Encapsulation of DNA in Macroscopic and ... - ACS Publications

Nov 27, 2013 - Department of Chemistry, University of Coimbra, 3004-535 Coimbra, Portugal. ∥. CR Competence AB, Center for Chemistry and Chemical ...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/Langmuir

Encapsulation of DNA in Macroscopic and Nanosized Calcium Alginate Gel Particles Alexandra H. E. Machado,*,†,‡ Dan Lundberg,§,†,∥ António J. Ribeiro,‡ Francisco J. Veiga,‡ Maria G. Miguel,§ Björn Lindman,§,† and Ulf Olsson† †

Division of Physical Chemistry, Center for Chemistry and Chemical Engineering, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden ‡ Center for Pharmaceutical Studies, Faculty of Pharmacy, University of Coimbra, 3000-548 Coimbra, Portugal § Department of Chemistry, University of Coimbra, 3004-535 Coimbra, Portugal ∥ CR Competence AB, Center for Chemistry and Chemical Engineering, Lund University, P.O. Box 124, 221 00 Lund, Sweden S Supporting Information *

ABSTRACT: Calcium alginate beads, which are biodegradable and biocompatible, have been widely employed as delivery matrices for biomacromolecules. In the present work, the feasibility of encapsulation of DNA (which is used as a model biomacromolecule) in calcium alginate nanobeads (sub-200 nm size), prepared using a recently developed protocol based on the phase inversion temperature (PIT) emulsification method [Machado et al. Langmuir 2012, 28, 4131−4141], was assessed. The properties of the nanobeads were compared to those of the corresponding macroscopic (millimeter sized) calcium alginate beads. It was found that DNA, representing a relatively stiff and highly charged polyanion (thus like-charged to alginate), could be efficiently encapsulated in both nanosized and macroscopic beads, with encapsulation yields in the range of 77− 99%. Complete release of DNA from the beads could be accomplished on dissolution of the gel by addition of a calcium-chelating agent. Importantly, the DNA was not denatured or fragmented during the preparation and collection of the nanobeads, which are good indicators of the mildness of the preparation protocol used. The calcium alginate nanobeads prepared by the herein utilized protocol thus show good potential to be used as carriers of sensitive biomacromolecules. modifications of the alginate molecule9,10 or by inclusion of additional components in the system.10,11 A majority of studies on calcium alginate have been performed on macroscopic gels. Gel particles with sizes of 1− 3 mm are easily prepared by dripping an aqueous solution of alginate into a solution containing calcium ions.8,12,13 However, many of the potential applications involving biomacromolecules, e.g., the preparation of vehicles that can be efficiently taken up by cells and certain mucosal tissues, require the use of particles with sizes in the nanometer to micrometer range.14−16 When designing a protocol for the preparation of such particles, it is crucial to consider the high sensitivity of biomolecules toward e.g. elevated temperatures and shear; hence, mild conditions should be employed. Recent reports have shown that calcium alginate nanobeads can be prepared using protocols where microemulsions17,18 or nanoemulsions19 are employed as templates. In a previous publication, the authors of this work have developed a method for the preparation of calcium alginate particles in the sub-200 nm size range, where the characteristic temperature depend-

1. INTRODUCTION Biomacromolecules, such as peptides, proteins, and nucleic acids, have received great attention over the past few decades because of their increasing potential in therapeutic applications.1 The formulation of these molecules is however often a challenge. First, these molecules are large and often highly charged. Second, biomolecules are typically susceptible to degradation during handling, storage, and/or after administration to the patient and thus require protection to ensure their integrity. Taken together, these factors can seriously compromise transport of the biomacromolecules to the site of action and ultimately limit bioavailability. To overcome these limitations, encapsulation or complexation of biomacromolecules in a suitable delivery system is thus of critical importance.2−4 A common strategy for the encapsulation of biomacromolecules is based on the use of hydrogels, such as calcium alginate gels. Calcium alginate gels are formed on cross-linking of the anionic polysaccharide by the divalent cations. These gels are biocompatible, biodegradable, nontoxic, show bioadhesive properties, and have been found to provide a good matrix for the entrapment of biomolecules as well as complete cells.5−8 Additionally, when appropriate for the desired application, the properties of calcium alginate gels can be modulated by direct © 2013 American Chemical Society

Received: January 16, 2013 Revised: November 21, 2013 Published: November 27, 2013 15926

dx.doi.org/10.1021/la4032927 | Langmuir 2013, 29, 15926−15935

Langmuir

Article

2.3. Macrobeads. 2.3.1. Preparation, Collection, and Determination of Size. Alginate macrobeads containing DNA were prepared by dripping, by means of a peristaltic pump with a 10 μL pipet tip at the open end, the solution containing alginate and DNA into a stirred solution of 0.1 M calcium chloride (CaCl2, Riedel-de Haën, Germany) at room temperature. After incubation in the CaCl2 solution for 24 h (without stirring), the macrobeads were separated by vacuum filtration and extensively washed with Milli-Q water. Freshly collected beads were used in the following investigations. For comparison, alginate macrobeads without DNA were prepared using the same procedure. The mean size of the macrobeads was estimated by measuring the diameter of 10 freshly collected macrobeads from each batch, using an electronic digital caliper (Sylvac, Switzerland). 2.3.2. Visualization of the DNA. Freshly prepared macrobeads were stained with a solution of Gelstar nucleic acid gel stain (10 000× concentrate, Cambrex; diluted to 1× in 0.1 M CaCl2); after at least 30 min equilibration, they were placed on a microscopic slide and irradiated with a 365 nm light, provided by an Axioplan microscope (Zeiss, Germany) equipped with an appropriate filter. The presence of DNA was assessed visually by the color of the irradiated macrobeads. 2.3.3. Quantification of the Encapsulated DNA. In order to release the DNA, the macrobeads were dissolved in 5 mL of 0.25 M disodium hydrogen phosphate (Na2HPO4, Merck, Germany). The samples were then diluted 20-fold with 10 mM Trizma (Tris) buffer (Sigma), pH 7.6, and filtered using a 0.80 μm Minisart syringe filter (Sartorius, Germany) to remove the calcium phosphate precipitate. The DNA in the filtrates (i.e., the encapsulated DNA) was quantified by measuring the absorbance of the samples at 260 nm. Alginate macrobeads without DNA were subjected to the same procedure, and the filtrates were used as blanks. The encapsulation efficiency (%) was calculated as the percentage of encapsulated DNA relative to the total amount of DNA initially added. 2.3.4. Assessment of the Effects of Exposure to Different Aqueous Media. These studies were performed in either 10 mM Tris-HCl buffer, pH 7.6, or 10 mM Tris-HCl with 150 mM sodium chloride (NaCl, Sigma), pH 7.6. In order to estimate the amount of DNA released from the alginate macrobeads, 10 macrobeads (∼90 mg) were placed in glass vials, each containing ∼4 mL of medium, and the samples were tilted back and forth at a constant speed at room temperature using a Mixer 440 (Swelab, Sweden). At specific time intervals (between 4 and 840 h), aliquots of the media were removed, and the amount of DNA in the supernatant was determined by measuring the absorbance at 260 nm. The media were put back in the respective vials, and samples were again placed on the mixer. As a control, alginate macrobeads without DNA were also subjected to the same conditions, and the removed media were used as blanks. 2.4. Nanobeads. 2.4.1. Preparation and Collection. Calcium alginate nanobeads were prepared and separated according to a method previously described by us,19 with the difference that mixtures of DNA and alginate were used instead of pure alginate solutions as the aqueous phase of emulsions containing tetraethylene glycol monododecyl ether (C12E4, Nikko Chemicals, Japan) and decane (Sigma) as the surfactant and the oil, respectively. The experimental details are presented in the Supporting Information. 2.4.2. Estimation of Average Sizes and Size Distributions of Nanoemulsions and Nanobeads. Aliquots of the samples were placed in low volume poly(methyl methacrylate) (PMMA) cuvettes and analyzed using dynamic light scattering (DLS) to assess the mean hydrodynamic radius (RH) and the polydispersity index (PDI). The experiments were performed on a Malvern Zetasizer Nano ZS (Malvern, UK) instrument with backscatter detection (173°), controlled by the Dispersion Technology Software (DTS 5.03, Malvern, UK). Mean RH and intensity-averaged size distributions were obtained from the raw data using the general purpose inverse Laplace transform method provided in the instrument software. The PDIs were estimated from cumulant analysis, which is also provided with the instrument software. The measurements were performed at 40 °C for the nanoemulsions and at 5 °C for the nanobead dispersions; the latter temperature, which is below the cloud point of the surfactant, i.e., 7 °C for 1 wt % C12E4,19 was used in order to avoid

ence of nonionic oligoethylene oxide (EO) surfactants properties was exploited.19 By using the phase inversion temperature (PIT) emulsification method,20,21 which involves rapid changes between carefully selected temperatures, waterin-oil (W/O) nanoemulsions containing alginate in the aqueous phase, with low size distributions, could be prepared. Calcium alginate gel particles were then formed in the aqueous pools of the emulsion on addition of a solution containing calcium ions.19 Since mild conditions (relatively low temperatures and low input of mechanical energy) are used throughout this protocol, there are good reasons to believe that it is appropriate for the encapsulation of sensitive biomacromolecules. In the present work, we explored the feasibility of applying the PIT-based protocol to produce calcium alginate nanobeads incorporating biomacromolecules. The biomacromolecule selected for the experiments was the challenging substance deoxyribonucleic acid (DNA). Since DNA is a stiff (persistence length of ∼40−50 nm)22,23 and highly negatively charged polyelectrolyte, it can occupy a large volume in solution if not, by some means, compacted. The fact that DNA is anionic presents an additional challenge due to the charge similarity with alginate. Furthermore, DNA chains are easily fragmented when subjected to shear,24,25 which implies that an assessment of the extent of fragmentation can be used as an indicator of “mildness” of the evaluated protocol. Finally, DNA can be easily visualized by the addition of an appropriate fluorescent probe molecule. The nanobeads produced were characterized with respect to different physicochemical properties, DNA encapsulation efficiency, and release profiles. Furthermore, their behavior was compared to that of corresponding macroscopic calcium alginate beads prepared by a dripping method.

2. EXPERIMENTAL PROCEDURES 2.1. Preparation of Polymer Solutions. Pharmaceutical grade sodium alginate Protanal LF 10/60, containing 65−75% guluronic acid (G) residues and 25−35% mannuronic acid (M) residues (kindly donated by FMC BioPolymer, Norway), was purified according to a previously reported procedure19 (details can be found in the Supporting Information). The pH of the stock alginate solution (3 wt %) was adjusted to 7−8, by addition of NaOH, prior to dilution to the desired concentrations. The average molecular weight of Protanal LF 10/60 alginate has been previously reported to be 126 000 g mol−1.26 Stock solutions of DNA sodium salt from salmon testes (Sigma) were dialyzed against water for 2 days, at room temperature, using dialysis tubing with a molecular weight cutoff of 6−8 kDa (Spectrum). The pH of the DNA solutions was adjusted to 7−8, by addition of NaOH, before dilution to the wanted concentrations. The concentration of DNA was assessed from the absorbance at 260 nm (measured using a Cary 300 Bio UV−vis spectrophotometer (Varian)), considering a molar extinction coefficient of 6600 M−1 cm−1 for double-stranded DNA.27 The average molecular weight of a nucleotide phosphate residue was assumed to be 330 g mol−1.28 The A260/A280 ratio was higher than 1.8, which suggests that the DNA is essentially free from protein contamination.27 The size of the herein used quality of DNA has been previously determined by agarose gel electrophoresis to be approximately 10 000 base pairs (bp).29 2.2. Aqueous Mixtures of Alginate and DNA. Mixtures of alginate and DNA were prepared by mixing the appropriate proportions of stock solutions of 3 wt % alginate and 0.32 wt % DNA at different ratios; the final concentrations in the mixtures were of 1 or 1.5 wt % alginate and 0.08 or 0.16 wt % DNA. The aqueous mixtures were stored in a refrigerator until further use. Before the particle preparation, the mixtures were allowed to attain room temperature and were well stirred to ensure homogeneity. 15927

dx.doi.org/10.1021/la4032927 | Langmuir 2013, 29, 15926−15935

Langmuir

Article

Figure 1. Photographs of macrobeads obtained by dripping a 1.5 wt % alginate solution containing (a) 0, (b) 0.08, or (c) 0.16 wt % DNA into a 100 mM CaCl2 solution.

Figure 2. Photographs of macrobeads stained with Gelstar (1×) and irradiated with 365 nm light provided by a fluorescence microscope. Beads were prepared from 1 wt % alginate solutions with (a) 0, (b) 0.08, or (c) 0.16 wt % DNA. the possible interference from the residual surfactant. The properties of the beads are not expected to be affected by this change in temperature.6 2.4.3. Quantification of the Encapsulated DNA. Dispersions of nanobeads were centrifuged at 40000g for ∼30 min. Since the pellet was often detached from the bottom of the flask, samples were further centrifuged at 10000g for 10 min. The supernatant was collected and diluted with 10 mM Tris buffer (pH 7.6), whereas the pellet was dissolved with 2 mL of 0.25 M Na2HPO4. The amount of DNA in both phases was determined by measuring the absorbance of the samples at 260 nm. Because of the slight variations in the yield of the nanobead fraction between batches, the total amount of DNA present in the nanobead dispersion was taken as the sum of the amount of DNA in the respective phases. For reference, the same procedure was performed on alginate nanobeads without DNA. The encapsulation efficiency (%) was calculated as the percentage of DNA in the nanobead fraction relative to the total amount of DNA in the dispersion. 2.4.4. Assessment of the Effects of Exposure to Different Aqueous Media. These studies were performed in either 10 mM Tris-HCl buffer, pH 7.6, or 10 mM Tris-HCl with 150 mM NaCl, pH 7.6. Nanobead dispersions were placed in Eppendorf tubes, each containing ∼0.6 mL of medium, and then tilted back and forth in the same way as described above for the macrobeads. Different samples were prepared for different mixing times. At specific time intervals (between 24 and 840 h), samples were transferred to centrifugation flasks and processed as described above. 2.5. Dissolution of the Beads. In order to promote the complete release of DNA from the macrobeads, as well as from the nanobeads, ethylenediaminetetraacetic acid (EDTA, Sigma) was added to the beads placed in 10 mM Tris-HCl buffer, pH 7.6, or in 10 mM TrisHCl with 150 mM NaCl, pH 7.6, media, which were prepared for the release assays mentioned above and had been agitated for ∼35 days. 0.2 mL of 0.5 M EDTA was added to the ∼4 mL of media containing the macrobeads, whereas 0.15 mL of 0.5 M EDTA was added to the ∼0.4 mL of media containing the nanobeads. Samples were agitated for ∼2 h to ensure complete dissolution of the beads. The solutions of the dissolved macrobeads or nanobeads were diluted 20- and 25-fold, respectively, with the same buffer where they had been incubated, after which the amount of phosphorus in the samples was determined by inductively coupled plasma mass

spectrometry (ICP-MS). The amount of DNA was then calculated from the amount of phosphorus. For the visualization of DNA release, one stained macrobead was placed in a glass vial containing 10 mM Tris buffer. 0.02 mL of 0.5 M EDTA was added to the 0.4 mL of media, and the process was monitored visually by the changes in fluorescence over time during irradiation with a 365 nm light, as described in section 2.3.2.

3. RESULTS AND DISCUSSION 3.1. Aqueous Mixtures of Alginate and DNA. Alginate and DNA are both negatively charged polymers, with linear charge densities, ξ, of 1.7 (∼0.23 negative charge/Å)30 and 4.2 (∼0.59 negative charge/Å),23,31 respectively. Mixtures of similarly charged polyelectrolytes usually show a net-repulsive interaction, which often results in so-called segregative phase separation, where two separated phases, each enriched in either of the polymers, are formed.32 The propensity for phase separation typically increases with the molecular weight as well as with the concentration of polymers.32 It was found that for the ranges of concentrations of alginate and DNA used in this study, i.e., 1−1.5 and 0.08−0.16 wt %, respectively, all mixtures were slightly turbid, in contrast to the clear solutions of either of the single polymers. This shows that phase separation does occur also at these relatively low concentrations. Importantly, however, separation into macroscopic layers was only observed after the mixtures had been left at rest for extended periods of time. 3.2. Encapsulation of DNA in Calcium Alginate Macrobeads. In the first stage of this work, we prepared and characterized macroscopic calcium alginate beads containing DNA. The macrosopic beads are easy to prepare and handle, and the results from these experiments constitute an important reference point when approaching the nanosized beads, which are significantly more demanding to produce and investigate. 3.2.1. Preparation of Macrobeads. The macrobeads were prepared using an external gelation method,33 by dripping, using a peristaltic pump, an aqueous mixture of 1 or 1.5 wt % alginate, and 0−0.16 wt % DNA into a solution of 100 mM 15928

dx.doi.org/10.1021/la4032927 | Langmuir 2013, 29, 15926−15935

Langmuir

Article

CaCl2. After formation, the beads were left for curing in the CaCl2 solution for ∼24 h in order to allow optimal cross-linking of the alginate chains by the calcium ions. As the Ca2+ propagates from the surface to the core of the beads, the gel beads shrink and release some water (syneresis).13,34 The weight loss on curing was of the order of ∼50% compared to the weight of the initial alginate and DNA droplets (Figure S1, Supporting Information), which is in accordance with previous findings.33 Figure 1 shows photographs of representative beads prepared from solutions of 1.5 wt % alginate and 0−0.16 wt % DNA. The beads were essentially spherical and showed a narrow size distribution, with diameters between 2.0 and 2.5 mm for all conditions used. It can be noted that macrobeads prepared from alginate only were transparent, whereas those also containing DNA showed a degree of cloudiness similar to that observed for the original aqueous polymer mixtures. Beads prepared from mixtures with 1 wt % alginate were very similar with respect to both appearance and size. 3.2.2. Visualization of DNA in the Macrobeads. Staining with the fluorescent probe Gelstar allowed visualization of the DNA entrapped in the beads. Figure 2 shows results for beads prepared from 1 wt % alginate solutions with or without DNA. Whereas beads containing alginate only, as expected, showed no fluorescence (Figure 2a), those containing DNA showed bright green fluorescence that increased with DNA concentration (Figures 2b,c). The fact that the emitted light is green confirms that the entrapped DNA retains the initial doublestranded structure. Similar results were obtained for the corresponding beads prepared from mixtures with 1.5 wt % alginate. 3.2.3. Encapsulation Efficiency. In order to quantify the DNA contained in the calcium alginate beads, these need to be dissolved, which can be accomplished by the addition of molecules or ions that chelate or preferentially bind to calcium ions. The substances most commonly used for this purpose, sodium citrate and EDTA, both show an absorbance cutoff wavelength very close to the absorbance maximum of DNA, which would interfere with the spectrophotometric quantification of the latter. Thus, an alternative approach was used, where Na2HPO4 was added to sequester the calcium ions by precipitation of calcium phosphate (Ca3(PO4)2; solubility product, Ksp, of 2.07 × 10−33 at 25 °C)35 and, consequently, disrupt the gel network. The precipitate formed was removed by filtration. It was shown in control experiments that the loss of DNA on filtration was insignificant. Table 1 shows the encapsulation efficiency for beads prepared from mixtures with different concentrations of alginate and DNA. It was found that DNA could be efficiently entrapped in the calcium alginate network, with yields ranging from 88 to 99%. These values are similar to those previously obtained for the encapsulation of calf thymus DNA in calcium alginate macrobeads prepared using a similar procedure.33 No distinct trends were found with respect to the variation in encapsulation efficiency with the composition of the initial aqueous mixture. Possibly, the encapsulation efficiency increases slightly with the initial concentration of DNA. The results clearly show that the previously discussed inherent propensity for segregative phase separation in the aqueous alginate−DNA mixtures does not interfere with the possibility to entrap DNA in a calcium alginate gel. First, since the approach of macroscopic phase separation in this system is slow, agitation of the polyelectrolyte mixtures shortly before

Table 1. Efficiency of Encapsulation of DNA in Calcium Alginate Macrobeads Prepared with Different Concentrations of Alginate and DNA alginatea (wt %)

DNAa (wt %)

1.0 1.0 1.5 1.5

0.08 0.16 0.08 0.16

encapsulation efficiencyb,c (%) 91.5 98.7 88.6 91.8

± ± ± ±

4.7 1.1 5.4 6.0

a

The concentrations in the initial aqueous mixtures of alginate and DNA. bThe encapsulation efficiency is given as the percentage of entrapped DNA relative to the total amount of DNA in the droplet of polymer mixture added to the CaCl2 solution. cThe presented values are average yields and standard deviations for n = 4.

preparation of the calcium alginate gel ensures homogeneity on a length scale well below that of the beads. Furthermore, it has been reported that calcium alginate macrobeads prepared with an initial concentration of 2% alginate present a molecular weight cutoff of 394 kDa for transport of linear molecules, such as DNA, through the gel network.36 Since the herein used DNA has a molecular weight of about 1 order of magnitude above this value (6600 kDa), one would expect only very small DNA fragments to be able to escape through the gel network during preparation of the beads. Additionally, it has been suggested that the specific electrostatic interactions between DNA and calcium ions37 might contribute to its retention within the calcium alginate network.36 It is noteworthy that the interactions with Ca2+ do not induce significant changes in DNA conformation.37 In order to assess whether there was a preferential inclusion of larger DNA molecules in the beads, which would result in a shift in the size distribution as compared to that of the DNA in the original solution, agarose gel electrophoresis was performed on the dissolved gels (the experimental details are given in the Supporting Information and the results shown in Figure S2). It was found that the entrapped DNA showed no appreciable difference to that of the DNA in the original solution, which indicates that no significant fractioning of the DNA occurred in the encapsulation process. 3.2.4. Compositions of the Final Macrobeads. Based on the findings discussed in the previous sections, it is possible to estimate the final compositions of the macrobeads. Data are shown in Table 2. One point to note is that the concentrations of alginate and DNA in the final beads were substantially higher than in the initial droplets. This increase in concentration can be ascribed to the syneresis during the curing step, as discussed above. 3.2.5. Swelling and Integrity of Macrobeads on Exposure to Different Aqueous Media. The integrity of a calcium alginate network depends on the balance between swelling of the gel, induced by differences in the chemical potential of water inside and outside the gel, and the elastic retraction force exerted by the physical cross-links.34,38 The composition of an aqueous medium, to which the calcium alginate gels are exposed, will thus influence their swelling and integrity. We assessed the influence on the calcium alginate macrobeads of two different aqueous media, 10 mM Tris buffer (pH 7.6) as well as a solution of the same buffer containing physiological concentration of NaCl, i.e., 0.9% or 150 mM. By visual inspection alone, it was obvious that the macrobeads showed a significant increase in volume when exposed to either of the aqueous media. The degree of swelling 15929

dx.doi.org/10.1021/la4032927 | Langmuir 2013, 29, 15926−15935

Langmuir

Article

Table 2. Final Compositions of the DNA-Loaded Calcium Alginate Macrobeads initial conc in droplets

final conc in beadsa b

alginate (wt %)

DNA (wt %)

alginate (wt %)

DNAc (wt %)

Rmbd (mm)

NA/Nmbe,f (×1014)

ND/Nmbg,h (×1012)

ND/NAi

1.0 1.0 1.5 1.5

0.080 0.16 0.080 0.16

2.2 2.2 2.7 2.7

0.16 0.35 0.13 0.27

1.1 1.1 1.2 1.2

∼5.9 ∼5.9 ∼9.4 ∼ 9.4

∼0.8 ∼1.8 ∼0.8 ∼1.8

∼0.0014 ∼0.0030 ∼0.0009 ∼0.0019

a Concentrations were calculated considering a weight loss during curing of ∼55% and ∼45% for macrobeads initially prepared from solutions of 1 and 1.5 wt % alginate, respectively. bThe alginate concentration was calculated assuming that no alginate was lost in the gelation process.33 cThe DNA concentration was calculated based on the mean DNA encapsulation yields presented in Table 1. dRmb is the mean radius of the final macrobeads, determined using an electronic digital caliper. eNA/Nmb is the number of alginate molecules per macrobead. fNA is calculated using a molecular weight of alginate of 126 000 g mol−1.26 gND/Nmb is the number of DNA molecules per macrobead. hND is calculated using a molecular weight of DNA of 6 600 000 g mol−1.29 iND/NA is the ratio of the number of DNA molecules to the number of alginate molecules in a macrobead.

Figure 3. Release of DNA from macrobeads in the presence of (a) 10 mM Tris-HCl buffer (pH 7.6) or (b) 10 mM Tris-HCl buffer containing 150 mM NaCl (pH 7.6) solutions. The beads were prepared from solutions with alginate and DNA concentrations as presented in the legend of the figure.

ions. In order to escape from the gel, the DNA molecules need to diffuse through the pores of the polymer network. The fact that a vast majority of the encapsulated DNA is retained in the gel suggests that the swelling of the gel is rather homogeneous, i.e., that no very large pores are formed and that no major restructuring of the gel occurs even after extended exposure to an aqueous solution. It can be noted, however, that a radial variation in the pore size over the bead may be expected. Calcium alginate macrobeads prepared by the external gelation method have been found to show a polymer concentration gradient within the gel, with a more compact outer layer and a less dense core, which can be ascribed to a rapid and virtually irreversible crosslinking of the alginate on exposure to calcium ions.8,40,41 The possible narrower pores of the outer layer of the beads may restrain diffusion of DNA out from the bead and thus enhance its retention in the beads. The fact that the release of DNA occurs mainly at an initial state and levels out after a certain period of time is probably a consequence of the swelling of the gels, and an increase in pore size, that occurs within the first hours after exposure to the external media. It is thus reasonable to expect that smaller fragments or not properly entangled DNA are released from the macrobeads during this time; possibly, a fraction of this DNA could be described as adsorbed to the surface of the beads rather than properly embedded in the gel structure. At both conditions investigated, the release seems to be somewhat more extensive for the beads with the higher concentrations of both polymers, i.e., those prepared from the mixture of 1.5 wt % alginate and 0.16 wt % DNA. This finding can tentatively be ascribed to a larger extent of segregative

was quantified from the increase in weight of the beads after incubation in the different media (the experimental details can be found in the Supporting Information). Whereas the variation in composition of the beads did not give any appreciable effect on the swelling degree, significant differences were found in the different aqueous media. For instance, after an incubation period of 24 h, the increase in weight was ∼20% for the beads placed in 10 mM Tris buffer and ∼85% for those placed in the same buffer also containing NaCl. The higher swelling degree observed in the latter can be explained by a competition exerted by Na+ on Ca2+ for the alginate.12,13,39 As sodium ions replace calcium ions, there is an increase in the osmotic pressure in the gel and a decrease in the effective cross-link density. Therefore, concomitant with the swelling of the gel, there is a more significant loosening of the gel structure and a consequent softening of the beads on exposure to the buffer with 150 mM NaCl. However, since Ca2+ has a higher affinity for the alginate (specifically for the G residues) than Na+, the replacement remains incomplete, and the gel network is retained also with an excess of sodium ions. The influence of the external aqueous medium on the internal structure of the calcium alginate network may also affect the rate and extent of a possible release of encapsulated DNA from the beads. Figure 3 shows the release of DNA over time from macrobeads of different composition when exposed to Tris buffer with or without NaCl. It can clearly be seen that the release of DNA from the beads to the surrounding media was very limited for all the investigated conditions. Since there is no electrostatic attraction between alginate and DNA, the retention of the latter in the beads is predominantly controlled by steric effects, possibly supported by the aforementioned interactions between the DNA and the calcium 15930

dx.doi.org/10.1021/la4032927 | Langmuir 2013, 29, 15926−15935

Langmuir

Article

compared to that of alginate, 57 nm.26 The fact that no populations of large droplets were found for the emulsions containing both polymers can be taken as proof that the DNA resides together with the alginate in the droplets. This coexistence of DNA and alginate can likely be ascribed to a combination of entanglements among the polymers and electrostatic screening resulting from the significant ionic strength provided by the alginate. To induce cross-linking of the alginate in the emulsion droplets, and thereby obtain the DNA-loaded calcium alginate nanobeads, a solution of CaCl2 was added to the W/O emulsions containing the polymer mixtures. The addition of calcium ions had no significant influence on emulsion stability or on the size distribution profiles. A minor decrease in the mean droplet volume, of up to ∼10% and independent of the presence of DNA, was observed over the first hour after the addition of CaCl2. A similar droplet shrinkage was previously observed for emulsions containing alginate only and can reasonably be attributed to the aforementioned syneresis effect.19 After curing, the nanobeads formed in the nanoemulsion droplets were collected using the previously described gentle extraction procedure.19 This separation step also results in removal of the majority of the oil and the surfactant, although traces of these components might still be present in the nanobead dispersion.19 3.3.2. Size and Charge of the Nanobeads. The DNAloaded calcium alginate nanobeads were characterized with respect to their mean size and size distributions using DLS. The results are shown in Table 3.

phase separation and thus to the presence of a larger fraction of DNA not entangled in the alginate network. It is noteworthy that the release of DNA is overall slightly higher from the beads that have been exposed to the medium containing 150 mM NaCl. These small differences can probably be attributed to a somewhat larger increase in the average pore size of the gel network as a consequence of the larger degree of swelling observed in this medium. 3.3. Encapsulation of DNA in Calcium Alginate Nanobeads. The findings on the macrobeads clearly show that DNA can be efficiently encapsulated in calcium alginate gels and remains entrapped within the gel network when the beads are exposed to aqueous solutions containing up to 150 mM of NaCl. In the following, our evaluation of the feasibility of encapsulating DNA in calcium alginate nanobeads is discussed. 3.3.1. Preparation of Calcium Alginate Nanobeads. As was discussed in the Introduction, we have recently developed a mild protocol for the preparation of calcium alginate beads in the sub-200 nm size range.19 In a different work, we have shown that the interactions between DNA and nonionic EO surfactants are weakly repulsive, suggesting that no notable changes in neither the conformation and stability of the DNA nor the surfactant phase behavior are expected.29 The first step toward the preparation of DNA-containing calcium alginate nanobeads was to ensure that DNA could be incorporated in alginate-containing aqueous droplets of a W/O nanoemulsion. Thus, aqueous mixtures of DNA and alginate were emulsified using the previously developed method.19 Figure 4 shows the influence of the inclusion of DNA on the

Table 3. Characterization of the Calcium Alginate Nanobeads after Collectiona

Figure 4. Size distributions (determined using DLS) of the nanoemulsions containing the concentrations of alginate and DNA presented in the legend of the figure. The samples were prepared with ϕW/ϕO = 0.07 and ϕW/ϕS = 5. All measurements were performed at 40 °C.

alginateb (wt %)

DNAb (wt %)

R Hc (nm)

PDI

1.0 1.0 1.0 1.5 1.5 1.5

0.00 0.080 0.16 0.00 0.080 0.16

89 91 92 98 98 97