Research paper
Carrier characteristics influence the kinetics of passive drug loading into lipid nanoemulsions

https://doi.org/10.1016/j.ejpb.2017.08.004Get rights and content

Abstract

Passive loading as a novel screening approach is a material-saving tool for the efficient selection of a suitable colloidal lipid carrier system for poorly water soluble drug candidates. This method comprises incubation of preformed carrier systems with drug powder and subsequent determination of the resulting drug load of the carrier particles after removal of excess drug. For reliable routine use and to obtain meaningful loading results, information on the kinetics of the process is required. Passive loading proceeds via a dissolution-diffusion-based mechanism, where drug surface area and drug water solubility are key parameters for fast passive loading. While the influence of the drug characteristics is mostly understood, the influence of the carrier characteristics remains unknown. The aim of this study was to examine how the lipid nanocarriers’ characteristics, i.e. the type of lipid, the lipid content and the particle size, influence the kinetics of passive loading. Fenofibrate was used as model drug and the loading progress was analyzed by UV spectroscopy. The saturation solubility in the nanocarrier particles, i.e. the lipid type, did not influence the passive loading rate constant. Low lipid content in the nanocarrier and a small nanocarrier particle size both increased passive loading speed. Both variations increase the diffusivity of the nanocarrier particles, which is the primary cause for fast loading at these conditions: The quicker the carrier particles diffuse, the higher is the speed of passive loading. The influence of the diffusivity of the lipid nanocarriers and the effect of drug dissolution rate were included in an overall mechanistic model developed for similar processes (A. Balakrishnan, B.D. Rege, G.L. Amidon, J.E. Polli, Surfactant-mediated dissolution: contributions of solubility enhancement and relatively low micelle diffusivity, J. Pharm. Sci. 93 (2004) 2064–2075). The resulting mechanistic model gave a good estimate of the speed of passive loading in nanoemulsions. Whilst the drug’s characteristics – apart from drug surface area – are basically fixed, the lipid nanocarriers can be customized to improve passive loading speed, e.g. by using small nanocarrier particles. The knowledge of the loading mechanism now allows the use of passive loading for the straightforward, material-saving selection of suitable lipid drug nanocarriers.

Introduction

Lipid nanoparticle dispersions are colloidal drug delivery systems under investigation for the formulation of poorly water soluble drugs. A broad range of such lipid particle-based systems exists: there are commercially available ones like drug-loaded nanoemulsions or liposomes [1], [2] as well as systems which are still in the research stage like solid lipid nanoparticles [3], supercooled smectic nanoparticles [4] or cubic phase nanoparticles [5]. Modifications of starting materials and preparation techniques have led to a great variety of carrier systems to choose from, but there is only little information on how to systematically select an appropriate carrier for a new drug. In the conventional preparation of drug-loaded lipid nanocarriers, the drug is dissolved in the lipid first and then, this drug-lipid mixture is processed to nanoparticles [6]. This method requires comparatively large batch sizes, may lead to contamination of the production equipment and might stress the drug thermally. However, the required large batch size and the correspondingly high consumption of drug substance is probably the most serious problem in early formulation where drug substance is scarce. To overcome this problem and to allow for the straightforward, material-saving selection of a suitable lipid nanocarrier for a given drug, a novel screening approach called passive loading has been developed [7], [8]: In this approach, the nanocarriers are prepared drug-free in advance and the respective drug crystals are then incubated with the preformed carrier. After a sufficiently long incubation time, undissolved drug is filtered off and the drug content solubilized in the lipid carrier particles is determined. This way, many potential carrier systems can be investigated in parallel with minimal input of carrier dispersion or drug, significantly reduced work input and minimized exposure to potentially toxic drug substances. The passive loading approach was tested in two formulation studies for drug candidates, where less than 400 mg of each drug substance were spent to find formulations suitable for efficacy studies in mice [8], [9]. Also, passive loading of various drug substances onto trimyristin nanoparticles allowed to draw conclusions on the impact of particle size and physical state on the drug loading capacity of the nanoparticles [10]. However, passive loading will only yield reliable results if the maximum drug load of the respective carrier is reached during the loading process. So the drug should not be filtered off before the passive loading process is complete. To ensure complete loading, the loading kinetics have to be known at least roughly. To approximate the required loading time and to gain knowledge on the predominant loading mechanism, the influence of drug characteristics on the course of passive loading was investigated in a previous study [11]: It was found that an increase in drug surface area led to a directly proportional increase in passive loading speed. Also, an increase in the molecular water solubility of the drug resulted in faster passive loading. Since drug surface area and drug water solubility are crucial factors in dissolution as described by the Noyes-Whitney equation, these findings pointed to a dissolution-diffusion-based passive loading mechanism rather than drug loading via a contact-based process. Accordingly, passive loading also occurred when drug and carrier particles were separated by a dialysis membrane. Increasing the drug solubility by solubilization into polysorbate 80 micelles did not increase passive loading speed. Also, there was no linear correlation between the increase in water solubility and passive loading speed, which is inconsistent with the Noyes-Whitney equation. Presumably, the presence of micelles or the lipid nanoparticles, which both act as acceptor particles, sets hurdles to the simple application of the Noyes-Whitney equation to dissolution in lipid nanoparticle dispersions. The data from the first study was not sufficient to calculate how the saturation solubility in the water phase and the solubility in the colloidal system interact and influence the course of passive loading. Moreover, this previous study could not explain why passive loading speed varied considerably between carrier systems without any apparent trend as had been observed in preliminary investigations. Additionally, it was surprising that concentrated lipid carriers seemed to be loaded more slowly than diluted ones (see Fig. S1 in the supporting information). In short, two essential questions on the course of passive loading remained after the first study. One applies to the influence of the nanoparticle characteristics on the course of passive loading. Do factors like the loading capacity of the carrier or the particle size influence the course of passive loading, and if so, to what extent? The second question addresses the wish for an overall mechanistic model that is able to describe the complete passive loading process and also allows predicting the speed of passive loading in advance. How do the parameters influencing passive loading have to be combined to describe the complete passive loading process?

The aim of this study was therefore to examine how the lipid nanocarriers’ characteristics, i.e. the type of lipid, the lipid content and the particle size, influence the speed of passive loading and to develop an overall mechanistic model for the process. The lipid nanoemulsions for passive loading were varied systematically either in lipid type, lipid content or particle size and the effect of the variation on the resulting loading speed was studied. Loading speed was analyzed by quantification of dissolved drug via UV spectroscopy. Like in the previous study on the passive loading mechanism [11], poloxamer 188 was used as emulsifier since it does not form micelles at the prevailing conditions [12]. Fenofibrate was employed as a model drug as it is poorly water soluble, chemically stable, shows high solubility in lipid droplets [10], has a comparatively high molecular extinction coefficient [13] and because it had shown suitable loading speed in a previous study [11].

Section snippets

Chemicals and reagents

The triglyceride trimyristin (Dynasan® 114) and acetylated monoglycerides (Dynacet® 285; both Sasol/Cremer Oleo, Witten, Germany) were a kind gift from the manufacturer as was poloxamer 188 (Kolliphor® P188, BASF, Ludwigshafen, Germany). Medium-chain triglycerides (MCT® 812, Ph.Eur.) were from Caesar & Loretz GmbH (Hilden, Germany). Sodium azide and glycerol were obtained from Roth (Karlsruhe, Germany) as were all syringe filters. Fenofibrate was from Sigma-Aldrich (Steinheim, Germany).

Preparation and characterization of lipid nanoemulsions

Eleven nanoemulsions differing in nanoparticle size or lipid type were produced (Table 1). All nanoemulsions had a narrow particle size distribution width as indicated by span values (measured by LD-PIDS) below 1.2 or a PdI value (measured by PCS) below 0.1. Two emulsions had not been freshly prepared but were 8 or 18 months old, respectively. Ostwald ripening of the comparatively polar medium-chain triglyceride nanodroplets had led to an increase in particle size and a decrease in particle size

Discussion

The saturation solubility in the nanocarrier particles, i.e. the lipid type, did not influence the rate constant k in our test of three lipids. This result is very valuable, as it enables selection of the best carrier even if passive loading is not complete at the time of analysis. The arrow in Fig. 1 illustrates this: If the drug loading process was stopped too early, e.g. after 5 h as indicated by the arrow, the acetylated monoglycerides nanoemulsion would have the highest fenofibrate content

Conclusion

The speed of passive loading is not only influenced by drug water solubility and drug surface area [11], but also by the diffusivity of the nanocarrier particles. The quicker the carrier particles diffuse, the higher is the speed of passive loading. The influence of the diffusivity of the lipid nanocarriers and the effect of drug dissolution rate were included in an overall mechanistic model developed for similar processes [22]. The resulting mechanistic model gave a good correlation between

Acknowledgement

The authors thank the Niedersächsisches Ministerium für Wissenschaft und Kultur (MWK) for financial support in the joint research project SynFoBiA – “Novel synthesis and formulation methods for poorly soluble drugs and sensitive biopharmaceuticals” of the Center of Pharmaceutical Engineering (PVZ).

References (24)

  • R. Finsy

    Particle sizing by quasi-elastic light scattering

    Adv. Colloid Interface Sci.

    (1994)
  • G.L. Flynn et al.

    Mass transport phenomena and models: theoretical concepts

    J. Pharm. Sci.

    (1974)
  • Cited by (10)

    • Preparation of Daidzein microparticles through liquid antisolvent precipitation under ultrasonication

      2021, Ultrasonics Sonochemistry
      Citation Excerpt :

      This was also consistent with the results of the fitting curve. The major characteristic of the nano-sized drug particles was the increase in the dissolution rate of the compound [53]. Consequently, we could increase the particle surface area by reducing the particle size, and ultimately improve the drug saturation solubility and the dissolution rate [54].

    • Platelet lysate-loaded PLGA nanoparticles in a thermo-responsive hydrogel intended for the treatment of wounds

      2020, European Journal of Pharmaceutical Sciences
      Citation Excerpt :

      As observed, although the %EE decreased when increasing the amount of PL, the LC increased by more than 60%. The reduction in %EE by increasing the amount of PL from 2 mg (drug:polymer ratio 1:75) to 8 mg (drug:polymer ratio 1:19) can be attributed to the saturation of the system, as reported for various nanometric carriers (Goke and Bunjes, 2018; Sun et al., 2012). One of the advantages of therapeutic proteins is that very small concentrations have significant pharmacological effects (Mitragotri et al., 2014).

    • Production of solid lipid nanoparticles with a supercritical fluid assisted process

      2019, Journal of Supercritical Fluids
      Citation Excerpt :

      SLNs were introduced in 1991 as an alternative to traditional drug transport systems [15–17]. They are very similar to lipid nanoemulsions [18–20] or liposomes [21–23], since they are both suspensions of lipid nanoparticles dispersed in an aqueous surfactant solution; whereas the main difference among SLNs and lipid nanoemulsions or liposomes is the aggregation status of lipids. The use of solid lipids for the production of nanocarriers is an extremely interesting approach to obtain drug controlled release, since the mobility of a molecule in a solid lipid is considerably lower than in a liquid oil [24].

    • Novel strategies for the formulation and processing of poorly water-soluble drugs

      2018, European Journal of Pharmaceutics and Biopharmaceutics
      Citation Excerpt :

      Therefore, the nanocarrier concentration during passive loading should be just as high as required for drug quantification after passive loading. As no significant influences by the matrix of the lipid nanocarrier or the saturation solubility within the overall dispersion on the loading speed have been observed [63], the best carrier for a given drug can be identified at a time point even before completion of the passive loading process. The knowledge on the loading mechanism now allows the use of passive loading for the straightforward, material-saving selection of suitable lipid nanocarriers.

    View all citing articles on Scopus
    View full text