- Email: [email protected]
Preparation of lipid nanoemulsions by premix membrane emulsification with disposable materials
International Journal of Pharmaceutics 511 (2016) 741–744
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Note
Preparation of lipid nanoemulsions by premix membrane emulsification with disposable materials Sandra Gehrmanna,b , Heike Bunjesa,b,* a b
Technische Universität Braunschweig, Institut für Pharmazeutische Technologie, Mendelssohnstraße 1, 38106 Braunschweig, Germany Zentrum für Pharmaverfahrenstechnik, Franz-Liszt-Straße 35a, 38106 Braunschweig, Germany
A R T I C L E I N F O
Article history: Received 2 June 2016 Received in revised form 26 July 2016 Accepted 27 July 2016 Available online 28 July 2016 Keywords: Nanoemulsion Small scale production Premix membrane emulsification Filter extrusion Phospholipids Polysorbate 80
A B S T R A C T
The possibility to prepare nanoemulsions as drug carrier systems on small scale was investigated with disposable materials. For this purpose premix membrane emulsification (premix ME) as a preparation method for nanoemulsions with narrow particle size distributions on small scale was used. The basic principle of premix ME is that the droplets of a coarse pre-emulsion get disrupted by the extrusion through a porous membrane. In order to implement the common preparation setup for premix ME with disposable materials, the suitability of different syringe filters (made from polyethersulfone, cellulose acetate, cellulose ester and nylon) and different pharmaceutically relevant emulsifiers (phospholipids, polysorbate 80 and sucrose laurate) for the preparation of nanoemulsions was investigated. Already the preparation of the premix could be realized by emulsification with the help of two disposable syringes. As shown for a phospholipid-stabilized emulsion, the polyethersulfone filter was the most appropriate one and was used for the study with different emulsifiers. With this syringe filter, the median particle size of all investigated emulsions was below 500 nm after 21 extrusion cycles through a 200 nm filter and a subsequent extrusion cycle through a 100 nm filter. Furthermore, the particle size distribution of the polysorbate 80- and sucrose laurate-stabilized emulsions prepared this way was very narrow (span value of 0.7). ã 2016 Elsevier B.V. All rights reserved.
At the beginning of formulation studies with new drug candidates, only small amounts of drug with unknown toxicology are available (Balbach and Korn, 2004). Therefore, formulation preparation methods requiring only small amounts of substances are highly desirable. Additionally, most of the newly developed drug substances are poorly water soluble (Buckley et al., 2013). One formulation method for poorly water soluble drugs is loading them into lipid nanoparticles (Bunjes, 2010). For the preparation of lipid nanoemulsions, a promising small scale preparation method is premix membrane emulsification (premix ME) (Joseph and Bunjes, 2012). In this process, a coarse pre-emulsion is extruded through the pores of a membrane yielding smaller emulsion droplets (Suzuki et al., 1996; Vladisavljevi c et al., 2004; Surh et al., 2008). With premix ME, emulsions having a narrow particle size distribution can be prepared (Nazir et al., 2010), which is, in
* Corresponding author at: Technische Universität Braunschweig, Institut für Pharmazeutische Technologie, Mendelssohnstraße 1, 38106 Braunschweig, Germany. E-mail address: [email protected] (H. Bunjes). http://dx.doi.org/10.1016/j.ijpharm.2016.07.067 0378-5173/ã 2016 Elsevier B.V. All rights reserved.
addition to small particle sizes, highly desirable for pharmaceutical formulations. To combine the advantages of premix ME with a low risk of contamination with highly potent or toxic drug substances and cross-contamination of subsequently prepared formulations, the use of disposable materials was tested in the present study. The use of disposable materials during the first formulation studies would also enable a simple and cost effective equipment setup and would avoid a cleaning step. For this purpose, the whole production process including the preparation of the premix was performed on small scale with single-use materials. The suitability of different syringe filters for the preparation of nanoemulsions via premix ME was investigated and the influence of cycle numbers on the resulting particle sizes of emulsions stabilized with different pharmaceutically relevant emulsifiers was analyzed. The emulsion formulations consisted of 10% Miglyol 8121 (MCT, Caesar & Loretz GmbH, Hilden, Germany) stabilized with 2.5% phospholipids or 5% other surfactants in double distilled water. The soybean phospholipids Lipoid S100 and Lipoid S75 (both Lipoid GmbH, Ludwigshafen, Germany) as well as Tween 80 (Fluka-Sigma-Aldrich, Steinheim, Germany) and sucrose laurate (SL) (Royoto Sugar Ester L-1695, Mitsubishi-Kagaku Food
742
S. Gehrmann, H. Bunjes / International Journal of Pharmaceutics 511 (2016) 741–744
Three measurements of 90 s each were carried out and used to calculate the median (d50 value), d10, d90 and the span (d90–d10/ d50), assuming a refraction index of 1.46 for the particles and 1.33 for the aqueous phase. As shown for the phospholipid-stabilized emulsions, the preparation of the pre-premix with two syringes led to similar particle size distributions as the preparation with an Ultra-Turrax1 (Fig. 2); it was therefore suitable for the preparation of the prepremix in the following. Compared to the process with an UltraTurrax1, the process with the syringes took less time and was freely scalable in volume depending on the size of the syringes without a required minimum volume. The median particle sizes of the pre-premixed emulsions prepared with the two-syringe method were between 10 and 24 mm (Fig. 2 for Lipoid S75 and Lipoid S100, other data not shown). Emulsions with median particle sizes that big could not be extruded directly through a 220 or 200 nm syringe filter. Therefore, a pre-extrusion step was introduced into the process. After the pre-extrusion step through the glass fiber filter, the median particle sizes of all emulsions were reduced to the lower micrometer range (1–6 mm) and the emulsions could subsequently be extruded through the syringe filters with smaller pore sizes. The extrusion of the Lipoid S100-stabilized emulsion through different syringe filters led to different particle sizes depending on the cycle number and the membrane material (Fig. 3). The main decrease in particle size was realized after the first cycle in all cases. With increasing cycle numbers, the median particle size generally decreased further. Only with the CE filter, the median particle size of the emulsion increased after the first cycle. The nylon filter did not lead to particle sizes smaller than 1 mm even after 21 cycles. With the CA filter, the median particle size was below 1 mm already after the first cycle but there was no further decrease in particle size and span. With the PES filter, the median particle size of the emulsion was also below 1 mm after the first cycle and decreased further. After 5 cycles the median particle size was already below 500 nm and after 21 cycles around 400 nm. Not every hydrophilic filter membrane with pore sizes around 200 nm was thus suitable for the preparation of emulsions with particle sizes in the nanometer range. This phenomenon has also been observed previously when investigating the influence of the membrane material during premix ME (Gehrmann and Bunjes, 2015). As the PES filter was the most appropriate one, it was used for the investigation of emulsion preparation with different emulsifiers (Fig. 4). The extrusion process differed in efficiency depending on the emulsifier as described before (Vladisavljevi c et al., 2006).
Particle size [µm]
Corporation, Tokyo, Japan) were used as surfactants. To prepare the aqueous phase, the soybean phospholipids were dispersed in the double distilled water with a magnetic stirrer over night, Tween 80 and SL were simply dissolved in water. The interfacial tension of the Tween 80 or the SL solution against MCT was measured with a tensiometer (K100, Krüss GmbH, Germany) according to the Wilhelmy method at 25 C in triplicate using densities of the Tween 80 and SL solution of 1.0026 and 1.0077 g/ml, respectively (determined with DMA 46, Anton Paar K.G., Austria, at 25 C). As a first step, a pre-premix was prepared using two 20 ml disposable syringes (B. Braun, Melsungen, Germany) connected with a Luer Lock connection (female-female) (CT67.1, Carl Roth, Karlsruhe, Germany). The MCT was filled into one syringe and the aqueous phase into the other syringe. By pushing the liquids 10 times back and forth between the two syringes, an emulsion was formed (Fig. 1a). This method was compared to the common preparation procedure with an Ultra-Turrax1 IKA 25 T (IKA1Werke GmbH & Co. KG, Staufen, Germany) at 13,000 rpm for 1 min with the two phospholipid-stabilized formulations. Prior to the main extrusion process, the pre-premixed emulsions were extruded once through a 1 mm glass fiber filter (, 25 mm, Acrodisc1, Pall1, Ann Arbor, USA) to lead to emulsions with smaller particle sizes. The resulting emulsions are called “premix” in the following. The pre-extrusion step was necessary in order to pass the emulsions through the following filters. Otherwise, the required pressure was too high. The actual premix ME was performed with two 1 ml syringes (B. Braun, Melsungen, Germany) with a syringe filter and a Luer Lock connection between them (Fig. 1b). By pushing the emulsion back and forth through the filter, the droplets were disrupted. To compare different disposable membrane filters, the emulsion stabilized with Lipoid S100 was extruded 21 times through a 220 nm filter out of polyethersulfone (PES) (, 33 mm, Rotilabo1) or a 200 nm filter out of nylon (PA, , 25 mm, both from Carl Roth, Karlsruhe, Germany), cellulose acetate or cellulose ester (both , 28 mm, Minisart1, Sartorius, Göttingen, Germany). The cycle number was originally derived from liposome preparation (Hope et al., 1985; Hunter and Frisken, 1998) and later successfully used in the preparation of colloidal emulsions (Joseph and Bunjes, 2012). To investigate the influence of different emulsifiers, the differently stabilized emulsions were processed 21 times through the 220 nm PES filter followed by one cycle through a 100 nm PES filter (, 28 mm, Minisart1, Sartorius, Göttingen, Germany). The last step was only possible after a 1:5 dilution with aqueous phase in most cases (only with the Lipoid S100-stabilized emulsion an extrusion without dilution was possible but for a better comparability, this formulation was also diluted). In the end, the lipid content of the formulations was thus around 2%. For particle size determination, the samples were measured after adequate dilution with demineralized water in a laser light diffractometer with polarization intensity differential scattering (PIDS) technology (LS 13320, Beckman-Coulter, Krefeld, Germany).
60 55 50 45 40 35 30 25 20 15 10 5 0
d90 d50 d10
ges ax® syrin -Turr ith 2 Ultra w h it w
x® ges Turra h 2 syrin it Ultra w h it w
Lipoid S100 Fig. 1. Principle of pre-premix (a) and nanoemulsion (b) preparation.
Fig. 2. Pre-premix preparation with Ultra-Turrax
Lipoid S75 1
vs. 2-syringe method.
S. Gehrmann, H. Bunjes / International Journal of Pharmaceutics 511 (2016) 741–744
PES CA Nylon CE
4.0
Median particle size [µm]
743
3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 2 Premix
4
6
8
10
12
14
16
18
20
22
Cycle number Fig. 3. Emulsion preparation with different filter materials, 10% MCT, 2.5% Lipoid S100.
Median particle size [µm]
6 2.5% Lipoid S100 2.5% Lipoid S75 5% Tween 80 5% SL * = 21x200 nm + 1x100 nm
5 4 3 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0 2 Premix
4
6
8
10
12
14
16
18
20
22*
Cycle number Fig. 4. Emulsion preparation with different emulsifiers, 10% MCT, 220 nm PES filter, mean SD of 2 independent experiments.
With Lipoid S100 and SL as emulsifier, the median particle size was in the nanometer range already after the first cycle, whereas the emulsions stabilized with Lipoid S75 and Tween 80 were still in the micrometer range at that point. With a molecular weight of 525 g/ mol SL was the smallest molecule within this investigation and may thus most rapidly adsorb to the newly formed droplet surface. Tween 80 had the highest molecular weight of 1310 g/mol and led to a higher interfacial tension than SL (SL: 2 mN/m, Tween 80: 4.4 mN/m), which may complicate the droplet disruption. The molecular weight of the phospholipids with around 760 g/mol is in between that of the two other emulsifiers, but the performance of phospholipids in the water phase is not comparable to SL and Tween 80. They tend to form vesicles or bilayers rather than existing as individual molecules or micelles (Klang and Valenta, 2011). The particle size of the Tween 80- stabilized emulsion decreased with increasing cycle number down to the value of the SL- and Lipoid S100- stabilized emulsions (between 400 and 440 nm after 21 cycles). The particle size of the Lipoid S75stabilized emulsion decreased more slowly and increased after the 21st cycle. While the median particle sizes of the Lipoid S100-, SLand Tween 80-stabilized emulsions were comparable after 21 cycles through the 200 nm PES filter, the particle size distributions of the Tween 80- and the SL-stabilized emulsions were more
Fig. 5. Particle size distribution of differently stabilized emulsions, 10% MCT, mean SD of 3 independent experiments.
narrow (span 0.5) than with Lipoid S100 (span 1) (Fig. 5). This may be due to the fact that Tween 80 and SL may diffuse more rapidly to the newly formed interface between water and oil phase because they do not form vesicles as phospholipids do (Kabalnov et al., 1995). Thus the stabilization of the newly formed droplets might be more homogeneous. The following extrusion step through the 100 nm PES filter led to a further decrease of the median particle size (Figs. 4 and 5). The particle size distributions of the resulting emulsions were slightly broader than after the 21 extrusion cycles through the 200 nm PES filter but still narrow, except for the Lipoid S75-stabilized emulsion. After the whole premix membrane emulsification process with both PES syringe filters (200 nm and 100 nm), the median particle sizes of all emulsions were below 500 nm. In the case of the Lipoid S100-stabilized emulsion, the d50-value was even around 200 nm. In conclusion, it was possible to prepare emulsions stabilized with different pharmaceutically relevant emulsifiers with median particle sizes below 500 nm only with the help of cost-effective single-use laboratory equipment. This way of premix ME is an easy and very fast method for the production of a small amount of emulsion with a narrow particle size distribution. References Balbach, S., Korn, C., 2004. Pharmaceutical evaluation of early development candidates the 100 mg-approach. Int. J. Pharm. 275, 1–12. Buckley, S.T., Frank, K.J., Fricker, G., Brandl, M., 2013. Biopharmaceutical classification of poorly soluble drugs with respect to enabling formulations. Eur. J. Pharm. Sci. 50, 8–16. Bunjes, H., 2010. Lipid nanoparticles for the delivery of poorly water-soluble drugs. J. Pharm. Pharmacol. 62, 1637–1645. Gehrmann, S., Bunjes, H., 2015. Interaction of emulsifier and membrane material during the preparation of nanoemulsions by premix membrane emulsification, Poster. 1st European Conference on Pharmaceutics – Drug Delivery, Reims, France. Hope, M.J., Bally, M.B., Webb, G., Cullis, P.R., 1985. Production of large unilamellar vesicles by a rapid extrusion procedure. Characterization of size distribution, trapped volume and ability to maintain a membrane potential. Biochim. Biophys. Acta 812, 55–65. Hunter, D.G., Frisken, B.J., 1998. Effect of extrusion pressure and lipid properties on the size and polydispersity of lipid vesicles. Biophys. J. 74, 2996–3002. Joseph, S., Bunjes, H., 2012. Preparation of nanoemulsions and solid lipid nanoparticles by premix membrane emulsification. J. Pharm. Sci. 101, 2479– 2489. Kabalnov, A., Weers, J., Arlauskas, R., Tarara, T., 1995. Phospholipids as emulsion stabilizers 1. Interfacial tensions. Langmuir 11, 2966–2974. Klang, V., Valenta, C., 2011. Lecithin-based nanoemulsions. J. Drug Deliv. Sci. Technol. 21, 55–76.
744
S. Gehrmann, H. Bunjes / International Journal of Pharmaceutics 511 (2016) 741–744
Nazir, A., Schroën, K., Boom, R., 2010. Premix emulsification: a review. J. Membr. Sci. 362, 1–11. Surh, J., Jeong, Y.G., Vladisavljevi c, G.T., 2008. On the preparation of lecithinstabilized oil-in-water emulsions by multi-stage premix membrane emulsification. J. Food Eng. 89, 164–170. Suzuki, K., Shuto, I., Hagura, Y., 1996. Characteristics of the membrane emulsification method combined with preliminary emulsification for preparing corn oil-in-water emulsions. Food Sci. Technol. Int. 2, 43–47.
Vladisavljevi c, G.T., Shimizu, M., Nakashima, T., 2004. Preparation of monodisperse multiple emulsions at high production rates by multi-stage premix membrane emulsification. J. Membr. Sci. 244, 97–106. Vladisavljevi c, G.T., Surh, J., McClements, J.D., 2006. Effect of emulsifier type on droplet disruption in repeated Shirasu Porous Glass membrane homogenization. Langmuir 22, 4526–4533.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Note
Preparation of lipid nanoemulsions by premix membrane emulsification with disposable materials Sandra Gehrmanna,b , Heike Bunjesa,b,* a b
Technische Universität Braunschweig, Institut für Pharmazeutische Technologie, Mendelssohnstraße 1, 38106 Braunschweig, Germany Zentrum für Pharmaverfahrenstechnik, Franz-Liszt-Straße 35a, 38106 Braunschweig, Germany
A R T I C L E I N F O
Article history: Received 2 June 2016 Received in revised form 26 July 2016 Accepted 27 July 2016 Available online 28 July 2016 Keywords: Nanoemulsion Small scale production Premix membrane emulsification Filter extrusion Phospholipids Polysorbate 80
A B S T R A C T
The possibility to prepare nanoemulsions as drug carrier systems on small scale was investigated with disposable materials. For this purpose premix membrane emulsification (premix ME) as a preparation method for nanoemulsions with narrow particle size distributions on small scale was used. The basic principle of premix ME is that the droplets of a coarse pre-emulsion get disrupted by the extrusion through a porous membrane. In order to implement the common preparation setup for premix ME with disposable materials, the suitability of different syringe filters (made from polyethersulfone, cellulose acetate, cellulose ester and nylon) and different pharmaceutically relevant emulsifiers (phospholipids, polysorbate 80 and sucrose laurate) for the preparation of nanoemulsions was investigated. Already the preparation of the premix could be realized by emulsification with the help of two disposable syringes. As shown for a phospholipid-stabilized emulsion, the polyethersulfone filter was the most appropriate one and was used for the study with different emulsifiers. With this syringe filter, the median particle size of all investigated emulsions was below 500 nm after 21 extrusion cycles through a 200 nm filter and a subsequent extrusion cycle through a 100 nm filter. Furthermore, the particle size distribution of the polysorbate 80- and sucrose laurate-stabilized emulsions prepared this way was very narrow (span value of 0.7). ã 2016 Elsevier B.V. All rights reserved.
At the beginning of formulation studies with new drug candidates, only small amounts of drug with unknown toxicology are available (Balbach and Korn, 2004). Therefore, formulation preparation methods requiring only small amounts of substances are highly desirable. Additionally, most of the newly developed drug substances are poorly water soluble (Buckley et al., 2013). One formulation method for poorly water soluble drugs is loading them into lipid nanoparticles (Bunjes, 2010). For the preparation of lipid nanoemulsions, a promising small scale preparation method is premix membrane emulsification (premix ME) (Joseph and Bunjes, 2012). In this process, a coarse pre-emulsion is extruded through the pores of a membrane yielding smaller emulsion droplets (Suzuki et al., 1996; Vladisavljevi c et al., 2004; Surh et al., 2008). With premix ME, emulsions having a narrow particle size distribution can be prepared (Nazir et al., 2010), which is, in
* Corresponding author at: Technische Universität Braunschweig, Institut für Pharmazeutische Technologie, Mendelssohnstraße 1, 38106 Braunschweig, Germany. E-mail address: [email protected] (H. Bunjes). http://dx.doi.org/10.1016/j.ijpharm.2016.07.067 0378-5173/ã 2016 Elsevier B.V. All rights reserved.
addition to small particle sizes, highly desirable for pharmaceutical formulations. To combine the advantages of premix ME with a low risk of contamination with highly potent or toxic drug substances and cross-contamination of subsequently prepared formulations, the use of disposable materials was tested in the present study. The use of disposable materials during the first formulation studies would also enable a simple and cost effective equipment setup and would avoid a cleaning step. For this purpose, the whole production process including the preparation of the premix was performed on small scale with single-use materials. The suitability of different syringe filters for the preparation of nanoemulsions via premix ME was investigated and the influence of cycle numbers on the resulting particle sizes of emulsions stabilized with different pharmaceutically relevant emulsifiers was analyzed. The emulsion formulations consisted of 10% Miglyol 8121 (MCT, Caesar & Loretz GmbH, Hilden, Germany) stabilized with 2.5% phospholipids or 5% other surfactants in double distilled water. The soybean phospholipids Lipoid S100 and Lipoid S75 (both Lipoid GmbH, Ludwigshafen, Germany) as well as Tween 80 (Fluka-Sigma-Aldrich, Steinheim, Germany) and sucrose laurate (SL) (Royoto Sugar Ester L-1695, Mitsubishi-Kagaku Food
742
S. Gehrmann, H. Bunjes / International Journal of Pharmaceutics 511 (2016) 741–744
Three measurements of 90 s each were carried out and used to calculate the median (d50 value), d10, d90 and the span (d90–d10/ d50), assuming a refraction index of 1.46 for the particles and 1.33 for the aqueous phase. As shown for the phospholipid-stabilized emulsions, the preparation of the pre-premix with two syringes led to similar particle size distributions as the preparation with an Ultra-Turrax1 (Fig. 2); it was therefore suitable for the preparation of the prepremix in the following. Compared to the process with an UltraTurrax1, the process with the syringes took less time and was freely scalable in volume depending on the size of the syringes without a required minimum volume. The median particle sizes of the pre-premixed emulsions prepared with the two-syringe method were between 10 and 24 mm (Fig. 2 for Lipoid S75 and Lipoid S100, other data not shown). Emulsions with median particle sizes that big could not be extruded directly through a 220 or 200 nm syringe filter. Therefore, a pre-extrusion step was introduced into the process. After the pre-extrusion step through the glass fiber filter, the median particle sizes of all emulsions were reduced to the lower micrometer range (1–6 mm) and the emulsions could subsequently be extruded through the syringe filters with smaller pore sizes. The extrusion of the Lipoid S100-stabilized emulsion through different syringe filters led to different particle sizes depending on the cycle number and the membrane material (Fig. 3). The main decrease in particle size was realized after the first cycle in all cases. With increasing cycle numbers, the median particle size generally decreased further. Only with the CE filter, the median particle size of the emulsion increased after the first cycle. The nylon filter did not lead to particle sizes smaller than 1 mm even after 21 cycles. With the CA filter, the median particle size was below 1 mm already after the first cycle but there was no further decrease in particle size and span. With the PES filter, the median particle size of the emulsion was also below 1 mm after the first cycle and decreased further. After 5 cycles the median particle size was already below 500 nm and after 21 cycles around 400 nm. Not every hydrophilic filter membrane with pore sizes around 200 nm was thus suitable for the preparation of emulsions with particle sizes in the nanometer range. This phenomenon has also been observed previously when investigating the influence of the membrane material during premix ME (Gehrmann and Bunjes, 2015). As the PES filter was the most appropriate one, it was used for the investigation of emulsion preparation with different emulsifiers (Fig. 4). The extrusion process differed in efficiency depending on the emulsifier as described before (Vladisavljevi c et al., 2006).
Particle size [µm]
Corporation, Tokyo, Japan) were used as surfactants. To prepare the aqueous phase, the soybean phospholipids were dispersed in the double distilled water with a magnetic stirrer over night, Tween 80 and SL were simply dissolved in water. The interfacial tension of the Tween 80 or the SL solution against MCT was measured with a tensiometer (K100, Krüss GmbH, Germany) according to the Wilhelmy method at 25 C in triplicate using densities of the Tween 80 and SL solution of 1.0026 and 1.0077 g/ml, respectively (determined with DMA 46, Anton Paar K.G., Austria, at 25 C). As a first step, a pre-premix was prepared using two 20 ml disposable syringes (B. Braun, Melsungen, Germany) connected with a Luer Lock connection (female-female) (CT67.1, Carl Roth, Karlsruhe, Germany). The MCT was filled into one syringe and the aqueous phase into the other syringe. By pushing the liquids 10 times back and forth between the two syringes, an emulsion was formed (Fig. 1a). This method was compared to the common preparation procedure with an Ultra-Turrax1 IKA 25 T (IKA1Werke GmbH & Co. KG, Staufen, Germany) at 13,000 rpm for 1 min with the two phospholipid-stabilized formulations. Prior to the main extrusion process, the pre-premixed emulsions were extruded once through a 1 mm glass fiber filter (, 25 mm, Acrodisc1, Pall1, Ann Arbor, USA) to lead to emulsions with smaller particle sizes. The resulting emulsions are called “premix” in the following. The pre-extrusion step was necessary in order to pass the emulsions through the following filters. Otherwise, the required pressure was too high. The actual premix ME was performed with two 1 ml syringes (B. Braun, Melsungen, Germany) with a syringe filter and a Luer Lock connection between them (Fig. 1b). By pushing the emulsion back and forth through the filter, the droplets were disrupted. To compare different disposable membrane filters, the emulsion stabilized with Lipoid S100 was extruded 21 times through a 220 nm filter out of polyethersulfone (PES) (, 33 mm, Rotilabo1) or a 200 nm filter out of nylon (PA, , 25 mm, both from Carl Roth, Karlsruhe, Germany), cellulose acetate or cellulose ester (both , 28 mm, Minisart1, Sartorius, Göttingen, Germany). The cycle number was originally derived from liposome preparation (Hope et al., 1985; Hunter and Frisken, 1998) and later successfully used in the preparation of colloidal emulsions (Joseph and Bunjes, 2012). To investigate the influence of different emulsifiers, the differently stabilized emulsions were processed 21 times through the 220 nm PES filter followed by one cycle through a 100 nm PES filter (, 28 mm, Minisart1, Sartorius, Göttingen, Germany). The last step was only possible after a 1:5 dilution with aqueous phase in most cases (only with the Lipoid S100-stabilized emulsion an extrusion without dilution was possible but for a better comparability, this formulation was also diluted). In the end, the lipid content of the formulations was thus around 2%. For particle size determination, the samples were measured after adequate dilution with demineralized water in a laser light diffractometer with polarization intensity differential scattering (PIDS) technology (LS 13320, Beckman-Coulter, Krefeld, Germany).
60 55 50 45 40 35 30 25 20 15 10 5 0
d90 d50 d10
ges ax® syrin -Turr ith 2 Ultra w h it w
x® ges Turra h 2 syrin it Ultra w h it w
Lipoid S100 Fig. 1. Principle of pre-premix (a) and nanoemulsion (b) preparation.
Fig. 2. Pre-premix preparation with Ultra-Turrax
Lipoid S75 1
vs. 2-syringe method.
S. Gehrmann, H. Bunjes / International Journal of Pharmaceutics 511 (2016) 741–744
PES CA Nylon CE
4.0
Median particle size [µm]
743
3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 2 Premix
4
6
8
10
12
14
16
18
20
22
Cycle number Fig. 3. Emulsion preparation with different filter materials, 10% MCT, 2.5% Lipoid S100.
Median particle size [µm]
6 2.5% Lipoid S100 2.5% Lipoid S75 5% Tween 80 5% SL * = 21x200 nm + 1x100 nm
5 4 3 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0 2 Premix
4
6
8
10
12
14
16
18
20
22*
Cycle number Fig. 4. Emulsion preparation with different emulsifiers, 10% MCT, 220 nm PES filter, mean SD of 2 independent experiments.
With Lipoid S100 and SL as emulsifier, the median particle size was in the nanometer range already after the first cycle, whereas the emulsions stabilized with Lipoid S75 and Tween 80 were still in the micrometer range at that point. With a molecular weight of 525 g/ mol SL was the smallest molecule within this investigation and may thus most rapidly adsorb to the newly formed droplet surface. Tween 80 had the highest molecular weight of 1310 g/mol and led to a higher interfacial tension than SL (SL: 2 mN/m, Tween 80: 4.4 mN/m), which may complicate the droplet disruption. The molecular weight of the phospholipids with around 760 g/mol is in between that of the two other emulsifiers, but the performance of phospholipids in the water phase is not comparable to SL and Tween 80. They tend to form vesicles or bilayers rather than existing as individual molecules or micelles (Klang and Valenta, 2011). The particle size of the Tween 80- stabilized emulsion decreased with increasing cycle number down to the value of the SL- and Lipoid S100- stabilized emulsions (between 400 and 440 nm after 21 cycles). The particle size of the Lipoid S75stabilized emulsion decreased more slowly and increased after the 21st cycle. While the median particle sizes of the Lipoid S100-, SLand Tween 80-stabilized emulsions were comparable after 21 cycles through the 200 nm PES filter, the particle size distributions of the Tween 80- and the SL-stabilized emulsions were more
Fig. 5. Particle size distribution of differently stabilized emulsions, 10% MCT, mean SD of 3 independent experiments.
narrow (span 0.5) than with Lipoid S100 (span 1) (Fig. 5). This may be due to the fact that Tween 80 and SL may diffuse more rapidly to the newly formed interface between water and oil phase because they do not form vesicles as phospholipids do (Kabalnov et al., 1995). Thus the stabilization of the newly formed droplets might be more homogeneous. The following extrusion step through the 100 nm PES filter led to a further decrease of the median particle size (Figs. 4 and 5). The particle size distributions of the resulting emulsions were slightly broader than after the 21 extrusion cycles through the 200 nm PES filter but still narrow, except for the Lipoid S75-stabilized emulsion. After the whole premix membrane emulsification process with both PES syringe filters (200 nm and 100 nm), the median particle sizes of all emulsions were below 500 nm. In the case of the Lipoid S100-stabilized emulsion, the d50-value was even around 200 nm. In conclusion, it was possible to prepare emulsions stabilized with different pharmaceutically relevant emulsifiers with median particle sizes below 500 nm only with the help of cost-effective single-use laboratory equipment. This way of premix ME is an easy and very fast method for the production of a small amount of emulsion with a narrow particle size distribution. References Balbach, S., Korn, C., 2004. Pharmaceutical evaluation of early development candidates the 100 mg-approach. Int. J. Pharm. 275, 1–12. Buckley, S.T., Frank, K.J., Fricker, G., Brandl, M., 2013. Biopharmaceutical classification of poorly soluble drugs with respect to enabling formulations. Eur. J. Pharm. Sci. 50, 8–16. Bunjes, H., 2010. Lipid nanoparticles for the delivery of poorly water-soluble drugs. J. Pharm. Pharmacol. 62, 1637–1645. Gehrmann, S., Bunjes, H., 2015. Interaction of emulsifier and membrane material during the preparation of nanoemulsions by premix membrane emulsification, Poster. 1st European Conference on Pharmaceutics – Drug Delivery, Reims, France. Hope, M.J., Bally, M.B., Webb, G., Cullis, P.R., 1985. Production of large unilamellar vesicles by a rapid extrusion procedure. Characterization of size distribution, trapped volume and ability to maintain a membrane potential. Biochim. Biophys. Acta 812, 55–65. Hunter, D.G., Frisken, B.J., 1998. Effect of extrusion pressure and lipid properties on the size and polydispersity of lipid vesicles. Biophys. J. 74, 2996–3002. Joseph, S., Bunjes, H., 2012. Preparation of nanoemulsions and solid lipid nanoparticles by premix membrane emulsification. J. Pharm. Sci. 101, 2479– 2489. Kabalnov, A., Weers, J., Arlauskas, R., Tarara, T., 1995. Phospholipids as emulsion stabilizers 1. Interfacial tensions. Langmuir 11, 2966–2974. Klang, V., Valenta, C., 2011. Lecithin-based nanoemulsions. J. Drug Deliv. Sci. Technol. 21, 55–76.
744
S. Gehrmann, H. Bunjes / International Journal of Pharmaceutics 511 (2016) 741–744
Nazir, A., Schroën, K., Boom, R., 2010. Premix emulsification: a review. J. Membr. Sci. 362, 1–11. Surh, J., Jeong, Y.G., Vladisavljevi c, G.T., 2008. On the preparation of lecithinstabilized oil-in-water emulsions by multi-stage premix membrane emulsification. J. Food Eng. 89, 164–170. Suzuki, K., Shuto, I., Hagura, Y., 1996. Characteristics of the membrane emulsification method combined with preliminary emulsification for preparing corn oil-in-water emulsions. Food Sci. Technol. Int. 2, 43–47.
Vladisavljevi c, G.T., Shimizu, M., Nakashima, T., 2004. Preparation of monodisperse multiple emulsions at high production rates by multi-stage premix membrane emulsification. J. Membr. Sci. 244, 97–106. Vladisavljevi c, G.T., Surh, J., McClements, J.D., 2006. Effect of emulsifier type on droplet disruption in repeated Shirasu Porous Glass membrane homogenization. Langmuir 22, 4526–4533.