Multivariate analysis for the optimization of polysaccharide-based nanoparticles prepared by self-assembly

Highlights

• Multivariate analysis facilitates the optimization of polysaccharide nanoparticles.

• Monodispersed particles are obtained by using a low polysaccharide concentration.

• Interaction between preparation factors was revealed through multivariate analysis.

• Factors⿿ interactions allow preparing particles at a high polysaccharide concentration.

Abstract

Polysaccharide-based nanoparticles are promising carriers for drug delivery applications. The particle size influences the biodistribution of the nanoparticles; hence size distributions and polydispersity index (PDI) are critical characteristics. However, the preparation of stable particles with a low PDI is a challenging task and is usually based on empirical trials. In this study, we report the use of multivariate evaluation to optimize the formulation factors for the preparation of alginate-zinc nanoparticles by ionotropic gelation. The PDI was selected as the response variable. Particle size, size distributions, zeta potential and pH of the samples were also recorded. Two full factorial (mixed-level) designs were analyzed by partial least squares regression (PLS). In the first design, the influence of the polysaccharide and the crosslinker concentrations were studied. The results revealed that size distributions with a low PDI were obtained by using a low polysaccharide concentrations (0.03⿿0.05%) and a zinc concentration of 0.03% (w/w). However, a high polysaccharide concentration can be advantageous for drug delivery systems. Therefore, in the second design, a high alginate concentration was used (0.09%) and a reduction in the PDI was obtained by simultaneously increasing the ionic strength of the solvent and the zinc concentration. The multivariate analysis also revealed the interaction between the factors in terms of their effects on the PDI; hence, compared to traditional univariate analyses, the multivariate analysis allowed us to obtain a more complete understanding of the effects of the factors scrutinized. In addition, the results are considered useful in order to avoid extensive empirical tests for future formulation studies.

Introduction

Alginate is a natural polysaccharide extracted from brown algae that is widely used in several different fields [1], [2], [3]. FDA classifies alginate among the generally regarded as safe (GRAS) food additives, and purified alginate is commonly used also for biomedical and pharmaceutical applications [4], [5], [6] due to its biocompatibility, biodegradability and low toxicity.

Alginate is a linear unbranched copolymer composed of β-d-mannuronic units (M) bound with units of the C-5 epimer α-l-guluronic acid (G) through (1 ⿿ 4) linkages [7] (Fig. 1). M and G residues, when deprotonated, are negatively charged and in this form they can bind divalent or multivalent cations, thus creating three-dimensional ionotropic gel networks [7], [8]. The gel-forming properties make alginate an interesting compound for drug delivery, in particular for formulations with sustained and controlled release [9], [10], [11]. In fact, the alginate gel network can create a viscous barrier that controls the diffusion of the entrapped molecules depending on the pH and ionic strength of the environment [12], [13]. Moreover, alginate has mucoadhesive properties that can also improve the bioavailability by facilitating prolonged delivery at the site of administration [14].

Depending on the desired application, alginate gels can be prepared in different forms, such as block gels, films, beads, fibers, microparticles, and nanoparticles. Alginate nanoparticles have mainly been prepared through two methods; the alginate-in-oil emulsification method and the self-assembly method [15]. The self-assembly method is particularly interesting since organic solvents are avoided and mild preparation conditions are used allowing for encapsulation of sensitive material [16], [17], [18], [19].

In the self-assembly method, solutions of one or two different cationic crosslinkers are added to an alginate solution under controlled conditions while mixing. This can be followed by further processing, such as sonication and centrifugation. Also other polysaccharide nanoparticles can be prepared through the self-assembly method, such as nanoparticles made with chitosan or pectin [20], [21]. This method of preparation of polysaccharide nanoparticles involves many factors that can affect important characteristics of the nanoparticles. For example, the concentration of the polysaccharide solution, the molecular weight of the polysaccharide and the preparation conditions can affect the size; the crosslinker concentration can modify the polydispersity and the zeta potential; the pH and the ionic strength of the solvent can change the compactness of the particles [17], [21], [22], [23], [24], [25], [26], [27], [28]. Therefore, by tuning different factors, it is possible to achieve the desired characteristics of the nanoparticles.

Nevertheless, empirically finding the optimal levels for all the factors is seldom an easy task since many factors are involved. Moreover, the variation of one factor at a time as in univariate approaches could be limiting since possible interactions between the factors can be hidden. Multivariate analysis can be a valuable tool for analyzing the data when more than one factor is involved in the process [29], [30]. In particular, the partial least squares regression analysis (PLS) is a statistical method that can be used to determine the interactions between the factors and their square effects, the significant factors that modify the response, and the trends of the effects on the response. In addition, PLS can also predict the values of the factors that, based on the model, can provide the optimal response value [31].

In a recent study [26], we have successfully prepared stable alginate nanoparticles by self-assembly using zinc as the crosslinker without the need of additional polycations, which are commonly used for stabilization. In the present study, a PLS was performed for the first time for the investigation of important factors influencing the characteristics of polysaccharide nanoparticles prepared by self-assembly. The formulation factors of the new nanoparticulate system made with alginate were optimized by the use of factorial design and multivariate evaluation for providing nanoparticles suitable for drug delivery purposes with a low polydispersity and monomodal size distributions. Two different designs were investigated. The first design was used to study the factors crosslinker (zinc) concentration and polysaccharide concentration. The second design was used to investigate the possibility of producing nanoparticles at the highest level of alginate concentration, since a high polysaccharide concentration can be preferred for drug delivery formulations. This was examined by varying the ionic strength of the solvent and the crosslinker concentration. The response variable investigated in the multivariate analysis was the polydispersity index (PDI) of the samples. In addition, the average size, the size distributions, the zeta potential and the pH were recorded.