Article pubs.acs.org/molecularpharmaceutics
Interaction with Mixed Micelles in the Intestine Attenuates the Permeation Enhancing Potential of Alkyl-Maltosides Kerstin Gradauer,†,‡ Ayano Nishiumi,† Kota Unrinin,† Haruki Higashino,† Makoto Kataoka,† Betty L. Pedersen,‡ Stephen T. Buckley,‡ and Shinji Yamashita*,† †
Faculty of Pharmaceutical Sciences, Setsunan University, Osaka 573-0101, Japan Global Research, Novo Nordisk A/S, DK-2760 Måløv, Denmark
‡
ABSTRACT: The purpose of the present study was to investigate the interaction of intestinal permeation enhancers with lipid and surfactant components present in the milieu of the small intestine. Maltosides of different chain lengths (decyl-, dodecyl-, and tetradecyl-maltoside; DM, DDM, TDM, respectively) were used as examples of nonionic, surfactant-like permeation enhancers, and their effect on the permeation of FD4 across Caco-2 monolayers was monitored. To mimic the environment of the small intestine, modified versions of fasted and fed state simulated intestinal fluid (FaSSIFmod, FeSSIFmod6.5, respectively) were used in addition to standard transport media (TM). Compared to the buffer control, 0.5 mM DDM led to a 200-fold permeation enhancement of FD4 in TM. However, this was dramatically decreased in FaSSIFmod, where a concentration of 5 mM DDM was necessary in order to elicit a moderate, 4-fold, permeation enhancement. Its capacity to promote permeation was diminished further when FeSSIFmod6.5 was employed. Even when cells were exposed to a concentration of 5 mM, no significant permeation enhancement of FD4 was observed. Analogous effects were observed in the case of DM and TDM, with slight deviations on account of differences in their critical micelle concentration (CMC). This observation was corroborated by calculating the amount of maltoside monomer versus micellar bound maltoside in FaSSIFmod and FeSSIFmod6.5, which demonstrated a reduced amount of free monomer in these fluids. To evaluate the in vivo significance of our findings, DDM solutions in TM, FaSSIFmod, and FeSSIFmod6.5 were used for closed intestinal loop studies in rats. Consistent with the results found in in vitro permeation studies, these investigations illustrated the overwhelming impact of sodium taurocholate/lecithin micelles on the permeation enhancing effect of DDM. While DDM led to a 20-fold increase in FD4 bioavailability when it was applied in TM, no significant permeation enhancement was seen in FaSSIFmod/FeSSIFmod6.5. Collectively, these investigations highlight the importance of using biorelevant media when evaluating the potency of permeation enhancers. In doing so, this ensures improved correlations between in vitro and in vivo studies and thus enables an early and more accurate assessment of promising permeation enhancers. KEYWORDS: maltoside, intestinal absorption, Caco-2, permeation enhancement, in vitro−in vivo correlation (IVIVC), biorelevant media
1. INTRODUCTION The oral route of drug delivery is often considered to be the most preferable by virtue of the fact that it represents an easy and painless route of administration, thus leading to high patient compliance. However, not all drugs show sufficient oral bioavailability. Poor oral absorption of drugs is caused by low solubility in the gastrointestinal (GI) fluid, insufficient permeation through the GI membrane, and/or limited stability (chemical/biological) in the GI tract or liver. One of the most extensively studied approaches to address the insufficient permeability of drugs is to combine the API with a permeation enhancer. This approach can facilitate increased uptake by altering the cell membrane, enhancing the paracellular or the transcellular permeation, or forming a noncovalent complex with the API.1−4 Maltosides, which were employed in this study, are a group of water-soluble, nonionic detergents that demonstrate high enhancer potency and low toxicity. Given their surfactant-like nature, it seems reasonable © 2015 American Chemical Society
to expect them to give rise to membrane perturbation effects to some extent;5,6 however, it has been demonstrated that at lower concentrations maltosides primarily promote absorption via the paracellular route by modulating tight junctions.7,8 Despite extensive investigations on permeation enhancers over the past decades, no oral product has yet reached the market.4 Conceivably, this may be attributed to a lack of efficacy in vivo, even though in vitro studies may suggest that an enhancer has high potency. In order to improve the predictability in screening permeation enhancers at an early research stage, in vitro studies need to be carried out under more biorelevant conditions. The media composition in a cellbased permeation assay may play a pivotal role in terms of Received: Revised: Accepted: Published: 2245
November 21, 2014 February 16, 2015 April 15, 2015 April 15, 2015 DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics Table 1. Overview of All Conditions Investigated within this Studya TM
FaSSIFmod
FeSSIFmod6.5
DM
DDM
TDM
DM
DDM
TDM
DM
DDM
TDM
− − − − − − 0.5 − − 5 − − 0.5 5
0.08 0.12 0.16 0.2 0.25 0.35 0.5 − − − − − 0.5 −
− − − − − − 0.5 − − − − − 0.5 −
− − − − − − − − − 5 − − 0.5 5
− − − − − − 0.5 2 3.5 5 0.08 0.16 0.5 5
− − − − − − 0.5 2 3.5 5 − − 0.5 5
− − − − − − − − − 5 − − 0.5 5
− − − − − − − − − 5 − − 0.5 5
− − − − − − − − − 5 − − 0.5 5
(A) Permeation of FD4 through a Caco-2 monolayer
(B) Permeation of maltoside monomer through a dialysis membrane
a
The D/P system was used to monitor both the permeability of FD4 through Caco-2 monolayers in the presence of different maltosides (A) as well as the interaction of maltosides with FaSSIFmod and FeSSIFmod6.5, where free and micellar bound maltosides were separated using a dialysis membrane (B). All values depict the concentration of enhancer in the apical compartment of the D/P system (mM).
Louis, MO). Nonessential amino acids (10 mM), fetal bovine serum (FBS), L-glutamine (200 mM), trypsin−EDTA (trypsin, 0.25%; EDTA, 1 mM), Hank’s balanced salt solution (HBSS), and antibiotic−antimycotic mixture (penicillin, 10 000 U/mL; streptomycin, 10 mg/mL; and amphotericin B, 25 μg/mL; dissolved in 0.85% (w/v) sodium chloride aqueous solution) were purchased from Gibco Laboratories (Lenexa, KS). Cell culture inserts with polyethylene terephthalate filters (pore size, 3.0 μm; growth area, 4.20 cm2) were obtained from Becton Dickinson Bioscience (Bedford, MA). Bovine serum albumin (BSA), phosphate pH standard equimolal solution (pH 6.86), sodium taurocholate (NaTC; purity > 90%), and purified egg yolk lecithin (containing >98% phosphatidylcholine) were obtained from Wako Pure Chemical Industries (Osaka, Japan). Fluorescein isothiocyanate−dextran (FD4) was purchased from Sigma-Aldrich (St. Louis, MO); n-decyl-β-D-maltopyranoside (DM), n-dodecyl-β-D-maltopyranoside (DDM), and n-tetradecyl-β-D-maltopyranoside (TDM) were purchased from Affymetrix (Santa Clara, CA). All other reagents used were of the highest purity available (95% or higher). 2.2. Caco-2 Cell Culture. Caco-2 cells were cultured in DMEM supplemented with 10% (v/v) FBS, 1% (v/v) nonessential amino acids, and 0.5% (v/v) antibiotic−antimycotic mixture in a flask of adequate volume (Nippon Becton Dickinson, Tokyo, Japan) at 95% humidity and 37 °C in a 95% air/5% CO2 atmosphere. Cells (passages 47−60) were harvested using trypsin−EDTA and seeded on cell culture inserts with a polyethylene terephthalate membrane (3.0 μm pores, 4.2 cm2 growth area) at a density of 3 × 105 cells/insert. The medium was changed every 48 h for 17−21 days, and the cell monolayer was then used for permeation studies. 2.3. Buffers Used for Transport Studies. Hank’s balanced salt solution (HBSS), containing 5.36 mM KCl, 136.89 mM NaCl, 0.34 mM Na2HPO4, 0.44 mM KH2PO4, 4.17 mM NaHCO3, 1.26 mM CaCl2, 0.49 mM MgCl2, 0.41 mM MgSO4, and 25 mM glucose, was used as standard buffer solution in this study (transport medium, TM). To monitor the effect of biorelevant media, modified versions of the fasted and fed state simulated intestinal fluids (FaSSIFmod and FeSSIFmod6.5; both pH 6.5) were prepared. Both media were based on TM, which was supplemented with 15 mM NaTC and 3.75
influencing the behavior of the API and associated excipients. While closely mimicking the salt concentration and osmotic pressure, such media typically lack other native components found in intestinal fluids, such as bile acids and lipids. To this end, biorelevant buffer systems have been developed that represent the fasted and fed states (FaSSIF and FeSSIF, respectively). While FaSSIF in its original composition, as defined by Dressman and co-workers,9 appears to be compatible with Caco-2 cells,10,11 FeSSIF causes severe cell damage.11 This was mainly attributed to its high osmolality and its high concentration of acetic acid. Consequently, the media have been modified in order to be suitable for applications in cellular-based investigations.11,12 The influence of such biorelevant media on the absorption of various drugs has received growing interest;10,13,14 however, to our knowledge, there is a scarcity of published data describing the behavior of permeation enhancers in these systems.15 Many permeation enhancers are amphiphilic in nature. By virtue of this, interactions with bile salt−phospholipid mixed micelles are very likely. To investigate this, the permeation of FD4, a hydrophilic macromolecule-sized surrogate marker compound, was monitored across Caco-2 cell monolayers in the presence or absence of three different maltosides: n-decyl-βD -maltopyranoside (DM), n-dodecyl-β- D -maltopyranoside (DDM), and n-tetradecyl-β-D-maltopyranoside (TDM), with alkyl chain lengths of 10, 12, and 14 C atoms, respectively. To simulate the effect of micelles, FaSSIFmod and FeSSIFmod6.5 were used in the apical compartment of the permeation system. Furthermore, the amount of maltoside monomer was calculated in the different media by separating it from micellar-bound maltoside using a dialysis membrane. In addition to in vitro examinations, in situ studies on rat intestine were performed to determine the in vivo relevance of the obtained results and to confirm the predictability of the in vivo behavior of a permeation enhancer based on in vitro studies.
2. MATERIALS AND METHODS 2.1. Materials. A human colorectal adenocarcinoma cell line, Caco-2, was purchased from American Type Culture Collection (Rockville, MD). Dulbecco’s modified Eagle’s medium (DMEM) was obtained from Sigma-Aldrich (St. 2246
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
Figure 1. Permeation enhancing effect of maltosides in different concentrations and buffers. The permeation of FD4 (2 mg/mL) through a Caco-2 cell monolayer was measured over 2 h with and without the addition of DM, DDM, or TDM in various concentrations: 0.5 mM in TM (●), 5 mM in TM (★), 0.5 mM in FaSSIFmod (△), 2 mM in FaSSIFmod (▽), 3.5 mM in FaSSIFmod (◊), 5 mM in FaSSIFmod (□), and 5 mM in FeSSIFmod6.5 (×). Indicated values are the mean ± SD of at least three experiments. Values above 8000 pmol/cm2 exceeded the measuring range and are therefore not depicted.
effect in the jejunum,8,16,17 colon was used throughout this study. All animal experiments were approved by the Ethical Review Committee of Setsunan University. Studies were carried out with male Sprague−Dawley rats with an average body weight of 250−300 g. Rats were fasted overnight and anaesthetised prior to surgery by inhalation of an isoflurane− oxygen mixture (1.5−3% isoflurane), which was maintained throughout the study by mask ventilation. A blood sample was taken from the jugular vein prior to the experiment (time point zero). Their abdominal cavity was opened, and a loop (5 cm) was created by inserting two silicone tubes (i.d., 3 mm; o.d., 5 mm) through small slits at the proximal and distal ends of the colon. Intestinal contents were removed by a slow infusion of saline followed by air. Test solution (0.5 mL) containing either solely FD4 or a mixture of FD4 and DDM, in TM, FaSSIFmod, or FeSSIFmod6.5, was introduced into the intestinal loop, and then both ends of the loop were closed. For a period of 2 h, blood samples (400 μL) were collected from the jugular vein and centrifuged, and the plasma was analyzed by fluorescence spectroscopy. Heparinized syringes were used for all blood samples taken. The plasma concentration of FD4 after intraintestinal administration was deconvoluted with the concentration after intravenous (i.v.) injection (100 μg/rat) to obtain the absorption rate of FD4 as a function of the input. 2.6. Analytical Methods. FD4 from in vitro permeation studies was analyzed at an excitation wavelength of 490 nm and an emission wavelength of 525 nm using the fluorescence detector of an HPLC system (mobile phase: 5 mM phosphate buffer, pH 7.4, mixed with acetonitrile, 88:12 v/v). Quantification was achieved by preparing a calibration curve of FD4 in the basolateral buffer. The amount of FD4 in rat plasma was determined using a fluorescence plate reader (Microplate reader MTP 601Lab, Corona Electric, Japan). One-hundred microliters of undiluted plasma was mixed with 100 μL of phosphate pH standard equimolal solution (pH 6.86), added to a 96-well plate, and measured at an excitation wavelength of 490 nm and an emission wavelength of 530 nm. A calibration curve was prepared from FD4 in a 1:1 mixture of the same buffer and blank plasma. Maltosides from in vitro permeation studies were analyzed using a UPLC system (Acquity UPLC, Waters, MA) equipped with a tandem mass spectrometer (Acquity TQD, Waters, MA). A reverse-phase analytical column (Asahipack NH2P-50 2D) of 150 mm length × 2.0 mm i.d. and 5 μm particle size (Shodex, Tokyo, Japan) at 40 °C was used with a mobile phase consisting of 0.028% (v/v) ammonia solution in water (solvent
mM lecithin to obtain FeSSIFmod6.5 and with 5 mM NaTC and 0.75 mM lecithin to obtain FaSSIFmod. As both media were adjusted to pH 6.5, FaSSIFmod retains the same pH as that of conventional FaSSIF, whereas the pH of FeSSIFmod6.5 differs from that of the conventionally prepared one, which has a pH of 5. 2.4. Transport Studies. Depending on the test setup, either a Caco-2 cell monolayer or a dialysis membrane (regenerated cellulose, cutoff, 1000 Da) was mounted between the side-by-side chambers of a combined dissolution/ permeation system (D/P system; effective surface area, 1.77 cm2), and both sides were filled with the appropriate buffer (see Table 1 for all test conditions). The volumes of the apical and basal sides were set to 8 and 5.5 mL, respectively, and were constantly stirred at 200 rpm with magnetic stirrers. All experiments were performed at 37 °C. The D/P system was then used to monitor the transport of FD4 through Caco-2 cells under different conditions (Table 1A) and to assess the amount/concentration of maltoside monomer (Table 1B). FD4 (final concentration of 2 mg/mL) and/or maltosides were added as a solution to the apical side after allowing the system to equilibrate for 10 min. Their permeation was monitored by taking samples from the basolateral compartment after 15, 30, 60, 90, and 120 min, which were subsequently analyzed for either FD4 or maltoside content. In the case of the latter, 4.5% w/v BSA was added to the TM in the basolateral chamber in order to prevent any interaction of the maltosides with the chamber. Apparent permeability coefficients (Papp; cm/s) for FD4 were calculated according to the following equation
Papp =
dCr V × r dt C0A
where dCr/dt is the increased drug concentration in the BL compartment over time, Vr is the buffer volume in the BL compartment, A is the diffusion area of the membrane (cm2), and C0 is the initial concentration of FD4 in the AP compartment (μg/cm3). As a second parameter reflecting the influence of maltosides on Caco-2 barrier integrity, the transepithelial electric resistance (TEER) was measured before and after the experiment. The TEER measured prior to each experiment was set as 100%, and all other values were calculated relative to this. 2.5. In Situ Intestinal Loop Studies on Rats. Intestinal absorption of FD4 from rat intestine was evaluated by an in situ closed loop method. Since it has already been demonstrated that DDM does not show a pronounced permeation enhancing 2247
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics Table 2. Comparison of Papp Values and TEER Reduction of All Enhancers Tested within This Studya enhancer
apical mediumb
no enhancer
TM pH 6.5 FaSSIFmod FeSSIFmod6.5
DM
TM pH 6.5 FaSSIFmod FeSSIFmod6.5
DDM
TM pH 6.5 FaSSIFmod
FeSSIFmod6.5 TDM
TM pH 6.5 FaSSIFmod
FeSSIFmod6.5
Papp × 10−6 (cm/s)
TEER reduction (%)
0 0 0
0.019 ± 0.002 0.017 ± 0.004 0.036 ± 0.007
11.9 ± 1.9 13.3 ± 1.1 17.3 ± 2.3
0.5 5 5 5
0.065 11.00 0.077 0.108
± ± ± ±
0.032 0.716 0.016* 0.040*
13.3 57.5 23.3 19.3
± ± ± ±
1.2 1.1 1.3* 2.9*
0.5 0.5 2 3.5 5 5
3.795 0.054 0.026 0.077 0.601 0.085
± ± ± ± ± ±
0.076 0.025* 0.007* 0.016* 0.215* 0.009*
51.9 10.5 17.2 27.4 43.7 24.4
± ± ± ± ± ±
0.9 1.3* 0.0* 0.2* 2.0* 0.4*
0.5 0.5 2 3.5 5 5
3.935 0.051 0.069 1.514 7.217 0.100
± ± ± ± ± ±
0.147 0.012* 0.016* 0.037* 1.931* 0.035*
57.0 15.9 26.5 53.6 59.6 25.0
± ± ± ± ± ±
1.1 0.5* 2.1* 3.3 0.8 2.3*
enhancer concentration (mM)
Indicated values are the mean ± SD of at least three experiments. bTM, pH 7.4, was used in the basolateral chamber in all cases. *p < 0.05 compared to their corresponding functional concentration(s) in TM, which was 5 mM for DM and 0.5 mM for DDM and TDM.
a
Figure 2. Decrease of the transepithelial electrical resistance (TEER) of Caco-2 cells after a 2 h incubation with maltosides in different concentrations and buffers. Indicated values are the mean ± SD of at least three experiments (*p < 0.05 compared to their corresponding functional concentration(s) in TM, which was 5 mM for DM and 0.5 mM for DDM and TDM).
2.7. Statistical Analysis. GraphPad Prism 6 was used to calculate mean values, standard deviations, and statistical parameters throughout the study. All data sets were compared using one-way ANOVA. Differences were considered significant at p < 0.05.
A) and 0.028% (v/v) ammonia solution in acetonitrile (solvent B) with a gradient time period. The initial mobile phase was 98% solvent A and 2% solvent B pumped at a flow rate of 0.3 mL/min. Between 0.5 and 1.5 min, the percentage of solvent B was increased linearly to 95%, where it was held for 1.3 min. Between 2.8 and 3.0 min, the percentage of solvent B was decreased linearly to 2%. This condition was maintained until 4 min, at which time the next sample was injected into the UPLC system. All treated samples were injected as 5 μL into the UPLC system. Ionization conditions for analysis of maltosides were as follows: electrospray ionization; negative mode; source temperature, 150 °C; desolvation temperature, 400 °C; cone voltage, 50 V; and collision energy, 30 eV. Precursor and production ions (m/z) for detection of maltosides were 481.40 and 89.16 for DM, 509.41 and 89.10 for DDM, and 537.63 and 89.17 for TDM, respectively. A standard curve was prepared for each maltoside, and linearity (r2 > 0.99) was obtained for a concentration range of 0.001−10 μg/mL (∼0.002−20 μM).
3. RESULTS 3.1. Permeation of FD4 through Caco-2 Cells. The influence of maltosides (0.5 to 5 mM) on the transport of FD4 in different buffers is summarized in Figure 1 and Table 2. When a concentration of 0.5 mM DDM or TDM was used in TM and added to the AP chamber of the D/P system, both exhibited a pronounced permeation enhancing effect, reflected by Papp values of approximately 3.80 × 10−6 and 3.94 × 10−6 cm/s, respectively. The reference value, measured for the same concentration of FD4 without maltosides, was 1.9 × 10−8 cm/s. On the basis of this, both enhancers, DDM and TDM, elicited a more than 100-fold increase in FD4 permeation. In contrast, 0.5 2248
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
Figure 3. Amount of permeated maltoside through a dialysis membrane. Different quantities of DM, DDM, or TDM were applied to the apical side of the D/P system, and their permeation was monitored in different buffers. Gray, filled symbols represent a concentration of 5 mM, whereas black, open symbols represent a concentration of 0.5 mM. The different shapes reflect the buffer that was used: triangles represent transport medium (TM), squares represent FaSSIFmod, and circles represent FeSSIFmod6.5. For DDM, two more conditions were included in the study: 0.08 (×) and 0.016 (*) mM maltoside in TM. Indicated values are the mean ± SD of at least three experiments.
Table 3. Calculation of the Amount of Maltoside Monomera amount of maltoside monomer (% of the maximum possible) 5 mM in TM 5 mM in FaSSIFmod 5 mM in FeSSIFmod6.5 0.5 mM in TM 0.5 mM in FaSSIFmod 0.5 mM in FeSSIFmod6.5 0.16 mM in TM 0.08 mM in TM
amount of maltoside monomer (mM)
DM
DDM
TDM
DM
DDM
100b 53.0 ± 1.6 15.3 ± 0.9 24.8 ± 2.5 10.9 ± 0.2 1.8 ± 0.2 − −
100b 54.4 ± 0.9 14.3 ± 1.1 97.8 ± 3.2 15.9 ± 0.6 0.7 ± 0.1 78.1 ± 9.5 41.9 ± 2.8
100b 63.4 ± 4.7 13.6 ± 2.7 100.4 ± 4.6 22.3 ± 2.2 6.0 ± 3.4 − −
2.015 1.068 0.308 0.500c 0.220 0.036 − −
0.198 0.108 0.028 0.194 0.031 0.001 0.155c 0.083c
a
Various amounts of maltosides were added to the apical compartment of the D/P system, and the permeation through a dialysis membrane was monitored in different buffers: transport medium (TM), FaSSIFmod, and FeSSIFmod6.5. This concentration was then used to calculate the amount of maltoside monomer in the apical compartment. bThe permeation of 5 mM of each maltoside in TM was set to 100%, and all other values were normalized to this. cThe calculation of the permeated amount of maltosides in millimolar was based on the permeation in TM, at a concentration lower than the CMC. All other values were assessed with respect to that.
sides, the TEER measurements recorded reflect the results obtained from the FD4 study. A marked decrease in TEER was measured at concentrations that led to a permeation enhancing effect in TM, i.e., 5 mM for DM and 0.5 mM for both DDM and TDM. In accordance with permeation studies, the impact on the TEER was discernibly lower for the same concentration in FaSSIFmod, but it increased again as the concentration of maltosides increased. When a concentration of 5 mM DDM or TDM was applied in FeSSIFmod6.5, the TEER decrease achieved was considerably less than that observed in its fasted state counterpart (FaSSIFmod); 24.39% versus 43.72% for DDM and 24.95% versus 59.56% for TDM. 3.2. Calculating the Amount of Maltoside Monomer. Figure 3 depicts the permeated amount of maltosides through a dialysis membrane (cutoff, 1000 Da) measured over 2 h. In order to remain in the same concentration range as that used in the FD4 permeation study, the permeation of 0.5 and 5 mM DM, DDM, or TDM was monitored in all three buffers (TM, FaSSIFmod, and FeSSIFmod6.5). In TM, the amount of maltoside monomer depends only on its critical micelle concentration (CMC). The concentration of maltoside monomer measured in the BL compartment was therefore highest for DM, a factor of 10 lower for DDM, and a further factor of 10 lower for TDM, corresponding to their CMCs in water, as provided by the supplier (1.8, 0.17, and 0.01 mM, respectively). As the CMC represents the maximal concentration of monomer, the amount of permeated maltoside should be the same in cases where the concentration in the AP chamber exceeds the CMC. In accordance with this consideration, there was no difference in the permeated amount of DDM or TDM between an apical
mM DM showed a markedly less potent effect, as reflected by a Papp of 6.5 × 10−8 cm/s. Therefore, the concentration of DM was augmented to 5 mM, which resulted in a noticeable permeation enhancing effect (Papp = 1.10 × 10−5 cm/s). To examine the influence of biorelevant fluids, FaSSIFmod was used instead of TM in the apical compartment. Maltosides were added at their functional concentration(s) in TM (5 mM DM, 0.5 mM DDM, and 0.5 mM TDM). For all three maltosides, the permeation enhancing effect seen in TM was abrogated. On account of this, for both DDM and TDM, the concentration of maltoside was increased stepwise to 5 mM to determine the capacity of FaSSIFmod to attenuate the permeation enhancement. For DDM, a concentration of 5 mM was required to provide enhanced FD4 permeation (Papp = 6.01 × 10−7 cm/s), whereas in the case of TDM, a concentration of 3.5 mM was sufficient to lead to a Papp of 1.51 × 10−6 cm/s. To mimic the situation after food intake, FeSSIFmod6.5 was used in the apical chamber, a buffer system composed of a 5fold higher concentration of lecithin and NaTC than that of FaSSIFmod. For all three maltosides tested in FeSSIFmod6.5 (5 mM), the resultant permeation rates measured for FD4 were found to lie close to the reference value of 3.6 × 10−8 cm/s. In parallel to monitoring the permeation of FD4, the barrier integrity of Caco-2 monolayers was evaluated by means of TEER measurements both prior to and at the end of each experiment (Figure 2 and Table 2). TEER values at the beginning of each study were between 300 and 400 Ω cm2. Control studies without any maltosides showed a TEER reduction of approximately 12−17% (Table 2). With malto2249
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
loop studies were carried out on rat colon (Figure 5). Owing to the fact that the in vivo potency of intestinal permeation enhancers is generally lower than their in vitro potency,3 a DDM concentration of 2 mM was used for all studies, leading to a maximum FD4 plasma concentration of about 1.3 μg/mL (reached after 45 min), as seen in Figure 5A. When using either FaSSIFmod or FeSSIFmod6.5, effects analogous to those seen in vitro were observed. The permeation enhancing potential of DDM was drastically reduced, with almost no FD4 being absorbed in the first hour. Only toward the end of the 2 h did the absorption of FD4 begin to gradually increase. In the case of FeSSIFmod6.5, this increase was observed to the same extent when omitting the maltoside, demonstrating that this effect is not caused by the permeation enhancer but rather by the interaction of bile acids with the colon. The situation was comparable for FaSSIFmod, even though in that case the effect of DDM in FaSSIFmod was slightly greater than that of the FaSSIFmod control. The bioavailability derived from deconvolution revealed an almost 20% absorption of FD4 in the presence of 2 mM DDM in TM after 2 h, compared to that of the control in TM, where less than 1% of applied FD4 was absorbed (Figure 5B). When FaSSIFmod or FeSSIFmod6.5 was used instead of TM, the bioavailability decreased to 6.6 and 5.0%, respectively. However, as mentioned previously, this was almost the same for the FaSSIFmod and FeSSIFmod6.5 control, where FD4 showed a bioavailability of 5.4 and 6.1%, respectively. Thus, while 2 mM DDM was able to enhance the bioavailability of FD4 in TM by a factor of 20, relative to the TM control, there was no significant difference for DDM in FaSSIFmod or FeSSIFmod6.5 when compared to their corresponding controls.
concentration of 0.5 and 5 mM, provided that TM was used as apical medium (Figure 3 and Table 3). In the case of DM, where the reported CMC of 1.8 mM lies between the two tested concentrations, the permeated amount in the case of an apical concentration of 0.5 mM was approximately 75% lower than that measured for a concentration of 5 mM. When FaSSIFmod was employed in the apical compartment, the permeated amount of maltoside was significantly reduced for all three compounds tested, and it was reduced to an even greater extent when FeSSIFmod6.5 was used. From the permeated amount of maltosides in TM, the amount of free maltoside, i.e., that corresponding to the maltoside monomer, was calculated (Table 3). The calculation of the percent values was based on a concentration above their CMC in TM, which was set to 100%. In this case, the permeated amount of maltoside monomer should correlate with the balance between monomer and micelles in the apical compartment. To calculate the permeated amount in millimolar, a concentration below the CMC was used, where only maltoside monomers are present. For this reason, two more concentrations of DDM were tested, 0.08 and 0.16 mM. On the basis of this calculation, the new CMC for DM and DDM was found to be 2.02 and 0.196 mM, respectively. Both values corresponded very well with the CMC measured in water, as reported from the manufacturer. For TDM, this was not possible, as the sensitivity of the measuring method did not permit a precise quantification of such low quantities. 3.3. Further Characterization of the Behavior of DDM. To further study the permeation enhancement mechanism of maltosides and its interaction with mixed micelles, DDM was chosen as a representative for all three and subjected to a more detailed investigation. The permeation of FD4 was monitored across a Caco-2 cell monolayer in the presence of different concentrations of DDM, ranging from 0 to 0.5 mM (all in TM; Figure 4). Analyzing the data showed no measurable
4. DISCUSSION Bringing a drug from the bench to the market is a long and cost-intensive process. Thus, it is crucial to be able to identify potent drug candidates in an efficient and timely manner. In this regard, predicting the in vivo performance of candidates from in vitro data is highly relevant and requires appropriate in vitro−in vivo correlations (IVIVC). In the development of drugs for oral use, it is not only the intrinsic pharmacological potency of candidates but also their capacity to be absorbed following oral administration that requires critical evaluation. Kataoka et al. have developed a combined in vitro dissolution/permeation system (D/P system)18 that has proven to be successful in predicting the in vivo absorption of several drugs.19,20 Even though dissolution was not an issue in the present study (FD4 is highly water-soluble), the D/P system was used nonetheless as a tool to mimic drug absorption in the intestine. In this study, the permeation enhancing effects of maltosides were tested in the D/P system in order to provide a more accurate assessment of how they behave and perform in vivo in the GI tract. The results obtained with two biorelevant media, FaSSIFmod and FeSSIFmod6.5, were compared with those determined in conventional TM. A dramatic reduction in the permeation enhancing potential of maltosides was observed when using FaSSIFmod or FeSSIFmod6.5, strongly suggestive of an interaction with NaTC/lecithin. However, additional characterization is likely necessary to confirm the hypothesis that maltosides are incorporated into NaTC/lecithin mixed micelles. In the human intestine, concentrations of bile salts are close to their CMCs, and different colloidal structures formed by NaTC and lecithin form spontaneously.21 Although simulated intestinal fluids mimicking the fasted and fed states are
Figure 4. Permeation enhancing effect of DDM in transport medium (TM). The permeation of FD4 through a Caco-2 cell monolayer was measured over 2 h with and without the addition of DDM at various concentrations (n = 3).
permeation enhancement up to a concentration of 0.08 mM (Papp = 4.5 × 10−8 cm/s). For the next concentration tested, 0.12 mM, a small permeation enhancement was measured, as reflected by a Papp value of 3.76 × 10−7 cm/s. However, from this concentration onward, the permeation enhancement increased sharply until a concentration of 0.2 mM, which corresponds to the CMC of DDM. Above this, the curve flattens and shows a lower slope until 0.5 mM, which was the highest concentration included in this study. 3.4. In Situ Intestinal Loop Studies in Rats. To assess the in vivo relevance of the results achieved in vitro, closed intestinal 2250
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
Figure 5. In situ intestinal loop studies. (A) Absorption of FD4 from a closed colon loop in the presence (filled symbols) and absence (open symbols) of 2 mM DDM in TM (●, ○), FaSSIFmod (■, □), or FeSSIFmod6.5 (▲, △). The uptake of FD4 was monitored by taking blood samples from the jugular vein over 2 h. Data are presented as the mean ± SD (n = 10 for DDM in TM, n = 6 for DDM in FaSSIFmod and FeSSIFmod6.5, and n = 3 for controls). The absorption enhancement achieved by 2 mM DDM in TM was determined to be statistically significant; all others did not show a statistical difference compared to the TM control. (B) FD4 bioavailability calculated based on the results obtained from (A).
simplified models of the more complex in vivo system, they have been successfully used for in vitro dissolution studies for several years.22 Considering that the concentration of lecithin and NaTC resembles that reported in the human small intestine in fasted and fed states, an interaction with colloidal structures in the intestine provides one potential reason for the often observed mismatch between in vitro potency of permeation enhancers and their actual performance in vivo.3,15 In addition, the impact of bile salts on the performance of permeation enhancers may well be multifaceted. While NaTC concentrations of 3 and 15 mM in fasted and fed states, respectively, are good approximations of the in vivo conditions,9 their concentrations change within the GI tract and also vary among individuals; values have been reported from 0 to 15 mM and from 0.5 to 37 mM21,23 for fasted and fed states, respectively. This subject-to-subject variability could conceivably give rise to a high variation in the permeation enhancing effect between individual patients and consequently in the bioavailability of the coformulated drug. In order to quantify the amount of lost maltosides due to interactions with mixed micelles, maltosides were mixed with FaSSIFmod/FeSSIFmod6.5 and the monomer was subsequently separated from the complex by mounting a dialysis membrane, instead of the Caco-2 monolayer, between the two chambers of the D/P system.24 In accordance with our hypothesis from the Caco-2 permeation study, this setup confirmed that the presence of NaTC/lecithin reduces the amount of maltoside monomer in the apical compartment in a concentrationdependent manner. Considering the situation in the intestine, three different forms of maltosides are theoretically possible: the monomer; a complex composed of bile acids, lipids, and maltosides; and a maltoside micelle consisting only of maltosides (Figure 6). As our test setup enables only the separation of monomers from all other possible colloidal structures, it was not possible to determine whether pure maltoside micelles exist at all or if all formed micelles contain NaTC/lecithin to some extent. It is most likely that their existence will depend on the concentration of maltoside or, more precisely, the ratio between NaTC/lecithin and the maltoside. In terms of permeation enhancement, it has been demonstrated that maltoside mixed micelles do not possess permeation enhancing properties. However, the question still
Figure 6. Scheme depicting the different forms of maltosides that could be present in the intestine.
remained as to whether only maltoside monomers or also maltoside micelles might act as permeation enhancers. To this end, the FD4 permeation through Caco-2 cells in the presence of numerous concentrations of DDM was investigated in TM. Up until the CMC, the permeated amount of FD4 increased rapidly with increasing concentrations of DDM. In this concentration range, only the monomer concentration increased, confirming that maltoside monomers do exert a permeation enhancing effect and that its strength is proportional to the concentration. However, also above the CMC the permeated amount of FD4 continued to increase with increasing concentration of DDM, albeit to a lesser extent. On the basis of this observation, it appears that both maltoside monomers and maltoside micelles contribute to the permeation enhancement, probably via different mechanisms. In this context, it is important to note that the potency of different permeation enhancers cannot be assessed solely by examining the amount of monomer/micelle. Permeation enhancement is a far more complex process that is also greatly influenced by the structure of the enhancer itself. Brayden et al. recently compared the permeation enhancing potential of mediumchain fatty acids of different chain lengths (C8−C12).25 They found that a longer fatty acid chain (and thus a lower CMC) confers higher permeation enhancing potential but also correlates with decreased in vitro cell viability. Conceivably, this association might be the same for maltosides, where we showed increasing enhancement potential with increasing chain length. Even though the use of maltosides as permeation enhancers has been investigated over several years,8,16,26,27 a precise 2251
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
importance of using biorelevant conditions, such as the appropriate media, should be emphasized in order to ensure a better correlation between in vitro and in vivo studies and to enable an early and more accurate assessment of promising substrates.
mechanism for this effect has yet to be elucidated. Measuring the TEER recovery after exposing tissues/cells to permeation enhancers can, however, give some fundamental understanding about the principal mode of action. A reversible reduction of the TEER generally indicates permeation enhancement via an interaction with tight junctions, whereas an irreversible TEER reduction may be suggestive of a more pronounced membrane solubilization component. Following this concept, Tirumalasetty et al. investigated the TEER of Caco-2 cells after a 6 h incubation with different concentrations of DDM (from 0.2 to 1 mM).7 They observed a very good recovery (>80%) up to a concentration of 0.6 mM, but the recovery was weak for concentrations exceeding this. This result was confirmed by our own studies, showing a TEER recovery of >90% after incubating the cells for 2−6 h with 0.5 mM DDM in TM. Given the surfactant nature of DDM, it should not be surprising that high concentrations lead to membrane perturbation effects. However, it could be demonstrated that lower concentrations enhance the permeation of drugs mainly via the paracellular route. Focusing on the results of in vitro permeation studies, where the highest concentration of DDM in TM was 0.5 mM, membrane solubilization is unlikely to be the predominant mechanism. We therefore hypothesized that maltoside micelles also enhance the permeation via tight junction modulation and not via a solubilization of the cell membrane. Recently, an interesting concept has been proposed for the permeation enhancing mechanism of sodium caprate, a medium-chain fatty acid, suggesting the displacement of specific tight junction proteins from lipid rafts.28 Lipid rafts, defined as small (10−200 nm), heterogeneous, highly dynamic, sterol- and sphingolipid-enriched domains that compartmentalize cellular processes,29 were suggested to stabilize the association of different tight junction-specific proteins.30,31 Considering the structural similarity between sodium caprate and alkyl-maltosides, a further investigation should examine an interaction with lipid rafts as a possible mechanism for the permeation enhancing effect of maltosides. To confirm the findings from the in vitro studies, in situ closed loop studies were carried out. Maltosides, like other permeation enhancers, were shown to exhibit a much greater effect in the colon than in the small intestine.8,16,17 The reason for this difference is not fully understood. However, after confirming these observations by preliminary experiments (data not shown), the colon was used for all loop studies. In accordance with in vitro studies, FaSSIFmod and FeSSIFmod6.5 both showed the same drastic effect on the permeation enhancing potential of DDM in situ. The 20-fold permeation enhancement seen in TM was completely abrogated in the presence of NaTC/lecithin, meaning that the bioavailability of a coadministered drug will depend to a great extent on the concentrations of NaTC/lecithin in the intestine. Given their large local and interpersonal variation and their change after food intake, ensuring a reliable and reproducible effect represents a challenge that needs to be addressed when using maltosides as intestinal permeation enhancers. To summarize the results presented in this work, we successfully demonstrated the impact of mixed micelles in the intestine on the permeation enhancing effect of maltosides. This study provides a systematic in vitro investigation of this effect, including media mimicking the fed state (FeSSIF) as well as an in situ study to demonstrate its relevance in vivo. Given the increasing interest in the application of permeation enhancers in the oral delivery of peptides and proteins, the
■
AUTHOR INFORMATION
Corresponding Author
*Tel/Fax: +81-72-866-3125/3126. E-mail: shinji@pharm. setsunan.ac.jp. Notes
The authors declare the following competing financial interest(s): K.G., S.T.B. and B.L.P. are employees of Novo Nordisk A/S.
■
ACKNOWLEDGMENTS The authors are grateful for the financial support received from Novo Nordisk A/S. The authors thank Naoya Hiroki, Kentaro Iwashige, and Noriaki Kashimoto for assisting with in situ loop studies and culturing Caco-2 cells.
■
REFERENCES
(1) Swenson, E. S.; Curatolo, W. J. (C) Means to enhance penetration: (2) Intestinal permeability enhancement for proteins, peptides and other polar drugs: mechanisms and potential toxicity. Adv. Drug Delivery Rev. 1992, 8, 39−92. (2) Walsh, E. G.; Adamczyk, B. E.; Chalasani, K. B.; Maher, S.; O’Toole, E. B.; Fox, J. S.; Leonard, T. W.; Brayden, D. J. Oral delivery of macromolecules: rationale underpinning gastrointestinal permeation enhancement technology (GIPET). Ther. Delivery 2011, 2, 1595− 1610. (3) Aungst, B. J. Absorption enhancers: applications and advances. AAPS J. 2012, 14, 10−18. (4) Maher, S.; Brayden, D. J. Overcoming poor permeability: translating permeation enhancers for oral peptide delivery. Drug Discovery Today: Technol. 2012, 9, 113−119. (5) Ahsan, F.; Arnold, J. J.; Yang, T.; Meezan, E.; Schwiebert, E. M.; Pillion, D. J. Effects of the permeability enhancers, tetradecylmaltoside and dimethyl-beta-cyclodextrin, on insulin movement across human bronchial epithelial cells (16HBE14o−). Eur. J. Pharm. Sci. 2003, 20, 27−34. (6) Yang, T.; Arnold, J. J.; Ahsan, F. Tetradecylmaltoside (TDM) enhances in vitro and in vivo intestinal absorption of enoxaparin, a low molecular weight heparin. J. Drug Targeting 2005, 13, 29−38. (7) Tirumalasetty, P. P.; Eley, J. G. Evaluation of dodecylmaltoside as a permeability enhancer for insulin using human carcinoma cells. J. Pharm. Sci. 2005, 94, 246−255. (8) Petersen, S. B.; Nolan, G.; Maher, S.; Rahbek, U. L.; Guldbrandt, M.; Brayden, D. J. Evaluation of alkylmaltosides as intestinal permeation enhancers: comparison between rat intestinal mucosal sheets and Caco-2 monolayers. Eur. J. Pharm. Sci. 2012, 47, 701−712. (9) Galia, E.; Nicolaides, E.; Horter, D.; Lobenberg, R.; Reppas, C.; Dressman, J. B. Evaluation of various dissolution media for predicting in vivo performance of class I and II drugs. Pharm. Res. 1998, 15, 698− 705. (10) Ingels, F.; Beck, B.; Oth, M.; Augustijns, P. Effect of simulated intestinal fluid on drug permeability estimation across Caco-2 monolayers. Int. J. Pharm. 2004, 274, 221−232. (11) Kataoka, M.; Masaoka, Y.; Sakuma, S.; Yamashita, S. Effect of food intake on the oral absorption of poorly water-soluble drugs: in vitro assessment of drug dissolution and permeation assay system. J. Pharm. Sci. 2006, 95, 2051−2061. (12) Markopoulos, C.; Thoenen, F.; Preisig, D.; Symillides, M.; Vertzoni, M.; Parrott, N.; Reppas, C.; Imanidis, G. Biorelevant media for transport experiments in the Caco-2 model to evaluate drug 2252
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics absorption in the fasted and the fed state and their usefulness. Eur. J. Pharm. Biopharm. 2014, 86, 438−448. (13) Fossati, L.; Dechaume, R.; Hardillier, E.; Chevillon, D.; Prevost, C.; Bolze, S.; Maubon, N. Use of simulated intestinal fluid for Caco-2 permeability assay of lipophilic drugs. Int. J. Pharm. 2008, 360, 148− 155. (14) Frank, K. J.; Westedt, U.; Rosenblatt, K. M.; Holig, P.; Rosenberg, J.; Magerlein, M.; Brandl, M.; Fricker, G. Impact of FaSSIF on the solubility and dissolution-/permeation rate of a poorly watersoluble compound. Eur. J. Pharm. Sci. 2012, 47, 16−20. (15) Tippin, T. K.; Thakker, D. R. Biorelevant refinement of the Caco-2 cell culture model to assess efficacy of paracellular permeability enhancers. J. Pharm. Sci. 2008, 97, 1977−1992. (16) Masahiro, M.; Yoko, K.; Kanji, T.; Shozo, M. Assessment of enhancing ability of medium-chain alkyl saccharides as new absorption enhancers in rat rectum. Int. J. Pharm. 1992, 79, 159−169. (17) Uchiyama, T.; Sugiyama, T.; Quan, Y. S.; Kotani, A.; Okada, N.; Fujita, T.; Muranishi, S.; Yamamoto, A. Enhanced permeability of insulin across the rat intestinal membrane by various absorption enhancers: their intestinal mucosal toxicity and absorption-enhancing mechanism of n-lauryl-beta-D-maltopyranoside. J. Pharm. Pharmacol. 1999, 51, 1241−1250. (18) Kataoka, M.; Masaoka, Y.; Yamazaki, Y.; Sakane, T.; Sezaki, H.; Yamashita, S. In vitro system to evaluate oral absorption of poorly water-soluble drugs: simultaneous analysis on dissolution and permeation of drugs. Pharm. Res. 2003, 20, 1674−1680. (19) Kataoka, M.; Itsubata, S.; Masaoka, Y.; Sakuma, S.; Yamashita, S. In vitro dissolution/permeation system to predict the oral absorption of poorly water-soluble drugs: effect of food and dose strength on it. Biol. Pharm. Bull. 2011, 34, 401−407. (20) Kataoka, M.; Sugano, K.; da Costa, M. C.; Wong, J. W.; Jones, K. L.; Masaoka, Y.; Sakuma, S.; Yamashita, S. Application of dissolution/permeation system for evaluation of formulation effect on oral absorption of poorly water-soluble drugs in drug development. Pharm. Res. 2012, 29, 1485−1494. (21) Holm, R.; Mullertz, A.; Mu, H. Bile salts and their importance for drug absorption. Int. J. Pharm. 2013, 453, 44−55. (22) Klein, S. The use of biorelevant dissolution media to forecast the in vivo performance of a drug. AAPS J. 2010, 12, 397−406. (23) Sugano, K. Biopharmaceutics Modeling and Simulations: Theory, Practice, Methods, and Applications, 1st ed.; John Wiley Sons, Inc.: Hoboken, NJ, 2012. (24) Yano, K.; Masaoka, Y.; Kataoka, M.; Sakuma, S.; Yamashita, S. Mechanisms of membrane transport of poorly soluble drugs: role of micelles in oral absorption processes. J. Pharm. Sci. 2010, 99, 1336− 1345. (25) Brayden, D. J.; Gleeson, J.; Walsh, E. G. A head-to-head multiparametric high content analysis of a series of medium chain fatty acid intestinal permeation enhancers in Caco-2 cells. Eur. J. Pharm. Biopharm. 2014, 88, 830−839. (26) Deshmukh, D. D.; Nagilla, R.; Ravis, W. R.; Betageri, G. V. Effect of dodecylmaltoside (DDM) on uptake of BCS III compounds, tiludronate and cromolyn, in Caco-2 cells and rat intestine model. Drug Delivery 2010, 17, 145−151. (27) Yamamoto, A.; Okagawa, T.; Kotani, A.; Uchiyama, T.; Shimura, T.; Tabata, S.; Kondo, S.; Muranishi, S. Effects of different absorption enhancers on the permeation of ebiratide, an ACTH analogue, across intestinal membranes. J. Pharm. Pharmacol. 1997, 49, 1057−1061. (28) Sugibayashi, K.; Onuki, Y.; Takayama, K. Displacement of tight junction proteins from detergent-resistant membrane domains by treatment with sodium caprate. Eur. J. Pharm. Sci. 2009, 36, 246−253. (29) Pike, L. J. Rafts defined: a report on the Keystone Symposium on lipid rafts and cell function. J. Lipid Res. 2006, 47, 1597−1598. (30) Nusrat, A.; Parkos, C. A.; Verkade, P.; Foley, C. S.; Liang, T. W.; Innis-Whitehouse, W.; Eastburn, K. K.; Madara, J. L. Tight junctions are membrane microdomains. J. Cell Sci. 2000, 113, 1771−1781. (31) Lambert, D.; O’Neill, C. A.; Padfield, P. J. Depletion of Caco-2 cell cholesterol disrupts barrier function by altering the detergent
solubility and distribution of specific tight-junction proteins. Biochem. J. 2005, 387, 553−560.
2253
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Interaction with Mixed Micelles in the Intestine Attenuates the Permeation Enhancing Potential of Alkyl-Maltosides Kerstin Gradauer,†,‡ Ayano Nishiumi,† Kota Unrinin,† Haruki Higashino,† Makoto Kataoka,† Betty L. Pedersen,‡ Stephen T. Buckley,‡ and Shinji Yamashita*,† †
Faculty of Pharmaceutical Sciences, Setsunan University, Osaka 573-0101, Japan Global Research, Novo Nordisk A/S, DK-2760 Måløv, Denmark
‡
ABSTRACT: The purpose of the present study was to investigate the interaction of intestinal permeation enhancers with lipid and surfactant components present in the milieu of the small intestine. Maltosides of different chain lengths (decyl-, dodecyl-, and tetradecyl-maltoside; DM, DDM, TDM, respectively) were used as examples of nonionic, surfactant-like permeation enhancers, and their effect on the permeation of FD4 across Caco-2 monolayers was monitored. To mimic the environment of the small intestine, modified versions of fasted and fed state simulated intestinal fluid (FaSSIFmod, FeSSIFmod6.5, respectively) were used in addition to standard transport media (TM). Compared to the buffer control, 0.5 mM DDM led to a 200-fold permeation enhancement of FD4 in TM. However, this was dramatically decreased in FaSSIFmod, where a concentration of 5 mM DDM was necessary in order to elicit a moderate, 4-fold, permeation enhancement. Its capacity to promote permeation was diminished further when FeSSIFmod6.5 was employed. Even when cells were exposed to a concentration of 5 mM, no significant permeation enhancement of FD4 was observed. Analogous effects were observed in the case of DM and TDM, with slight deviations on account of differences in their critical micelle concentration (CMC). This observation was corroborated by calculating the amount of maltoside monomer versus micellar bound maltoside in FaSSIFmod and FeSSIFmod6.5, which demonstrated a reduced amount of free monomer in these fluids. To evaluate the in vivo significance of our findings, DDM solutions in TM, FaSSIFmod, and FeSSIFmod6.5 were used for closed intestinal loop studies in rats. Consistent with the results found in in vitro permeation studies, these investigations illustrated the overwhelming impact of sodium taurocholate/lecithin micelles on the permeation enhancing effect of DDM. While DDM led to a 20-fold increase in FD4 bioavailability when it was applied in TM, no significant permeation enhancement was seen in FaSSIFmod/FeSSIFmod6.5. Collectively, these investigations highlight the importance of using biorelevant media when evaluating the potency of permeation enhancers. In doing so, this ensures improved correlations between in vitro and in vivo studies and thus enables an early and more accurate assessment of promising permeation enhancers. KEYWORDS: maltoside, intestinal absorption, Caco-2, permeation enhancement, in vitro−in vivo correlation (IVIVC), biorelevant media
1. INTRODUCTION The oral route of drug delivery is often considered to be the most preferable by virtue of the fact that it represents an easy and painless route of administration, thus leading to high patient compliance. However, not all drugs show sufficient oral bioavailability. Poor oral absorption of drugs is caused by low solubility in the gastrointestinal (GI) fluid, insufficient permeation through the GI membrane, and/or limited stability (chemical/biological) in the GI tract or liver. One of the most extensively studied approaches to address the insufficient permeability of drugs is to combine the API with a permeation enhancer. This approach can facilitate increased uptake by altering the cell membrane, enhancing the paracellular or the transcellular permeation, or forming a noncovalent complex with the API.1−4 Maltosides, which were employed in this study, are a group of water-soluble, nonionic detergents that demonstrate high enhancer potency and low toxicity. Given their surfactant-like nature, it seems reasonable © 2015 American Chemical Society
to expect them to give rise to membrane perturbation effects to some extent;5,6 however, it has been demonstrated that at lower concentrations maltosides primarily promote absorption via the paracellular route by modulating tight junctions.7,8 Despite extensive investigations on permeation enhancers over the past decades, no oral product has yet reached the market.4 Conceivably, this may be attributed to a lack of efficacy in vivo, even though in vitro studies may suggest that an enhancer has high potency. In order to improve the predictability in screening permeation enhancers at an early research stage, in vitro studies need to be carried out under more biorelevant conditions. The media composition in a cellbased permeation assay may play a pivotal role in terms of Received: Revised: Accepted: Published: 2245
November 21, 2014 February 16, 2015 April 15, 2015 April 15, 2015 DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics Table 1. Overview of All Conditions Investigated within this Studya TM
FaSSIFmod
FeSSIFmod6.5
DM
DDM
TDM
DM
DDM
TDM
DM
DDM
TDM
− − − − − − 0.5 − − 5 − − 0.5 5
0.08 0.12 0.16 0.2 0.25 0.35 0.5 − − − − − 0.5 −
− − − − − − 0.5 − − − − − 0.5 −
− − − − − − − − − 5 − − 0.5 5
− − − − − − 0.5 2 3.5 5 0.08 0.16 0.5 5
− − − − − − 0.5 2 3.5 5 − − 0.5 5
− − − − − − − − − 5 − − 0.5 5
− − − − − − − − − 5 − − 0.5 5
− − − − − − − − − 5 − − 0.5 5
(A) Permeation of FD4 through a Caco-2 monolayer
(B) Permeation of maltoside monomer through a dialysis membrane
a
The D/P system was used to monitor both the permeability of FD4 through Caco-2 monolayers in the presence of different maltosides (A) as well as the interaction of maltosides with FaSSIFmod and FeSSIFmod6.5, where free and micellar bound maltosides were separated using a dialysis membrane (B). All values depict the concentration of enhancer in the apical compartment of the D/P system (mM).
Louis, MO). Nonessential amino acids (10 mM), fetal bovine serum (FBS), L-glutamine (200 mM), trypsin−EDTA (trypsin, 0.25%; EDTA, 1 mM), Hank’s balanced salt solution (HBSS), and antibiotic−antimycotic mixture (penicillin, 10 000 U/mL; streptomycin, 10 mg/mL; and amphotericin B, 25 μg/mL; dissolved in 0.85% (w/v) sodium chloride aqueous solution) were purchased from Gibco Laboratories (Lenexa, KS). Cell culture inserts with polyethylene terephthalate filters (pore size, 3.0 μm; growth area, 4.20 cm2) were obtained from Becton Dickinson Bioscience (Bedford, MA). Bovine serum albumin (BSA), phosphate pH standard equimolal solution (pH 6.86), sodium taurocholate (NaTC; purity > 90%), and purified egg yolk lecithin (containing >98% phosphatidylcholine) were obtained from Wako Pure Chemical Industries (Osaka, Japan). Fluorescein isothiocyanate−dextran (FD4) was purchased from Sigma-Aldrich (St. Louis, MO); n-decyl-β-D-maltopyranoside (DM), n-dodecyl-β-D-maltopyranoside (DDM), and n-tetradecyl-β-D-maltopyranoside (TDM) were purchased from Affymetrix (Santa Clara, CA). All other reagents used were of the highest purity available (95% or higher). 2.2. Caco-2 Cell Culture. Caco-2 cells were cultured in DMEM supplemented with 10% (v/v) FBS, 1% (v/v) nonessential amino acids, and 0.5% (v/v) antibiotic−antimycotic mixture in a flask of adequate volume (Nippon Becton Dickinson, Tokyo, Japan) at 95% humidity and 37 °C in a 95% air/5% CO2 atmosphere. Cells (passages 47−60) were harvested using trypsin−EDTA and seeded on cell culture inserts with a polyethylene terephthalate membrane (3.0 μm pores, 4.2 cm2 growth area) at a density of 3 × 105 cells/insert. The medium was changed every 48 h for 17−21 days, and the cell monolayer was then used for permeation studies. 2.3. Buffers Used for Transport Studies. Hank’s balanced salt solution (HBSS), containing 5.36 mM KCl, 136.89 mM NaCl, 0.34 mM Na2HPO4, 0.44 mM KH2PO4, 4.17 mM NaHCO3, 1.26 mM CaCl2, 0.49 mM MgCl2, 0.41 mM MgSO4, and 25 mM glucose, was used as standard buffer solution in this study (transport medium, TM). To monitor the effect of biorelevant media, modified versions of the fasted and fed state simulated intestinal fluids (FaSSIFmod and FeSSIFmod6.5; both pH 6.5) were prepared. Both media were based on TM, which was supplemented with 15 mM NaTC and 3.75
influencing the behavior of the API and associated excipients. While closely mimicking the salt concentration and osmotic pressure, such media typically lack other native components found in intestinal fluids, such as bile acids and lipids. To this end, biorelevant buffer systems have been developed that represent the fasted and fed states (FaSSIF and FeSSIF, respectively). While FaSSIF in its original composition, as defined by Dressman and co-workers,9 appears to be compatible with Caco-2 cells,10,11 FeSSIF causes severe cell damage.11 This was mainly attributed to its high osmolality and its high concentration of acetic acid. Consequently, the media have been modified in order to be suitable for applications in cellular-based investigations.11,12 The influence of such biorelevant media on the absorption of various drugs has received growing interest;10,13,14 however, to our knowledge, there is a scarcity of published data describing the behavior of permeation enhancers in these systems.15 Many permeation enhancers are amphiphilic in nature. By virtue of this, interactions with bile salt−phospholipid mixed micelles are very likely. To investigate this, the permeation of FD4, a hydrophilic macromolecule-sized surrogate marker compound, was monitored across Caco-2 cell monolayers in the presence or absence of three different maltosides: n-decyl-βD -maltopyranoside (DM), n-dodecyl-β- D -maltopyranoside (DDM), and n-tetradecyl-β-D-maltopyranoside (TDM), with alkyl chain lengths of 10, 12, and 14 C atoms, respectively. To simulate the effect of micelles, FaSSIFmod and FeSSIFmod6.5 were used in the apical compartment of the permeation system. Furthermore, the amount of maltoside monomer was calculated in the different media by separating it from micellar-bound maltoside using a dialysis membrane. In addition to in vitro examinations, in situ studies on rat intestine were performed to determine the in vivo relevance of the obtained results and to confirm the predictability of the in vivo behavior of a permeation enhancer based on in vitro studies.
2. MATERIALS AND METHODS 2.1. Materials. A human colorectal adenocarcinoma cell line, Caco-2, was purchased from American Type Culture Collection (Rockville, MD). Dulbecco’s modified Eagle’s medium (DMEM) was obtained from Sigma-Aldrich (St. 2246
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
Figure 1. Permeation enhancing effect of maltosides in different concentrations and buffers. The permeation of FD4 (2 mg/mL) through a Caco-2 cell monolayer was measured over 2 h with and without the addition of DM, DDM, or TDM in various concentrations: 0.5 mM in TM (●), 5 mM in TM (★), 0.5 mM in FaSSIFmod (△), 2 mM in FaSSIFmod (▽), 3.5 mM in FaSSIFmod (◊), 5 mM in FaSSIFmod (□), and 5 mM in FeSSIFmod6.5 (×). Indicated values are the mean ± SD of at least three experiments. Values above 8000 pmol/cm2 exceeded the measuring range and are therefore not depicted.
effect in the jejunum,8,16,17 colon was used throughout this study. All animal experiments were approved by the Ethical Review Committee of Setsunan University. Studies were carried out with male Sprague−Dawley rats with an average body weight of 250−300 g. Rats were fasted overnight and anaesthetised prior to surgery by inhalation of an isoflurane− oxygen mixture (1.5−3% isoflurane), which was maintained throughout the study by mask ventilation. A blood sample was taken from the jugular vein prior to the experiment (time point zero). Their abdominal cavity was opened, and a loop (5 cm) was created by inserting two silicone tubes (i.d., 3 mm; o.d., 5 mm) through small slits at the proximal and distal ends of the colon. Intestinal contents were removed by a slow infusion of saline followed by air. Test solution (0.5 mL) containing either solely FD4 or a mixture of FD4 and DDM, in TM, FaSSIFmod, or FeSSIFmod6.5, was introduced into the intestinal loop, and then both ends of the loop were closed. For a period of 2 h, blood samples (400 μL) were collected from the jugular vein and centrifuged, and the plasma was analyzed by fluorescence spectroscopy. Heparinized syringes were used for all blood samples taken. The plasma concentration of FD4 after intraintestinal administration was deconvoluted with the concentration after intravenous (i.v.) injection (100 μg/rat) to obtain the absorption rate of FD4 as a function of the input. 2.6. Analytical Methods. FD4 from in vitro permeation studies was analyzed at an excitation wavelength of 490 nm and an emission wavelength of 525 nm using the fluorescence detector of an HPLC system (mobile phase: 5 mM phosphate buffer, pH 7.4, mixed with acetonitrile, 88:12 v/v). Quantification was achieved by preparing a calibration curve of FD4 in the basolateral buffer. The amount of FD4 in rat plasma was determined using a fluorescence plate reader (Microplate reader MTP 601Lab, Corona Electric, Japan). One-hundred microliters of undiluted plasma was mixed with 100 μL of phosphate pH standard equimolal solution (pH 6.86), added to a 96-well plate, and measured at an excitation wavelength of 490 nm and an emission wavelength of 530 nm. A calibration curve was prepared from FD4 in a 1:1 mixture of the same buffer and blank plasma. Maltosides from in vitro permeation studies were analyzed using a UPLC system (Acquity UPLC, Waters, MA) equipped with a tandem mass spectrometer (Acquity TQD, Waters, MA). A reverse-phase analytical column (Asahipack NH2P-50 2D) of 150 mm length × 2.0 mm i.d. and 5 μm particle size (Shodex, Tokyo, Japan) at 40 °C was used with a mobile phase consisting of 0.028% (v/v) ammonia solution in water (solvent
mM lecithin to obtain FeSSIFmod6.5 and with 5 mM NaTC and 0.75 mM lecithin to obtain FaSSIFmod. As both media were adjusted to pH 6.5, FaSSIFmod retains the same pH as that of conventional FaSSIF, whereas the pH of FeSSIFmod6.5 differs from that of the conventionally prepared one, which has a pH of 5. 2.4. Transport Studies. Depending on the test setup, either a Caco-2 cell monolayer or a dialysis membrane (regenerated cellulose, cutoff, 1000 Da) was mounted between the side-by-side chambers of a combined dissolution/ permeation system (D/P system; effective surface area, 1.77 cm2), and both sides were filled with the appropriate buffer (see Table 1 for all test conditions). The volumes of the apical and basal sides were set to 8 and 5.5 mL, respectively, and were constantly stirred at 200 rpm with magnetic stirrers. All experiments were performed at 37 °C. The D/P system was then used to monitor the transport of FD4 through Caco-2 cells under different conditions (Table 1A) and to assess the amount/concentration of maltoside monomer (Table 1B). FD4 (final concentration of 2 mg/mL) and/or maltosides were added as a solution to the apical side after allowing the system to equilibrate for 10 min. Their permeation was monitored by taking samples from the basolateral compartment after 15, 30, 60, 90, and 120 min, which were subsequently analyzed for either FD4 or maltoside content. In the case of the latter, 4.5% w/v BSA was added to the TM in the basolateral chamber in order to prevent any interaction of the maltosides with the chamber. Apparent permeability coefficients (Papp; cm/s) for FD4 were calculated according to the following equation
Papp =
dCr V × r dt C0A
where dCr/dt is the increased drug concentration in the BL compartment over time, Vr is the buffer volume in the BL compartment, A is the diffusion area of the membrane (cm2), and C0 is the initial concentration of FD4 in the AP compartment (μg/cm3). As a second parameter reflecting the influence of maltosides on Caco-2 barrier integrity, the transepithelial electric resistance (TEER) was measured before and after the experiment. The TEER measured prior to each experiment was set as 100%, and all other values were calculated relative to this. 2.5. In Situ Intestinal Loop Studies on Rats. Intestinal absorption of FD4 from rat intestine was evaluated by an in situ closed loop method. Since it has already been demonstrated that DDM does not show a pronounced permeation enhancing 2247
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics Table 2. Comparison of Papp Values and TEER Reduction of All Enhancers Tested within This Studya enhancer
apical mediumb
no enhancer
TM pH 6.5 FaSSIFmod FeSSIFmod6.5
DM
TM pH 6.5 FaSSIFmod FeSSIFmod6.5
DDM
TM pH 6.5 FaSSIFmod
FeSSIFmod6.5 TDM
TM pH 6.5 FaSSIFmod
FeSSIFmod6.5
Papp × 10−6 (cm/s)
TEER reduction (%)
0 0 0
0.019 ± 0.002 0.017 ± 0.004 0.036 ± 0.007
11.9 ± 1.9 13.3 ± 1.1 17.3 ± 2.3
0.5 5 5 5
0.065 11.00 0.077 0.108
± ± ± ±
0.032 0.716 0.016* 0.040*
13.3 57.5 23.3 19.3
± ± ± ±
1.2 1.1 1.3* 2.9*
0.5 0.5 2 3.5 5 5
3.795 0.054 0.026 0.077 0.601 0.085
± ± ± ± ± ±
0.076 0.025* 0.007* 0.016* 0.215* 0.009*
51.9 10.5 17.2 27.4 43.7 24.4
± ± ± ± ± ±
0.9 1.3* 0.0* 0.2* 2.0* 0.4*
0.5 0.5 2 3.5 5 5
3.935 0.051 0.069 1.514 7.217 0.100
± ± ± ± ± ±
0.147 0.012* 0.016* 0.037* 1.931* 0.035*
57.0 15.9 26.5 53.6 59.6 25.0
± ± ± ± ± ±
1.1 0.5* 2.1* 3.3 0.8 2.3*
enhancer concentration (mM)
Indicated values are the mean ± SD of at least three experiments. bTM, pH 7.4, was used in the basolateral chamber in all cases. *p < 0.05 compared to their corresponding functional concentration(s) in TM, which was 5 mM for DM and 0.5 mM for DDM and TDM.
a
Figure 2. Decrease of the transepithelial electrical resistance (TEER) of Caco-2 cells after a 2 h incubation with maltosides in different concentrations and buffers. Indicated values are the mean ± SD of at least three experiments (*p < 0.05 compared to their corresponding functional concentration(s) in TM, which was 5 mM for DM and 0.5 mM for DDM and TDM).
2.7. Statistical Analysis. GraphPad Prism 6 was used to calculate mean values, standard deviations, and statistical parameters throughout the study. All data sets were compared using one-way ANOVA. Differences were considered significant at p < 0.05.
A) and 0.028% (v/v) ammonia solution in acetonitrile (solvent B) with a gradient time period. The initial mobile phase was 98% solvent A and 2% solvent B pumped at a flow rate of 0.3 mL/min. Between 0.5 and 1.5 min, the percentage of solvent B was increased linearly to 95%, where it was held for 1.3 min. Between 2.8 and 3.0 min, the percentage of solvent B was decreased linearly to 2%. This condition was maintained until 4 min, at which time the next sample was injected into the UPLC system. All treated samples were injected as 5 μL into the UPLC system. Ionization conditions for analysis of maltosides were as follows: electrospray ionization; negative mode; source temperature, 150 °C; desolvation temperature, 400 °C; cone voltage, 50 V; and collision energy, 30 eV. Precursor and production ions (m/z) for detection of maltosides were 481.40 and 89.16 for DM, 509.41 and 89.10 for DDM, and 537.63 and 89.17 for TDM, respectively. A standard curve was prepared for each maltoside, and linearity (r2 > 0.99) was obtained for a concentration range of 0.001−10 μg/mL (∼0.002−20 μM).
3. RESULTS 3.1. Permeation of FD4 through Caco-2 Cells. The influence of maltosides (0.5 to 5 mM) on the transport of FD4 in different buffers is summarized in Figure 1 and Table 2. When a concentration of 0.5 mM DDM or TDM was used in TM and added to the AP chamber of the D/P system, both exhibited a pronounced permeation enhancing effect, reflected by Papp values of approximately 3.80 × 10−6 and 3.94 × 10−6 cm/s, respectively. The reference value, measured for the same concentration of FD4 without maltosides, was 1.9 × 10−8 cm/s. On the basis of this, both enhancers, DDM and TDM, elicited a more than 100-fold increase in FD4 permeation. In contrast, 0.5 2248
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
Figure 3. Amount of permeated maltoside through a dialysis membrane. Different quantities of DM, DDM, or TDM were applied to the apical side of the D/P system, and their permeation was monitored in different buffers. Gray, filled symbols represent a concentration of 5 mM, whereas black, open symbols represent a concentration of 0.5 mM. The different shapes reflect the buffer that was used: triangles represent transport medium (TM), squares represent FaSSIFmod, and circles represent FeSSIFmod6.5. For DDM, two more conditions were included in the study: 0.08 (×) and 0.016 (*) mM maltoside in TM. Indicated values are the mean ± SD of at least three experiments.
Table 3. Calculation of the Amount of Maltoside Monomera amount of maltoside monomer (% of the maximum possible) 5 mM in TM 5 mM in FaSSIFmod 5 mM in FeSSIFmod6.5 0.5 mM in TM 0.5 mM in FaSSIFmod 0.5 mM in FeSSIFmod6.5 0.16 mM in TM 0.08 mM in TM
amount of maltoside monomer (mM)
DM
DDM
TDM
DM
DDM
100b 53.0 ± 1.6 15.3 ± 0.9 24.8 ± 2.5 10.9 ± 0.2 1.8 ± 0.2 − −
100b 54.4 ± 0.9 14.3 ± 1.1 97.8 ± 3.2 15.9 ± 0.6 0.7 ± 0.1 78.1 ± 9.5 41.9 ± 2.8
100b 63.4 ± 4.7 13.6 ± 2.7 100.4 ± 4.6 22.3 ± 2.2 6.0 ± 3.4 − −
2.015 1.068 0.308 0.500c 0.220 0.036 − −
0.198 0.108 0.028 0.194 0.031 0.001 0.155c 0.083c
a
Various amounts of maltosides were added to the apical compartment of the D/P system, and the permeation through a dialysis membrane was monitored in different buffers: transport medium (TM), FaSSIFmod, and FeSSIFmod6.5. This concentration was then used to calculate the amount of maltoside monomer in the apical compartment. bThe permeation of 5 mM of each maltoside in TM was set to 100%, and all other values were normalized to this. cThe calculation of the permeated amount of maltosides in millimolar was based on the permeation in TM, at a concentration lower than the CMC. All other values were assessed with respect to that.
sides, the TEER measurements recorded reflect the results obtained from the FD4 study. A marked decrease in TEER was measured at concentrations that led to a permeation enhancing effect in TM, i.e., 5 mM for DM and 0.5 mM for both DDM and TDM. In accordance with permeation studies, the impact on the TEER was discernibly lower for the same concentration in FaSSIFmod, but it increased again as the concentration of maltosides increased. When a concentration of 5 mM DDM or TDM was applied in FeSSIFmod6.5, the TEER decrease achieved was considerably less than that observed in its fasted state counterpart (FaSSIFmod); 24.39% versus 43.72% for DDM and 24.95% versus 59.56% for TDM. 3.2. Calculating the Amount of Maltoside Monomer. Figure 3 depicts the permeated amount of maltosides through a dialysis membrane (cutoff, 1000 Da) measured over 2 h. In order to remain in the same concentration range as that used in the FD4 permeation study, the permeation of 0.5 and 5 mM DM, DDM, or TDM was monitored in all three buffers (TM, FaSSIFmod, and FeSSIFmod6.5). In TM, the amount of maltoside monomer depends only on its critical micelle concentration (CMC). The concentration of maltoside monomer measured in the BL compartment was therefore highest for DM, a factor of 10 lower for DDM, and a further factor of 10 lower for TDM, corresponding to their CMCs in water, as provided by the supplier (1.8, 0.17, and 0.01 mM, respectively). As the CMC represents the maximal concentration of monomer, the amount of permeated maltoside should be the same in cases where the concentration in the AP chamber exceeds the CMC. In accordance with this consideration, there was no difference in the permeated amount of DDM or TDM between an apical
mM DM showed a markedly less potent effect, as reflected by a Papp of 6.5 × 10−8 cm/s. Therefore, the concentration of DM was augmented to 5 mM, which resulted in a noticeable permeation enhancing effect (Papp = 1.10 × 10−5 cm/s). To examine the influence of biorelevant fluids, FaSSIFmod was used instead of TM in the apical compartment. Maltosides were added at their functional concentration(s) in TM (5 mM DM, 0.5 mM DDM, and 0.5 mM TDM). For all three maltosides, the permeation enhancing effect seen in TM was abrogated. On account of this, for both DDM and TDM, the concentration of maltoside was increased stepwise to 5 mM to determine the capacity of FaSSIFmod to attenuate the permeation enhancement. For DDM, a concentration of 5 mM was required to provide enhanced FD4 permeation (Papp = 6.01 × 10−7 cm/s), whereas in the case of TDM, a concentration of 3.5 mM was sufficient to lead to a Papp of 1.51 × 10−6 cm/s. To mimic the situation after food intake, FeSSIFmod6.5 was used in the apical chamber, a buffer system composed of a 5fold higher concentration of lecithin and NaTC than that of FaSSIFmod. For all three maltosides tested in FeSSIFmod6.5 (5 mM), the resultant permeation rates measured for FD4 were found to lie close to the reference value of 3.6 × 10−8 cm/s. In parallel to monitoring the permeation of FD4, the barrier integrity of Caco-2 monolayers was evaluated by means of TEER measurements both prior to and at the end of each experiment (Figure 2 and Table 2). TEER values at the beginning of each study were between 300 and 400 Ω cm2. Control studies without any maltosides showed a TEER reduction of approximately 12−17% (Table 2). With malto2249
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
loop studies were carried out on rat colon (Figure 5). Owing to the fact that the in vivo potency of intestinal permeation enhancers is generally lower than their in vitro potency,3 a DDM concentration of 2 mM was used for all studies, leading to a maximum FD4 plasma concentration of about 1.3 μg/mL (reached after 45 min), as seen in Figure 5A. When using either FaSSIFmod or FeSSIFmod6.5, effects analogous to those seen in vitro were observed. The permeation enhancing potential of DDM was drastically reduced, with almost no FD4 being absorbed in the first hour. Only toward the end of the 2 h did the absorption of FD4 begin to gradually increase. In the case of FeSSIFmod6.5, this increase was observed to the same extent when omitting the maltoside, demonstrating that this effect is not caused by the permeation enhancer but rather by the interaction of bile acids with the colon. The situation was comparable for FaSSIFmod, even though in that case the effect of DDM in FaSSIFmod was slightly greater than that of the FaSSIFmod control. The bioavailability derived from deconvolution revealed an almost 20% absorption of FD4 in the presence of 2 mM DDM in TM after 2 h, compared to that of the control in TM, where less than 1% of applied FD4 was absorbed (Figure 5B). When FaSSIFmod or FeSSIFmod6.5 was used instead of TM, the bioavailability decreased to 6.6 and 5.0%, respectively. However, as mentioned previously, this was almost the same for the FaSSIFmod and FeSSIFmod6.5 control, where FD4 showed a bioavailability of 5.4 and 6.1%, respectively. Thus, while 2 mM DDM was able to enhance the bioavailability of FD4 in TM by a factor of 20, relative to the TM control, there was no significant difference for DDM in FaSSIFmod or FeSSIFmod6.5 when compared to their corresponding controls.
concentration of 0.5 and 5 mM, provided that TM was used as apical medium (Figure 3 and Table 3). In the case of DM, where the reported CMC of 1.8 mM lies between the two tested concentrations, the permeated amount in the case of an apical concentration of 0.5 mM was approximately 75% lower than that measured for a concentration of 5 mM. When FaSSIFmod was employed in the apical compartment, the permeated amount of maltoside was significantly reduced for all three compounds tested, and it was reduced to an even greater extent when FeSSIFmod6.5 was used. From the permeated amount of maltosides in TM, the amount of free maltoside, i.e., that corresponding to the maltoside monomer, was calculated (Table 3). The calculation of the percent values was based on a concentration above their CMC in TM, which was set to 100%. In this case, the permeated amount of maltoside monomer should correlate with the balance between monomer and micelles in the apical compartment. To calculate the permeated amount in millimolar, a concentration below the CMC was used, where only maltoside monomers are present. For this reason, two more concentrations of DDM were tested, 0.08 and 0.16 mM. On the basis of this calculation, the new CMC for DM and DDM was found to be 2.02 and 0.196 mM, respectively. Both values corresponded very well with the CMC measured in water, as reported from the manufacturer. For TDM, this was not possible, as the sensitivity of the measuring method did not permit a precise quantification of such low quantities. 3.3. Further Characterization of the Behavior of DDM. To further study the permeation enhancement mechanism of maltosides and its interaction with mixed micelles, DDM was chosen as a representative for all three and subjected to a more detailed investigation. The permeation of FD4 was monitored across a Caco-2 cell monolayer in the presence of different concentrations of DDM, ranging from 0 to 0.5 mM (all in TM; Figure 4). Analyzing the data showed no measurable
4. DISCUSSION Bringing a drug from the bench to the market is a long and cost-intensive process. Thus, it is crucial to be able to identify potent drug candidates in an efficient and timely manner. In this regard, predicting the in vivo performance of candidates from in vitro data is highly relevant and requires appropriate in vitro−in vivo correlations (IVIVC). In the development of drugs for oral use, it is not only the intrinsic pharmacological potency of candidates but also their capacity to be absorbed following oral administration that requires critical evaluation. Kataoka et al. have developed a combined in vitro dissolution/permeation system (D/P system)18 that has proven to be successful in predicting the in vivo absorption of several drugs.19,20 Even though dissolution was not an issue in the present study (FD4 is highly water-soluble), the D/P system was used nonetheless as a tool to mimic drug absorption in the intestine. In this study, the permeation enhancing effects of maltosides were tested in the D/P system in order to provide a more accurate assessment of how they behave and perform in vivo in the GI tract. The results obtained with two biorelevant media, FaSSIFmod and FeSSIFmod6.5, were compared with those determined in conventional TM. A dramatic reduction in the permeation enhancing potential of maltosides was observed when using FaSSIFmod or FeSSIFmod6.5, strongly suggestive of an interaction with NaTC/lecithin. However, additional characterization is likely necessary to confirm the hypothesis that maltosides are incorporated into NaTC/lecithin mixed micelles. In the human intestine, concentrations of bile salts are close to their CMCs, and different colloidal structures formed by NaTC and lecithin form spontaneously.21 Although simulated intestinal fluids mimicking the fasted and fed states are
Figure 4. Permeation enhancing effect of DDM in transport medium (TM). The permeation of FD4 through a Caco-2 cell monolayer was measured over 2 h with and without the addition of DDM at various concentrations (n = 3).
permeation enhancement up to a concentration of 0.08 mM (Papp = 4.5 × 10−8 cm/s). For the next concentration tested, 0.12 mM, a small permeation enhancement was measured, as reflected by a Papp value of 3.76 × 10−7 cm/s. However, from this concentration onward, the permeation enhancement increased sharply until a concentration of 0.2 mM, which corresponds to the CMC of DDM. Above this, the curve flattens and shows a lower slope until 0.5 mM, which was the highest concentration included in this study. 3.4. In Situ Intestinal Loop Studies in Rats. To assess the in vivo relevance of the results achieved in vitro, closed intestinal 2250
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
Figure 5. In situ intestinal loop studies. (A) Absorption of FD4 from a closed colon loop in the presence (filled symbols) and absence (open symbols) of 2 mM DDM in TM (●, ○), FaSSIFmod (■, □), or FeSSIFmod6.5 (▲, △). The uptake of FD4 was monitored by taking blood samples from the jugular vein over 2 h. Data are presented as the mean ± SD (n = 10 for DDM in TM, n = 6 for DDM in FaSSIFmod and FeSSIFmod6.5, and n = 3 for controls). The absorption enhancement achieved by 2 mM DDM in TM was determined to be statistically significant; all others did not show a statistical difference compared to the TM control. (B) FD4 bioavailability calculated based on the results obtained from (A).
simplified models of the more complex in vivo system, they have been successfully used for in vitro dissolution studies for several years.22 Considering that the concentration of lecithin and NaTC resembles that reported in the human small intestine in fasted and fed states, an interaction with colloidal structures in the intestine provides one potential reason for the often observed mismatch between in vitro potency of permeation enhancers and their actual performance in vivo.3,15 In addition, the impact of bile salts on the performance of permeation enhancers may well be multifaceted. While NaTC concentrations of 3 and 15 mM in fasted and fed states, respectively, are good approximations of the in vivo conditions,9 their concentrations change within the GI tract and also vary among individuals; values have been reported from 0 to 15 mM and from 0.5 to 37 mM21,23 for fasted and fed states, respectively. This subject-to-subject variability could conceivably give rise to a high variation in the permeation enhancing effect between individual patients and consequently in the bioavailability of the coformulated drug. In order to quantify the amount of lost maltosides due to interactions with mixed micelles, maltosides were mixed with FaSSIFmod/FeSSIFmod6.5 and the monomer was subsequently separated from the complex by mounting a dialysis membrane, instead of the Caco-2 monolayer, between the two chambers of the D/P system.24 In accordance with our hypothesis from the Caco-2 permeation study, this setup confirmed that the presence of NaTC/lecithin reduces the amount of maltoside monomer in the apical compartment in a concentrationdependent manner. Considering the situation in the intestine, three different forms of maltosides are theoretically possible: the monomer; a complex composed of bile acids, lipids, and maltosides; and a maltoside micelle consisting only of maltosides (Figure 6). As our test setup enables only the separation of monomers from all other possible colloidal structures, it was not possible to determine whether pure maltoside micelles exist at all or if all formed micelles contain NaTC/lecithin to some extent. It is most likely that their existence will depend on the concentration of maltoside or, more precisely, the ratio between NaTC/lecithin and the maltoside. In terms of permeation enhancement, it has been demonstrated that maltoside mixed micelles do not possess permeation enhancing properties. However, the question still
Figure 6. Scheme depicting the different forms of maltosides that could be present in the intestine.
remained as to whether only maltoside monomers or also maltoside micelles might act as permeation enhancers. To this end, the FD4 permeation through Caco-2 cells in the presence of numerous concentrations of DDM was investigated in TM. Up until the CMC, the permeated amount of FD4 increased rapidly with increasing concentrations of DDM. In this concentration range, only the monomer concentration increased, confirming that maltoside monomers do exert a permeation enhancing effect and that its strength is proportional to the concentration. However, also above the CMC the permeated amount of FD4 continued to increase with increasing concentration of DDM, albeit to a lesser extent. On the basis of this observation, it appears that both maltoside monomers and maltoside micelles contribute to the permeation enhancement, probably via different mechanisms. In this context, it is important to note that the potency of different permeation enhancers cannot be assessed solely by examining the amount of monomer/micelle. Permeation enhancement is a far more complex process that is also greatly influenced by the structure of the enhancer itself. Brayden et al. recently compared the permeation enhancing potential of mediumchain fatty acids of different chain lengths (C8−C12).25 They found that a longer fatty acid chain (and thus a lower CMC) confers higher permeation enhancing potential but also correlates with decreased in vitro cell viability. Conceivably, this association might be the same for maltosides, where we showed increasing enhancement potential with increasing chain length. Even though the use of maltosides as permeation enhancers has been investigated over several years,8,16,26,27 a precise 2251
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics
importance of using biorelevant conditions, such as the appropriate media, should be emphasized in order to ensure a better correlation between in vitro and in vivo studies and to enable an early and more accurate assessment of promising substrates.
mechanism for this effect has yet to be elucidated. Measuring the TEER recovery after exposing tissues/cells to permeation enhancers can, however, give some fundamental understanding about the principal mode of action. A reversible reduction of the TEER generally indicates permeation enhancement via an interaction with tight junctions, whereas an irreversible TEER reduction may be suggestive of a more pronounced membrane solubilization component. Following this concept, Tirumalasetty et al. investigated the TEER of Caco-2 cells after a 6 h incubation with different concentrations of DDM (from 0.2 to 1 mM).7 They observed a very good recovery (>80%) up to a concentration of 0.6 mM, but the recovery was weak for concentrations exceeding this. This result was confirmed by our own studies, showing a TEER recovery of >90% after incubating the cells for 2−6 h with 0.5 mM DDM in TM. Given the surfactant nature of DDM, it should not be surprising that high concentrations lead to membrane perturbation effects. However, it could be demonstrated that lower concentrations enhance the permeation of drugs mainly via the paracellular route. Focusing on the results of in vitro permeation studies, where the highest concentration of DDM in TM was 0.5 mM, membrane solubilization is unlikely to be the predominant mechanism. We therefore hypothesized that maltoside micelles also enhance the permeation via tight junction modulation and not via a solubilization of the cell membrane. Recently, an interesting concept has been proposed for the permeation enhancing mechanism of sodium caprate, a medium-chain fatty acid, suggesting the displacement of specific tight junction proteins from lipid rafts.28 Lipid rafts, defined as small (10−200 nm), heterogeneous, highly dynamic, sterol- and sphingolipid-enriched domains that compartmentalize cellular processes,29 were suggested to stabilize the association of different tight junction-specific proteins.30,31 Considering the structural similarity between sodium caprate and alkyl-maltosides, a further investigation should examine an interaction with lipid rafts as a possible mechanism for the permeation enhancing effect of maltosides. To confirm the findings from the in vitro studies, in situ closed loop studies were carried out. Maltosides, like other permeation enhancers, were shown to exhibit a much greater effect in the colon than in the small intestine.8,16,17 The reason for this difference is not fully understood. However, after confirming these observations by preliminary experiments (data not shown), the colon was used for all loop studies. In accordance with in vitro studies, FaSSIFmod and FeSSIFmod6.5 both showed the same drastic effect on the permeation enhancing potential of DDM in situ. The 20-fold permeation enhancement seen in TM was completely abrogated in the presence of NaTC/lecithin, meaning that the bioavailability of a coadministered drug will depend to a great extent on the concentrations of NaTC/lecithin in the intestine. Given their large local and interpersonal variation and their change after food intake, ensuring a reliable and reproducible effect represents a challenge that needs to be addressed when using maltosides as intestinal permeation enhancers. To summarize the results presented in this work, we successfully demonstrated the impact of mixed micelles in the intestine on the permeation enhancing effect of maltosides. This study provides a systematic in vitro investigation of this effect, including media mimicking the fed state (FeSSIF) as well as an in situ study to demonstrate its relevance in vivo. Given the increasing interest in the application of permeation enhancers in the oral delivery of peptides and proteins, the
■
AUTHOR INFORMATION
Corresponding Author
*Tel/Fax: +81-72-866-3125/3126. E-mail: shinji@pharm. setsunan.ac.jp. Notes
The authors declare the following competing financial interest(s): K.G., S.T.B. and B.L.P. are employees of Novo Nordisk A/S.
■
ACKNOWLEDGMENTS The authors are grateful for the financial support received from Novo Nordisk A/S. The authors thank Naoya Hiroki, Kentaro Iwashige, and Noriaki Kashimoto for assisting with in situ loop studies and culturing Caco-2 cells.
■
REFERENCES
(1) Swenson, E. S.; Curatolo, W. J. (C) Means to enhance penetration: (2) Intestinal permeability enhancement for proteins, peptides and other polar drugs: mechanisms and potential toxicity. Adv. Drug Delivery Rev. 1992, 8, 39−92. (2) Walsh, E. G.; Adamczyk, B. E.; Chalasani, K. B.; Maher, S.; O’Toole, E. B.; Fox, J. S.; Leonard, T. W.; Brayden, D. J. Oral delivery of macromolecules: rationale underpinning gastrointestinal permeation enhancement technology (GIPET). Ther. Delivery 2011, 2, 1595− 1610. (3) Aungst, B. J. Absorption enhancers: applications and advances. AAPS J. 2012, 14, 10−18. (4) Maher, S.; Brayden, D. J. Overcoming poor permeability: translating permeation enhancers for oral peptide delivery. Drug Discovery Today: Technol. 2012, 9, 113−119. (5) Ahsan, F.; Arnold, J. J.; Yang, T.; Meezan, E.; Schwiebert, E. M.; Pillion, D. J. Effects of the permeability enhancers, tetradecylmaltoside and dimethyl-beta-cyclodextrin, on insulin movement across human bronchial epithelial cells (16HBE14o−). Eur. J. Pharm. Sci. 2003, 20, 27−34. (6) Yang, T.; Arnold, J. J.; Ahsan, F. Tetradecylmaltoside (TDM) enhances in vitro and in vivo intestinal absorption of enoxaparin, a low molecular weight heparin. J. Drug Targeting 2005, 13, 29−38. (7) Tirumalasetty, P. P.; Eley, J. G. Evaluation of dodecylmaltoside as a permeability enhancer for insulin using human carcinoma cells. J. Pharm. Sci. 2005, 94, 246−255. (8) Petersen, S. B.; Nolan, G.; Maher, S.; Rahbek, U. L.; Guldbrandt, M.; Brayden, D. J. Evaluation of alkylmaltosides as intestinal permeation enhancers: comparison between rat intestinal mucosal sheets and Caco-2 monolayers. Eur. J. Pharm. Sci. 2012, 47, 701−712. (9) Galia, E.; Nicolaides, E.; Horter, D.; Lobenberg, R.; Reppas, C.; Dressman, J. B. Evaluation of various dissolution media for predicting in vivo performance of class I and II drugs. Pharm. Res. 1998, 15, 698− 705. (10) Ingels, F.; Beck, B.; Oth, M.; Augustijns, P. Effect of simulated intestinal fluid on drug permeability estimation across Caco-2 monolayers. Int. J. Pharm. 2004, 274, 221−232. (11) Kataoka, M.; Masaoka, Y.; Sakuma, S.; Yamashita, S. Effect of food intake on the oral absorption of poorly water-soluble drugs: in vitro assessment of drug dissolution and permeation assay system. J. Pharm. Sci. 2006, 95, 2051−2061. (12) Markopoulos, C.; Thoenen, F.; Preisig, D.; Symillides, M.; Vertzoni, M.; Parrott, N.; Reppas, C.; Imanidis, G. Biorelevant media for transport experiments in the Caco-2 model to evaluate drug 2252
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
Article
Molecular Pharmaceutics absorption in the fasted and the fed state and their usefulness. Eur. J. Pharm. Biopharm. 2014, 86, 438−448. (13) Fossati, L.; Dechaume, R.; Hardillier, E.; Chevillon, D.; Prevost, C.; Bolze, S.; Maubon, N. Use of simulated intestinal fluid for Caco-2 permeability assay of lipophilic drugs. Int. J. Pharm. 2008, 360, 148− 155. (14) Frank, K. J.; Westedt, U.; Rosenblatt, K. M.; Holig, P.; Rosenberg, J.; Magerlein, M.; Brandl, M.; Fricker, G. Impact of FaSSIF on the solubility and dissolution-/permeation rate of a poorly watersoluble compound. Eur. J. Pharm. Sci. 2012, 47, 16−20. (15) Tippin, T. K.; Thakker, D. R. Biorelevant refinement of the Caco-2 cell culture model to assess efficacy of paracellular permeability enhancers. J. Pharm. Sci. 2008, 97, 1977−1992. (16) Masahiro, M.; Yoko, K.; Kanji, T.; Shozo, M. Assessment of enhancing ability of medium-chain alkyl saccharides as new absorption enhancers in rat rectum. Int. J. Pharm. 1992, 79, 159−169. (17) Uchiyama, T.; Sugiyama, T.; Quan, Y. S.; Kotani, A.; Okada, N.; Fujita, T.; Muranishi, S.; Yamamoto, A. Enhanced permeability of insulin across the rat intestinal membrane by various absorption enhancers: their intestinal mucosal toxicity and absorption-enhancing mechanism of n-lauryl-beta-D-maltopyranoside. J. Pharm. Pharmacol. 1999, 51, 1241−1250. (18) Kataoka, M.; Masaoka, Y.; Yamazaki, Y.; Sakane, T.; Sezaki, H.; Yamashita, S. In vitro system to evaluate oral absorption of poorly water-soluble drugs: simultaneous analysis on dissolution and permeation of drugs. Pharm. Res. 2003, 20, 1674−1680. (19) Kataoka, M.; Itsubata, S.; Masaoka, Y.; Sakuma, S.; Yamashita, S. In vitro dissolution/permeation system to predict the oral absorption of poorly water-soluble drugs: effect of food and dose strength on it. Biol. Pharm. Bull. 2011, 34, 401−407. (20) Kataoka, M.; Sugano, K.; da Costa, M. C.; Wong, J. W.; Jones, K. L.; Masaoka, Y.; Sakuma, S.; Yamashita, S. Application of dissolution/permeation system for evaluation of formulation effect on oral absorption of poorly water-soluble drugs in drug development. Pharm. Res. 2012, 29, 1485−1494. (21) Holm, R.; Mullertz, A.; Mu, H. Bile salts and their importance for drug absorption. Int. J. Pharm. 2013, 453, 44−55. (22) Klein, S. The use of biorelevant dissolution media to forecast the in vivo performance of a drug. AAPS J. 2010, 12, 397−406. (23) Sugano, K. Biopharmaceutics Modeling and Simulations: Theory, Practice, Methods, and Applications, 1st ed.; John Wiley Sons, Inc.: Hoboken, NJ, 2012. (24) Yano, K.; Masaoka, Y.; Kataoka, M.; Sakuma, S.; Yamashita, S. Mechanisms of membrane transport of poorly soluble drugs: role of micelles in oral absorption processes. J. Pharm. Sci. 2010, 99, 1336− 1345. (25) Brayden, D. J.; Gleeson, J.; Walsh, E. G. A head-to-head multiparametric high content analysis of a series of medium chain fatty acid intestinal permeation enhancers in Caco-2 cells. Eur. J. Pharm. Biopharm. 2014, 88, 830−839. (26) Deshmukh, D. D.; Nagilla, R.; Ravis, W. R.; Betageri, G. V. Effect of dodecylmaltoside (DDM) on uptake of BCS III compounds, tiludronate and cromolyn, in Caco-2 cells and rat intestine model. Drug Delivery 2010, 17, 145−151. (27) Yamamoto, A.; Okagawa, T.; Kotani, A.; Uchiyama, T.; Shimura, T.; Tabata, S.; Kondo, S.; Muranishi, S. Effects of different absorption enhancers on the permeation of ebiratide, an ACTH analogue, across intestinal membranes. J. Pharm. Pharmacol. 1997, 49, 1057−1061. (28) Sugibayashi, K.; Onuki, Y.; Takayama, K. Displacement of tight junction proteins from detergent-resistant membrane domains by treatment with sodium caprate. Eur. J. Pharm. Sci. 2009, 36, 246−253. (29) Pike, L. J. Rafts defined: a report on the Keystone Symposium on lipid rafts and cell function. J. Lipid Res. 2006, 47, 1597−1598. (30) Nusrat, A.; Parkos, C. A.; Verkade, P.; Foley, C. S.; Liang, T. W.; Innis-Whitehouse, W.; Eastburn, K. K.; Madara, J. L. Tight junctions are membrane microdomains. J. Cell Sci. 2000, 113, 1771−1781. (31) Lambert, D.; O’Neill, C. A.; Padfield, P. J. Depletion of Caco-2 cell cholesterol disrupts barrier function by altering the detergent
solubility and distribution of specific tight-junction proteins. Biochem. J. 2005, 387, 553−560.
2253
DOI: 10.1021/mp500776a Mol. Pharmaceutics 2015, 12, 2245−2253
