Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles

G Model

ARTICLE IN PRESS

COLSUB-7057; No. of Pages 10

Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles Justas Barauskas a,b , Hanna Anderberg c,d , Allan Svendsen e , Tommy Nylander c,∗ a

Department of Biomedical Science, Faculty of Health and Society, Malmö University, SE-20506 Malmö, Sweden Camurus AB, Ideon Science Park, Gamma Building, Sölvegatan 41, SE-22370 Lund, Sweden Physical Chemistry, Lund University, P.O. Box 124, SE-22100 Lund, Sweden d Department of Biochemistry and Structural Biology, Lund University, P.O. Box 124, SE-22100 Lund, Sweden e Novozymes A/S, Smørmosevej 25, DK-2880 Bagsvaerd, Denmark b c

a r t i c l e

i n f o

Article history: Received 22 December 2014 Received in revised form 24 March 2015 Accepted 26 April 2015 Available online xxx Keywords: Lipase Cubic phase Lipid liquid crystalline nanoparticles Hydrolysis X-ray diffraction pH-stat titration

a b s t r a c t In this study well-ordered glycerol monooleate (GMO)-based cubic liquid crystalline nanoparticles (LCNPs) have been used as substrates for Thermomyces lanuginosus lipase in order to establish the relation between the catalytic activity, measured by pH-stat titration, and the change in morphology and nanostructure determined by cryogenic transmission electron microscopy and synchrotron small angle X-ray diffraction. The initial lipase catalyzed LCNP hydrolysis rate is approximately 25% higher for large 350 nm nanoparticles compared to the small 190 nm particles, which is attributed to the increased number of structural defects on the particle surface. At pH 8.0 and 8.4 bicontinuous Im3m cubic LCNPs transform into “sponge”-like assemblies and disordered multilamellar onion-like structures upon exposure to lipase. At pH 6.5 and 7.5 lipolysis induced phase transitions of the inner core of the particles, following the sequence Im3m cubic → reversed hexagonal → reversed micellar Fd3m cubic → reversed micelles. These transitions to the liquid crystalline phases with higher negative curvature of the lipid/water interface were found to trigger protonation of the oleic acid produced during lipase catalyzed reaction. The increase curvature of the reversed discrete micellar cubic phase was suggested to cause an increase in the oleic acid pKa to a larger value observed by pH-stat titration. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Lipolytic enzymes or lipases have important biological functions in the lipid metabolism, but are also used in numerous applications [1]. The substrates for lipolytic enzymes are self-assembled structures or aggregates of different lipid and fat molecules. Most natural substrates have low aqueous solubility and are dispersed in or exposed to an aqueous solution containing the enzyme. The enzymes therefore act at the oil–water interface and hence the term “interfacial activation” has been used to describe lipase action, which implies that the interfacial structure of the substrate is important [2]. The surface properties of the substrate are dependent on the lipid composition as well as the conditions in the aqueous phase. Another important factor to consider is that the lipolytic enzyme is generally small in size in comparison to the composite substrate assembly.

∗ Corresponding author. Tel.: +46 46 2228 158; fax: +46 46 2224 413. E-mail address: [email protected] (T. Nylander).

In the pioneering in vitro study of lipolysis of triglyceride droplets in an intestine-like environment, Patton and Carey observed a sequence of liquid crystalline phases depending on the solution conditions, among them a viscous isotropic phase composed of monoglycerides and fatty acids, which is identical to the one formed in monoglyceride systems [3]. The lipolysis products formed transiently what was later defined as the cubic phase, after which they rapidly solubilized in mixed micelles of fatty acids and bile salts if present in excess. However, after a fat rich meal, the bile acid amounts in vivo are not always sufficient to solubilize all lipids, and therefore it has been argued that the cubic liquid crystalline phases participate during in vivo digestion [4]. This study is therefore focused on the glycerol monooleate (GMO) cubic liquid crystalline phases as a substrate for the lipolytic activity. Such a substrate can also be prepared as nanoparticles with narrow size distribution and with a well-defined liquid crystalline internal structure. Here, the bicontinuity as well as the ability of the cubic monoglyceride phases to solubilize hydrophobic and amphiphilic compounds are thought to be important features for the lipolysis process [5]. These structural features will make it

http://dx.doi.org/10.1016/j.colsurfb.2015.04.052 0927-7765/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10 2

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

possible for e.g. lipase and water to freely diffuse through the phases formed by the lipolysis products, surrounding the diminishing fat droplet. Although the study of lipases has increased very sharply during the last years, most studies concern either simple bulk studies, using less-defined substrates, or only concern the interfacial reactions, using monolayers [6,7]. Recent interfacial studies of lipase activities on monolayers have provided some leads on how to control the lipase activity by modulating the lipid composition [7]. In this study we will demonstrate that lipase activity can also be controlled by the substrate assembly structure. In early studies, we investigated effects of lipase action on the GMO liquid crystalline phase as well as other self-assemble structures such as vesicles and dispersions of cubic phases [8–10]. We showed that the observed changes in self-assembled structures could be mapped by following either the (i) GMO/oleic acid aqueous ternary phase diagram (low pH), where the lipolysis gives rise to a transition of cubic → reversed hexagonal → micellar cubic → reversed micellar phase + dispersion or (ii) GMO/sodium oleate aqueous ternary phase diagram (high pH), where the corresponding sequence is lamellar → normal hexagonal. These differences in structural changes could be related to the degree of protonation of the fatty acids [8]. In these studies we found a similar specific activity of Thermomyces lanuginosus lipase (TLL) on the cubic as well as on the reversed hexagonal GMO-based liquid crystalline phases [8]. Using small-angle X-ray diffraction, cryogenic transmission electron microscopy, and dynamic light scattering, Salentinig et al. looked at the impact of lipolysis products, i.e. oleic acid [11]. They found that the system undergoes structural transitions from dispersions of bicontinuous cubic phases (cubosomes) through dispersions of reversed hexagonal phases (hexosomes) and micellar cubic phases (Fd3m symmetry) to emulsified microemulsions occurring with increasing oleic acid concentration. As expected and previously reported by Borne et al., the internal structure of the dispersed particles depended strongly on the pH, where at a high pH it tends to favor vesicles instead of reversed phases [9]. They discuss their findings in terms of an apparent pKa for oleic acid in the cubic phase, which could be estimated from the change in structure with pH. In a follow-up study by the same group they followed the in vitro digestion of ␤-lactoglobulin and ␤casein stabilized triglyceride emulsions, by pancreatic lipase [12]. The structural analysis showed a transition from oil emulsion to emulsified microemulsion, micellar cubic, inverse hexagonal, and bicontinuous cubic liquid crystalline droplets as the lipolysis progresses. They also observed strong effects on the lipolysis reaction of solution properties such as bile-juice concentration and pH as well as of hydrophobic additives. We have previously studied lipase-catalyzed hydrolysis of cubic nanoparticles formed from GMO, which is the final step in the lipolysis of glycerol trioleate, leads to drastic changes in the liquid crystalline structure [13]. For that study we used lipase, TLL, conjugated to gold nanoparticles to visualize the enzyme location and the enzymatic digestion of lipid aggregates by means of cryogenic transmission electron microscopy. We showed that the use of lipase–gold nanoparticle conjugates provide a handle on single lipase molecules, but compared to lipase alone the manner in which enzymatic digestion occurs at the single molecule level is affected in terms of the lipid nanostructure. In the present studies we used well-defined GMO-based cubic phase nanoparticles as substrates for TLL with an aim to directly correlate the structural and morphological changes to the lipase activity and progression of the lipolysis. The main object of the present study is not to mimic the physiological conditions in the gastrointestinal tract, but rather to present and discuss in detail an assay for testing technologically relevant lipolytic enzymes that we previously outlined in 2008 [14].

2. Materials and methods Chemicals. RYLOTM MG19 Glycerol Monooleate (GMO) was produced and provided by Danisco Ingredients (Brabrand, Denmark) with the following fatty acid composition (Lot No. 2119/651): 89.3% oleic, 4.6% linoleic, 3.4% stearic, and 2.7% palmitic acid. The poly(ethylene oxide) (PEO)–poly(propylene oxide) (PPO)–poly(ethylene oxide) triblock copolymer with the trade name Lutrol® F127 and an approximate formula of PEO98 PPO57 PEO98 (average molecular weight of 12,600 g/mol) was obtained from BASF Svenska AB (Helsingborg, Sweden). Wild-type TLL and its inactive mutant (iTLL, in which the catalytic Ser146 is changed to Gly) were kindly supplied by M. Skjøt (Novozymes A/S, Denmark). Protein concentration was determined spectrophotometrically at 280 nm. A molar extinction coefficient of 43,000 M−1 cm−1 with Mw of 32 kDa was used [15]. Milli-Q purified water was used for all experiments. All other solvents and reagents were of analytical grade and were used as received. Preparation of lipid nanoparticles. Nanoparticle dispersions were prepared by adding appropriate amounts of melted GMO (40 ◦ C) into an aqueous F127 solution. In all experiments the GMO/F127 ratio was fixed to 9/1 (w/w) and the total amphiphile (GMO + F127) concentration in water was either 2 or 5 wt%. The total sample volume was usually 200–300 mL. The samples were immediately sealed, shaken, and mixed for 24–48 h on a mechanical mixing table at 350 rpm and room temperature. The resulting coarse dispersions were homogenized by passing 5 times through a Microfluidizer 110S (Microfluidics Corp., Newton, USA) at 345 bar and 25 ◦ C. Homogenized dispersions were then exposed to heat treatment in order to improve dispersion properties in terms of reducing the amount of metastable vesicular aggregates. Heat treatment was performed using a bench-type autoclave (CertoClav CV-EL, Certoclav Sterilizer GmbH, Traun, Austria) operated at 125 ◦ C and 1.4 bar. The samples were subjected to heat treatment for 20 min at 125 ◦ C and allowed to cool to room temperature before analysis. In all cases this nanoparticle preparation procedure resulted in homogenous milky dispersions with narrow and monomodal size distributions (polydispersity index of 0.2) with the mean particle size of 190 and 350 nm for dispersions containing 2 and 5 wt% of amphiphile, respectively. Particle size measurements. Particle size distributions were measured using a Coulter LS230 laser diffraction particle size analyzer (Beckman-Coulter, Inc., Miami, U.S.A.), which operates on the principles of Fraunhofer diffraction for large particles (0.4–2000 ␮m) and uses the polarization intensity differential scattering (PIDS) method for small particles (0.04–0.5 ␮m). The instrument was fitted with a 125 mL volume module. Data were collected during 90 s. A standard model based on homogenous oil spheres with a refractive index (RI) of 1.46 was used for the particle size calculations. The change in RI to either side only shifts the obtained particle size distributions within a few percents. Note that the model is based on spherical particles and the measured mean particle size is calculated based on this assumption. pH-stat titration. pH-stat titration experiments were performed on a computer-controlled TitraLab 856 titration workstation (Radiometer Analytical SAS, France) equipped with an automatic 10 ml burrete (volume controlled down to 0.6 ␮l for each injection step) and a general purpose Red Rod-combined pH electrode. During the lipolytic reaction the GMO within the cubic liquid crystalline nanoparticles is hydrolyzed and oleic acid is formed, which lowers the pH of the reaction mixture. The aim of the pH-stat titration method is therefore to follow the lipolytic reaction kinetics by continuously titrating generated oleic acid as a function of time while keeping the pH constant. In all experiments the reaction was carried out in a Teflon vessel (total reaction volume 20 ml) and reaction media contained 1.5 mM potassium phosphate buffer

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

with 30 mM KCl. In these experiments low buffering capacity was deliberately used to ensure the pH drop caused by liberated oleic acid during lipolysis. The lipolitic reaction was initiated by injecting 20 ␮l 0.1 mM TLL stock solution. For control experiments, 100 ␮l 0.1 mM stock solution of iTLL was used. After the lipase injection the reaction mixture was titrated with either 2.5 (for determining the initial reaction rate) or 25 mM KOH (for following the reaction progress). Cryogenic transmission electron microscopy (cryo-TEM). The samples were prepared in a controlled environment vitrification system. The climate chamber temperature was 25–28 ◦ C and the relative humidity was kept close to saturation to prevent evaporation from the sample during preparation. A 5 ␮L sample drop was placed on a carbon-coated holey film supported by a copper grid and gently blotted with a filter paper to obtain a thin liquid film (20–400 nm) on the grid. The grid was then rapidly plunged into a liquid ethane at −180 ◦ C and transferred into liquid nitrogen (−196 ◦ C). The vitrified specimens were stored in liquid nitrogen and transferred into a Philips CM120 BioTWIN microscope equipped with a post-column energy filter (Gatan GIF 100) using an Oxford CT 3500 cryo-holder and its workstation. The acceleration voltage was 120 kV and the working temperature was kept below −182 ◦ C. The images were recorded digitally with a CCD camera (Gatan MSC 791) under low-dose conditions with an underfocus of approximately 1 ␮m. Synchrotron small angle X-ray diffraction (SAXD). Synchrotron SAXD measurements were performed at either I711 or I911 beamline at MAX-lab (Lund University, Sweden), using a Marresearch 165 mm CCD detector mounted on a Marresearch Desktop Beamline baseplate [16,17]. 0.5 mL of nanoparticle dispersion was mixed with 0.5 mL 200 mM of either potassium phosphate or TRIS buffer solution before an appropriate small amount of lipase stock solution was added to the reaction mixture to initiate hydrolysis. The samples were immediately filled into 1 mm (i.d.) glass capillary, sealed and put into the instrument. A period of about 2 min was required between sample mixing and the actual start of the X-ray exposure. Diffractograms were recorded at 25 ◦ C with a wavelength, , of 0.91 A˚ at the sample-to-detector distance of 2304 mm. This gives an available q-range of about 0.25–3 nm, where q = 4␲sin()/ and the scattering angle is 2. No beam damage could be detected after the accumulated exposure time each sample was exposed to as the diffractograms of the sample, which was not incubated with lipase, were identical even after repeated exposure. Furthermore the lipase concentration was adopted so that the complete lipolysis process took less than 1 h. Temperature control within 0.1 ◦ C was achieved using a computer-controlled Julabo heating circulator F12-MC (Julabo Labortechnik GMBH, Seelbach, Germany). The resulting CCD images were integrated and analyzed using the Fit2D software provided by Dr. A. Hammersley (http://www.esrf.fr/computing/scientific/FIT2D) using silver behenate to determine the absolute detector position. The lipolytic reaction was followed by continuously measuring diffraction in 60 s intervals allowing 24 s between exposures.

3

order as a function of lipase-induced nanoparticle hydrolysis have been investigated by means of pH-stat titration, cryogenic transmission electron microscopy (cryo-TEM), and synchrotron small angle X-ray diffraction (SAXD). 3.1. LCNP properties One of the aims of this study was to reveal the effects of the lipid LCNP size and hence the specific surface area on the lipase catalyzed reaction kinetics. As shown in our previous study, LCNP dispersions with different mean size can easily be prepared by changing the total amphiphile concentration in the dispersion while keeping the same GMO/F127 weight ratio [18,19]. With increasing amphiphile concentration the developed preparation procedure results in larger LCNPs. After preparation, particles remain stable and do not change in size upon dilution with water. Here, LCNP dispersions were prepared at 2 and 5 wt% of the total amphiphile concentration (GMO + F127) in water. Fig. 1a shows the measured size distributions of two different GMO/F127 LCNPs used in this study. The results show that the LCNP dispersions have monomodal size distributions with two different mean sizes, 190 (for 2 wt% dispersion) and 350 nm (for 5 wt% dispersion). For simplicity in the text below they are denoted as small and large particles, respectively. No change in particle size was observed when LCNP dispersions were diluted with buffer solution in the pH range between 6.5 and 8.4. The SAXD patterns in Fig. 1b for both dispersions reveal clear √ Bragg diffraction peaks located at relative positions in ratios 2, √ √ 4, and 6, which can be indexed as (1 1 0), (2 0 0), and (2 1 1) reflections characteristic of a body-centered cubic phase of the Im3m space group. Higher order reflections (3 1 0), (2 2 2), and (3 2 1) are only visible for 5 wt% dispersion where larger particles have more structural repetition units per particle. Although (220)

3. Results and discussion In this study well-ordered and mono-dispersed GMO-based cubic liquid crystalline nanoparticles (LCNPs) have been used as substrates for the fungal lipase catalytic activity investigations. The enzyme used in this study was a lipase (triacylglycerol hydrolase) from fungus TLL that catalyses the ester bond hydrolysis of tri-, di-, and monoglycerides. In this study we deliberately used single monoglyceride to form liquid crystalline nanoparticles as in this case only single catalytic event is possible, i.e. lipase catalyzed hydrolysis of GMO forming oleic acid. The initial reaction kinetics, reaction progress, and the changes in lipid nanoparticle structural

Fig. 1. Particle size distributions (a) and SAXD data (b) of GMO LCNP dispersions prepared at 2 (1) and 5 wt% (2) of the total amphiphile concentration (GMO + F127) in water.

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

4

peak is missing, previous studies have shown that this reflection from the Im3m cubic phase formed in GMO/F127 is usually not detectable. The calculated lattice parameter (a) for both dispersions is about 13 nm and is similar to the previously determined for GMO-based LCNPs [17,18]. Note that dilution with either water or buffer solution did not affect the LCNP nanostructure and unit cell dimensions.

spheres containing 50% of water by volume, which is based on the phase boundary for the fully swollen cubic phase in the bulk phase diagram of the GMO/F127 system [20]. Here, the rate of the reaction (volume of titrant added per second) is expressed in turnover numbers, kcat (s−1 ), which is the moles of product per second and moles of enzyme using a simple relation, that is: kcat =

3.2. Initial lipolytic reaction rate The pH-stat titrations have been used to follow lipolytic reaction as a function of lipid nanoparticle size at pH 8.4 and 6.5. A summary of the obtained results of the initial lipolytic reaction rate on LCNPs as a function of GMO concentration is presented in Fig. 2. The GMO concentration in Fig. 2 is expressed as a surface concentration, i.e. the calculated number of GMO molecules on the nanoparticle surface, which are exposed for immediate lipase action. This was done to give a correct comparison of the effect of the lipolytic activity on particles with different sizes. The TLL concentration in the reaction vessel was kept constant at 1 × 10−4 mM, which means for the highest surface lipid concentration there are 45,000 surface substrate molecules per lipase molecule. It should be noted that the fact that at least for the larger particles, i.e. the highest total lipid substrate concentration, the initial lipolysis reaction rate increases linearly with the substrate concentration showing that the concentration of lipase is not the rate-limiting factor. The GMO surface concentration was calculated by assuming the GMO molecular volume and the cross sectional area of 593 A˚ 3 and 35 A˚ 2 , respectively [20]. For simplicity, it was also assumed that LCNPs are

Fig. 2. The initial lipase catalyzed reaction rate on GMO cubic LCNPs with the mean size of 190 nm (1) and 350 nm (2) as a function of surface GMO concentration at pH 8.4 (a) and 6.5 (b) obtained from pH-stat titration. In all cases the reaction media consisted of 1.5 mM potasium phosphate buffer, 30 mM KCl, and various concentrations of LCNPs. The total reaction volume was 20 ml. The lipolytic reaction was initiated with 20 ␮l 0.1 mM lipase solution and titrated with 2.5 mM KOH. Lines are drawn to guide the eye.

V

titrant

s



∗

Ctitrant , Venzyme ∗ Cenzyme

where Vtitrant is the volume of titrant solution (NaOH) added, Ctitrant is the concentration of titrant (in our case 2.5 or 25 mM NaOH), Venzyme is the volume of enzyme stock solution, and Cenzyme is the enzyme stock solution concentration. In order to prove that pHstat titration data is the result of the TLL catalytic action, control experiments with its inactive mutant iTLL were also performed. For the same reaction conditions the observed turnover numbers were about 200 times lower when compared to wild-type TLL. As can be immediately seen from the pH-stat titration data shown in Fig. 2, the initial enzymatic reaction rate for the GMO cubic nanoparticle is size dependent. At both pH values initial lipolysis rate is approximately 25% higher for large 350 nm LCNPs when compared to the small 190 nm particles. We believe that this difference can mainly be attributed to the different surface accessibility of the substrate to the lipase. As shown in previous cryo-TEM studies the particle size influences the shape and how GMO-based cubic nanoparticles are assembled [18,19]. Small nanoparticles with a mean size of about 190 nm are essentially made up of a single monocrystalline domain and these particles therefore have more of a cubic shape. If we would consider that the smaller particles were cubes instead of spheres, the differences in the lipolysis rate would be even larger. The large cubic LCNPs with a mean size of about 350 nm are rather polycrystalline particles composed of few smaller monocrystalline domains. In other words, larger particles have considerably more structural defects when compared to the small LCNPs. It is now generally accepted that the enzymatic activity of lipases is strongly affected by the nanostructure of the lipid–water interface. The most susceptible sites for the initial enzyme interaction with the substrate are the nanostructural defects of the interface [21–24]. At the same time, even large areas without structural defects are surprisingly stable upon enzyme action. Taking this into account we think that the larger number of surface defects of the 350 nm lipid nanoparticles is responsible for the increased initial enzymatic activity when compared to mostly smaller 190 nm monocrystalline particles. However, as shown by Boyd and co-workers the amount of stabilizer incorporated can vary with the structure and the lipid composition of the particles [25]. It cannot be ruled out that this would affect the lipase action on the particles. However, one would expect to see the largest effects on the larger particles as the F127 load per surface area unit on these particles are expected to be larger. We do not see any reduction in lipolytic activity on these particles, but rather the contrary. However detailed knowledge on how the stabilizer affects the lipolytic activity requires further investigation beyond the scope of this study. The lipase activity on GMO cubic nanoparticles appears to be very much pH dependent and somewhat surprisingly, the enzyme is almost twice more active at pH of 6.5 when compared to the pH value of 8.4. Previously it has been report that TLL have an optimal lipolytic activity at about pH 8 [26], as will be discussed further below. Note that the pH effects observed in the present study are in fact even a bit underestimated. For simplicity, the lipase turnover number values shown in Fig. 2 are calculated as a proportion of the amount of KOH titrant used to titrate released oleic acid assuming a stoichiometric reaction. On the other hand, it is well known that the dissociation constants, pKa , for long-chain fatty acids are much higher than for soluble carboxylic acids for which pKa values are

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

in the order of 5.0 [27]. For oleic acid, pKa values even higher than 9.0 are reported in the literature [11,28,29]. Salentinig et al. used a semi-empirical relation to connect the lattice parameter to the pH and pKa for the liquid crystalline phase consisting of GMO/oleic acid in excess water [11]. They reported pKa of about 6.1 for a H2 phase consisting of 44 wt% oleic acid and a higher content of oleic acid gives an increase in pKa . Their findings are consistent with our experimental results which show that during the lipolytic reaction on GMO cubic nanoparticles without using the pH control the pH decreases as oleic acid is formed to values even lower than 6.0 after the reaction had reached a steady state. For this reason in our pH-stat titration experiments at pH 6.5 the oleic acid is more protonated when compared to pH of 8.4. This means that real turnover number values are even higher than those presented in Fig. 2b, where simple stoichiometric reaction between oleic acid and KOH is assumed. On the other hand, the lipase used in this study displays its pH optimum in the alkaline pH range [26,30]. Studies of the Thermomyces lanuginosus lipase activity measurements on glycerol trioleate even show that the highest lipase activity is at neutral pH, and the activity practically stays the same in the pH range between 5 and 10 [31]. The reported pH optimum of TLL activity appears to contradict our data for the TLL activity measured with GMO cubic nanoparticles as substrates (Fig. 2). However it cannot be ruled out that the surface of the lipid nanoparticles changes slightly with the pH and hence contributes to the increased lipolytic activity at lower pH. It is known that GMO preparation used to make nanoparticles contains some amount of oleic acid [18]. At lower pH mainly protonated oleic acid may locally modify the lipid nanoparticle interface by lowering its curvature and thus making it slightly

5

more favorable for the lipase attachment and activation [32]. At higher pH oleic acid is predominantly deprotonated making the nanoparticle surface slightly negatively charged. At this pH the lipase itself is negatively charged which can make an additional electrostatic barrier for proper enzyme attachment to lipid nanoparticles and interfacial activation [15]. Another important factor is that the protonated oleic acid formed due to the hydrolysis of GMO at low pH can be more easily solubilized by a cubic phase at a low water content [33], i.e. in the interior of the nanoparticle. 3.3. Progress of the lipase catalyzed hydrolysis of LCNPs Further progress of the lipase catalyzed hydrolysis of GMObased LCNPs was followed by pH-stat titration for several hours. Fig. 3 shows pH-stat titration data of the lipase catalyzed hydrolyses versus time on small (Fig. 3a) and large (Fig. 3b) GMO cubic LCNPs as a function of time at pH 8.4. In both cases the hydrolysis of particles, measured as the amount of KOH added to keep the pH constant, is continuous with characteristic deceleration at longer reaction times. As with the initial hydrolysis rate data (Fig. 2), the data shown in Fig. 3a and b support the fact that the large GMO cubic LCNPs are hydrolyzed faster than the small particles. Although the surface concentration of GMO in large particles was only 1.4 times higher, the total amount of GMO molecules hydrolyzed (volume of KOH added) for large particles is about 1.7 times higher than for small LCNPs after 15 h of reaction. The obtained difference 1.7/1.4 = 22% is very similar to the observed difference in GMO molecules hydrolyzed for the initial reaction (25%) (Fig. 2a). As also evidenced from cryo-TEM images in Fig. 3, the morphology and nanostructure of initially well-ordered cubic LCNPs

Fig. 3. (Top) The lipase-catalyzed lipolysis reaction for small (a) and large (b) GMO cubic LCNPs as a function of time in 1.5 mM potassium phosphate buffer with 30 mM KCl at pH 8.4 as observed by pH-stat titration. The bulk LCNP concentration was 0.5 wt% (1.35 mM in terms of GMO surface concentration) and 1.25 wt% (1.86 mM in terms of GMO surface concentration) for small (a) and large particles (b), respectively. The total reaction volume was 20 ml. The lipolytic reaction was initiated with 20 ␮l 0.1 mM lipase solution and titrated with 25 mM KOH. Continuous lines denote the reaction pH and dashed lines show the volume of titrant added. (Bottom) cryo-TEM images of small LCNPs taken 15 h after lipolytic reaction initiation.

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10 6

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

becomes radically different upon exposure to lipase. After 15 h of reaction only rather disorganized “sponge”-like aggregates and disordered multilamellar onion-like structures are found in the reaction mixture. The appearance of discrete aggregates suggests that the formation and accumulation of oleic acid within the lipid particle during lipase catalyzed hydrolysis of GMO is responsible for the observed structural change. Other phase behavior studies of mixtures of GMO and oleic acid also show that oleic acid promotes the formation of lamellar type aggregates at pH values higher than 8 where the fatty acid is mostly deprotonated [34]. The lipase catalyzed hydrolysis of cubic LCNPs at pH 6.5 versus time is dramatically different to the reaction evolution at pH 8.4 (Fig. 4). An added amount of KOH to the reaction mixture is no longer continuous with time and the pH temporarily exceeds the set value of 6.5. As a response to the lipolytic reaction two spikes and one broad increase in pH are observed in the titration curve. Here it should be noted that our experimental pH-stat setup allows only one-directional titration, i.e. in this case titration with the base solution as response to a lowering of pH. An increase in pH above the set value stops the dispensing of KOH and it is only resumed once the set value of pH is reached again. The trend of pH increases is identical for both small and large LCNPs and is reproducible. Considering catalytic reaction being rather simple with only oleic acid raised the question is why at some points the pH of reaction media increases and what is responsible for such rapid consumption of protons. We believe that the only cause of the observed effect is an increase in the oleic acid pKa to larger value at particular stages of the reaction. This results in rapid protonation of the molecule and removal of fraction of free protons from the solution with subsequent slight increase in pH of the reaction media. This is apparent when we consider the lipolysis reaction together with dissociation of the fatty acid

R − COOCH(CH2 OH)2 + H2 O Lipase

−→ R − COO− + H+ + (CH2 )2 CH(OH)3 Lipase

R − COOCH(CH2 OH)2 + H2 O −→ R − COOH + (CH2 )2 CH(OH)3

(1) (2)

The protonated and soap form of the fatty acid depends on pH and is related by R − COOH ↔ H+ + R − COO−

(3)

OH− + H+ ↔ H2 O

(4)

Two things are important to bear in mind here: 1. The system considered here consists of a substrate dispersed in a large excess of water so that the reaction equilibrium is pushed toward the forward lipolytic reaction (Schemes 1 and 2) catalyzed by the lipase. 2. The pKa for the acid of the fatty acid is as discussed above higher in a self-assembly structure than for the corresponding fatty acid in solution. In fact the apparent pKa,app depends on electrostatic effects (pKel ), determined by the surface charge density, pH and ionic strength, as well as local effects (pKp ) include effects associated with the dielectric discontinuity at the interface as well as the curvature [35]. The apparent pKa,app can be expressed as: pKa,app = pKa + pKel + pKp

(5)

Here it should be noted that pKp is challenging to determine as it includes a number of factors that are not directly measureable [35]. In the present study, the lipids were dispersed in 30 mM

Fig. 4. (Top) The lipase catalyzed lipolysis reaction for small (a) and large (b) GMO cubic LCNPs as a function of time as observed by pH-stat titration. LCNP concentration and reaction conditions as in Fig. 3. The continuous line denotes the reaction media pH and the dashed line shows the volume of titrant added. (Bottom) cryo-TEM images of the sample taken after 12 h of lipase-catalyzed hydrolysis at pH 6.5 of GMO cubic small LCNPs.

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

KCl and 1.5 mM phosphate buffer. At such a relatively high ionic strength one would expect that pKel is relatively low, in particular, initially during the lipolysis. Consequently, pKp is expected to be relatively more important. The large pH change occurs when 1.1 or 2.2 ml of 25 mM KOH was added for small and large particles, respectively. This corresponds to a fatty acid concentration arising from the lipolytic activity of 1.38 and 2.75 mM, respectively, in relation to an initial GMO concentration of 12.6 and 31.5 mM, respectively. This means that the lipid in the dispersed phase at the pH increase would contain about 10 mol% (or about 8 wt%) oleic acid/oleate. The GMO/oleic acid/aqueous phase diagram suggests that this would correspond to a phase transition to a H2 or a cubic micellar phase [33]. This is supported by the cryo-TEM images of the lipolysis product shown in Fig. 4. After 12 h of lipolysis at pH 6.5, the nanostructure and morphology of the LCNPs is very different than after reaction at pH 8.4. Only nicely spherically shaped particles with evenly dense inner cores are found suggesting emulsion-like structured particles which are composed of disordered reversed micelles (L2 phase). At this pH oleic acid is more protonated (i.e. more hydrophobic) and its accumulation within the particles during GMO hydrolysis naturally induces the formation of more reversed phases than the initial substrate bicontinuous cubic phase. In this case, the lipolytic reaction is expected to transform the cubic LCNPs to L2 phase dispersions as the end product based on the phase behavior of the system [33]. The interior of these particles

7

is expected to be dominated by protonated oleic acid. In order for this kind of transition to occur, initially continuous bilayerbased cubic phase has to transform into discrete reversed micelles, which is associated with a drastic increase in the negative curvature. Such a transition also creates molecular packing frustration in the hydrophobic regions between the reversed micelles. Since the oleic acid may exist either in the protonated or deprotonated form the easiest way to relief this packing stress is to protonate the molecule and localize it in the hydrophobic region between micelles. In other words, the release of the packing stress within the lipid aggregate is associated with an increase in pKa,app . In addition, to the apparent increase in the negative curvature and other local effects at the interface, this will also contribute to the increase in pKp . Hence, the oleic acid pKa,app value in the reversed micellar phase is higher when compared to that within the bicontinuous cubic phase. SAXD experiments were further employed to probe the progress of nanostructural changes in cubic LCNPs upon lipase action (Fig. 5). For practical reasons the lipase concentration was adjusted to be able to perform full transformation of LCNPs into reaction products within or less than 1 h. Due to limitations in the experimental setup at the beamlines I711 and I911, it was not possible to perform simultaneous titration and diffraction experiments in a controlled way and one should note that slight differences in mixing protocol for the two studies might give slightly different kinetics. We note that the sequence of phases should be the same independently

Fig. 5. SAXD data of the progress of lipase-induced hydrolysis of GMO cubic small LCNPs as a function of time at pH 8.4 (a), 8.0 (b), 7.5 (c), and 6.5 (d). 0.5 mL of LCNP dispersion was mixed with 0.5 mL 200 mM of either potassium phosphate or TRIS (at pH 8.4) buffer solution before an appropriate small amount of 0.01 mM lipase stock solution was added to initiate hydrolysis. The amounts of lipase added were: 30 ␮l in (a), 50 ␮l in (b), 30 ␮l in (c), and 20 ␮l in (d).

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10 8

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

of the lipase concentration. Recently, Boyd et al. demonstrated that it was possible to combine the pH-stat experiment with synchrotron SAXD, although they noted that precautions have to be taken to avoid beam damage [36,37]. As shown in Fig. 5a, at pH 8.4 the initially ordered bicontinuous Im3m structure of LCNPs gradually transforms into less ordered sponge- and onion-like liposomes and possibly also vesicles, which is evidenced by the gradual disappearance of the (110), (200), and (211) Bragg peaks and their transformation into a weak diffraction/scattering pattern at low q. Such a rather disordered SAXD pattern is in line with cryo-TEM images of hydrolyzed LCNPs in Fig. 3 showing only some distance correlation between bilayers. As reaction progresses, the lattice parameter of the Im3m cubic phase increases from 13 to about 17 nm until structural order from this phase is lost. As expected from titration experiments, a different scenario of LCNP transformation catalyzed by lipase activity is observed at pH 6.5 (Fig. 5d). At first, bicontinuous Im3m LCNPs quite rapidly transform into particles with an internal reversed hexagonal nanostructure, H2 , which is verified by the typical appearance of three strong (10), (11), and (20) Bragg peaks in the diffractograms (Fig. 6 top). Cryo-TEM images taken at this stage of the reaction confirms the hexagonal nanostructure of LCNPs by their typical interior arrangement of bent long strings of rod-like micellar aggregates (Fig. 6 bottom left). During lipolysis reaction the lattice parameter of this phase decreases slightly from about 6.3 to 5.5 nm before transforming into yet another liquid crystalline structure. It shows typical (1 1 1), (2 2 0), (3 1 1), (2 2 2), (4 0 0), and (3 3 1) reflections of √ √ the face-centered Fd3m cubic phase positioned at ratios 3: 8: √ √ √ √ 11: 12: 16: 19 (Fig. 6 top). The fact that it appears after the reversed hexagonal phase suggests that this cubic phase is composed of discrete reversed micelles. It is not very surprising that the cryo-TEM images of LCNPs do not show a nice pattern

Fig. 6. Representative SAXD patterns (top) and cryo-TEM images (bottom) of the intermediate nanoparticles with internal hexagonal (1, bottom left) and reversed micellar Fd3m cubic (2, bottom right) nanostructure obtained during the progress of the lipase-induced hydrolysis of GMO cubic small LCNPs at pH 6.5 shown in Fig. 5d.

(Fig. 6 bottom right) since micelles in the particle are small, discrete, and arranged in a ordered but rather complicated cubic pattern. In contrast to bicontinuous and hexagonal structures, reversed micellar cubic phase aqueous compartments are discontinuous in all directions. It would require ideal orientation of spherical particles in order to get be able to view them in such a direction that the cubic arrangement is clearly visible. As LCNP hydrolysis advances, the lattice parameter of the Fd3m cubic phase shrinks from about 15.9 to 13.6 nm. Finally, further hydrolysis transforms the Fd3m LCNP into particles composed of a disordered reversed micellar phase, which in SAXD displays only a diffuse broad scattering feature at q value corresponding to the location of most intensive (3 1 1) and (2 2 2) Bragg peaks of the preceding Fd3m phase. For comparison, Fig. 5 also includes results from the progress of lipolytic reaction at pH 7.5 and 8.0. The lipid hydrolysis at pH 8.0 induced transformation of LCNPs is similar to that at pH 8.4 (Fig. 5b). The cubic Im3m structure gradually increases its lattice dimensions, loses long-range order, and transforms into particles with a spongelike interior as shown by widening of the Bragg peaks. In contrast, at pH 7.5 the progress of the lipase-induced LCNP phase transitions follows a similar sequence as at pH 6.5: bicontinuous cubic Im3m, reversed hexagonal, reversed micellar cubic Fd3m, and disordered reversed micelles (Fig. 5c). The obtained results clearly show that with increasing pH somewhere between pH 7.5 and 8.0, oleic acid, generated by the lipolytic process, is deprotonated enough to change the transformation pathway of the initially bicontinuous GMO cubic LCNPs. At slightly higher pH than 7.5 oleic acid is sufficiently deprotonated to reduce the negative curvature lipid–water interfaces of the particle nanostructure. For lower pH, oleic acid is protonated and hydrophobic enough to be able to induce the whole sequence of LCNP phase transitions toward the higher negative curvature of the lipid–water interface. Although sequences of phase transformations at pH 7.5 and 6.5 are the same, the progress of the phase transition appearances is clearly different (Fig. 5c and d). Even considering the fact that for reaction at pH 7.5, when 1.5 times larger amount of lipase was used all corresponding phase transitions occur much later as compared to the lipolysis at pH 6.5. For example, at pH 7.5 the phase transition from reversed hexagonal to reversed micellar cubic Fd3m occurs approximately 1300 s after the lipolytic reaction was initiated (Fig. 5c) whereas at pH 6.5 the same phase transition takes place approximately 700 s after adding the lipase (Fig. 5d). Similarly, the transformation to disordered reversed micelles occurs at about 2700 and 1400 s after the start of reaction at pH 7.5 and 6.5, respectively. Considering 1.5 times difference in the lipase amount used both transitions may be roughly estimated to take place 3 times later at pH 7.5 when compared to pH 6.5. Such differences can be explained by the differences in oleic acid protonation and that more or less protonated oleic acid is needed to accumulate during the reaction for transition to the more reversed phase to occur. Further employed pH-stat titration was performed at pH 7.5 to probe if the differences in the kinetics of the phase transformations between pH 7.5 and 6.5 observed by SAXD could be related to differences in the reaction progress. Fig. 7 shows pH-stat titration data for the lipase catalyzed lipolyses for small GMO cubic LCNPs as a function of time at pH 7.5. The observed pH increase during titration at pH 7.5 (Fig. 7) is curiously similar to the results presented in Fig. 4a. Although an initial small increase in pH is not observed the second spike followed by a broad increase in pH has exactly the same shape as at pH 6.5. More importantly, this feature appears at pH 7.5 approximately 6.5 h after the reaction initiation while at pH 6.5 it starts about 2 h after reaction was initiated by adding the lipase. The observed difference in kinetics is again about 3 times, which remarkably well coincides with the differences in time when the phase transition occurs at the two pH.

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

Fig. 7. The lipase-catalyzed lipolysis reaction on small GMO cubic LCNPs as a function of time as observed by pH-stat titration. LCNP concentration and reaction conditions as in Fig. 3. The continuous line denotes the reaction media pH and the dashed line shows the volume of titrant added.

This correlation clearly confirms that the phase transitions of the LCNPs are responsible for the sudden increase in pH during lipasecatalyzed reaction during which oleic acid is produced. Although the experimental condition during pH-stat titration and SAXD are very different we can conclude that the large broad pH increase (due to changes in an apparent pKa value of oleic acid) during the pH-stat titration with KOH is most likely associated with LCNP phase transition from the reversed hexagonal phase to the reversed Fd3m micellar cubic followed by its transformation to disordered reversed micelles. This phase transition would as discussed above, create larger packing stress when compared to the transformation from the bicontinuous cubic to reversed hexagonal phase. In addition to the complex arrangement of reversed micelles, the Fd3m phase is much less hydrated, which further promotes the oleic acid protonation and hence facilitate the incorporation into the hydrophobic regions of the nanostructure. It interesting to note that more complex mixtures of lipids show a similar evolution of structures, even though you might start with a different liquid crystalline phase. For instance Wadsäter et al. showed how the structure of the glyceroldioleate/Soy-PC LCNPs evolves during the exposure to a triacylglycerol lipase (TGL) at pH 7.5 [38]. It should be noted here that TGL catalyzes the degradation of only glycerodioleate to monoglycerides, glycerol, and free fatty acids. As for the GMO system the internal liquid crystalline structure of the nanoparticles changes continuously from the reversed Fd3m structure to structures with a less negative curvature (hexagonal, bicontinuous cubic, and sponge phases) and finally results in the formation of multilamellar liposomes. 4. Conclusions Lipase-catalyzed reactions are of large industrial importance in many technical applications, including detergency and food processing. Water-soluble lipases are rather complex enzymes for which the action takes place at the aqueous–lipid interface and the activity is extremely dependent on the structural organization of the lipid/water boundary. During the reaction hydrolysis products continuously change the interface characteristics, and alter cleavage mechanisms from the complex lipid aggregate. The challenge is to understand how the nanostructural changes in the substrate can control the activity of the enzyme. Here, well-defined glycerol monooleate-based liquid crystalline nanoparticles have been successfully employed to study lipolytic reaction. The system represents a simple and straightforward example of how the lipase reactions can be studied by a combination of various techniques. It has been shown there is possibility to probe the kinetic properties of the catalysis and simultaneously follow the nanostructural changes in the self-assembled lipid substrate. Notably, the surface

9

organization, nanostructural features, and external factors as pH affect the progress of the lipolysis together with changes in the lipid liquid crystalline phase structure. We have observed that not only the pH optimum of the catalytic reaction controls the lipolysis but also how well the self-assembled substrate can accommodate the product. We have shown that the lipid aggregate phase transformation sequence can indeed modulate the activity of the enzyme. It is also interested to note that pH changes during pHstat titration can be used to detect changes in pKa , which reflects liquid crystalline phase changes. Today many of lipases are used in a range of applications including detergency, materials science (new lipid-based materials and structures), food technology (functional foods with less fat but a maintained or increased structure) as well as biotechnology including lipid synthesis/modification from pharmaceuticals to biofuel-based natural oils and fats. The challenge is often to be able to screen the lipases developed with certain activity and specificity. The type of liquid crystalline self-assembly particles studied here, for which the internal structure and substrate composition can be varied to mimic a large range of relevant type of substrate, can be a very useful and versatile tool. We believe that some of the findings presented here will open new possibilities to provide a handle in understanding physiological processes and controlling technical applications.

Acknowledgements This work was financed by the Camurus Lipid Research Foundation and Swedish Foundation for Strategic Research (SSF) via frame work grant RMA08-0056 as well as EUSTREP FP6 project BIOSCOPE (Contract No. NMP4-CT-2003-505211). We are grateful to Gunnel Karlsson for her help with cryo-TEM instrumentation. The authors also thank the Swedish synchrotron X-ray facility MAX-lab for allocated beamtime at the I711 and I911 beamlines and Ana Labrador, Sylvio Haas, and Tomás Plivelic for technical support during experiments.

References [1] R.D. Schmid, R. Verger, Lipases interfacial enzymes with attractive applications, Angew. Chem. Int. Ed. 37 (1998) 1608–1633. [2] R. Verger, ‘Interfacial activation’ of lipases: facts and artifacts, Tibtech 15 (1997) 32–38. [3] J.S. Patton, M.C. Carey, Watching fat digestion, Science 204 (1979) 145–148. [4] M. Lindström, H. Ljusberg-Wahren, K. Larsson, B. Borgström, Aqueous lipid phases of relevance to intestinal fat digestion and absorption, Lipids 16 (1981) 749–754. [5] J.S. Patton, R.D. Vetter, M. Hamosh, B. Borgström, M. Lindström, M.C. Carey, The light-microscopy of triglyceride digestion, Food Microstruct. 4 (1985) 29–41. [6] M. Golding, T.J. Wooster, The influence of emulsion structure and stability on lipid digestion, Curr. Opin. Colloid Interf. Sci. 15 (2010) 90–101. [7] P. Reis, K. Holmberg, H. Watzke, M.E. Leser, R. Miller, Lipases at interfaces: a review, Adv. Colloid Interf. Sci. 147-148 (2009) 237–250. [8] J. Borné, T. Nylander, A. Khan, Effect of lipase on different lipid liquid crystalline phases formed by oleic acid based acyl glycerols in aqueous systems, Langmuir 18 (2002) 8972–8981. [9] J. Borné, T. Nylander, A. Khan, Effect of lipase on monoolein-based cubic phase dispersion (cubosomes) and vesicles, J. Phys. Chem. B 106 (2002) 10492–10500. [10] F. Caboi, J. Borné, T. Nylander, A. Khan, A. Svendsen, S. Patkar, Lipase action on a monoolein/sodium oleate aqueous cubic liquid crystalline phase—a NMR and X-ray diffraction study, Colloids Surf. B 26 (2002) 159–171. [11] S. Salentinig, L. Sagalowicz, O. Glatter, Self-assembled structures and pKa value of oleic acid in systems of biological relevance, Langmuir 26 (2010) 11670–11679. [12] S. Salentinig, L. Sagalowicz, M.E. Leser, C. Tedeshi, O. Glatter, Transitions in the internal structure of lipid droplets during fat digestion, Soft Matter 7 (2011) 650–661. [13] J.L. Brennan, A.G. Kanaras, P. Nativo, T.R. Tshikhudo, C. Rees, L.C. Fernandez, N. Dirvianskyte, V. Razumas, M. Skjot, A. Svendsen, C.I. Jorgensen, R. Schweins, M. Zackrisson, T. Nylander, M. Brust, J. Barauskas, Enzymatic activity of lipasenanoparticle conjugates and the digestion of lipid liquid crystalline assemblies, Langmuir 26 (2010) 13590–13599. [14] J. Barauskas, T. Nylander, Lyotropic liquid crystals as delivery vehicles for food ingredients, in: N. Garti (Ed.), Delivery and Controlled Release of Bioactives in

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052

G Model COLSUB-7057; No. of Pages 10 10

[15]

[16]

[17]

[18]

[19] [20] [21]

[22]

[23]

[24]

[25]

[26]

ARTICLE IN PRESS J. Barauskas et al. / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx

Foods and Nutraceuticals, Woodhead Publishing Series in Food Science, Technology and Nutrition, Cambridge, 2008, pp. 107–131. M. Martinelle, M. Holmquist, K. Hult, On the interfacial activation of Candida antarctica lipase A and B as compared with Humicola lanuginosa lipase, Biochim. Biophys. Acta 1258 (1995) 272–276. Y. Cerenius, K. Ståhl, L.A. Svensson, T. Ursby, Å. Oskarsson, J. Albertsson, A. Liljas, The crystallography beamline I711 at MAX II, J. Synchrotron Rad. 7 (2000) 203–208. M. Knaapila, C. Svensson, J. Barauskas, M. Zackrisson, S.S. Nielsen, K.N. Toft, L. Arleth, B. Vestergaard, Y. Cerenius, A new small-angle X-ray scattering set-up on the crystallography beamline I711 at MAX-lab, J. Synchrotron Rad. 16 (2009) 498–504. J. Barauskas, M. Johnsson, F. Joabsson, F. Tiberg, Cubic phase nanoparticles (Cubosome): principles for controlling size, structure, and stability, Langmuir 21 (2005) 2569–2577. J. Barauskas, M. Johnsson, F. Tiberg, Self-assembled lipid superstructures: beyond vesicles and liposomes, Nano Lett. 5 (2005) 1615–1619. T. Landh, Phase behavior in the system pine needle oil monoglyceridesPoloxamer 407-water at 20 degrees C, J. Phys. Chem. 98 (1994) 8453–8467. K. Balashev, N.J. DiNardo, T.H. Callisen, A. Svendsen, T. Bjørnholm, Atomic force microscope visualization of lipid bilayer degradation due to action of phospholipase A(2) and Humicola lanuginosa lipase, Biochim. Biophys. Acta 1768 (2007) 90–99. K. Balashev, M. Gudmand, L. Iversen, T.H. Callisen, A. Svendsen, T. Bjørnholm, Humicola lanuginosa lipase hydrolysis of mono-oleoyl-rac-glycerol at the lipidwater interface observed by atomic force microscopy, Biochim. Biophys. Acta 1615 (2003) 93–102. O.G. Berg, Y. Cajal, G.L. Butterfoss, R.L. Grey, M.A. Alsina, B.Z. Yu, M.K. Jain, Interfacial activation of triglyceride lipase from Thermomyces (Humicola) lanuginosa: kinetic parameters and a basis for control of the lid, Biochemistry 37 (1998) 6615–6627. M. Grandbois, H. Clausen-Schaumann, H.E. Gaub, Atomic force microscope imaging of phospholipid bilayer degradation by phospholipase A(2), Biophys. J. 74 (1998) 2398–2404. A.J. Tilley, C.J. Drummond, B.J. Boyd, Disposition and association of the steric stabilizer Pluronic F127 in lyotropic liquid crystalline nanostructured particle dispersions, J. Colloid Interf. Sci. 392 (2013) 288–296. M.L.M. Fernandes, N. Krieger, A.M. Baron, P.P. Zamora, L.P. Ramos, D.A. Mitchell, Hydrolysis and synthesis reactions catalysed by Thermomyces lanuginosa lipase in the AOT/Isooctane reversed micellar system, J. Mol. Catal. B 30 (2004) 43–49.

[27] M. Ptak, M. Egret-Charlier, A. Sanson, O. Bouloussa, A NMR study of the ionization of fatty acids, fatty amines and N-acylamino acids incorporated in phosphatidylcholine vesicles, Biochim. Biophys. Acta 600 (1980) 387–397. [28] D.P. Cistola, J.A. Hamilton, D. Jackson, D.M. Small, Ionization and phase behavior of fatty acids in water: application of the Gibbs phase rule, Biochemistry 27 (1988) 1881–1888. [29] J.R. Kanicky, D.O. Shah, Effect of degree, type, and position of unsaturation on the pK(a) of long-chain fatty acids, J. Colloid Interf. Sci. 256 (2002) 201–207. [30] G.E. Crooks, G.D. Rees, B.H. Robinson, M. Svensson, G.R. Stephenson, Comparison of hydrolysis and esterification behavior of Humicola lanuginosa and Rhizomucor miehei lipases in AOT-stabilized water-in-oil microemulsions: I. Effect of pH and water content on reaction kinetics, Biotechnol. Bioeng. 48 (1995) 78–88. [31] K.R. Poulsen, T. Snabe, E.I. Petersen, P. Fojan, M.T. Neves-Petersen, R. Wimmer, S.B. Petersen, Quantization of pH: evidence for acidic activity of triglyceride lipases, Biochemistry 44 (2005) 11574–11580. [32] Y. Aota-Nakano, S.J. Li, M. Yamazaki, Effects of electrostatic interaction on the phase stability and structures of cubic phases of monoolein/oleic acid mixture membranes, Biochim. Biophys. Acta 1461 (1999) 96–102. [33] J. Borné, T. Nylander, A. Khan, Phase behavior and aggregate formation for the aqueous monoolein system mixed with sodium oleate and oleic acid, Langmuir 17 (2001) 7742–7751. [34] D.A. Ferreira, M.V.L.B. Bentley, G. Karlsson, K. Edwards, Cryo-TEM investigation of phase behaviour and aggregate structure in dilute dispersions of monoolein and oleic acid, Int. J. Pharm. 310 (2006) 203–212. [35] G. Cevc, J.M. Sedon, R. Hartung, W. Eggert, Phosphatidylcholine-fatty acid membranes. I. Effects of protonation, salt concentration, temperature and chain-length on the colloidal and phase properties of mixed vesicles, bilayers and nonlamellar structures, Biochim. Biophys. Acta 940 (1988) 219–240. [36] S. Salentinig, S. Phan, A. Hawley, B.J. Boyd, Self-assembly structure formation during the digestion of human breast milk, Angew. Chem. 126 (2014) 1–5. [37] D. Warren, M. Anby, A. Hawley, B. Boyd, Real time evolution of liquid crystalline nanostructure during the digestion of formulation lipids using synchrotron small-angle X-ray scattering, Langmuir 27 (2011) 9258–9534. [38] M. Wadsäter, J. Barauskas, T. Nylander, F. Tiberg, Formation of highly structured cubic micellar lipid nanoparticles of soy phosphatidylcholine and glycerol dioleate and their degradation by triacylglycerol lipase, ACS Appl. Mater. Interf. 6 (2014) 7063–7069.

Please cite this article in press as: J. Barauskas, et al., Thermomyces lanuginosus lipase-catalyzed hydrolysis of the lipid cubic liquid crystalline nanoparticles, Colloids Surf. B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.04.052