood flow, local endocrine conditions and membrane integrity are maintained.

58

In addition, blood sampling enables accurate and more sensitive pharmacokinetic assessment of the

59

absorption rate, permeability, and first-pass extraction 21, 22. In this method, a segment of the rat

60

intestine is perfused (single-pass) with a drug solution and the effective permeability (Peff) is

61

determined based on the difference between the drug concentration entering and that leaving the

62

segment, by applying the parallel tube model 22, 23. It is common to add established absorptive marker

63

compounds, such as metoprolol, to account for intra- and interlaboratory variability 24-26. These two

64

methods have previously been compared with respect to human Peff, but only for jejunal segments and

65

jejunal tissue specimens, which means that further investigations are warranted with regard to the

66

other intestinal regions 27.

67 68

Four model drugs known to be transported across the intestinal membranes by passive lipoidal

69

diffusion were investigated in this study: atenolol, enalaprilat, ketoprofen, and metoprolol; their

70

physicochemical properties are described in Table 1. Atenolol and metoprolol are both basic drugs

71

(pKa 9.6 and 9.7, respectively) and β1-receptor antagonists that are used to treat e.g. hypertension 28, 29.

72

Metoprolol is considered by some to be a marker for the cut-off point between BCS high- and low-

73

permeability drugs 30. This has, however, been challenged, as the permeability may change along the



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 35

4 (29)

74

small intestine for some drugs, even if metoprolol is not affected specifically 3, 31, 32. Enalaprilat, the

75

active metabolite of the ACE inhibitor enalapril, which is used for the treatment of hypertension, is an

76

ampholyte (basic pKa 3.17 and acidic pKa 7.84) 29, 33. Ketoprofen is a nonsteriodal anti-inflammatory

77

drug (NSAID) with acidic properties (pKa 3.89) 29. Atenolol and enalaprilat are classified as BCS

78

class III drugs while ketoprofen and metoprolol are classified as BCS class II and BCS class I drugs,

79

respectively 3, 34. Ketoprofen is, however, readily soluble in the intestinal lumen as it is completely

80

charged at all normal intestinal pH values.

81 82

The primary objective of this study was to determine the permeability of the jejunum, ileum and colon

83

to a set of model drugs, using two intestinal transport methods, in the same laboratory, and to compare

84

the results with corresponding human regional intestinal permeability data. A unique aspect of this

85

study was that all the animals used, in both the Ussing chamber and the SPIP model, were of the same

86

genetic rat strain (male Wistar Han rats, strain 273), were from the same breeder, and had the same

87

diet. Furthermore all experiments were conducted in the same laboratory. The secondary objective was

88

to investigate the influence of different luminal (donor side) buffers with different pHs on the rate of

89

intestinal transport in bothmethods.

90


Page 5 of 35


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


5 (29)

91

Methods

92

Drugs and other chemicals

93

Atenolol and metoprolol tartrate were provided by AstraZeneca AB (Mölndal, Sweden). Enalaprilat,

94

ketoprofen, phenol red, sodium pyruvate, sodium fumarate dibasic, L-glutamic acid, D-glucose,

95

sodium bicarbonate (NaHCO3), magnesium sulfate heptahydrate (MgSO4 • 7H2O), potassium chloride

96

(KCl), and calcium chloride (CaCl2) were purchased from Sigma-Aldrich (St. Louis, MO, US).

97

Sodium phosphate dibasic dihydrate (Na2HPO4 • 2H2O), potassium dihydrogen phosphate (KH2PO4)

98

and sodium chloride (NaCl) were purchased from Merck KGaA (Darmstadt, Germany). Water used in

99

the buffer was purified in an ELGA Maxima Prima USF system (Elga Labwater, Lane End,

100

Buckinghamshire, UK) in the Ussing experiments and a Millipore Milli-Q Advantage A10 system

101

(Millipore Corporation, Billerica, MA) in the SPIP experiments.

102 103

Buffer and perfusate preparations

104

Ussing chamber experiments

105

Buffers with pHs of 6.5 (buffer A) and 7.4 (buffer B) were prepared for use as donor chamber media.

106

Buffer B was used as receiver chamber medium in all experiments. Buffer A consisted of 6.3 g NaCl,

107

0.35 g KCl, 0.30 g MgSO4 • 7H2O, 0.32 g Na2HPO4 • 2H2O, 0.082 g KH2PO4, 1.3 g NaHCO3, 0.86 g

108

sodium fumarate dibasic, 0.83 g L-glutamic acid, 0.54 g sodium pyruvate, 2.1 g D-glucose and 0.69

109

mg CaCl2 per liter. Buffer B consisted of 6.3 g NaCl, 0.35 g KCl, 0.30 g MgSO4 • 7H2O, 0.19 g

110

Na2HPO4 • 2H2O, 0.25 g KH2PO4, 1.3 g NaHCO3, 0.86 g sodium fumarate dibasic, 0.83 g L-glutamic

111

acid, 0.54 g sodium pyruvate, 2.1 g D-glucose, 5.25 ml 2M HCl and 0.69 mg CaCl2 per liter.

112

Atenolol, metoprolol, ketoprofen and enalaprilat were added from DMSO stock solutions to the donor

113

chamber medium (mucosal side) to achieve a final concentration of 25 µM for each drug. The final

114

concentration of DMSO never exceeded 0.5% (w/w), which has been validated previously with respect

115

to intestinal tissue integrity 35.

116 117

Single-pass intestinal perfusion (SPIP)



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 35

6 (29)

118

Three buffers, I, II, and III, were prepared for the rat SPIP experiments. Buffer I was of low buffer

119

strength (6.7 mM) at pH 6.5, buffer II was of high buffer strength (67 mM) at pH 6.5, and buffer III

120

was of high buffer strength (100 mM) at pH 7.4. Buffer I consisted of 0.59 g KH2PO4, 0.41 g

121

Na2HPO4 • H2O, and 8 g NaCl per liter. Buffer II consisted of 6.35 g KH2PO4, 3.56 g Na2HPO4 • H2O,

122

and 5.00 g NaCl per liter. Buffer III consisted of 1.81 g KH2PO4, 9.50 g Na2HPO4 • H2O, and 4 g NaCl

123

per liter. Prior to the single-pass perfusion of each segment, atenolol, metoprolol, ketoprofen, or

124

enalaprilat were added from DMSO stocks to each buffer to achieve a final perfusate concentration of

125

100 µM. Phenol red was added as a fluid flux marker, to a final perfusate concentration of 25 µM. The

126

final DMSO concentration in the buffer never exceeded 0.5% (w/w), which has been validated

127

previously with respect to intestinal tissue integrity 35.

128 129 130

Animals and study design

131

This study was approved by the local ethics committee for animal research (no: 66-2014 for SPIP and

132

no: 48-2013 for Ussing chamber) in Gothenburg, Sweden. Male Wistar Han rats from Charles River

133

(strain 273), aged 8-10 weeks, were used in all experiments. The animals arrived at the animal lab

134

facility at least one week prior to the experiment and were allowed water and food ad libitum prior to

135

the experiments. They were kept in a 12-hour light/dark cycle, at 21°C and 50% relative humidity.

136

Animals of the same genetic strain were used in both intestinal transport models.

137 138

Ussing method

139

Male Wistar Han rats weighing 250-350g were anesthetized using Attane isoflurane (Piramal

140

Healthcare, Hallbergmoos, Germany). The abdominal cavity was opened and specimens from the

141

jejunum, ileum and colon were identified and carefully removed. Any debris was removed by rinsing

142

with ice-cold Krebs-Ringer bicarbonate buffer (KRB). The intestinal specimens were placed in a

143

beaker containing ice-cold KRB, which was bubbled continuously with carbogen gas (95% O2/5%

144

CO2). The specimens used for the jejunum were excised 10-30 cm distal to the ligament of Treitz, the


Page 7 of 35


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


7 (29)

145

ileal specimens were excised 5-25 cm proximal to the ileocecal junction, and the colonic specimens

146

were excised 2-7 cm proximal to the rectum. The specimens were cut into 2 cm pieces and placed in a

147

preparation basin containing cold KRB which was bubbled continuously with carbogen gas. Before

148

mounting the specimens in the Ussing chamber, they were cut open along the mesenteric border. The

149

exposed tissue area for drug permeation in the Ussing chamber setup was 1.14 cm2 36. Specimens

150

containing Peyer’s patches were identified by visual inspection and avoided. 25°C KRB was added

151

simultaneously to both the mucosal and serosal sides and the intestinal specimen was allowed to

152

equilibrate for 20 minutes while being warmed to 37°C by heating jackets. The viability of the tissue

153

during the experiment was monitored by measuring the potential difference (PD) and electrical

154

resistance (R) across the specimen. Small intestinal specimens with initial PD and R values below 4

155

mV and 30 Ohm • cm2, respectively, were replaced with another specimen prior to the start of the

156

experiment. The corresponding PD and R cut-off values for large intestinal specimens were 6 mV and

157

70 Ohm • cm2, respectively. At time zero, the KRB in the chambers was replaced with 37°C KRB

158

containing the drug mix on the mucosal (donor) side and 37°C KRB on the serosal (receiver) side.

159

Throughout the course of the experiment there was stirring present in both the mucosal and the serosal

160

chambers, by form of rotors operating at approximately 300 rpm. 200 µl samples were taken from the

161

mucosal side at 0 minutes and 150 minutes, and from the serosal side at 0, 30, 60, 90, 120 and 150

162

minutes for drug quantification. Equal volumes of fresh KRB at 37°C was added to compensate for the

163

sample volume taken. Samples were immediately frozen and stored at -20 °C awaiting analysis.

164 165


166

Male Wistar Han rats weighing approximately 270-370g were anesthetized using Attane isoflurane

167

(Piramal Healthcare, Hallbergmoos, Germany) and placed on a heating table, to maintain their body

168

temperature at 37°C. The abdomen was opened by a 3-5 cm longitudinal incision along the midline.

169

An intestinal segment of 6-12 cm was located and cannulated with polypropylene tubing (O.D. 4 mm,

170

I.D. 2mm). The jejunal segments were located approximately 10 cm distal to the ligament of Treitz,

171

the ileal segments approximately 10 cm proximal to the ileocecal junction, and the colonic segments

172

approximately 1 cm distal to the cecum. Each segment was carefully rinsed with 20-30 ml 25°C saline



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 35

8 (29)

173

solution for 1-2 minutes to remove nonadherent mucus and debris, until a clear outlet perfusate was

174

attained. The intestinal segments were placed into the abdominal cavity along with a section of ~10 cm

175

of the tubing to ensure that the perfusate was at body temperature before entering the cannulated

176

intestinal segment. The animals were sutured, to minimize heat and fluid loss, leaving the inlet and

177

outlet tubing accessible from the outside. A 50 ml syringe (Becton Dickinson, Franklin Lakes, NJ)

178

containing the perfusate drug mix was attached to the inlet tubing and mounted in a syringe pump

179

(Perfusor® Space, Braun Melsungen AG, Germany). At time 0 a 5 ml bolus dose of 25 °C perfusate

180

was infused (10 ml/min) to fill the entire segment. The subsequent perfusion rate was set at 0.2

181

ml/min. The experiment lasted for 105 minutes, with perfusate outlet samples quantitatively collected

182

at 45, 60, 75, 90 and 105 minutes. The samples were immediately frozen and stored at -20 °C awaiting

183

analysis.

184 185

Analytical method

186

Sample preparation

187

Ussing experiments

188

The samples were thawed at 25°C while being shaken at 500 rpm. When completely thawed, the

189

Ussing donor-side samples were diluted 10-fold with KBR. 30 µl of the diluted donor-side samples

190

were then transferred to a 96-well plate and mixed with 420 µl KBR and 150 µl internal standard

191

solution [acetonitrile (ACN):H2O 60:40, 200 nM warfarin]. 60 µl of the receiver-side samples were

192

transferred to a 96-well plate and mixed with 20 µl internal standard solution. The plate was shaken

193

for 20 minutes at 1000 rpm, and then centrifuged at 3500 rpm at 4°C for 20 min prior to analysis on

194

UPLC-MS.

195 196


197

Perfusate samples were thawed at 25°C, while being shaken at 500 rpm. When the samples were

198

completely thawed, 30 µl of sample was taken to a 96-well plate and mixed with 420 µl KBR and 150

199

µl internal standard solution (ACN:H2O 60:40, 200 nM warfarin). The 96-well plate was shaken for 20


Page 9 of 35


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


9 (29)

200

minutes at 1000 rpm, and then centrifuged at 3500 rpm at 4°C for 20 min prior to analysis with the

201

UPLC-MS.

202 203

UPLC-MS

204

Samples from the Ussing chamber and SPIP experiments were analyzed on a UPLC I-Class binary

205

solvent manager, with a Waters Acquity flow through a needle sample manager (Waters Corporation,

206

Milford, MA). Chromatographic separation was achieved using a Waters BEH-C18 column (I.D. 2.1

207

mm, length: 50 mm, particle size 1.7 µm) and a pre-column (VanGuard™ HSS T3 2.1 x 5 mm id,

208

particle size 1.8µm, Waters Corporation, Milford, MA) kept at 60 ˚C. The mobile phase consisted of

209

(A) 95% water, 5% ACN, 0.1% formic acid (FA) and (B) ACN, 0.1% FA. A gradient was run as

210

follows: initially 5% B, 5-95% B for 1.5 min, 95% B for 0.5 min, and 95-5% B for 0.2 min. The total

211

run time was 2.2 min, the flow rate was 600 µL/min, and the injection volume was 5 µL for all

212

samples. All compounds were quantified simultaneously in the same chromatographic run, on a QDa

213

Mass Detector (Waters Corporation, Milford, MA) in selected ion recording (SIR) positive

214

electrospray ionization mode. The cone voltage was 10 kV, the source block temperature was 120°C,

215

and the source probe temperature was 600°C. The ratio of analyte to internal standard as a function of

216

the concentration of the analyte was calibrated using a linear curve fit model, with a weighing of 1/x2

217

for all compounds. The calibration curve was linear in the range 10-300 nM for all compounds. The

218

limit of quantification (LOQ) for all compounds was 10 nM, with a relative standard deviation of

219

Regional Intestinal Permeability in Rats: A Comparison of Methods

Recommend Documents