Tripolyphosphate

Oct 10, 2012 - Self-assembled micro- and nanogels are frequently prepared by mixing tripolyphosphate (TPP) with dilute chitosan solutions. Upon its ad...
1 downloads 6 Views 1MB Size
Download PDF
Article pubs.acs.org/Biomac

Salt-Assisted Mechanistic Analysis of Chitosan/Tripolyphosphate Micro- and Nanogel Formation Yan Huang† and Yakov Lapitsky*,†,‡ †

Department of Chemical and Environmental Engineering, University of Toledo, Toledo, Ohio 43606, United States School of Green Chemistry and Engineering, University of Toledo, Toledo, Ohio 43606, United States

‡

S Supporting Information *

ABSTRACT: Self-assembled micro- and nanogels are frequently prepared by mixing tripolyphosphate (TPP) with dilute chitosan solutions. Upon its addition, the TPP ionically cross-links the chitosan molecules into gel-like colloids that range from tens of nanometers to micrometers in diameter. These particles are biocompatible, mucoadhesive and, because they are easy to prepare under very mild conditions, attract widespread interest in the encapsulation of drugs, neutraceuticals, and other bioactive payloads. Despite their broad use, however, their formation mechanism has remained largely obscured by the very fast kinetics of their self-assembly. To this end, we have tuned the TPP and monovalent salt (NaCl) concentrations to dramatically slow down this process (to occur on the time scale of days instead of milliseconds), and then probed the evolution in the size and morphology of micro- and nanogels during their formation. This revealed that the micro- and nanogel formation rates are extremely sensitive to NaCl and TPP concentrations, and that the formation process occurs in two stages: (1) formation of small primary nanoparticles and (2) aggregation of primary particles into larger, higher-order colloids that are obtained at the end of the ionotropic gelation process.

■

INTRODUCTION When anionic tripolyphosphate (TPP) is titrated into dilute chitosan solutions, the TPP ionically cross-links the cationic chitosan chains into micro- and nanogels that range from tens of nanometers to micrometers in diameter.1−4 These microand nanogels are biocompatible,5,6 mucoadhesive,7,8 and stabilize proteins against denaturation.9,10 Due to these properties, and their ability to form in mild aqueous environments (which prevents payload degradation11), they attract widespread interest as vectors for the delivery of drugs,4,12−14 genes,11,15 and food additives.16,17 The suitability of these chitosan/TPP particles to their various applications depends on their size distributions and swelling characteristics.18,19 To this end, the effects of various process and formulation parameters (e.g., chitosan and TPP concentrations,1,3,20,21 chitosan molecular weight and degree of deacetylation,2,4 ionic strength,19,20 pH18,19 and temperature22,23) on their formation and structure have been investigated. This revealed that stable micro- and nanogel dispersions form within a limited range of chitosan and TPP concentrations,1 and that the dispersions coagulate when the TPP concentration is in excess to the cationic chitosan binding sites.1,20 Similarly, the chitosan and TPP concentrations used during micro- and nanogel preparation affect particle size.3,20,21 The particle size increases with the chitosan concentration3,21 and, depending on the ionic strength, can either increase or decrease with the TPP concentration.20 At low ionic strengths, particle size decreases with the addition of TPP due to the neutralization of chitosan charge and the higher cross-link density.3,20 At high ionic strengths, however, the particle size slightly increases with the TPP concentration.20 This is © 2012 American Chemical Society

attributed to the lower degree of swelling at higher ionic strengths,20,24 where the effect of TPP-induced network collapse is outweighed by the higher degree of aggregation that occurs at higher TPP concentrations.20 Variations in particle size have likewise been linked to the chitosan molecular weight4 and solution pH.19 Moreover, the particle size is sensitive to the procedure by which the chitosan and TPP are mixed,25−28 which indicates that their size distributions are kinetically controlled. Despite this interest in controlling particle size, the ionotropic gelation mechanism, which underlies the ultimate micro- and nanogel size distribution and morphology, remains poorly understood. Consequently, the formulation of chitosan/ TPP micro- and nanogels (as well as other ionically cross-linked biopolymer micro- and nanoparticles29,30) currently relies on trial and error. This lack of detailed understanding reflects (at least in part) the very rapid kinetics of ionotropic gelation, which make the formation and evolution in structure of newly formed particles difficult to study experimentally.31 Recently, we have shown that the rate of TPP binding to chitosan can be reduced through the addition of monovalent salt (e.g., NaCl).20 On the basis of this result, we hypothesized that NaCl can also be used to slow down the micro- and nanogel formation process, thereby enabling its experimental study by conventional methods (e.g., light scattering and electron microscopy). Here, for the first time, we show how monovalent salt (and low TPP concentration) can be used to Received: September 11, 2012 Revised: October 8, 2012 Published: October 10, 2012 3868

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

Figure 1. Stopped-flow light scattering data showing (a) the evolution in light scattering intensity over time in (red square) 75 mM, (blue triangle) 125 mM and (green triangle) 150 mM NaCl solutions (the inset shows the data obtained without added NaCl), where the solid lines are the doubleexponential fits; and (b) the τ1-values obtained from the double exponential curve fitting at each NaCl concentration (the solid line shows the power law scaling of τ1 with [NaCl] and the error bars are standard deviations). released after each TPP injection over time and subtracting the heat of dilution (obtained by titrating TPP into chitosan-free water at a matching pH), the binding heat absorbed by the sample cell per mole of added TPP was obtained as a function of the TPP:glucosamine molar ratio.20 This binding heat was used to infer the effect of TPP concentration on chitosan/TPP binding. The ITC experiment was repeated thrice to ensure reproducibility. Micro- and Nanogel Size and Morphology. The evolutions in the micro- and nanogel size distributions were probed by DLS, using a Zetasizer Nano ZS (Malvern, UK) dynamic and electrophoretic light scattering instrument. To trigger particle formation, 0.40 wt % TPP was slowly titrated into 0.10 wt % chitosan solutions (to various TPP:glucosamine molar ratios), where both the parent chitosan and TPP solutions were charged with 150 mM NaCl. To ensure uniform mixing during the titrations, the receiving solutions were stirred at 800 rpm with a cylindrical magnetic stir bar (12 mm × 4 mm). DLS measurements were then made 1 h, 1 day, and 7 days after mixing, whereupon the micro- and nanogel size distributions were obtained from the intensity autocorrelation functions using the multiple narrow modes algorithm.32 In addition to the particle size distributions, the DLS measurements tracked the variations in the light scattering intensity (at the 173° detector angle), which was expressed as a derived count rate. This derived count rate indicated the scattering rate that would be obtained using the maximum laser power, and enabled the comparison of light scattering intensities between strongly- and weakly scattering dispersions (the analysis of which required different intensities of incident light). Three replicate samples were used for each DLS measurement. In addition to measuring the size distributions, the evolution in the micro- and nanogel morphology was probed by STEM, using a Hitachi HD-2300 STEM at a 200 kV acceleration voltage. To prepare the TEM grids, one drop of microgel dispersion was placed on the Formvar film side of a carbon grid (carbon type A; Ted Pella, Redding, CA) and allowed to absorb for 3 min. Then, to remove the excess sample and salt, the grid was washed by consecutively placing it onto the surface of four DI water drops (sample side down) for approximately 3 s each. The grid was then air-dried overnight and imaged. Two replicate samples were used for each experimental condition.

reduce the rate of chitosan/TPP micro- and nanogel formation by multiple orders of magnitude, and exploit this phenomenon to elucidate the evolution in their size distributions and morphology during their formation process. The effect of monovalent salt and TPP concentrations on the particle formation kinetics was investigated by stopped-flow light scattering. Then, the temporal evolution in micro- and nanogel size was tracked by dynamic light scattering (DLS) and scanning transmission electron microscopy (STEM). Furthermore, to relate the structural evolution to the molecular binding events, the chitosan/TPP binding was probed by isothermal titration calorimetry (ITC). Using this salt-assisted approach, we identified some of the key colloidal phenomena underlying chitosan/TPP micro- and nanogel formation.

■

MATERIALS AND METHODS

Materials. Millipore Direct-Q 3 deionized water (18.2 MΩ·m) was used in all experiments. Chitosan (viscosity-average molecular weight of 120 kDa), sodium tripolyphosphate (TPP), sodium chloride (NaCl), and acetic acid were purchased from Sigma-Aldrich (St. Louis, MO) and used as received without further purification. The degree of chitosan deacetylation was determined at 90% by pH titration.20 To prepare the parent chitosan and TPP solutions for the micro- and nanogel preparation, the chitosan and TPP were dissolved in water, whereupon acetic acid was added to adjust the pH to 4.00 ± 0.02. Micro- and Nanogel Formation Kinetics. The particle formation kinetics were probed by stopped-flow light scattering, using a DM 45K stopped-flow spectrophotometer (Olis Inc., Bogart, GA). Here, the parent chitosan (0.10 wt %) and TPP (0.06 − 0.15 wt %) solutions were rapidly mixed and settled in the sample cell (within 2 ms) at a 1:5 TPP:chitosan volume ratio. When particles formed inside the sample cell, the light scattering intensity (detected with a photomultiplier tube at a 90° scattering angle) increased. Accordingly, the variations in microgel formation rates were characterized by tracking the temporal evolution in the light scattering intensity (over 4 h) at various NaCl and TPP concentrations. The differences in the rates of scattering intensity increase revealed the effects of NaCl and TPP on the microgel formation kinetics. Each stopped-flow light scattering measurement was repeated three times. Molecular Chitosan/TPP Binding. To relate the micro- and nanogel formation process to its underlying molecular interactions, the chitosan/TPP binding in the limit of low TPP concentrations was probed by ITC (MicroCal VP-ITC, GE Healthcare, USA). Here, fifty 5-μL injections of 0.10 wt % TPP (2.7 mM) were added to a sample cell filled with 1.48 mL of 0.10 wt % (5.5 mM in its cationic glucosamine monomer units) chitosan solution. The calorimeter measured the heat supplied to (or removed from) the sample cell to keep the cell at a constant temperature. By integrating the heat

■

RESULTS AND DISCUSSION Micro- and Nanogel Formation Kinetics. To determine whether NaCl can slow down micro- and nanogel formation, the effects of NaCl on ionotropic gelation rates were investigated by stopped-flow light scattering. Here, 0.10 wt % chitosan and 0.10 wt % TPP solutions were prepared at various NaCl concentrations (ranging between 0 and 150 mM), and rapidly mixed inside the stopped-flow sample cell (such that the chitosan and TPP were mixed at a 0.1:1 TPP:glucosamine 3869

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

Figure 2. Stopped-flow light scattering data showing (a) the temporal evolution in light scattering intensity from chitosan/TPP mixtures (in 100 mM NaCl) containing (black square) 0.27 mM, (green triangle) 0.32 mM, (red circle) 0.45 mM, and (blue triangle) 0.57 mM TPP, where the solid lines are the double exponential fits; and (b) the τ1 values obtained from the double exponential curve fitting at each NaCl concentration (the solid line shows the power law scaling of τ1 with [TPP], and the error bars are standard deviations).

with R2 ≥ 0.95 (see Supporting Information, Table S1). In 75 mM NaCl, the initial particle formation process had a characteristic time (τ1) of 18 s, whereupon further aggregation took place over thousands of seconds. When the NaCl concentration was raised further (from 75 to 150 mM), the increase in the light scattering intensity became even slower, indicating further reduction in micro- and nanoparticle formation rates. These differences in kinetics were quantified by comparing the τ1-values (fitted to the data in Figure 1a) obtained at various NaCl concentrations (see Figure 1b). The decision to use τ1 rather than τ2 as the basis of rate comparison was made with the view of probing the initial rates of micro- and nanogel formation, instead of the slow, higherorder microgel aggregation (discussed later) that occurred late in the process. The use of τ1 was also the reason for limiting the double-exponential fitting to I(t)/I∞ values below 0.8. This was done to prevent the data at the end of the process, which ultimately deviated from the double-exponential function, from reducing the fit quality at the early time points (which were essential for fitting τ1). As shown in Figure 1b, τ1 increased dramatically with the NaCl concentration, from 18 to 400 s as the NaCl concentration was increased from 75 mM to 150 mM. The power-law fit to this data indicated that τ1 scales approximately with [NaCl]5. Because the micro- and nanogel formation rate is inversely related to the formation time, this τ1scaling means that the particle formation rate decreases very dramatically with the addition of NaCl. This results from the slower chitosan/TPP binding,20 which likely reflects the competitive binding of monovalent ions to the chitosan.34,35 Similar experiments were performed to investigate the effect of TPP concentration on the ionotropic gelation rates. Here, the chitosan and NaCl concentrations were held constant (at 0.10 wt % and 100 mM respectively), and the concentration of the parent TPP solution was varied from 0.060 wt % (1.6 mM) to 0.13 wt % (3.4 mM). Opposite to the monovalent salt effect, where the gelation became slower with increasing NaCl concentration, the particle formation time decreased significantly with increasing TPP concentration (Figure 2a). Indeed, the scaling of τ1 with the final TPP concentration in the stopped-flow light scattering cell (see Figure 2b) was inverse to that of the NaCl effect (i.e., τ1 approximately scaled with [TPP]−5). This indicates that the particle formation rate increases sharply with the TPP concentration. The scaling illustrated in Figures 1b and 2b shows that the rates of chitosan/TPP micro- and nanogel formation are

molar ratio). In the absence of micro- and nanogel formation, the light scattering intensity from the sample remained constant. This was confirmed by control measurements where chitosan solutions were mixed with TPP-free water at matching pH and NaCl concentrations (data not shown). When the micro- and nanogels formed, however, the light scattering intensity increased with time, as shown in Figure 1a (where the evolution in the normalized light scattering intensity, I(t)/I∞, is plotted). To obtain I∞ for each salt concentration, I(t) was also tracked for several days by DLS (which allowed for longer-term data collection than stopped-flow light scattering) until it stopped changing. The scattering data measured by DLS, from chitosan/TPP samples prepared via stopped-flow mixing, was nearly proportional to that measured by stopped-flow light scattering (data not shown). Thus, by renormalizing the I(t) curves measured by DLS to overlap those obtained by stoppedflow light scattering, DLS was used to estimate the theoretical I∞-value for the stopped-flow light scattering experiment. Without added NaCl, there was a sharp increase in the light scattering intensity (see inset in Figure 1a), which mostly occurred during the 2-ms mixing process. These rapid selfassembly kinetics made the micro- and nanogel formation process difficult to analyze. When 75 mM NaCl was added, however, the aggregation process became much slower, occurring over thousands of seconds. Similarly, when the NaCl concentration was raised from 75 to 150 mM, the increase in the light scattering intensity became even more gradual, with aggregation taking place over several days. To provide a semiquantitative comparison between the aggregation rates, the evolutions in light scattering intensities were fitted (up to a I(t)/I∞ value of 0.8 for each experimental condition) to an empirical double-exponential function:33 I(t )/I∞ = A1(1 − e−t / τ1) + A 2 (1 − e−t / τ2)

(1)

where t is the time, A1 and A2 are pre-exponential constants (where A1 + A2 = 1), and τ1 and τ2 are the characteristic time constants of particle formation. At lower NaCl concentrations, such as 0 mM NaCl (see inset of Figure 1a), the increase in the scattering intensity was too rapid (due to the significant chitosan/TPP association occurring during mixing) to obtain reliable double-exponential fits, and was therefore omitted from the analysis. When NaCl was raised to 75 mM, however, the aggregation process became much slower, and the stopped-flow light scattering data (where the I(t)/I∞ values ranged between 0 and 0.8) could be fitted to the double-exponential equation 3870

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

Figure 3. The effect of TPP:glucosamine ratio in 150 mM NaCl on (a,b) molecular binding, characterized by (a) raw and (b) integrated ITC data, and on (c,d) the resulting particle formation, as analyzed by DLS. The DLS analysis shows (c) the volume-weighted particle size distributions for the (black square) 0.03:1, (red circle) 0.06:1, and (green triangle) 0.09:1 TPP:glucosamine ratios, and (d) the variations in the derived light scattering intensity. The solid curves are guides to the eye, while the dashed lines indicate the transitions between the three binding/particle formation behaviors. The error bars indicate the standard deviations.

within 20 min of each injection. As shown in the integrated data, when the TPP:glucosamine ratio was low (below 0.03:1), the binding heat, after subtracting the heat of dilution, was near zero (see Figure 3b, region i), suggesting negligible chitosan/ TPP interaction. Above the 0.03:1 TPP:glucosamine ratio, an exothermic binding signal was detected (Figure 3b region ii). This exothermic signal increased steadily from ∼0 kJ/mol of added TPP to ∼2 kJ/mol of added TPP at the TPP:glucosamine ratio of ∼0.07:1, whereupon it increased sharply to >12 kJ/mol of added TPP at the 0.09:1 TPP:glucosamine ratio (Figure 3b, region iii). This suggested that the binding became stronger as more TPP was added (i.e., that the binding was cooperative), and was qualitatively consistent with the ITC results obtained using more-concentrated (0.40 wt %) TPP solutions.20 Yet, when the more-concentrated TPP solution was titrated into the sample cell, the binding signal at low TPP:glucosamine ratios was significantly stronger (e.g., ∼5 kJ/mol at < 0.03:1 TPP:glucosamine ratios),20 despite having the same overall chitosan and TPP compositions. Thus, although the heat flow returns to the baseline after each TPP addition (see Figure 3a), there can be kinetic artifacts in the ITC analysis of chitosan/TPP binding in the presence of added salt. These artifacts likely reflect the extremely slow binding at low TPP and high NaCl concentrations (e.g., when dilute, 0.1 wt % TPP is titrated in 150 mM NaCl), which makes some of the binding events too slow to significantly perturb the baseline rate at which heat is added to the sample cell. When the added TPP solution is more-concentrated, however, the binding is faster and (even at low TPP:glucosame ratios) produces a strong enthalpic signal.20 To relate the molecular binding to the formation of microand nanogels, TPP was titrated into chitosan solutions to

extremely sensitive to the NaCl and TPP concentrations. Additionally, these apparent kinetics are sensitive to the procedure by which the chitosan and TPP are mixed. When the parent TPP and chitosan solutions are mixed in a 1:1 volumetric ratio, instead of the 1:5 used in the above experiments, the micro- and nanogel formation tends to be slower (despite the same overall composition), especially under conditions where particle formation is fast (data not shown). This suggests that the quantitative scaling relationships illustrated in Figures 1b and 2b are specific to the mixing protocol and should not be interpreted as fundamental rate laws for chitosan/TPP binding. Despite their limitations, however, these trends clearly show that dramatically reduced micro- and nanogel formation rates, which could facilitate the experimental analysis of their self-assembly mechanism, can be achieved using high NaCl and low TPP concentrations. Micro- and Nanogel Formation at High NaCl and Low TPP Concentrations. We hypothesized that the slow ionotropic gelation kinetics achieved at high NaCl and low TPP concentrations can facilitate the elucidation of the microand nanogel formation mechanism. To this end, particles were prepared in 150 mM NaCl using TPP:glucosamine molar ratios below 0.1:1. The first step in this work was to probe the evolution in the molecular chitosan/TPP binding and microgel size distributions during the titration of the chitosan solution with TPP. The molecular binding was tested by ITC, by titrating 0.10 wt % TPP solution into 0.10 wt % chitosan solution (at 150 mM NaCl; with a 20-min equilibration time after each addition). Consistent with the previous work using higher TPP concentrations (where 0.40 wt % TPP solution was titrated),20 the raw data (shown in Figures 3a) reveals that the heat supplied to the sample cell mostly returns to the baseline 3871

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

Figure 4. DLS data comparing (a) volume-weighted size distributions at (black square) 0.03:1, (red circle) 0.06:1 and (green triangle) 0.09:1 TPP:glucosamine ratios, measured (i) 1 h, (ii) 1 d and (iii) 7 d after mixing; and (b) temporal evolutions in size between (red diamond) 1 h, (blue square) 1 d and (green circle) 7 d after mixing at (i) 0.03:1, (ii) 0.06:1 and (iii) 0.09:1 TPP:glucosamine ratios. The lines are guides to the eye.

prepare mixtures with varying TPP:glucosamine ratios, whereupon (after 1 h of equilibration) the samples were characterized by DLS. The volume-weighted particle size distributions obtained at 0.03, 0.06, and 0.09:1 TPP:glucosamine ratios (Figure 3c) suggest that each region in the ITC data corresponds to a different stage in the micro- and nanogel formation process. At the 0.03:1 TPP:glucosamine ratio (i.e., the end of region i), the multiple narrow modes size distribution reveals a single peak with a maximum around 10 nm. This peak matches the apparent size distribution obtained from the chitosan solution without TPP (see Supporting Information, Figure S1), and confirms that the microgels do not form in the absence of chitosan/TPP binding. At the 0.06:1 TPP:glucosamine ratio (which is within region ii, where the chitosan/TPP binding begins to occur), the apparent particle size increases and suggests the onset of microand nanogel formation. Here, the apparent size distribution becomes bimodal, where some of the particles are in the 20−50 nm range (shown by the first large peak), while others are much larger (hundreds of nanometers in diameters, as shown by the second, smaller peak). This coexistence of small nanoparticles with a few larger colloids was confirmed by STEM (discussed later). Because the autocorrelation functions (not shown) did not yield two clearly-defined relaxation modes, however, it is inconclusive whether the particle size distribution is truly bimodal; instead, the particles in region ii might have broad monomodal distributions36 that are dominated by small, 20−50 nm nanogels. Finally, at the 0.09:1 TPP:glucosamine ratio (in region iii, where the ITC binding signal becomes dramatically stronger), all of the 20−50 nm nanogels appear to grow (or aggregate) into larger microgels, hundreds of nanometers in diameter. This interpretation of the size distributions is supported by the evolution in the light scattering intensity (Figure 3d). Below the 0.03:1 TPP:glucosamine ratio (in region i), the addition of TPP has little impact on the light scattering intensity, which suggests limited molecular aggregation. In region ii, the slope in the scattering intensity curve begins to increase. This suggests the formation of progressively larger chitosan/TPP aggregates, and agrees with the size distributions in Figure 3c.37 Finally, in region iii, the slope of the curve becomes linear, which is consistent with increasing concentrations of particles at a roughly constant size (i.e., hundreds of nanometers in diameter, such as shown for the 0.09:1 TPP:glucosamine ratio in Figure 3c).37

There are two potential explanations for the sudden transition points in the aggregation behavior. The first is the existence of thermodynamically controlled critical aggregation concentrations, similar to those seen in surfactant mixtures.38 The second is that, as suggested by the very strong dependence of the ionotropic gelation rates on the TPP concentration (see Figure 2), the binding at lower TPP concentrations (and at high NaCl concentrations) is too slow to occur over the short equilibration times used in the above experiments. In the latter case, the sudden onset of chitosan/TPP binding and micro- and nanogel formation might reflect the highly nonlinear scaling of the ionotropic gelation rates with the TPP concentration (see Figure 2b). Thus, micro- and nanogels might ultimately form even at very low TPP concentrations (below the 0.03:1 TPP:glucosamine ratio). To determine which of these explanations is correct, the formation and structure of chitosan/TPP micro- and nanogels was tracked over longer time scales. Temporal Evolution in the Micro- and Nanogel Structure. To probe how the micro- and nanogel structure evolves over longer time scales, samples were prepared at various TPP:glucosamine ratios and characterized by DLS and STEM over one week. The DLS data was analyzed using the multiple narrow modes algorithm to obtain volume-weighted particle size distributions. The temporal changes in particle size (see Figure 4, which compares size distributions 1 h, 1 day, and 7 days after mixing) confirm that the micro- and nanogel size at low TPP and high NaCl concentrations continues to evolve over several days. Figure 4a shows how the size distributions of micro- and nanogels prepared at 0.03, 0.06, and 0.09:1 evolve as their equilibration time is varied between (i) 1 h, (ii) 1 day, and (iii) 7 days. It reveals that, although the formation and size distribution of micro- and nanogels at short time scales (e.g., after 1 h; see Figure 4ai) are very sensitive to the TPP concentration, all three TPP:glucosamine ratios (even at 0.03:1, where no binding could be detected by ITC) ultimately yield similarly sized particles (mostly >100 nm in diameter) when equilibrated for 7 days (Figure 4aiii). The temporal changes at each TPP:glucosamine ratio are further illustrated in Figure 4b, which shows the evolution in particle size in mixtures with (i) 0.03:1, (ii) 0.06:1 and (iii) 0.09:1 TPP:glucosamine ratios. At the 0.03:1 TPP:glucosamine ratio (Figure 4bi), the ionotropic gelation is very slow, with no particles forming after 1 h of equilibration (the peak at ∼10 nm reflects the hydrodynamic diameter of molecular chitosan). After a day, however, the chitosan chains are cross-linked into 3872

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

Figure 5. STEM images of particles prepared in 150 mM solution at a 0.06:1 TPP:glucosamine ratio taken (a) 1 h, (b) 1 day, and (c) 7 days after preparation.

0.03:1 TPP:glucosamine ratio). At the 0.09:1 TPP:glucosamine ratio (Figure 4biii), the ionotropic gelation is accelerated further yet, and the size distribution is composed entirely of larger secondary colloids, even after 1 h of equilibration. This faster aggregation at higher TPP:glucosamine ratios is consistent with the kinetic analysis performed by stoppedflow light scattering (see Figure 2), where the τ1-values diminished sharply with increasing TPP concentrations. Interestingly, as shown in Figures 4bii and 4biii, the secondary aggregate peaks become sharper and shift slightly to larger particle sizes (between the 1-day and 7-day time points). This suggests that the particles continue to aggregate even after the

particles with a broad (and possibly bimodal) size distribution, composed of many 20−50 nm nanogels and a few larger microgels, hundreds of nanometers in diameter. When these samples are equilibrated for a full week, the smaller nanogel peak disappears, while the larger microgel peak grows. This suggests that the small primary nanogels aggregate over time into larger secondary microgels. At the 0.06:1 TPP:glucosamine ratio (Figure 4bii), the ionotropic gelation process becomes faster, and yields an apparently-bimodal particle size distribution after only 1 h of equilibration. At longer time scales (e.g., ≥ 1 day), the primary nanogels once again aggregate into larger secondary microgels (but over a shorter time scale than at the 3873

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

which lead to irreversible coagulation each time that the primary nanoparticles collide.38,43,44 Conversely, the chitosan/ TPP particles are water-swollen lyophilic colloids, whose aggregation is mediated by the ionic bridging of their surfaces by TPP.20 Unlike their lyophobic counterparts, their collisions do not appear to result in coagulation unless they are bridged by TPP ions.20 This has recently been shown through the addition of high concentrations of monovalent salt (which was added to screen the electrostatic interactions and yield very rapid, diffusion-limited particle collisions) to preformed chitosan/TPP microgels, where, despite the frequent collisions, no rapid microgel coagulation occurred.20 Thus, although the ionotropic gelation process occurs through primary and secondary aggregation, the dominant interactions underlying micro- and nanogel aggregation (i.e., ionic bridging) are likely distinct from those governing the secondary aggregation of most other classes of colloids. Although the present study was performed at high NaCl concentrations, the formation mechanism shown here also likely extends to micro- and nanogels prepared at lower ionic strengths. Microscopy of chitosan/TPP particles prepared without added NaCl often reveals the presence of small (20− 30 nm) nanoparticles, similar to the primary nanoparticles observed in the present study.26,46−49 The existence of such nanoparticles at low ionic strengths is also evident in the bimodal size distributions measured by DLS (where these nanoparticles coexist with larger microgels, 100−1000 nm in diameter).2,20,50,51 To the best of our knowledge, however, these nanoparticles were never considered as the primary building blocks for the chitosan/TPP micro- and nanogels. Instead, the presence of these small particles was attributed to unassociated chitosan, particle dehydration (during electron microscopy sample preparation)47 or the self-assembly of the low-molecular-weight fraction of the parent chitosan.2 While both dehydration and chitosan molecular weight can influence particle size, the temporal evolutions in Figures 4 and 5 suggest these particles to be unaggregated primary nanoparticles, which ultimately assemble into the larger micro- and nanogels that typically dominate the final product. Further evidence of this secondary aggregation appears in TEM images of larger, submicrometer chitosan/TPP microgels (prepared at low ionic strengths), which appear to be composed of smaller nanoparticles that are 20−30 nm in diameter.3,27,48 These observations in low-ionic-strength mixtures suggest that the ionotropic gelation mechanism proposed herein likely applies to both high and low ionic strength dispersions.

primary nanoparticles are consumed. When the samples are equilibrated for three weeks, however, little additional aggregation occurs, suggesting that the particles are fully formed within 7 days. To confirm the DLS data and explore the particle morphology, micro- and nanogels formed at the 0.06:1 TPP:glucosamine ratio (in 150 mM NaCl) were imaged by STEM over time. Chitosan solutions without added TPP (which were imaged by STEM as a control) yielded ring-like or branched fractal structures, but virtually no dispersed particles (Supporting Information, Figure S2). The chitosan/TPP mixtures, however, produced numerous spherical particles that aggregated over time. In the early stages of micro- and nanogel formation (e.g., at 1 h after mixing), the dispersions were primarily composed of nanoparticles, tens of nanometers in diameter (see Figure 5a). Most of these colloids were dispersed as single particles, but (as suggested by the second, smaller peak in the DLS data; see Figure 4bii) there were also a few aggregates such as the doublets in Figures 5aii and 5aiii. When these mixtures were equilibrated for a whole day (see Figure 5b), most of the primary nanoparticles aggregated into larger secondary colloids that were mostly 100 − 200 nm in diameter, which was again consistent with the DLS data in Figure 4bii. These secondary colloids had irregular shapes and were composed of many nanoscale subunits (thus suggesting the microgels to be assembled from multiple primary nanoparticles). When the mixture was equilibrated further yet (for 7 d), additional aggregation took place and yielded even larger microgels with more-globular morphologies (see Figure 5c). This interpretation is supported by the DLS data (in Figures 4bii and 4biii) where the left side of the secondary microgel peak (which corresponds to smaller microgels) shrinks over time, while the right side (which corresponds to larger microgels) grows, and is consistent with the microgel aggregation reported by Gan et al. in the absence of added NaCl.3 Thus, the temporal evolution in particle size and morphology indicates that chitosan/TPP micro- and nanogels form through a two-step process: (1) primary nanoparticle formation, and (2) secondary aggregation of primary nanoparticles into larger micro- and nanogels, which can thereupon undergo further (higher-order) aggregation over longer time scales. Moreover, the eventual formation of microgels at the very low, 0.03:1 TPP:glucosamine ratio indicates that the sharp transitions in the short term binding and particle formation behavior (illustrated in Figure 3) have kinetic rather than thermodynamic origins, i.e., they result from the strong, nonlinear dependence of chitosan/TPP binding rates on the TPP concentration. Further Mechanistic Discussion. The ionotropic microand nanogel formation mechanism indicated by this study (i.e., primary and secondary aggregation) is consistent with the current view on the formation of colloidal particles from other constituents, such as from poorly soluble inorganic salts39−41 or polyelectrolyte complexes (formed between oppositely charged polymer species).28,42,43 The interactions underlying these aggregation events in chitosan/TPP mixtures, however, are likely different from those in most other colloidal systems. The formation and aggregation of the primary nanoparticles formed from polyelectrolyte complexes or inorganic salts are typically attributed to their lyophobicity; their associated ion pairs assemble into hydrophobic primary nanoparticles, which then aggregate into larger secondary colloids.39,44,45 Here, the aggregation results from strong van der Waals interactions,

■

CONCLUSIONS The kinetics of chitosan/TPP micro- and nanogel formation are very sensitive to monovalent salt (NaCl) and ionic crosslinker (TPP) concentrations. By using high NaCl and low TPP concentrations, chitosan/TPP micro- and nanogel formation rates can be diminished by several orders of magnitude (likely by reducing the rate of chitosan/TPP binding). This effect facilitates the mechanistic analysis of the ionotropic gelation process, which is otherwise too fast to be investigated by most experimental techniques. Using this approach, the mechanism of chitosan/TPP microgel formation was probed by DLS and STEM and was shown to occur in two stages: first, the formation of small primary nanoparticles; second, the aggregation of primary nanoparticles into larger, higher-order colloids. The salt-assisted approach used in this study could facilitate future analyses of ionotropic gelation processes. 3874

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

■

Article

(15) Peng, J.; Xing, X.; Wang, K.; Tan, W.; He, X.; Huang, S. Influence of anions on the formation and properties of chitosan−DNA nanoparticles. J. Nanosci. Nanotechnol. 2005, 5, 713−7. (16) Desai, K. G. H.; Park, H. J. Encapsulation of vitamin C in tripolyphosphate cross-linked chitosan microspheres by spray drying. J. Microencapsulation 2005, 22, 179−192. (17) Hu, B.; Pan, C.; Sun, Y.; Hou, Z.; Ye, H.; Zeng, X. Optimization of fabrication parameters to produce chitosan−tripolyphosphate nanoparticles for delivery of tea catechins. J. Agric. Food Chem. 2008, 56, 7451−7458. (18) Zhang, H.; Mardyani, S.; Chan, W. C. W.; Kumacheva, E. Design of biocompatible chitosan microgels for targeted pH-mediated intracellular release of cancer therapeutics. Biomacromolecules 2006, 7, 1568−1572. (19) Lopez-Leon, T.; Carvalho, E. L. S.; Seijo, B.; Ortega-Vinuesa, J. L.; Bastos-Gonzalez, D. Physicochemical characterization of chitosan nanoparticles: Electrokinetic and stability behavior. J. Colloid Interface Sci. 2005, 283, 344−351. (20) Huang, Y.; Lapitsky, Y. Monovalent salt enhances colloidal stability during the formation of chitosan/tripolyphosphate microgels. Langmuir 2011, 27, 10392−10399. (21) Gan, Q.; Wang, T. Chitosan nanoparticle as protein delivery carrier-systematic examination of fabrication conditions for efficient loading and release. Colloids Surf., B 2007, 59, 24−34. (22) Fan, W.; Yan, W.; Xu, Z.; Ni, H. Formation mechanism of monodisperse, low molecular weight chitosan nanoparticles by ionic gelation technique. Colloids Surf., B 2012, 90, 21−27. (23) Morris, G. A.; Castile, J.; Smith, A.; Adams, G. G.; Harding, S. E. The effect of prolonged storage at different temperatures on the particle size distribution of tripolyphosphate (TPP)−chitosan nanoparticles. Carbohydr. Polym. 2011, 84, 1430−1434. (24) Jonassen, H.; Kjoniksen, A.-L.; Hiorth, M. Effects of ionic strength on the size and compactness of chitosan nanoparticles. Colloid Polym. Sci. 2012, 290, 919−929. (25) Tsai, M. L.; Bai, S. W.; Chen, R. H. Cavitation effects versus stretch effects resulted in different size and polydispersity of ionotropic gelation chitosan−sodium tripolyphosphate nanoparticle. Carbohydr. Polym. 2008, 71, 448−457. (26) Li, J.; Huang, Q. Rheological properties of chitosan− tripolyphosphate complexes: From suspensions to microgels. Carbohydr. Polym. 2012, 87, 1670−1677. (27) Ur-Rehman, T.; Tavelin, S.; Grobner, G. Chitosan in situ gelation for improved drug loading and retention in poloxamer 407 gels. Int. J. Pharm. 2011, 409, 19−29. (28) Nasti, A.; Zaki, N. M.; de, L. P.; Ungphaiboon, S.; Sansongsak, P.; Rimoli, M. G.; Tirelli, N. Chitosan/TPP and chitosan/TPPhyaluronic acid nanoparticles: Systematic optimization of the preparative process and preliminary biological evaluation. Pharm. Res. 2009, 26, 1918−1930. (29) Rajaonarivony, M.; Vauthier, C.; Couarraze, G.; Puisieux, F.; Couvreur, P. Development of a new drug carrier made from alginate. J. Pharm. Sci. 1993, 82, 912−17. (30) Fang, Y.; Al-Assaf, S.; Phillips, G. O.; Nishinari, K.; Funami, T.; Williams, P. A.; Li, L. Multiple steps and critical behaviors of the binding of calcium to alginate. J. Phys. Chem. B 2007, 111, 2456−2462. (31) Kabanov, V. A.; Zezin, A. B. Soluble interpolymeric complexes as a new class of synthetic polyelectrolytes. Pure Appl. Chem. 1984, 56, 343−54. (32) Nobbmann, U.; Morfesis, A. Characterization of nanoparticles by light scattering. Mater. Res. Soc. Symp. Proc. 2008, 1074E, 1074-I1045. (33) Zhu, Z.; Armes, S. P.; Liu, S. pH-Induced micellization kinetics of ABC triblock copolymers measured by stopped-flow light scattering. Macromolecules 2005, 38, 9803−9812. (34) Dautzenberg, H.; Kriz, J. Response of polyelectrolyte complexes to subsequent addition of salts with different cations. Langmuir 2003, 19, 5204−5211. (35) Dautzenberg, H.; Rother, G. Response of polyelectrolyte complexes to subsequent addition of sodium chloride: Time-

ASSOCIATED CONTENT

S Supporting Information *

Pre-exponential and time constants fitted to the stopped-flow light scattering data; DLS analysis of TPP-free chitosan solutions; STEM images of TPP-free chitosan solutions. This information is available free of charge via the Internet at http:// pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to the National Science Foundation (CBET1133795) for supporting this work.

■

REFERENCES

(1) Calvo, P.; Remunan-Lopez, C.; Vila-Jato, J. L.; Alonso, M. J. Novel hydrophilic chitosan-polyethylene oxide nanoparticles as protein carriers. J. Appl. Polym. Sci. 1997, 63, 125−132. (2) Zhang, H.; Oh, M.; Allen, C.; Kumacheva, E. Monodisperse chitosan nanoparticles for mucosal drug delivery. Biomacromolecules 2004, 5, 2461−2468. (3) Gan, Q.; Wang, T.; Cochrane, C.; McCarron, P. Modulation of surface charge, particle size and morphological properties of chitosanTPP nanoparticles intended for gene delivery. Colloids Surf., B 2005, 44, 65−73. (4) Janes, K. A.; Alonso, M. J. Depolymerized chitosan nanoparticles for protein delivery: Preparation and characterization. J. Appl. Polym. Sci. 2003, 88, 2769−2776. (5) Csaba, N.; Koping-Hoggard, M.; Alonso, M. J. Ionically crosslinked chitosan/tripolyphosphate nanoparticles for oligonucleotide and plasmid DNA delivery. Int. J. Pharm. 2009, 382, 205−214. (6) Janes, K. A.; Calvo, P.; Alonso, M. J. Polysaccharide colloidal particles as delivery systems for macromolecules. Adv. Drug Delivery Rev. 2001, 47, 83−97. (7) Sogias, I. A.; Williams, A. C.; Khutoryanskiy, V. V. Why is chitosan mucoadhesive? Biomacromolecules 2008, 9, 1837−1842. (8) Takeuchi, H.; Yamamoto, H.; Kawashima, Y. Mucoadhesive nanoparticulate systems for peptide drug delivery. Adv. Drug Delivery Rev. 2001, 47, 39−54. (9) Colonna, C.; Conti, B.; Perugini, P.; Pavanetto, F.; Modena, T.; Dorati, R.; Genta, I. Chitosan glutamate nanoparticles for protein delivery: Development and effect on prolidase stability. J. Microencapsulation 2007, 24, 553−564. (10) Hou, Y.; Hu, J.; Park, H.; Lee, M. Chitosan-based nanoparticles as a sustained protein release carrier for tissue engineering applications. J. Biomed. Mater. Res., A 2012, 100A, 939−947. (11) Katas, H.; Alpar, H. O. Development and characterization of chitosan nanoparticles for siRNA delivery. J. Controlled Release 2006, 115, 216−225. (12) Pan, Y.; Li, Y.; Zhao, H.; Zheng, J.; Xu, H.; Wei, G.; Hao, J.; Cui, F. Bioadhesive polysaccharide in protein delivery system: chitosan nanoparticles improve the intestinal absorption of insulin in vivo. Int. J. Pharm. 2002, 249, 139−147. (13) Fernandez-Urrusuno, R.; Romani, D.; Calvo, P.; Vila-Jato, J. L.; Alonso, M. J. Development of a freeze-dried formulation of insulinloaded chitosan nanoparticles intended for nasal administration. STP Pharma Sci. 1999, 9, 429−436. (14) Vila, A.; Sanchez, A.; Janes, K.; Behrens, I.; Kissel, T.; Jato, J. L. V.; Alonso, M. J. Low molecular weight chitosan nanoparticles as new carriers for nasal vaccine delivery in mice. Eur. J. Pharm. Biopharm. 2004, 57, 123−131. 3875

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876

Biomacromolecules

Article

dependent static light scattering studies. Macromol. Chem. Phys. 2004, 205, 114−121. (36) Madani, H.; Kaler, E. W. Measurement of polydispersed colloidal suspensions with quasi-elastic light-scattering. Part. Part. Syst. Charact. 1991, 8, 259−266. (37) Schmitz, K. S. An Introduction to Dynamic Light Scattering by Macromolecules; Academic Press: Boston, 1990. (38) Hiemenz, P. C. Principles of Colloid and Surface Chemistry, 3rd ed.; Marcel Dekker, Inc.: New York, 1977. (39) Privman, V.; Goia, D. V.; Park, J.; Matijevic, E. Mechanism of formation of monodispersed colloids by aggregation of nanosize precursors. J. Colloid Interface Sci. 1999, 213, 36−45. (40) Lee, S.-H.; Her, Y.-S.; Matijevic, E. Preparation and growth mechanism of uniform colloidal copper oxide by the controlled double-jet precipitation. J. Colloid Interface Sci. 1997, 186, 193−202. (41) Matijevic, E. Nanosize precursors as building blocks for monodispersed colloids. Colloid J. 2007, 69, 29−38. (42) Tsuchida, E.; Kokufuta, E.; Dubin, P. L. Formation of polyelectrolyte complexes and their structures. J. Macromol. Sci., Pure Appl. Chem. 1994, A31, 1−15. (43) Mueller, M.; Kessler, B.; Richter, S. Preparation of monomodal polyelectrolyte complex nanoparticles of PDADMAC/poly(maleic acid-alt-α-methylstyrene) by consecutive centrifugation. Langmuir 2005, 21, 7044−7051. (44) Starchenko, V.; Mueller, M.; Lebovka, N. Growth of polyelectrolyte complex nanoparticles: Computer simulations and experiments. J. Phys. Chem. C 2008, 112, 8863−8869. (45) Shovsky, A.; Varga, I.; Makuska, R.; Claesson, P. M. Formation and stability of water-soluble, molecular polyelectrolyte complexes: Effects of charge density, mixing ratio, and polyelectrolyte concentration. Langmuir 2009, 25, 6113−21. (46) Xu, Y.; Du, Y. Effect of molecular structure of chitosan on protein delivery properties of chitosan nanoparticles. Int. J. Pharm. 2003, 250, 215−226. (47) Wu, Y.; Yang, W.; Wang, C.; Hu, J.; Fu, S. Chitosan nanoparticles as a novel delivery system for ammonium glycyrrhizinate. Int. J. Pharm. 2005, 295, 235−245. (48) Yoksan, R.; Jirawutthiwongchai, J.; Arpo, K. Encapsulation of ascorbyl palmitate in chitosan nanoparticles by oil-in-water emulsion and ionic gelation processes. Colloids Surf., B 2010, 76, 292−297. (49) Chen, F.; Zhang, Z.-R.; Huang, Y. Evaluation and modification of N-trimethyl chitosan chloride nanoparticles as protein carriers. Int. J. Pharm. 2007, 336, 166−173. (50) Vllasaliu, D.; Exposito-Harris, R.; Heras, A.; Casettari, L.; Garnett, M.; Illum, L.; Stolnik, S. Tight junction modulation by chitosan nanoparticles: Comparison with chitosan solution. Int. J. Pharm. 2010, 400, 183−193. (51) Dudhani, A. R.; Kosaraju, S. L. Bioadhesive chitosan nanoparticles: Preparation and characterization. Carbohydr. Polym. 2010, 81, 243−251.

3876

dx.doi.org/10.1021/bm3014236 | Biomacromolecules 2012, 13, 3868−3876