Original papers
Preparation of a nanostructured multi-phase lipid carrier for iron encapsulation: A lipase-triggered release of ferric ions
2021 年 27 巻 4 号 p. 559-565
詳細
2021 年 27 巻 4 号 p. 559-565
A multi-phase lipid particle, a novel iron carrier, was prepared using a mixture of wax (derived from rice bran) and coconut oil. A high shear homogenization method was employed for particle processing. The main body of the prepared particle consisted of a waxy matrix containing nanostructured oil phases. When these particles were dispersed in an aqueous solution with lipase, the oil phases were digested and free fatty acids were released. Water droplets containing dissolved ferric ammonium sulfate were emulsified in coconut oil, and this W/O emulsion was used in particle preparation. This particle was designed for a triggered release of ferric ions due to the action of a digestive enzyme. We successfully demonstrated the preparation of the multi-phase lipid particle ranging in size from 10–160 μm and the lipase-induced release of ferric ions.
Lipid-based particulate systems are increasingly being recognized as potential delivery vehicles for nutraceutical and pharmaceutical agents (Babazadeh et al., 2016, 2017; Tamjidi et al., 2013). These particles are designed for targeted delivery of active ingredients to various regions in the gastrointestinal tract, enhancing their bioavailability. Solid nanoparticles made from a single solid lipid core are called solid lipid nanoparticles (SLNs). Nanostructured lipid carriers (NLCs) are lipid-based systems that contain nanostructures prepared by blending lipids. Numerous NLCs have been proposed in recent studies that illustrate the relationships between preparation methods, nanostructures, and biological functionalities (Gordillo-Galeano and Mora-Huertas, 2018). NLCs can be classified into three types based on the lipid phase: imperfect crystal, multiple type, and amorphous type (Sharma and Baldi, 2018). The preparation methods of NLCs are also classified into three categories (Gordillo-Galeano and Mora-Huertas, 2018): high energy methods (homogenization, sonication, microwave assisted, etc.), low energy methods (emulsification, phase inversion, membrane, etc.), and organic solvent-based methods (solvent evaporation, solvent diffusion, etc.). Formulations and preparation methods contribute to the resultant particle properties, such as particle geometry, phase transition property, encapsulation capacity, and bioavailability. It is desirable to develop a technique to manufacture NLCs with high capability and a characteristic design, using natural excipients that are biodegradable and biocompatible (Gasco, 2007).
Several studies have recognized NLCs as carriers for lipophilic compounds such as polyphenols, carotenoids, and phytosterols (Bagherpour et al., 2017; da Silva Santos et al., 2019; Ezhilarasi et al., 2016; Gomes et al., 2019; Hejri et al., 2013; Hentschel et al., 2008; Liu and Wu, 2010; Ni et al., 2015; Pimentel-Moral et al., 2019; Tamjidi et al., 2013). The oral administration of these NLCs is expected to deliver entrapped compounds via the gastrointestinal tract region that is associated with lipid digestion (Jannin et al., 2015). Thus, it is an important challenge to design NLCs capable of triggered release of entrapped material by the action of digestive enzymes. NLCs can also be employed as an effective carrier for water soluble materials. Zariwala et al. demonstrated the preparation of an iron loaded SLN, in which ferrous sulfate was incorporated in the chitosan matrix that coated the lipid particle (Zariwala et al., 2013). Singh et al. developed SLNs loaded with a water soluble drug (zidovudine) by applying the solvent evaporation method to a W/O/W emulsion (Singh et al., 2010). Water droplets containing zidovudine were dispersed in the lipid matrix. These reports showed the potential use of lipid-based matrices as delivery vehicles of hydrophilic compounds. In this context, we were motivated to develop a practical technique for the preparation of an iron delivery system, using a simplified protocol with reasonable functionality and food system compatibility.
In this study, we attempted to entrap ferric ions in a multiphase lipid carrier. Considering the human nutrient absorption mechanism, the ferrous ion should be selected for investigation. However, in the present study, which focuses on food processing, ferric ions were selected in order to avoid unfavorable loss of ions (e.g., due to oxidization) during processing. A mixture of wax (derived from rice bran) and triglyceride (coconut oil) was used to develop a nano-phase separation structure. Water droplets containing dissolved ferric ions were stabilized in the triglyceride phase. This particle design was expected to induce triggered release of ferric ions by lipase-induced lipolysis of the triglyceride phase. A high shear homogenization method was used to produce the nanostructured multi-phase lipid carriers (NMLCs).
Chemicals Waxy residue obtained via oil fractionation of rice bran was provided by Tsuno Co., Ltd. (Wakayama, Japan). Lipase from porcine pancreas (Type II, 0.79 unit/mg) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Coconut oil, sorbitan monooleate, ammonium iron (III) sulfate 12-water, citric acid, potassium hydrogen sulfate, thioglycolic acid, and all other chemicals were purchased from Fujifilm Wako Pure Chemical (Osaka, Japan).
Purification of rice bran wax The waxy residue was purified to remove traces of the triacylglyceride fraction. It was first dissolved in toluene to obtain a final concentration of 20 % (w/w). The wax was then precipitated by adding a threefold volume of 2-propanol as an anti-solvent. The precipitate was collected via centrifugation and vacuum evaporation. The yield of wax was around 65 %. The purified rice bran wax was used in this study.
Preparation of multi-phase lipid particles Powdered rice bran (0.15 g) was weighed in a 10 mL test tube and dissolved in 5 mL of water. This mixture was heated up to 95 °C using an electric heating device equipped with an aluminum heat block (DTU-18, Taitec Co., Saitama, Japan). The molten wax floated on the water. Further, 0.04 g of coconut oil was added to the molten phase and subsequently homogenized for 5 sec at 15 000 rpm using a mechanical homogenizer (T25 Ultra-Turrax, IKA Works Inc., Wilmington, NC, USA). The homogenized mixture was then immediately poured into a beaker containing water precooled at 0 °C (400 mL). The dispersed mixture of molten wax and oil rapidly solidified in the cold water while maintaining a spherical shape. These particles were used as the dispersion after removing excess amounts of water by pipetting. The particle concentration in the dispersion was around 0.05 g/mL.
A 10 % (w/w) sorbitan monooleate preparation was generated by dissolving it in coconut oil. An aqueous solution of ammonium iron (III) sulfate (2 mol/L) was separately prepared. These solutions were mixed at a 9:1 ratio (volume of oil phase:volume of aqueous phase) and emulsified by homogenization for 1 min at 15 000 rpm using the mechanical homogenizer. The obtained W/O emulsion was mixed with molten wax, and the particles were prepared using precooled water as previously mentioned. A schematic representation of this procedure is shown in Figure 1.
Preparation scheme.
Estimation of free fatty acids (FFAs) and iron released during lipid hydrolysis A lipase solution (0.2 mg/mL) was prepared in 38.2 mmol/L of maleic acid buffer solution with a pH of 6.5 (137 mmol/L of NaCl and 20 mmol/L of CaCl2). In a 50 mL glass vial, 10 mL of lipase solution and 10 mL of the prepared lipid nanoparticle dispersion were aliquoted. These vials were incubated in a water bath at 37 °C with shaking. The vials were removed at selected time points and 20 mL of 2-propanol was immediately added to stop the enzymatic reaction. Next, 100 µL of phenolphthalein solution (1 % (w/v)) was added to 10 mL of test solution, and titrated against 50 mmol/L of sodium hydroxide to estimate the amount of FFAs produced by hydrolysis of coconut oil via the action of lipase. Each titration was performed in triplicate using more than 3 different prepared nanoparticle dispersions. The remaining 10 mL of test solution was used to determine the amount of iron released. Significant amounts of ferric ions are adsorbed on the surface of the prepared particles. Therefore, the rest of the slurry was washed using 0.1 M HCl solution and mixed with the test solution for the analysis of iron.
Estimation of ferric ions The amount of ferric ions was analyzed via a colorimetric method using thioglycolic acid. A mixture of 0.25 mL of citric acid (5 % (w/v)), 0.75 mL of potassium hydrogen sulfate aqueous solution (15 % (w/v)), 0.25 mL of aqueous thioglycolic acid (4.3 % (v/v)), and a drop of ammonium solution was added to 3.5 mL of test solution. The absorbance of the solution was immediately measured at a wavelength of 530 nm using a UV-Vis spectrophotometer (U-5100, Hitachi High-Technologies Corporation, Tokyo, Japan). These measurements were taken in triplicate.
It should be noted that hydrolysis products (FFAs) affect the resulting absorbance of the test solution. Therefore, multiple calibration curves were obtained using different amounts of pre-mixed FFAs (obtained by lipolysis of coconut oil by lipase). This was used as a standard for the analyses of test solutions.
Estimation of iron encapsulation efficiency The particle dispersion was first dried in an oven at 30 °C overnight. The dried powder was then incinerated at 800 °C. Weight loss due to the calcination process was measured using a thermogravimetric device (STA6000, PerkinElmer Inc., Shelton, CT, USA). Assuming that the remaining ash content corresponded to the iron oxide (III) derived from entrapped iron, the iron encapsulation efficiency was estimated.
SEM observation The microstructure of the prepared particles was observed using scanning electron microscopy (SEM) (SU-6600, Hitachi High-Technologies, Tokyo, Japan). Observation was carried out at low vacuum mode (ca 50 Pa) to minimize vaporization of lipid components.
Particle observation Prepared particles were observed using SEM (Figure 2). The spherical particles exhibited a rough surface with the presence of void pores. Close-up images of the particle voids showed the presence of channel-like structures. This could be attributed to phase separation between rice bran wax and coconut oil during solidification. The channel diameters measured approximately 1 to 2 µm. Figure 3 shows the particle size distribution. The particle size ranged from 10 to 160 µm, with a mean value of 54 µm. The size of particles in this process was determined by the homogenization step of molten wax and the oil mixture. Droplets of the dispersed mixture solidified in the cooling step and the original droplet sizes were maintained. It was also possible that the droplets fused with each other to produce larger particles. The non-spherical particles observed in the SEM images could be due to fusing.
SEM images of the prepared NMLCs.
Particle size distribution of the prepared NMLC.
Release of FFAs via the action of lipase Particles without iron load (prepared using rice bran wax and coconut oil) and particles without oil were prepared to investigate the triggered release of FFAs by the action of lipase (Figure 4). When lipase was added to the suspension of particles prepared only with rice bran wax, only small amounts of FFAs were released. The crude rice bran wax was washed, as mentioned in the protocol for separation of the waxy components. Components that could undergo hydrolysis by the action of lipase were retained in the fraction, and they produced FFAs during the test. When the test was carried out with the particles prepared using wax and oil, larger amounts of FFAs were released over time. The total amount of FFAs released from 1 g of particles in 1 h was estimated to be 0.23 mmol. Assuming that the amount of FFAs released from the wax alone is 0.08 mmol/g, nearly 0.15 mmol/g of FFAs could be released from the oil by the action of lipase. This corresponds to around 4.8 % of the added oil.
Release of free fatty acids due to the action of lipase.
The amount of FFAs produced was estimated in particles loaded with iron (Figure 4). The amount of FFAs produced and the associated kinetics were almost identical to those of particles prepared without iron load. The iron was entrapped in the particle in the form of an aqueous solution stabilized as a W/O emulsion. The aqueous phase in the emulsion could leach out to the external aqueous phase due to hydrolysis of the oil phase by the action of lipase. In this process, as suggested by the release curve, the dispersed aqueous phase did not disturb the hydrolysis of the continuous phase (oil phase).
Release of Fe3+ due to the action of lipase As discussed in the previous section, FFAs were released from the prepared particles by the action of lipase. The prepared particles were designed with the function of triggered release of entrapped materials by the action of a digestive enzyme. The results of the iron encapsulation efficiency measurement confirmed that around 94–99 % of iron could be loaded in the prepared particles. It is suggested that this entrapped iron (stabilized in the water droplets) is released with the progress of FFAs release. The amount of released iron without the action of the lipase was preliminary tested by measuring the iron concentration in the supernatant of the particle suspension, which was stabilized at an ambient condition for days. As a result, iron elution from the particles could not be detected (data not shown).
As addressed in the previous section, FFAs release showed contributions by both the pre-mixed oil phase and the waxy phase. The release curve of the FFAs released from the W/O emulsion phase can be obtained by subtracting the values on the release curve of particles without oil from the corresponding values on the curve of particles with rice bran and oil (Figure 5A). The amount of released Fe3+ was measured and plotted (Figure 5B). As expected, a significant amount of Fe3+ was released from particles by the action of lipase over time. In the first 5 min after the addition of the lipase solution, Fe3+ could not be detected; however, it was detected after 10 min. The amount of Fe3+ released from 1 g of particles in 1 h reached around 0.16 mg, which corresponded to 1.2 % of the total loading amount. This value appears to be quite low as a carrier matrix. However, this does not indicate a deficiency in the present technique, as the release efficiency could be controlled by adjusting the loading amounts. In the present study, a certain amount of iron must be loaded to detect the release by the colorimetric method (i.e., thioglycolic acid method). A future challenge is to optimize the loading amounts of iron to maximize the release efficiency and improve iron absorption through the intestinal tract.
Released amount of FFA excluded the release from waxy phase (A); Release of ferric ion due to the action of lipase (B).
The amount of released Fe3+ was plotted as a function of the amount of FFAs released (Figure 6). The linear relationship shown in Figure 6 suggested that the release of Fe3+ occurred coincidently with the release of FFAs. As noted above, this particulate system was designed to release materials entrapped in triglyceride in response to enzymatic digestion. Furthermore, the aqueous phase in the emulsion could leach out to the external aqueous phase as a result of hydrolysis of the oil phase by the action of lipase. This study demonstrated the preparation of a multi-phase lipid carrier. The oil phase was embedded in the wax phase, and the water droplets entrapped in the oil phase could be released by enzymatic hydrolysis. This technique is useful for designing a particulate system that delivers water soluble active ingredients to the intestinal system inside a lipid matrix. Further studies are required to optimize release control via refinement of formulations and processing conditions.
Correlation between the amount of free fatty acids and ferric ion released.
This study demonstrated that a multi-phase lipid particle ranging in size from 10–160 µm could be prepared using a mixture of wax and triglyceride via a high shear homogenization method. A channel-like structure of the triglyceride phase was embedded in the wax matrix. When an intestinal enzyme (lipase) was added to the suspension of particles, FFAs produced as a result of hydrolysis of the triglyceride phase were released from the particles. Water droplets could be stabilized in the triglyceride phase (i.e., as a W/O emulsion). A compound dissolved in the aqueous phase could be released by the action of lipase due to hydrolysis of the triglyceride. In this study, the triggered release of ferric ions dissolved in the aqueous phase was successfully confirmed by the action of lipase. The release of ferric ions showed a linear relationship with the release of FFAs generated via the hydrolysis of triglyceride by lipase.
