Exploring the fate of liposomes in the intestine by dynamic in vitro lipolysis

International Journal of Pharmaceutics 437 (2012) 253–263

Contents lists available at SciVerse ScienceDirect

International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm

Exploring the fate of liposomes in the intestine by dynamic in vitro lipolysis Johannes Parmentier a,1 , Nicky Thomas b,1 , Anette Müllertz c , Gert Fricker a , Thomas Rades b,d,∗ a

Department of Pharmaceutical Technology and Biopharmacy, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Germany New Zealand’s National School of Pharmacy, University of Otago, Dunedin, New Zealand c Bioneer:FARMA, Department of Pharmacy, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark d Department of Pharmacy, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark b

a r t i c l e

i n f o

Article history: Received 5 June 2012 Received in revised form 6 August 2012 Accepted 9 August 2012 Available online 19 August 2012 Keywords: Scanning ion occlusion sensing Dynamic light scattering Danazol Bile salts Liposome stability In vitro lipolysis

a b s t r a c t Liposomes are generally well tolerated drug delivery systems with a potential use for the oral route. However, little is known about the fate of liposomes during exposure to the conditions in the gastro-intestinal tract (GIT). To gain a better understanding of liposome stability in the intestine, a dynamic in vitro lipolysis model, which so far has only been used for the in vitro characterisation of other lipid-based drug delivery systems, was applied to different liposomal formulations. Liposome size and phospholipid (PL) digestion were determined as two markers for liposome stability. In addition, the effect of PL degradation on the ability to maintain liposomally incorporated danazol in solution during lipolysis was evaluated in order to address the feasibility of liposomes designed for oral administration. Rate and extend of hydrolysis of PLs mediated by pancreatic enzymes was determined by titration and HPLC. Size of liposomes was determined by dynamic light scattering during incubation in lipolysis medium (LM) and during lipolysis. SPC-based (soy phosphatidylcholine) liposomes were stable in LM, whereas for EPC-3-based (hydrated egg phosphatidylcholine) formulations the formation of aggregates of around 1 ␮m in diameter was observed over time. After 60 min lipolysis more than 80% of PLs of the SPC-liposomes were digested, but dependent on the liposome concentration only a slight change in size and size distribution could be observed. Although EPC-3 formulations did form aggregates during lipolysis, the lipids exhibited a higher stability compared to SPC and only 30% of the PLs were digested. No direct correlation between liposome integrity assessed by vesicle size and PL digestion was observed. Danazol content in the liposomes was around 5% (mol/mol danazol/total lipid) and hardly any precipitation was detected during the lipolysis assay, despite pronounced lipolytic degradation and change in vesicle size. In conclusion, the tested dynamic in vitro lipolysis model is suitable for the assessment of liposome stability in the intestine. Furthermore, liposomes might be a useful alternative to other lipid based delivery systems for the oral delivery of poorly soluble drugs. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Soon after their discovery by Bangham et al. (1965) liposomes were proposed as suitable carriers for the oral administration of drugs with low oral bioavailability (Gregoriadis, 1976). Their high versatility with respect to composition, preparation method and physicochemical properties, their good biocompatibility and ability to encapsulate both hydrophilic and lipophilic drugs are evident advantages of liposomes as drug delivery systems (Szoka and Papahadjopoulos, 1980; Jesorka and Orwar, 2008). Orally applied liposomes have mostly been used to improve the bioavailability

∗ Corresponding author at: Department of Pharmacy, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark. Tel.: +45 35 33 60 00. E-mail address: [email protected] (T. Rades). 1 Both authors contributed equally to the work. 0378-5173/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijpharm.2012.08.018

of BCS class III drugs, like proteins, by protecting them against degradation in the gastro-intestinal tract (GIT) and by improving their intestinal absorption (Patel and Ryman, 1976; Rowland and Woodley, 1981; Fricker and Drewe, 1996; Singh et al., 2008; Werle and Takeuchi, 2009). However, more recently, the possibility of liposomes for the oral delivery of small, lipophilic drugs, e.g. curcumin and progesterone, was investigated to increase drug solubility and, eventually, bioavailability (Bayomi et al., 1998; Potluri and Betageri, 2006; Takahashi et al., 2009; Fricker et al., 2010). One of the major drawbacks of conventional liposomes used for oral delivery is their instability in the GIT due to hydrolysis of phospholipid (PL) by lipases and micelle formation with bile salts (Chiang and Weiner, 1987a). Several approaches have been undertaken to overcome these problems, including polymer coating of the vesicles, use of membrane-spanning tetraether lipids or lipids with a phase transition above body temperature (Takeuchi et al., ´ c´ et al., 2001; Werle 1994; Muramatsu et al., 1996; Filipovic-Grci

254

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

and Takeuchi, 2009; Parmentier et al., 2011). Liposomes with a PL membrane in the gel state are considered as less susceptible against lipolytic degradation and micellisation by bile salts. Indeed, in vitro results of several studies suggest a higher resistance of liposomes composed of PLs with only fully saturated fatty acids (FAs) against both bile salts and pancreatin, compared to liposomes containing PLs with unsaturated FAs (Rowland and Woodley, 1980; Aramaki et al., 1993; Kokkona et al., 2000). However, liposomes composed of PLs with a relatively high phase transition temperature did not show necessarily advantages in in vivo studies compared to liposomes in the fluid state (Patel et al., 1982; Chiang and Weiner, 1987b). Since the mid 1970s, along with the oral use of liposomes, stability of liposomal formulations in simulated GI fluids has been investigated. In the beginning, research focussed on the interaction of bile salts with PL vesicles, but later the influence of low pH values, phospholipases alone (Rowland and Woodley, 1980; Taira et al., 2004) and in combination with bile salts (Patel et al., 2000) was also examined. To facilitate experimental procedures and to guarantee reproducibility of the results, purified bile salts and phospholipases were often used in in vitro stability assays. These models however, do not represent the physiological mixture of bile salts, PLs and other enzymes present in the human intestine and might lead to an inaccurate estimation of liposome stability in the human GIT. Even the use of animal models might lead to a false prediction of drug bioavailability, since intestinal anatomy and physiology from common laboratory animals differ from those in humans. As an example, bile is not stored in a gall bladder in rats, but is released continuously compared to the release of bile observed in the presence of food in minipigs, dogs and humans (Grove et al., 2007; Kararli, 1995). Consequently, physiological differences between species need to be considered when trying to extrapolate animal data to those obtained in humans. Ethical considerations and costs of animal studies further complicate their use for example for drug formulation screenings, especially if dogs and monkeys are used. In vitro models simulating the influence of the physiological conditions in the human gut on the fate of lipid-based drug delivery systems containing BCS class II drugs have gained more interest in recent years, since the number of these compounds is increasing as a result of high-throughput screening utilised in drug development programs (Lipinski et al., 1997; Fatouros and Mullertz, 2008; Chakraborty et al., 2009). To study digestion of the lipidic drug carrier and its influence on drug solubilisation, in vitro lipolysis models, simulating intestinal digestion, have been developed (Larsen et al., 2011). These comprise lipolytic enzymes and co-enzymes representing the physiological conditions in the intestine, and are carried out in a pH-stat, at a constant pH in order to assess the amount of hydrolyzed FAs. Zangenberg et al. (2001a,b) developed a dynamic in vitro lipolysis model employing a porcine bile extract and porcine pancreatin as lipase source. The continuous addition of Ca2+ used in this dynamic in vitro lipolysis model allows controlling the rate of lipolysis. It seems reasonable to apply this dynamic lipolysis model to the assessment of liposomes intended for oral use, since it resembles the conditions in the human gut closer than the previous static approaches for liposome stability testing. Porcine pancreatin also contains phospholipases, which are the relevant enzymes for PL hydrolysis. The present study, to our knowledge, is the first to investigate liposome stability, with respect to size and size distribution, during dynamic in vitro lipolysis. Influence of the phase transition temperature of PLs used for liposome preparation and of the presence of cholesterol on size and size distribution of liposomes was tested in the model. Additionally, digestion of phosphatidylcholine (PC) during lipolysis was monitored by HPLC. To conclude on the benefit of liposomes for the oral delivery of poorly water-soluble

Table 1 Composition in % (mol/mol) of the tested liposomal formulations. SPC SPC SPC/Chol EPC-3 SPC/Dan EPC-3/Dan

EPC-3

DPPG-Na

95

5 5 5

100 60

Cholesterol

Danazol

40

85 75

10 20

drugs, danazol, a BCS class II compound, was incorporated into liposomes and its distribution between the aqueous phase and the precipitate formed during lipolysis was investigated. Beside size determination by dynamic light scattering (DLS), Scanning Ion Occlusion Sensing (SIOS) was used. SIOS is based on single particle counting, whereby charged particles are driven by an electrical current through a tuneable nanopore. SIOS records a discrete blockade signal, which is proportional to the volume of the particle (GarzaLicudine et al., 2010; Willmott and Bauerfeind, 2010). Since the SIOS technology determines the size of single particles unlike DLS, where a size distribution is calculated from the intensity fluctuation of light scattered by particles undergoing Brownian motion, the techniques can be seen as complementary. 2. Materials and methods 2.1. Materials Soy PC (SPC) (EpikuronTM 200) containing phosphatidylcholine (min. 92%), lyso-phosphatidylcholine (max. 3%), other PLs (max. 2%) and FA (approx. 1%) was obtained from Cargill (Hamburg, Germany). Hydrated egg PC (EPC-3) (purity > 98%) and dipalmitoyl phosphatidylglycerol sodium salt (DPPG-Na) (purity > 98%) were provided by Lipoid (Ludwigshafen, Germany). Danazol was obtained from Sanofi Aventis (Fawdon, UK). Cholesterol, pancreatin from porcine pancreas (P-1625), bile extract from porcine mucosa (B-8631), 4-bromobenzeneboronic acid (4-BBBA), trizma maleate, calcium chloride, and sodium hydroxide were purchased from Sigma–Aldrich (St. Louis, MO, US). Acetonitril, ethanol, methanol and sodium chloride were obtained from Merck (Darmstadt, Germany). All other chemicals were obtained in the highest purity from the usual commercial sources. 2.2. Liposome preparation Composition of the tested liposomal formulations is displayed in Table 1. Liposomes were prepared by the film method followed by extrusion. Briefly, lipids and danazol were dissolved in chloroform at the desired molar ratio and mixed in a round bottom flask. The solvent was then evaporated under vacuum. To remove any solvent traces, lipid films were dried under high vacuum for an additional 12 h. Subsequently tris buffer pH 7.5 (TB) (tris base 50 mM and NaCl 120 mM) was added to the film to achieve a final total lipid concentration of 100 mM and lipids were hydrated for 24 h at room temperature. The resulting multilamellar vesicles (MLVs) were extruded five times through an 800-nm followed by 20 times through a 200-nm membrane either at room temperature (SPC-based liposomes) or at 55 ◦ C (EPC-3-based liposomes) using a LipexTM extruder (Northern Lipids, Burnaby, BC, Canada). 2.3. Liposome characterisation 2.3.1. Particle size of liposomes Size and size distribution of the liposomal formulation were determined after appropriate dilution in TB by DLS with a Zetasizer

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263 Table 2 Composition of lipolysis medium (LM) with and without pancreatin. Component

LM without pancreatin

LM with pancreatin

Bile salts SPC NaCl Trizma maleate Porcine pancreatin

5 mM 1.25 mM 150 mM 2 mM

5 mM 1.25 mM 150 mM 2 mM 2.5 g/100 ml

Nano ZS (Malvern, Worcestershire, UK) set at 37 ◦ C in automatic mode. 2.3.2. Determination of danazol in liposomes Final danazol content in the corresponding formulations after extrusion was examined by HPLC as described below. 2.4. Size stability of liposomes in lipolysis medium Liposomes were added to pre-warmed lipolysis medium (LM) without pancreatin in a clear cuvette to obtain a final lipid concentration of 1 mM and 5 mM, respectively. Both size measurements and incubation were carried out in a Zetasizer Nano ZS (Malvern, Worcestershire, UK) set at 37 ◦ C. Immediately and after 5, 10, 25, 40 and 60 min incubation liposome size distribution was determined and graphs were generated using intensity weighted size distribution data. 2.5. Dynamic in vitro lipolysis LM (without pancreatin) was prepared according to Zangenberg et al. (2001a) with minor adjustments. The composition of the LM is summarised in Table 2. Bile salts (5 mM) and SPC (1.25 mM) were added to a trizma maleate/saline buffer (2 mM/150 mM, pH 6.5) and stirred over night at room temperature to form micelles. The micelles were allowed to equilibrate at 37 ◦ C and subsequently, the pH was again adjusted to 6.5. Dynamic in vitro digestion of the liposomal formulations was carried out by a Titrando 842 system (Metrohm, Herisau, Switzerland) equipped with a Metrohm Titrino 744 pH-meter and the corresponding Tiamo 1.3 software. 83.3 ml LM were maintained at 37 ◦ C under stirring in a temperature controlled vessel. 3.0 g pancreatin were dispersed in 20.0 ml LM and the mixture was centrifuged (Centrifuge 5810 R, Eppendorf, Hamburg, Germany) at 3250 rcf (at rmax ) for 10 min at 37 ◦ C. 16.7 ml of the supernatant were added to the pre-warmed LM in the reaction vessel and the pH was adjusted to 6.5. Lipolysis was started by the injection of 5 ml liposomal dispersion adjusted to pH 6.5 with a total lipid concentration of 10 mM or 50 mM. The digestion was controlled by the continuous addition of CaCl2 solution (0.015 mmol/min) and the pH was kept constant at 6.5 throughout the 60 min experimental time by titration with 1 M NaOH. As blank control 5.0 ml buffer at pH 6.5 were injected instead of the liposomal dispersion (blank lipolysis). Samples of 1.0 ml were withdrawn immediately after injection of the liposomes and after 5, 10, 25, 40 and 60 min of digestion and evenly split into two aliquots. To 0.5 ml of the sample 2.5 ␮l 1 M 4BBBA in methanol were added to stop lipase activity. Samples were immediately frozen at −20 ◦ C and then kept at −80 ◦ C until HPLC analysis. The remaining 0.5 ml were used for DLS size measurement without any further dilution as described below. This time, no 4BBBA was added to avoid interference of precipitates, which are formed after addition of 4-BBBA, with the DSL measurement. Since size was determined directly after sample taking and measurement time was less than 2 min no significant influence on the results was expected. Additionally, for experiments with 5.0 mM lipid concentration, size was also determined by SIOS using a qNanoTM

255

(Izon, Christchurch, NZ). For the present study, elastomeric, resizable nanopores with a nominal resolution of 100–400 nm were used. A Delta-stretch of 45.00 mm was applied at least 2 h prior to the experiments to allow equilibration and was kept constant throughout the experiment. Voltage was set to 0.22 mV resulting in a baseline current around 120 nA. For each nanopore a calibration curve was established using polystyrene particles (100 nm, 220 nm, 400 nm) provided by Izon. For each sample a minimum of 200 counting events was collected and the data was processed by qNano software (version V2.0). To investigate the influence of PL digestion on the allocation of danazol distributed between the aqueous liposomal phase and the precipitate, 5.0 ml samples were withdrawn after 0, 30 and 60 min. Samples were mixed with 2.5 ␮l of 1 M 4-BBBA in methanol and used for density centrifugation. The samples were underlayed with 3.0 ml of 20% (m/v) sucrose solution in water in an Ultra-ClearTM centrifuge tube (14 mm × 89 mm) (Beckmann, Palo Alto, CA, US). The tubes were centrifuged in a L-80 Beckmann Ultracentrifuge (Beckmann, Palo Alto, CA, US) equipped with a SW 41 Ti swinging bucket rotor at 107,000 rcf (at rmax ) for 25 min at 37 ◦ C. Pellet and supernatant were separated and used for HPLC analysis of PC and danazol as described below. In order to validate the centrifugation method, blank SPC liposomes were centrifuged as described above and the SPC content was determined in the pellet after centrifugation. The SPC content in the pellet determined by HPLC was below the detection limit. Additionally, danazol was dispersed in TB and the particle size was reduced by sonication for 20 min at 900 W with a 60 s on/off pulse by the use of a JY92-II D Ultrasonicator (Scientz Biotechnology, Ningbo, China). Subsequently, particles were filtered through an 800-nm polycarbonate membrane and 1.0 ml of 0.125 mg/ml danazol were centrifuged as described above. Recovery rate of danazol in the pellet was 93.2% (±1.5% (SEM)). 2.6. Determination of PC and danazol by HPLC For determination of precipitated danazol, pellets from the ultracentrifugation were redispersed in 1.0 ml ethanol and dissolved by 10 min sonication in a bath-type sonicator (Sonorex Super RK 106, Bandelin, Berlin, Germany). Insoluble components were separated prior to HPLC analysis by centrifugation at 12,100 rcf (at rmax ) in a MiniSpin® centrifuge (Eppendorf, Hamburg, Germany). For determination of danazol in the supernatant and of PLs in the LM, samples were diluted 1:10 with ethanol, incubated for 30 min at 37 ◦ C to ensure full dissolution of the PLs and danazol and subsequently centrifuged as described above. 50 ␮l of the samples were injected with an autosampler in an Agilent 1200 series system (Agilent Technologies, Palo Alto, CA, US) equipped with a SecurityGuardTM (4 mm × 3 mm) pre-column (Phenomenex, Torrance, CA, US) and a 5 ␮m C18 Gemini® -NX column (4.6 mm × 250 mm). For concentration determination the area of the PC and danazol peak, respectively, were compared against calibration curves obtained either by UV (SPC and danazol) or by evaporative light scattering (EPC-3) detection. An overview of the HPLC conditions for the three different substances is given in Table 3. 3. Results 3.1. Liposome properties An overview of the properties of the tested liposomal formulations is given in Table 4. Most of the formulations had a Z-average between 130 nm and 160 nm and showed a narrow size distribution. However, SPC/Dan liposomes were smaller, with a size of

256

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

Fig. 1. Size distribution of liposomes during incubation in LM at 5 mM (solid, black line) or 1 mM (dotted, grey line) total lipid concentration measured by DLS. Values represent means with n = 3.

Table 3 HPLC conditions for SPC, EPC-3 and danazol assays. Flow (ml/min)

SPC EPC-3 Danazol

2.0 1.5 1.0

Temp (◦ C)

45 45 45

Eluent (%) (v/v)

Detector

ACN

MeOH

80

20 100 80

around 105 nm. In general, liposomes containing the hydrated EPC had a broader size distribution compared to liposomes based on SPC. Danazol content in the liposomes was independent from the lipid composition around 5% of the total lipid content.

H2 O

20

TFA

Type

Settings

0.05

UV ELSD UV

205 nm N2 /60 ◦ C/50 psi 280 nm

the size stability of the vesicles. EPC-3 based liposomes exhibited a slight decrease in size at the higher concentration, but an increase over time at lower lipid concentration. 3.3. Characterisation of lipolysis medium during lipolysis

3.2. Size stability of liposomes in lipolysis medium Before the fate of liposomes during lipolysis was investigated, the influence of LM without pancreatin on the different liposomal formulations was assessed. As can be seen in Fig. 1 liposomes containing either only SPC or SPC and cholesterol in a mixture behaved similar and the total lipid concentration had only little influence on

PL content and size characteristics of LM during in vitro lipolysis without the addition of any liposomes were evaluated. Over 60 min the SPC content in the medium was reduced to less than 10% of the initial amount (Fig. 2 A). Directly from the beginning, but more clearly with progressing lipolysis, larger particles with a size around 1 ␮m were detectable by DLS (Fig. 2B).

Table 4 Z-Average, PDI and danazol content determined by HPLC in % (mol/mol) of the tested liposomal formulations and final danazol content in the reaction vessel. Values represent means (±SEM) with n = 3.

SPC SPC/Chol EPC-3 SPC/Dan EPC-3/Dan

Z-Average (nm)

PDI

141.6 (±6.40) 157.0 (±0.20) 134.5 (±12.37) 104.9 (±0.50) 143.69 (±16.50)

0.126 (±0.0015) 0.093 (±0.0035) 0.194 (±0.0345) 0.117 (±0.0080) 0.197 (±0.0885)

Danazol cont. (%)

5.62 (±0.015) 6.75 (±0.045)

Final danazol amount in LM (mg/100 ml) 1 mM

5 mM

1.28 1.26

6.42 6.32

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

257

Fig. 2. (A) SPC content in % (black, solid line, right axis) in the LM and NaOH consumption in ml (grey, dashed line, left axis) during lipolysis of LM without addition of liposomes. Values represent means (±SEM) with n = 3. Mean values of three runs from NaOH consumption were fitted by a LOWESS Smooth with five points per window. (B) Size distribution of LM during lipolysis without addition of liposomes measured by DLS. Values represent means with n = 3.

3.4. Lipolysis of liposomes – influence of lipid composition HPLC analysis of remaining PL and NaOH consumption, compared to blank lipolysis, during in vitro lipolysis revealed that hydrolysis of PLs occurred for all tested liposomal formulations (Fig. 3A). The final PC amount remaining after 60 min lipolysis was around or below 20% for the two SPC-based formulations. The corresponding NaOH consumption was almost linear during the entire lipolysis. Formation of lyso-PC from EPC-3 was slower compared to SPC, with less than 50% of the PLs hydrolysed after 60 min. As shown in Fig. 3B both SPC based liposomal formulations were rather stable in size over 60 min. Lipid digestion of the EPC-3 liposomes did not lead to substantial formation of aggregates. However, the liposomes were more heterogeneous in size compared to SPC liposomes. In general, size determination with SIOS displayed a narrower size distribution. Whereas particle size measured by SIOS showed bigger particles for the SPC liposomes compared to DLS, the median size of EPC-3 liposomes was comparable for both size determination techniques. 3.5. Lipolysis of liposomes – influence of lipid concentration In Fig. 4 influence of the lipid concentration is shown exemplarily for the SPC and EPC-3 formulations. Whereas for SPC-based liposomes the degree of hydrolysis after 60 min lipolysis was comparable for both tested concentrations (1 mM: 4.7% (±0.18%) and 5 mM: 16.3% (±0.58) of final lipid concentration), hydrolysis of the saturated PL turned out to be concentration dependent (1 mM: 7.0% (±0.47%) and 5 mM: 49.7% (±5.82) final lipid concentration). Interestingly, NaOH consumption at 1 mM lipid concentration was similar to the consumption at 5 mM lipid concentration for both types of PLs although the absolute quantity of PL hydrolysed was higher for the higher lipid concentration. Formation of bigger particles over time could be observed at 1 mM PL concentration for both formulations (Fig. 4B). EPC-3 liposomes showed the same trend also at the higher concentration,

but to a lower extent. However, the 5 mM SPC liposomes hardly exhibited any change in their size distribution measured by DLS. 3.6. Lipolysis of liposomes – kinetics evaluation NaOH consumption rate and digestion rate of PLs to lyso PLs determined by HPLC were fitted with both pseudo-zero order (Eqs. (1) and (2)) and pseudo-first order kinetics (Eqs. (3) and (4)): FA(t) = kt,

(1)

PL(t) = PL(0) − kt, FA(t) = FA(end) · (1 − e PL(t) = PL(0) · e−kt ,

(2) −kt

),

(3) (4)

where FA(t) is the concentrations of fatty acids in the lipolysis medium in mM at time t determined by the NaOH consumption, FA(end) is the concentration of fatty acids at the end of the lipolysis, PL(t) is the concentration of PL in the lipolysis medium in mM at time t, PL(0) is the concentration of PL at the beginning of the lipolysis, t is the time of lipolysis in minutes and k is the rate constant. An overview of the kinetics data is given in Table 5. Additionally, the percentage of lyso-PL, determined by the reduction of PLs in the medium, and the percentage of lipid-digestion, i.e. difference of actual amount of free FAs to maximum possible amount of free FAs, is summarised in Table 5. In general, PL digestion to lyso-PL followed a pseudo-first order kinetic with higher rate constants for the lower lipid concentrations. Digestion of EPC-3 liposomes at 5 mM concentration followed a pseudo-zero order kinetic. Also for the total lipolysis process measured by NaOH consumption first order kinetics was observed for most of the formulations except for the SPC liposomes at 5 mM. For this formulation the pseudo-zero order model showed a better fit. Whereas for the higher lipid concentrations the digestion process measured by the amount of free FAs was only progressed to around 50% and less, it was for the lower lipid concentrations around 80% and above. The NaOH consumption

258

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

Fig. 3. (A) PL content in % (solid lines, right axis) in the LM and NaOH consumption in ml (dashed lines, left axis) during lipolysis of liposomes at 5 mM total lipid concentration. Values represent means (±SEM) with n = 3. Mean values of three runs from NaOH consumption were fitted by a LOWESS Smooth with five points per window. (B) Size distribution of liposomes during lipolysis at 5 mM total lipid concentration measured by DLS (solid, black line) or SIOS (dotted, grey line). Values represent means with n = 3.

over 100% for SPC 1 mM liposomes was also observed for blank liposomes (data not shown) and can be explained by unspecific reaction, such as the reaction of CO2 with hydroxide ions. 3.7. Lipolysis of liposomes – influence of danazol As shown in Fig. 5A influence of danazol on hydrolysis of SPC liposomes was negligible, but for EPC-3 liposomes a slightly lower hydrolysis rate was observed in the presence of danazol (49.75% vs. 69.53% final lipid concentration). Danazol promoted, independent of the lipid type, the formation of larger particles (Fig. 5B). This effect was more pronounced and appeared earlier with EPC-3 liposomes compared to SPC liposomes. 3.8. Danazol precipitation Percentages of danazol recovered in the precipitate at different times during lipolysis are summarised in Table 6. Danazol incorporated into liposomes hardly showed any precipitation during

lipolysis and no distinct changes over time were observed. Furthermore, neither lipid type nor lipid concentration seemed to influence danazol precipitation. Detailed DLS measurements in lipolysis medium and during lipolysis at 1 mM and 5 mM lipid concentration and of the SIOS measurements during lipolysis at 5 mM lipid concentration are given in the supplement figures of the online version of this article for each of the five formulations.

4. Discussion In the present study, we investigated the fate of PC-based liposomes in the dynamic in vitro lipolysis model. This model was developed to simulate intestinal digestion and its impact on drug solubilisation with the aim of achieving better prediction of the oral bioavailability of poorly soluble drugs, when dosed in lipidbased drug delivery systems (Larsen et al., 2011). However, it is also suitable to study the digestion of orally administered liposomes in the intestine, since all important parameters having an

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

259

Fig. 4. (A) PL content in % (descending curves, right axis) in the LM and NaOH consumption in ml (ascending curves, left axis) during lipolysis of liposomes at 5 mM (solid lines) and 1 mM (dashed lines) total lipid concentration. Values represent means (±SEM) with n = 3. Mean values of three runs from NaOH consumption were fitted by a LOWESS Smooth with five points per window. (B) Size distribution of liposomes during lipolysis at 5 mM (solid, black line) and 1 mM (grey, dotted line) total lipid concentration measured by DLS. Values represent means with n = 3.

impact on liposome stability are subsumed in the model (Kokkona et al., 2000; Schubert and Schmidt, 1988). Bile salts interact with the PL membrane of liposomes initially causing small membrane defects, eventually leading to the formation of mixed micelles (Schubert et al., 1986; Hildebrand et al., 2004). Pancreatic juices contain amylases, proteases, lipases and also phospholipases and lysophospholipases that can hydrolyse PLs to lyso-PLs and further to glycerophosphocholine and other degradation products (Dennis, 1983; Nacka et al., 2001; Anderson and Omri, 2004). To observe formation of micellar structures, the size and size distribution of the liposomal formulations during incubation in bio-relevant medium and during in vitro digestion was investigated. Additionally, content of undigested PLs during the lipolysis process was determined by HPLC. The bile salt concentration of 5 mM present in the LM is representative of the physiological concentration in the human intestine in the fasted state and is comparable to the bile salt concentration in other simulated intestinal fluids (Kleberg et al., 2010). The presence of PLs in the LM should increase the stability of liposomes against the bile salt mixture, because the bile salts are already incorporated into mixed micelles before the addition of liposomes. Indeed, SPC-based

liposomes did only show negligible changes in size during incubation in LM at both lipid concentrations (Fig. 1). In contrast to SPC, EPC-3 is still in the gel state at body temperature and has more saturated and shorter alkyl chains. It has been shown that bile salts can incorporate more easily into PL-membranes of liposomes composed of saturated lipids in the liquid-crystalline state compared to those of unsaturated lipids and that these vesicles are also more easily transformed into micelles (Garidel et al., 2007). However, it is reasonable to assume that the tight packing and low flexibility of the lipid molecules in the gel state leading to a higher stability of these membranes also hamper the incorporation of surfactants, such as bile salts, into lipid bilayers. In the present study, no micellisation, which should be visible as a size decrease in the DLS measurement, of the EPC-3 liposomes at 5 mM concentration was observed. In contrast, a strong size growth for EPC-3 liposomes at 1 mM concentration was detectable by DLS during incubation in LM. In several studies it was shown by differential scanning calorimetry that a system with dipalmitoyl phosphatidylcholine (DPPC) or DPPG and sodium cholate (NaC) exhibits a coexistence of the gel and liquid-crystalline phase for certain PL/NaC ratios over a broad temperature range (Polozova et al., 1995; Hildebrand et al., 2002, 2004). Moreover, this phase separation was associated with

Table 5 Percentage of lyso-PL and FAs formed after 60 min lipolysis and digestion rate constant for digestion of PL to lyso-PL and formation of FAs. In the table either first or zero order kinetic rate constant is given, depending on which kinetic was more probable to explain the data. Values represent means (±SEM) with n = 3. Presence after 60 min lipolysis

SPC 5 mM SPC 1 mM SPC/Chol 5 mM SPC/Chol 1 mM EPC-3 5 mM EPC-3 1 mM

Lipolysis PL to lyso-PL

Formation FAs 2

lyso-PL

FA

k (1/min or mM/min)

Order

R

k (1/min or mM/min)

Order

R2

83.7 (±0.10) 95.3 (±0.34) 83.4 (±0.36) 97.2 (±0.9) 50.3 (±3.36) 93.0 (±2.7)

42.5 (±1.97) 114.1 (±1.96) 52.9 (±2.39) 90.8 (±2.18) 23.1 (±1.41) 79.4 (±9.55)

0.0288 (±0.0003) 0.0504 (±0.0008) 0.0299 (±0.0012) 0.0659 (±0.0011) 0.0503 (±0.0052) 0.0478 (±0.0006)

1

0.9965

0

0.9933

1

0.9967

1

0.9669

1

0.9934

1

0.9962

1

0.9976

1

0.9829

0

0.9710

1

0.9440

1

0.9990

0.0089 (±0.0008) 0.0178 (±0.0004) 0.0074 (±0.0016) 0.0200 (±0.0024) 0.0030 (±0.0002) 0.0102 (±0.0007)

1

0.9782

260

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

Fig. 5. (A) PL content in % (descending curves, right axis) in the LM and NaOH consumption in ml (ascending curves, left axis) during lipolysis of liposomes with or without danazol. Values represent means (±SEM) with n = 3. Mean values of three runs from NaOH consumption were fitted by a LOWESS Smooth with five points per window. (B) Size distribution of liposomes during lipolysis without (black, solid line) or with danazol (grey, dotted line) measured by DLS. Values represent means with n = 3.

the formation of network-like structures together with membrane fragments. Polozova et al. (1995) argue that the initial formation of mixed micelles and their association to network-like structures is followed by a transformation to multi-layer membranes. Furthermore, fusion processes of the mixed micelles to larger fragments can contribute to the formation of aggregates. In the present study, incorporation of bile salt molecules in the bilayer of EPC-3 liposomes might have led to a phase separation, which was more pronounced at lower lipid concentrations. This eventually resulted in the observable size change for EPC-3 liposomes at 1 mM concentration during incubation. PLs in the gut are mainly hydrolysed by pancreatic phospholipase A2 but also phospholipase A1 , C and D contribute to the overall hydrolysis. Lysophospholipases have also been described to be present in the pancreas (Hostetler and Hall, 1980; Dennis, 1983; Tanaka et al., 1999; Laine et al., 2000; Sonoda et al., 2002; Gabor et al., 2009). LM without the addition of liposomes (blank lipolysis) already contains PLs and hydrolysis of SPC occurred as evidenced by the decrease of SPC concentration and the increase of NaOH consumption over time (Fig. 2). Moreover, larger particles were visible in the LM immediately after the addition of pancreatin to the medium. These particles are likely impurities from the pancreatin, which were not separated in the centrifugation step before the addition to the LM. It was described in previous studies that liposomes are more susceptible to hydrolysis mediated by phospholipase A2 , when the PLs are at or near their phase transition temperature and that hardly any hydrolysis occurred below or above the transition temperature (Op den Kamp et al., 1974; Rowland and Woodley, 1980). However, bile salts can lead to membrane defects and can disturb the tight packing of PLs in the lipid bilayer and make them more susceptible to lipolytic degradation. Indeed, stimulating effects on phospholipase activity have been reported (Gjone et al., 1967). By comparison of the three different liposome compositions (Fig. 3) the higher resistance of EPC-3 towards digestion by phospholipases is in good agreement with other studies, where a higher phase transition temperature of PLs was associated with a lower

susceptibility to lipolytic degradation (Rowland and Woodley, 1980; Kokkona et al., 2000). Considering the high degree of hydrolysis of SPC-liposomes during lipolysis, they exhibited a remarkable stability in size. It appears that vesicles can be formed with lipid mixtures containing a considerable amount of lyso-PLs and a less substantial quantity of non-hydrolysed PLs. In the literature, mixtures containing up to 30% (mol/mol) of lyso-PLs forming vesicles were described (Ralston et al., 1980; Kumar et al., 1989), but this might differ inter alia with the type of PL, temperature and buffer concentration used. Size stability of SPC-based liposomes was not improved by the addition of cholesterol. From the lipolysis data one would expect the EPC-3 liposomes to be superior in their size stability. However, in agreement with the data obtained upon incubation of the liposomes in LM, EPC-3 based liposomes seemed to be more susceptible towards a change in size. Size distributions measured by the SIOS technique were narrower and the median differed slightly with the two methods. Nevertheless, data generated with SIOS were generally in agreement with DLS measurements. The size range that can be determined by SIOS is limited without changing or stretching the nanopore, since particles too small in size do not cause any occlusion signal whereas bigger particles might get stuck in the pore. Structures bigger than 1 ␮m were most likely too big to pass the membrane. Therefore, only liposomes in a size range between 100 nm and 1 ␮m were detected. In general, the SIOS technique displayed particles slightly bigger compared to DLS. To calculate absolute size values, a calibration curve had to be established using polystyrene particles of a defined size. Since the voltage drop caused by translocation of a particle is not only dependent on the particle volume but also on the surface charge, the signal amplitude might differ for particles of the same size that are composed of different materials (Willmott and Bauerfeind, 2010; Gaumet et al., 2008). Regarding the influence of lipid concentration on liposome stability during lipolysis, it was found that SPC liposomes were less stable in size at the lower lipid concentration, where the formation of particles bigger than 1 ␮m was detected by DLS (Fig. 4B). This is in contrast to the findings for the incubation of liposomes in LM

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

261

Table 6 Percentage of danazol precipitated during lipolysis of liposomes. Values represent means (±SEM) with n = 3.

SPC/Dan EPC-3/Dan

Lipid conc. (mM)

Precipitated danazol (%) (m/m) 0 min

30 min

60 min

1 5 1 5

3.95 (±0.57) 4.20 (±0.63) 5.38 (±0.87) 9.47 (±2.6)

4.31 (±1.00) 4.54 (±1.15) 3.93 (±0.53) 6.73 (±4.98)

3.17 (±1.10) 3.32 (±1.26) 2.92 (±0.65) 3.30 (±1.51)

without lipolysis, where no impact of SPC concentration on size stability was observed. As described above, in LM without any addition of liposomes, particles comparable in size to those found for 1 mM liposomes were seen. At least partly, the signal of still intact vesicles might have been superimposed by the observed aggregates. Considering the small percentage (<10%, m/m) of PL from the original 1 mM SPC and EPC-3 liposomal dispersions still present after 60 min of lipolysis (Fig. 4A), it is likely that also the quantity of intact vesicles is reduced and PLs and lyso-PLs exist in mixed micelles, which give a light scattering signal of lower intensity compared to vesicles in a size around 100 nm. Whilst at 5 mM lipid concentration EPC-3 liposomes were substantially more resistant against lipolytic degradation, at 1 mM they were not superior compared to SPC-based liposomes. Thus, the above described higher resistance of liposomes containing lipids with a high phase transition temperature against hydrolysis by pancreatic enzymes is probably more related to the tighter packing of the lipid bilayer, which hinders substrate-enzyme contact, and less to a lower phospholipase activity towards lipids with saturated or longer alkyl chains. Considering the impact of bile salts on EPC-3 liposomes at lower concentration, the packing of the bilayer might be disturbed during incubation in LM, which eventually results in increased contact of phospholipase and PLs. This is also in agreement with the fact that the relative degradation of SPC was not considerably different at both concentrations. SPC liposomes at body temperature, independent of the influence of bile salts and lipid concentration, are in the liquid crystalline state and exhibit a better accessibility for pancreatic enzymes. In conclusion, whereas for EPC-3 liposomes bile salt concentration and total lipid concentration are of high importance for their stability, these factors are far less important in case of SPC-based liposomes. Although pseudo-first order kinetics was for most formulations the better fitting model regarding the NaOH consumption during lipolysis the rather poor regression coefficient and the almost linear shape of the curve indicate that at least partially the lipolysis reaction was not concentration dependent (Table 5 and Fig. 3). In the case of the highest PL concentration in the medium (SPC 5 mM) the reaction followed clearly a pseudo-zero order reaction. It can be hypothesised that at high PL concentration Ca2+ is the limiting factor for the enzymatic hydrolysis, since it is a cofactor for phospho- and lyso-phospholipases and its amount was limited during the experiment due to the constant addition and the continuous removal by precipitation with FAs (Figarella et al., 1971; DeBose and Roberts, 1983; Dennis, 1983). Surprisingly, NaOH consumption for 1 mM and 5 mM liposomes was comparable – independent of the lipid type. Lyso-PLs can be further hydrolysed to glycerophosphocholine and a second FA, which eventually is deprotonated under NaOH consumption. This second reaction was promoted in terms of the lower PL concentration, as can be seen from the percentage of digestion at the end of the lipolysis summarised in Table 5. In case of the higher lipid concentrations the total lipolysis process (digestion to two FAs) was only to around 50% or below progressed, although for SPC containing liposomes the first reaction step (formation of lyso-PL from PL) was almost completed. For the lower lipid concentrations both reaction steps were completed almost entirely. This makes NaOH

an incomplete surrogate for the quantity of hydrolysis of PLs occurring during lipolysis, since it is not exactly known, which reaction is preferred, when lyso-PLs are present. Furthermore, hydrolysis of the PLs present in the LM might occur at different rates because of the different physical properties of liposomes and micelles. This becomes even more evident, when different PLs are used in liposomes and the LM. In the case of EPC-3 liposomes the NaOH consumption was not substantially different from the blank LM without the addition of liposomes. This could be caused by partial inhibition of the phospholipases by the fully saturated PLs, which leads eventually also to reduced hydrolysis of SPC present in the LM. Furthermore, EPC-3 contains mostly saturated FAs that a have a very low water solubility and might precipitate as acids before they are deprotonated or form poorly soluble salts with Ca2+ -ions. In a study, where lipolytic digestion of dipalmitoyl glycerophosphocholine liposomes was investigated, the authors found a pseudo-first order kinetics for the lipid digestion (Mohanraj et al., 2010). Whilst this is in general agreement with our findings, the type of kinetics for EPC-3 was concentration dependent. It should be noted, that the lipolysis model employed in the previous study varied from the model used in the present study. For example, the calcium addition added to the lipolysis medium as a bolus has been known to have profound effects on the kinetics of lipolysis (Thomas et al., 2012). To our knowledge we are the first to encapsulate danazol into liposomes. The rather low encapsulation rate of 5% is typical for drugs with a log P ≈ 5 and was already found for simvastatin (Afergan et al., 2010). The dose of liposomal encapsulated danazol applied to the lipolysis model was dependent on the formulation and liposome concentration and was found to be between 1.62 mg/100 ml and 6.85 mg/100 ml (Table 4). This is comparable to the amount used in a study from Larsen et al. (2008), where lipid-based drug delivery systems containing danazol were tested in vivo and in the dynamic in vitro lipolysis model. Danazol had only a minor influence on hydrolysis rate of SPC and EPC-3 based liposomes (Fig. 5A). The slightly lower hydrolysis rate for EPC-3 liposomes containing danazol could be related to the observed strong size increase. The smaller surface of these aggregates exhibit less contact area for the phospholipases, which eventually slows down the hydrolysis. The tendency to form aggregates during lipolysis of danazol containing liposomes based on SPC was less evident. The unsaturated FAs in SPC have a higher flexibility at body temperature and a more hydrophilic character compared to the fully saturated FAs in EPC-3. This might reduce vesicle aggregation due to hydrophobic interactions of the incorporated danazol. Despite the fact that most of the PLs are hydrolysed after 60 min lipolysis, no significant change in the solubilisation of danazol was found over time (Table 6). In the study by Larsen et al. (2008), danazol exhibited a solubility of around 2.2 mg/100 ml in LM and the solubility hardly changed over time. The danazol concentration in the LM in the present study was around (1 mM liposomes) or three times higher (5 mM liposomes) than its saturation solubility in the medium. With ongoing lipolysis a precipitation at least for the higher liposome concentration could be expected. Since danazol solubility in the cited study did not change with proceeding

262

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263

lipolysis, it seems plausible that the same applied for the present study for liposomal danazol. Probably, mixed micelles and vesicles formed by bile salts, lyso-PLs and PLs can prevent the crystallisation of danazol keeping the drug in the aqueous phase. However, the destabilising effect of danazol on EPC-3 liposomes leading to the formation of larger aggregates could reduce its availability for intestinal uptake compared to danazol incorporated into vesicles smaller in size or micelles. Thus, the higher resistance of EPC-3 towards lipolytic degradation might be not necessarily advantageous in terms of a higher bioavailability of danazol. This should be further investigated in in vivo studies. Although, incorporation of danazol in liposomes is low, they could be advantageous for oral delivery, since results of the present study suggest, that the drug does not precipitate despite proceeding digestion of the delivery system. The in vitro lipolysis model employed in this study to predict the fate of liposomes in the GIT can be seen as a progress from the first developed models to assess stability of liposomes for oral delivery. Initially, the interaction of bile salts with PL vesicles, later also the influence of phospholipases with and without bile salts was examined (Patel et al., 2000; Rowland and Woodley, 1980; Taira et al., 2004). In these studies liposomes composed of PLs with their phase transition above body temperature were found to be more resistant under the tested conditions. Nevertheless, in previous in vivo studies no difference in bioavailability for hydrophilic substances encapsulated in liposomes with their phase transition either below or above body temperature could be found (Patel et al., 1982; Chiang and Weiner, 1987b). A similar process to the one we could observe for EPC-3 based vesicles might have led to a phase separation in the PL membrane and eventually to a destabilisation of the liposomes. The use of a physiological bile salt mixture, addition of PLs, employment of a pH-stat and the continuous supply of Ca2+ -ions are all attempts to mirror the in vivo situation in the in vitro lipolysis model as close as possible. However, the model still has some shortcomings, like the neglecting of brush border enzymes and intestinal microflora and use of bile salts and pancreatin derived from animals. Although studies exist investigating enzyme composition and activity in humans, less information is found on the enzyme composition of animal models (Ihse et al., 1977). In general, mammalian pancreatic phospholipases are highly homologous enzymes and similar binding mechanisms have been identified for human and porcine phospholipase justifying the use of porcine pancreatin for liposomal digestion (Dennis, 1983; Snitko et al., 1999). To assess how important the drawbacks of the model are, in vitro–in vivo correlation studies have to be performed. Still, the model can give valuable information to rank different PL based formulations according to their stability.

5. Conclusions We could show that the dynamic in vitro lipolysis model can be applied to study the lipolytic digestion of liposomal delivery systems. SPC-based liposomes show only little changes in size during in vitro lipolysis, although most of the PLs are hydrolysed during the assay. Liposomes containing the hydrogenated PL EPC-3, which is in the gel phase at body temperature, exhibited an increase in size during lipolysis. Nevertheless, digestion of EPC-3 occurred to a lower extent compared to the unsaturated PL mixture. Danazol incorporation into liposomes was moderate, but the drug did not precipitate during lipolysis. In the case of danazol, liposome degradation as indicated by size and lipid digestion had no influence on the solubilised amount of danazol. However, for other drugs, the structural integrity of the liposomes might have an important effect on drug bioavailability. The in vitro lipolysis model should, thus, be tested in future studies to assess the stability of other liposomally

encapsulated drugs, including peptides and proteins drugs for oral delivery. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.ijpharm.2012.08.018. References Afergan, E., Ben David, M., Epstein, H., Koroukhov, N., Gilhar, D., Rohekar, K., Danenberg, H.D., Golomb, G., 2010. Liposomal simvastatin attenuates neointimal hyperplasia in rats. AAPS J. 12, 181–187. Anderson, M., Omri, A., 2004. The effect of different lipid components on the in vitro stability and release kinetics of liposome formulations. Drug Deliv. 11, 33–39. Aramaki, Y., Tomizawa, H., Hara, T., Yachi, K., Kikuchi, H., Tsuchiya, S., 1993. Stability of liposomes in vitro and their uptake by rat Peyer’s patches following oral administration. Pharm. Res. 10, 1228–1231. Bangham, A.D., Standish, M.M., Watkins, J.C., 1965. Diffusion of univalent ions across the lamellae of swollen phospholipids. J. Mol. Biol. 13, 238–252. Bayomi, M., Al-Angary, A., Al-Meshal, M., Al-Dardiri, M., 1998. In vivo evaluation of arteether liposomes. Int. J. Pharm. 175, 1–7. Chakraborty, S., Shukla, D., Mishra, B., Singh, S., 2009. Lipid – an emerging platform for oral delivery of drugs with poor bioavailability. Eur. J. Pharm. Biopharm. 73, 1–15. Chiang, C., Weiner, N., 1987a. Gastrointestinal uptake of liposomes. I: in vitro and in situ studies. Int. J. Pharm. 37, 75–85. Chiang, C., Weiner, N., 1987b. Gastrointestinal uptake of liposomes. II: in vivo studies. Int. J. Pharm. 40, 143–150. DeBose, C.D., Roberts, M.F., 1983. The interaction of dialkyl ether lecithins with phospholipase A2 (Naja naja naja). J. Biol. Chem. 258, 6327–6334. Dennis, E.A., 1983. Phospholipases – Phospholipid Enzymology. The Enzymes, vol. 3. Academic Press, New York, pp. 307–353. Fatouros, D.G., Mullertz, A., 2008. In vitro lipid digestion models in design of drug delivery systems for enhancing oral bioavailability. Expert Opin. Drug Metab. 4, 65–76. Figarella, C., Clemente, F., Guy, O., 1971. A zymogen of phospholipase A in human pancreatic juice. Biochim. Biophys. Acta 227, 213–217. ´ ´ J., Skalko-Basnet, N., Jalsenjak, I., 2001. Mucoadhesive chitosanFilipovic-Grci c, coated liposomes: characteristics and stability. J. Microencapsul. 18, 3–12. Fricker, G., Drewe, J., 1996. Current concepts in intestinal peptide absorption. J. Pept. Sci. 2, 195–211. Fricker, G., Kromp, T., Wendel, A., Blume, A., Zirkel, J., Rebmann, H., Setzer, C., Quinkert, R.-O., Martin, F., Müller-Goymann, C., 2010. Phospholipids and lipidbased formulations in oral drug delivery. Pharm. Res. 27, 1469–1486. Gabor, F., Fillafer, C., Neutsch, L., Ratzinger, G., Wirth, M., 2009. Improving oral delivery. In: Schäfer-Korting, M. (Ed.), Drug Delivery, Handbook of Experimental Pharmacology, vol. 197. Springer, Berlin/Heidelberg, Germany, pp. 345–398. Garidel, P., Hildebrand, A., Knauf, K., Blume, A., 2007. Membranolytic activity of bile salts: influence of biological membrane properties and composition. Molecules 12, 2292–2326. Garza-Licudine, E., Deo, D., Yu, S., Uz-Zaman, A., Dunbar, W.B., 2010. Portable Nan particle unitization using a resizable nanopore instrument – the IZON qNanoTM . In: Conf Proc IEEE Eng Med Boil Soc 2010, pp. 5736–5739. Gaumet, M., Vargas, A., Gurny, R., Delie, F., 2008. Nan particles for drug delivery: the need for precision in reporting particle size parameters. Eur. J. Pharm. Biopharm. 69, 1–9. Gjone, E., Ofstad, E., Marton, P., Amundsen, E., 1967. Phospholipase activity in pancreatic exudates in experimental acute pancreatitis. Scand. J. Gastroenterol. 2, 181–185. Gregoriadis, G., 1976. The carrier potential of liposomes in biology and medicine (second of two parts). N. Engl. J. Med. 295, 765–770. Grove, M., Pedersen, G.P., Nielsen, J.L., Mullertz, A., 2007. Bioavailability of seocalcitol (III): administration of lipid-based formulations to minipigs in the fasted and fed state. Eur. J. Pharm. Sci. 31, 8–15. Hildebrand, A., Beyer, K., Neubert, R., Garidel, P., Blume, A., 2004. Solubilization of negatively charged DPPC/DPPG liposomes by bile salts. J. Colloid Interface Sci. 279, 559–571. Hildebrand, A., Neubert, R., Garidel, P., Blume, A., 2002. Bile salt induced solubilization of synthetic phosphatidylcholine vesicles studied by isothermal titration calorimetry. Langmuir 18, 2836–2847. Hostetler, K.Y., Hall, L.B., 1980. Phospholipase C activity of rat tissues. Biochem. Biophys. Res. Commun. 96, 388–393. Ihse, I., Arnesjö, B., Kugelberg, C., Lilja, P., 1977. Intestinal activities of trypsin, lipase, and phospholipase after a test meal. An evaluation of 474 examinations. Scand. J. Gastroenterol. 12, 663–668. Jesorka, A., Orwar, O., 2008. Liposomes: technologies and analytical applications. Annu. Rev. Anal. Chem. 1, 801–832. Kararli, T.T., 1995. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm. Drug Dispos. 16, 351–380.

J. Parmentier et al. / International Journal of Pharmaceutics 437 (2012) 253–263 Kleberg, K., Jacobsen, F., Fatouros, D.G., Mullertz, A., 2010. Biorelevant media simulating fed state intestinal fluids: colloid phase characterization and impact on solubilization capacity. J. Pharm. Sci. 99, 3522–3532. Kokkona, M., Kallinteri, P., Fatouros, D., Antimisiaris, S.G., 2000. Stability of SUV liposomes in the presence of cholate salts and pancreatic lipases: effect of lipid composition. Eur. J. Pharm. Sci. 9, 245–252. Kumar, V.V., Malewicz, B., Baumann, W.J., 1989. Lysophosphatidylcholine stabilizes small unilamellar phosphatidylcholine vesicles. Phosphorus-31 NMR evidence for the wedge effect. Biophys. J. 55, 789–792. Laine, J., Bourgoin, S., Bourassa, J., Morisset, J., 2000. Subcellular distribution and characterization of rat pancreatic phospholipase D isoforms. Pancreas 20, 323–336. Larsen, A., Holm, R., Pedersen, M.L., Muellertz, A., 2008. Lipid-based Formulations for danazol containing a digestible surfactant, Labrafil M2125CS: in vivo bioavailability and dynamic in vitro lipolysis. Pharm. Res. 25, 2769–2777. Larsen, A.T., Sassene, P., Müllertz, A., 2011. In vitro lipolysis models as a tool for the characterization of oral lipid and surfactant based drug delivery systems. Int. J. Pharm. 417, 245–255. Lipinski, C., Lombardo, F., Dominy, B., Feeney, P., 1997. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev. 23, 3–25. Mohanraj, V.J., Barnes, T.J., Prestidge, C.A., 2010. Silica Nan particle coated liposomes: a new type of hybrid nanocapsule for proteins. Int. J. Pharm. 392, 285–293. Muramatsu, K., Maitani, Y., Nagai, T., 1996. Dipalmitoylphosphatidylcholine liposomes with soybean-derived sterols and cholesterol as a carrier for the oral administration of insulin in rats. Biol. Pharm. Bull. 19, 1055–1058. Nacka, F., Cansell, M., Entressangles, B., 2001. In vitro behavior of marine lipid-based liposomes. Influence of pH, temperature, bile salts, and phospholipase A2. Lipids 36, 35–42. Op den Kamp, J.A., de Gier, J., van Deenen, L.L., 1974. Hydrolysis of phosphatidylcholine liposomes by pancreatic phospholipase A2 at the transition temperature. Biochim. Biophys. Acta 345, 253–256. Parmentier, J., Becker, M.M., Heintz, U., Fricker, G., 2011. Stability of liposomes containing bio-enhancers and tetraether lipids in simulated gastro-intestinal fluids. Int. J. Pharm. 405, 210–217. Patel, G.B., Agnew, B.J., Deschatelets, L., Fleming, L.P., Sprott, G.D., 2000. In vitro assessment of archaeosome stability for developing oral delivery systems. Int. J. Pharm. 194, 39–49. Patel, H., Ryman, B.E., 1976. Oral administration of insulin by encapsulation within liposomes. FEBS Lett. 62, 60–63. Patel, H.M., Stevenson, R.W., Parsons, J.A., Ryman, B.E., 1982. Use of liposomes to aid intestinal absorption of entrapped insulin in normal and diabetic dogs. Biochim. Biophys. Acta 716, 188–193. Polozova, A.I., Dubachev, G.E., Simonova, T.N., Barsukov, L.I., 1995. Temperatureinduced micellar–lamellar transformation in binary mixtures of saturated phosphatidylcholines with sodium cholate. FEBS Lett. 358, 17–22. Potluri, P., Betageri, G., 2006. Mixed-micellar proliposomal systems for enhanced oral delivery of progesterone. Drug Deliv. 13, 227–232.

263

Ralston, E., Blumenthal, R., Weinstein, J.N., Sharrow, S.O., Henkart, P., 1980. Lysophosphatidylcholine in liposomal membranes: enhanced permeability but little effect on transfer of a water-soluble fluorescent marker into human lymphocytes. Biochim. Biophys. Acta 597, 543–551. Rowland, R.N., Woodley, J.F., 1980. The stability of liposomes in vitro to pH, bile salts and pancreatic lipase. Biochim. Biophys. Acta 620, 400–409. Rowland, R.N., Woodley, J.F., 1981. Uptake of free and liposome-entrapped insulin by rat intestinal sacs in vitro. Biosci. Rep. 1, 345–352. Schubert, R., Beyer, K., Wolburg, H., Schmidt, K.H., 1986. Structural changes in membranes of large unilamellar vesicles after binding of sodium cholate. Biochemistry 25, 5263–5269. Schubert, R., Schmidt, K.H., 1988. Structural changes in vesicle membranes and mixed micelles of various lipid compositions after binding of different bile salts. Biochemistry 27, 8787–8794. Singh, R., Singh, S., Lillard, J.W., 2008. Past, present, and future technologies for oral delivery of therapeutic proteins. J. Pharm. Sci. 97, 2497–2523. Snitko, Y., Han, S., Lee, B., Cho, W., 1999. Differential interfacial and substrate binding modes of mammalian pancreatic phospholipases A(2): a comparison among human, bovine, and porcine enzymes. Biochemistry-Us 38, 7803– 7810. Sonoda, H., Aoki, J., Hiramatsu, T., Ishida, M., Bandoh, K., Nagai, Y., Taguchi, R., Inoue, K., Arai, H., 2002. A novel phosphatidic acid-selective phospholipase A1 that produces lysophosphatidic acid. J. Biol. Chem. 277, 34254–34263. Szoka, F., Papahadjopoulos, D., 1980. Comparative properties and methods of preparation of lipid vesicles (liposomes). Annu. Rev. Biophys. Bioeng. 9, 467–508. Taira, M., Chiaramoni, N., Pecuch, K., Alonso-Romanowski, S., 2004. Stability of liposomal formulations in physiological conditions for oral drug delivery. Drug Deliv. 11, 123–128. Takahashi, M., Uechi, S., Takara, K., Asikin, Y., Wada, K., 2009. Evaluation of an oral carrier system in rats: bioavailability and antioxidant properties of liposomeencapsulated curcumin. J. Agric. Food Chem. 57, 9141–9146. Takeuchi, H., Yamamoto, H., Niwa, T., Hino, T., Kawashima, Y., 1994. Mucoadhesion of polymer-coated liposomes to rat intestine in vitro. Chem. Pharm. Bull. 42, 1954–1956. Tanaka, H., Mierau, I., Ito, F., 1999. Purification and characterization of bovine pancreatic bile salt activated lipase. J. Biochem. – Tokyo 125, 883–890. Thomas, N., Holm, R., Rades, T., Müllertz, A., 2012. In vitro lipolysis and its implication on the development of lipid-based drug delivery systems. AAPS J., http://dx.doi.org/10.1208/s12248-012-9398-6. Werle, M., Takeuchi, H., 2009. Chitosan–aprotinin coated liposomes for oral peptide delivery: development, characterisation and in vivo evaluation. Int. J. Pharm. 370, 26–32. Willmott, G., Bauerfeind, L., 2010. Detection of polystyrene sphere translocations using resizable elastomeric nanopores. arXiv:1002.0611. Zangenberg, N., Mullertz, A., Kristensen, H., Hovgaard, L., 2001a. A dynamic in vitro lipolysis model I. Controlling the rate of lipolysis by continuous addition of calcium. Eur. J. Pharm. Sci. 14, 115–122. Zangenberg, N., Mullertz, A., Kristensen, H., Hovgaard, L., 2001b. A dynamic in vitro lipolysis model II: evaluation of the model. Eur. J. Pharm. Sci. 14, 237–244.