Article pubs.acs.org/Langmuir
The Role of Porous Nanostructure in Controlling Lipase-Mediated Digestion of Lipid Loaded into Silica Particles Paul Joyce, Angel Tan, Catherine P. Whitby, and Clive A. Prestidge* Ian Wark Research Institute, University of South Australia, Mawson Lakes Campus, Mawson Lakes, South Australia 5095, Australia S Supporting Information *
ABSTRACT: The rate and extent of lipolysis, the breakdown of fat into molecules that can be absorbed into the bloodstream, depend on the interfacial composition and structure of lipid (fat) particles. A novel method for controlling the interfacial properties is to load the lipid into porous colloidal particles. We report on the role of pore nanostructure and surface coverage in controlling the digestion kinetics of medium-chain and long-chain triglycerides loaded into porous silica powders of different particle size, porosity, and hydrophobicity/ hydrophilicity. An in vitro lipolysis model was used to measure digestion kinetics of lipid by pancreatic lipase, a digestive enzyme. The rate and extent of lipid digestion were significantly enhanced when a partial monolayer of lipid was loaded in porous hydrophilic silica particles compared to a submicrometer lipid-in-water emulsion or a coarse emulsion. The inhibitory effect of digestion products was clearly evident for digestion from a submicrometer emulsion and coarse emulsion. This effect was minimal, however, in the two silica−lipid systems. Lipase action was inhibited for lipid loaded in the hydrophobic silica and considered due to the orientation of lipase adsorption on the methylated silica surface. Thus, hydrophilic silica promotes enhanced digestion kinetics, whereas hydrophobic silica exerts an inhibitory effect on hydrolysis. Evaluation of digestion kinetics enabled the mechanism for enhanced rate of lipolysis in silica−lipid systems to be derived and detailed. These investigations provide valuable insights for the optimization of smart food microparticles and lipidbased drug delivery systems based on lipid excipients and porous nanoparticles.
■
INTRODUCTION Understanding and controlling the enzymatic hydrolysis of lipids are of great importance for the rational design of functional foods and drug formulations. Lipid digestion is fundamental to the delivery of nutrients and fat from the gastrointestinal (GI) tract to the blood and surrounding tissues. Consequently, the bioavailability of bioactive lipophilic compounds can be manipulated by engineering lipid-based delivery systems that have controlled stability and digestibility in the GI tract.1−3 Processing of lipids is initiated in the stomach, where digestive enzymes (lipases) partially hydrolyze triglycerides into diglycerides and fatty acids. Crude emulsions are formed due to the amphiphilic nature of these digestion products and gastric agitation and are emptied into the small intestine. Lipid hydrolysis, e.g. in the small intestine, consists of two steps, shown schematically in Figure 1: (i) the physical adsorption of pancreatic lipase to the lipid−droplet interface, which leads to the activation of lipase, and (ii) the formation of an enzyme/substrate complex, whereby triglycerides are hydrolyzed to free fatty acids (FFA) and monoglycerides and the enzyme desorbs from the interface (eq 1).4,5
interfacial process of lipase action is unique as its active site is occluded by a α-helical lid or a polypeptide flap in aqueous environments. The enzyme is inactive and unable to bind to isolated substrate molecules due to this conformation and is activated through the adsorption to an interface whereby the lid domain opens. This “interfacial activation” promotes the exposure of the catalytic residues and increases the nonpolarity of the exposed surface, increasing the affinity of the enzyme for the substrate.8,9 Since lipolysis is an interfacial process, enzymatic activity is controlled by the quality and quantity of the substrate on the interface,10,11 and thereby the binding rate constant (kB) may be manipulated by microstructure to increase or decrease the lipid digestion rate, whereas the hydrolysis rate constant (kH), which can be described by a pseudo-Michaelis−Menten mechanism,12 is dependent on the ability for lipase to dissociate from the substrate/product interface. The desorption rate of lipase from the emulsion interface can be controlled in a number of ways including altering interfacial tension conditions.13 The desorption of digestion products (free fatty acids and monoglycerides) from the lipid-in-water interface to the aqueous media is the rate-limiting process for lipase-catalyzed reactions as they are highly surface active and compete with
kB
lipid + lipase → lipid:lipase complex kH
→ FFA + monoglycerides + lipase
(1)
Received: January 17, 2014 Revised: February 17, 2014 Published: February 19, 2014
Lipid hydrolysis is predominantly controlled by the ability for pancreatic lipase to bind to the emulsified lipid interface.6,7 The © 2014 American Chemical Society
2779
dx.doi.org/10.1021/la500094b | Langmuir 2014, 30, 2779−2788
Langmuir
Article
Figure 1. Compositional changes to the emulsion interface during the lipid digestion of a phospholipid-stabilized emulsion. As lipid digestion progresses, digestion products partition to the oil-in-water interface or are dispersed within the aqueous phase, forming mixed micelles or existing as FFA and monoglycerides.
lipase to adsorb at the emulsion interface (Figure 1).14 High FFA concentrations cause compositional changes to the interface and restrict the ability for lipase to access the substrate.15 Under physiological conditions, bile salts and calcium ions act to displace lipolytic products from the emulsion interface by micellar solubilization and precipitation, respectively.16,17 In doing so, hydrolysis of the emulsified lipid by lipase can be treated as self-catalytic at sufficient bile salt and calcium concentrations. The bioaccessibility of the lipid can be manipulated through a number of colloidal engineering methods. Examples include controlling specific surface area of lipid,18,19 lipid and interfacial composition,20−23 and the encapsulation of lipid within a matrix or hydrogel structure.24−26 Differences in chain length between medium-chain-length triglycerides (MCT) and long-chainlength triglycerides (LCT) facilitate differences in lipase action due to the effect of steric hindrance of the substrate on the enzyme and the higher interference effect on lipase of longchain digestion products. FFA released from the lipasemediated digestion of MCT are more rapidly dispersed in the aqueous environment than FFA released from LCT due to their higher polarity and amphiphilic behavior.27 Consequently, longchain FFAs accumulate at oil−water interfaces more readily and thereby have a greater inhibitory effect on lipase.28 The rate and extent of lipid hydrolysis are also dependent on the emulsifier type and concentration used to stabilize triglyceride emulsions.29−31 Emulsifiers can be used to control lipolysis due to their ability to alter the interfacial composition of the lipid interface, thereby restricting the adsorption of lipase. Vinarov et al.32 demonstrated that lipid hydrolysis is enhanced at intermediate emulsifier concentrations due to the increased solubilization capacity of micelles formed by bile salts and emulsifier molecules. As a result, digestion products are transferred into the aqueous phase at a greater rate, preventing their precipitation at the lipid-in-water interface. At low emulsifier concentrations, lipid hydrolysis proceeds as if there was a bare interface, and for high emulsifier concentrations, hydrolysis is inhibited due to the restricted ability for lipase to adsorb to the interface. A dry emulsion delivery system developed by our group, known as silica−lipid hybrid (SLH) microparticles, has shown
enhanced in vitro lipid digestion and in vivo drug bioavailability compared to conventional emulsions and lipid solutions.33−36 SLH microparticles are prepared by spray drying an oil-in-water emulsion stabilized with silica nanoparticles to produce a highly porous three-dimensional structure whereby lipid droplets are encapsulated within a silica matrix.37,38 It has been postulated that some combination of the high surface area of lipid and lipase binding to hydrophilic silica are responsible for SLH action. However, the exact mechanism of lipase action within SLH microparticles remains unknown. Gustafsson et al.39,40 demonstrated the importance of pore size for catalytic activity of lipases encapsulated in mesoporous silica. Lipase was immobilized in mesoporous silica particles of three different pore sizes: 5.0, 6.0, and 8.9 nm. Lipase activity was found to increase with pore diameter. It was suggested that the confined environment of the smaller pores restricted the ability for the lid domain of the lipase to open upon interfacial activation. The approximate diameter of lipase is 4.5 nm. Therefore, when immobilized in 5.0 and 6.0 nm pores, lipase has restricted ability for movement. For pores sufficiently large for lipase action (i.e., 8.9 nm pores), the product yield was more than twice that for free lipase in solution due to the pore walls acting as an interface for lipase activation. This paper investigates lipolysis kinetics when mesoporous silica materials are used as hosts for lipid molecules. The study was designed to systematically investigate a number of critical parameters that have an effect on the rate of lipolysis. First, changes to the substrate were investigated by altering (i) the loading concentration of lipid to determine the impact of multilayer binding within the pores and (ii) the lipid type loaded in the porous silica particles, that is, MCT versus LCT. Second, changes to the carrier particles were investigated by loading lipid into porous silica particles with different porous characteristics including pore size and volume, porosity, and particle size along with different interfacial surface chemistries. By determining the role of nanostructure and surface chemistry on lipase activity within porous silica particles, further insights into the interfacial mechanism of lipase action will be derived which will contribute to the optimization of digestion performance of lipid-based drug formulations and nutraceuticals/functional foods. 2780
dx.doi.org/10.1021/la500094b | Langmuir 2014, 30, 2779−2788
Langmuir
Article
Table 1. Properties of Three Porous Silica Particles, PS-1, PS-2, and PS-3, Used To Investigate the Effect on Lipase-Mediated Digestion of Triglycerides PS-1 PS-2 PS-3
surf. chemistry
surf. area (m2/g)
primary particle size (nm)
hydrophilic hydrophilic hydrophobic
380 ± 30 311 ± 14c 220 ± 25a
7.0 ± 1.0 2500−3700c 7.0 ± 1.0a
a
av pore diam (nm)
a
b
2.0−7.0 19.0 ± 1.1d 2.0−7.0b
pore vol (cm3/g) 0.77a 1.60c 0.68a
a
Product specifications, Evonik Industries AG.42 bDetermined theoretically based on hexagonally and tetrahedrally packed particles.41 cProduct specifications, Grace Division Discovery Sciences, Grace GmbH & Co. KG.43 dKinnari et al.44
■
medium for 4 h. The powder was removed and dried at 30 °C for 24 h, and then the lipid content was measured (TGA). In Vitro Lipolysis Studies. Preparation of Lipid Digestion Medium. Fasted state mixed micelles, i.e., phospholipid/bile salt (1.25 mM PC/5 mM NaTDC), were prepared in the following sequence: egg lecithin was dissolved in chloroform (4 mL) followed by evaporation of chloroform under vacuum (Rotavapor RE, Buchi, Switzerland) to form a thin film of lecithin in a 50 mL round-bottom flask; NaTDC and digestion buffer [50 mM Trizma maleate (pH 7.5), 150 mM NaCl, and 5 mM CaCl2·2H2O] were added, and the mixture stirred for approximately 12 h to produce a transparent (light yellow) micellar solution. Pancreatin extracts (containing pancreatic lipase, colipase, and other nonspecific lipolytic enzymes such as phospholipase A2) were freshly prepared each day by stirring 1 g of porcine pancreatin powder in 5 mL of digestion buffer for 15 min, followed by centrifugation (at ∼5000 rpm, 4 °C) for 20 min. The supernatant phase was collected and stored on ice until use. Lipid Digestion Kinetics Studies. The progress of lipid digestion was monitored for 60 min by using a pH-stat titration unit (TIM854 Titration Manager, Radiometer, Copenhagen, Denmark) according to the lipolysis protocol as described by Sek et al.45 Briefly, a known quantity of sample (equivalent to approximately 200 mg of lipid) was dispersed in 18 mL of buffered micellar solution by stirring continuously for 10 min in a thermostated glass reaction vessel (37 °C). The pH of digestion medium was readjusted with 0.1 M NaOH or HCl to 7.50 ± 0.01. Lipolysis was initiated by addition of 2 mL of pancreatin extract (containing approximately 2000 TBU of pancreatic lipase activity) into the digestion medium. Free fatty acids (FFA) produced in the reaction vessel were immediately titrated with NaOH via an autoburet to maintain a constant pH in the digestion medium at the preset value of 7.50 ± 0.01 throughout the experiment. A solution of 0.6 M NaOH was used for MCT and LCT as per the established experiment protocol. All lipid digestion studies were repeated three times and showed good reproducibility. It is noteworthy that the silica particles do not alter the pH through ionization during digestion and thereby do not interfere with the pH-stat titration. Since an excess amount of lipase is commonly added to the intestinal fluid, the hydrolysis reaction can be treated as a pseudo-firstorder process, and the rate of reaction is dependent on the concentration of the substrate. The pseudo-first-order model can be used to develop a simple rate expression but ignores the effect of interfacial composition on lipase activity:
MATERIALS AND METHODS
Materials. Porous hydrophilic silica particles, PS-1 (Aerosil 380), and porous hydrophobic silica particles, PS-3 (Aerosil R812), were supplied by Degussa (Essen, Germany), and micronized synthetic amorphous silica microparticles, PS-2 (Syloid 244P), were supplied by Grace Davison Discovery Sciences (Rowville, Australia). The pore properties of the particles are listed in Table 1. The silica nanoparticles in PS-1 and PS-3 cluster into 50 nm aggregates in a disordered, random fashion in the dry powder. The size of the voids in the aggregates was estimated theoretically as ranging between the pore size in hexagonally and tetrahedrally packed particles.41 The size of the pores in PS-2 was determined by the manufacturer. MCT (Miglyol 812) and LCT (soybean oil) were obtained from Hamilton Laboratories (Adelaide, Australia) and soybean lecithin (containing >94% phosphatidycholine and
