Visualizing the Interaction between Poly-l-lysine and Poly(acrylic acid

Visualizing the Interaction between Poly-L-lysine and Poly(acrylic acid). Microgels Using Microscopy Techniques: Effect of Electrostatics and. Peptide...
0 downloads 5 Views 1MB Size
Download PDF
5476

Langmuir 2006, 22, 5476-5484

Visualizing the Interaction between Poly-L-lysine and Poly(acrylic acid) Microgels Using Microscopy Techniques: Effect of Electrostatics and Peptide Size Helena Bysell* and Martin Malmsten Department of Pharmacy, Uppsala UniVersity, P.O. Box 580, SE-751 23 Uppsala, Sweden ReceiVed February 16, 2006. In Final Form: March 29, 2006 The interaction between lightly cross-linked poly(acrylic acid) (pAA) microgels (50-150 µm in diameter) and poly-L-lysine (pLys) was studied as a function of pH, ionic strength, peptide size, and concentration. The swelling response and distribution of polypeptides within microgel particles was monitored by micromanipulator-assisted light microscopy and confocal laser scanning microscopy, while binding isotherms of pLys in the microgels were determined spectrophotometrically. Conformational changes of pLys were investigated by circular dichroism. The molecular weight of pLys was found to influence the degree of peptide-induced microgel deswelling, largely due to limitation of peptides larger than the effective network mesh size to penetrate the entire gel. Large peptides were concentrated within a surface layer of the gel particles, and at low ionic strength this dense surface layer was shown to act as a largely steric barrier for further penetration of compounds into the gel core. Small peptides, however, distributed evenly throughout the microgel particles and were able to create large microgel volume reductions. The deswelling of microgels increased with decreasing pH, while the uptake of pLys was significantly reduced at low pH. The effect of ionic strength on the interactions of pLys and oppositely charged pAA microgels was moderate and only pronounced for deswelling of gels at high pH. A significant increase in the R-helix content of pLys interacting with the oppositely charged microgels was observed for high molecular weight peptides, and the extent of R-helix formation was as expected more pronounced at high pH, i.e., at high charge density of the microgels but reduced charge density of the peptides.

Introduction In the past decade the number of peptide and protein drugs on the market and in clinical trial has increased strongly. Such substances can be difficult to administer due to enzymatic and/or chemical degradation, poor bioavailability due to their large size, and change or loss of activity due to conformational changes and aggregation resulting from interaction with substances in the biological system and/or in the drug formulation. There is therefore a need to develop functional transporter systems for such substances. One interesting system, playing an important role in nature, e.g., in cellular secretion of hormones and other transmitter substances, is based on polyelectrolyte networks that are able to store large amounts of oppositely charged substances in a small volume and release them upon swelling.1 Analogously, microgels show a highly cooperative and fast volume transition in response to external stimuli, such as temperature, pH, excess electrolyte concentration, and specific metabolites,2-5 which makes them interesting for storing and administering protein/ polypeptide drugs in a protected environment, followed by triggered release at the site of action with an appropriate release profile.6,7 Complexation of polyelectrolytes with oppositely charged macroions, such as proteins and peptides, has been studied extensively both experimentally8,9 and theoretically.10-12 The * To whom correspondence should be addressed. (1) Rahamimoff, R.; Fernandez, J. M. Neuron 1997, 18, 17-27. (2) Hoare, T.; Pelton, R. Macromolecules 2004, 37, 2544-2550. (3) Bromberg, L.; Temchenko, M.; Hatton, T. A. Langmuir 2002, 18, 49444952. (4) Eichenbaum, G. M.; Kiser, P. F.; Simon, S. A.; Needham, D. Macromolecules 1998, 31, 5084-5093. (5) Miyata, T.; Uragami, T.; Nakamae, K. AdV. Drug DeliVery ReV. 2002, 54, 79-98. (6) Kiser, P. F.; Wilson, G.; Needham, D. Nature 1998, 394, 459-462. (7) Kiser, P. F.; Wilson, G.; Needham, D. J. Controlled Release 2000, 68, 9-22.

complexation behavior has been found to be highly dependent on the pH and ionic strength, as well as on the charge densities, the concentration, and the size of the interacting species.12-14 Generally, associative phase separation occurs when the polyelectrolyte to macromolecule charge ratio is 1, i.e., at stoichiometric charge conditions. In the case of chemically cross-linked microgels, however, the macroions bind electrostatically to oppositely charged groups in the polymer network, causing osmotic deswelling of the polymer network rather than macroscopic phase separation in the traditional sense. Such stoichiometric binding reactions were reported for cationic surfactants interacting with anionic macrogels15,16 and microgels17-19 and have also been investigated for positively charged polymers and proteins binding to anionic macrogels.20-23 For both surfactants and proteins, the complex formation is initiated at the surface (8) Park, J. M.; Muhoberac, B. B.; Dubin, P. L.; Xia, J. Macromolecules 1992, 25, 290-295. (9) Chen, W.; Walker, S.; Berg, J. C. Chem. Eng. Sci. 1992, 47, 1039-1045. (10) Carlsson, F.; Linse, P.; Malmsten, M. J. Phys. Chem. B 2001, 105, 90409049. (11) Skepo¨, M.; Linse, P. Macromolecules 2003, 36, 508-519. (12) Carlsson, F.; Linse, P.; Malmsten, M. J. Am. Chem. Soc 2003, 125, 31403149. (13) Jiang, J.; Prausnitz, J. M. J. Phys. Chem. B 1999, 103, 5560-5569. (14) Chen, W.; Berg, J. C. Chem. Eng. Sci 1993, 48, 1775-1784. (15) Hansson, P.; Schneider, S.; Lindman, B. J. Phys. Chem. B 2002, 106, 9777-9793. (16) Khokhlov, A. R.; Kramarenko, E. Y.; Makhaeva, E. E.; Starodubtzev, S. G. Macromolecules 1992, 25, 4779-4783. (17) Andersson, M.; Råsmark, P.-J.; Elvingson, C.; Hansson, P. Langmuir 2005, 21, 3773-3781. (18) Nilsson, P.; Hansson, P. J. Phys. Chem. B 2005, 109, 23843-23856. (19) Go¨ransson, A.; Hansson, P. J. Phys. Chem. B 2003, 107, 9203-9213. (20) Rogacheva, V. B.; Prevysh, V. A.; Zezin, A. B.; Kabanov, V. A. Polym. Sci. 1988, 30, 2262-2270. (21) Zezin, A.; Rogacheva, V.; Skobeleva, V.; Kabanov, V. Polym. AdV. Technol. 2002, 13, 919-925. (22) Kabanov, V. A.; Skobeleva, V. B.; Rogacheva, V. B.; Zezin, A. B. J. Phys. Chem. B 2004, 108, 1485-1490. (23) Karabanova, V. B.; Rogacheva, V. B.; Zezin, A. B.; Kabanov, V. A. Polym. Sci. 1995, 37, 1138-1143.

10.1021/la060452a CCC: $33.50 © 2006 American Chemical Society Published on Web 05/09/2006

Interaction between pLys and pAA Microgels

of the polymer network, creating a dense surface phase which propagates inward at the expense of the swollen core as long as there are sufficient interacting compounds present.22 However, in contrast to surfactants, which form self-assembly aggregates and hexagonal or cubic phases, depending on properties such as the chain length and type of surfactant used,15,24 no ordered structures have to our knowledge been reported for proteinpolyelectrolyte interacting complexes. In the present investigation, the deswelling response of poly(acrylic acid) (pAA) microgels when including an oppositely charged linear peptide, poly-L-lysine (pLys), was studied. The electrostatic interaction of oppositely charged proteins or peptides with lightly cross-linked polyelectrolyte microgels has not, to the best of our knowledge, been extensively studied regarding the effects of polypeptide size and electrostatics on the deswelling response, and these aspects will therefore be reported on, and correlated to the peptide distribution within the microgel particles. By focusing on microgel particles in the diameter range 50-150 µm, various parameters of interest, notably the deswelling response, the distribution of the polypeptide within the microgel particles, and the nature of the shell structures formed under some conditions, can be straightforwardly monitored by micromanipulator-assisted light microscopy and confocal laser scanning microscopy. Experimental Section Materials. N,N′-Methylenebisacrylamide and N,N,N′,N′-tetramethylethylenediamine (TEMED) were obtained from Sigma (Germany), ammonium persulfate and acrylic acid were obtained from Aldrich (Germany), and sorbitan monostearate (Span 60) was obtained from Carl ROTH (Germany). Poly-L-lysine of different molecular weights (pLys 1000, pLys 10000, pLys 28000, pLys 58000, pLys 84000, and pLys 170000) and poly(aspartic acid) (pAsp 11000) were obtained from Sigma-Aldrich (Germany) and used without further purification. Fluorescein isothiocyanate (FITC) and N,Ndimethylformamide were obtained from Acros Organics (Belgium), while Alexa 633 was purchased from Invitrogen (Eugene, OR). The bisinchoninic acid (BCA) assay kit was purchased from Pierce (Rockford, IL). All other chemicals used in the present investigation were of analytical grade. Purified Milli-Q water was used throughout. To control the pH, buffer solutions (5 mM) of sodium acetate/acetic acid, sodium phosphate monobasic/sodium phosphate dibasic, and sodium carbonate/sodium bicarbonate were used for pH 4.5, 7, and 9.5, respectively. Sodium chloride was added to obtain the appropriate ionic strength. Preparation of Microgels. Microgel particles of cross-linked poly(acrylic acid) were synthesized by inverse suspension polymerization using cyclohexane as the continuous phase and sorbitan monostearate (Span 60) as the nonionic polymeric surfactant, a combination of two methods described previously.19,25 In brief, 0.05 g of Span 60 was dissolved in 20 mL of cyclohexane. The solution was preheated to 45 °C and stirred at 1000 rpm in a nitrogen atmosphere. A reaction mixture of 2.6 g of acrylic acid, 0.1 g of N,N′-methylenebisacrylamide (cross-linking agent), 20 g of NaOH (2 M), 4 g of NaCl, and 60 µL of TEMED was then prepared. A 10 mL sample of the reaction mixture was mixed with 0.5 mL of 0.18 M ammonium persulfate solution and added to the preheated cyclohexane. The polymerization reaction was carried out at 65 °C in a nitrogen atmosphere to prevent quenching by oxygen. Gelation took place within a few minutes, and the reaction was stopped after 30 min by addition of 40 mL of methanol. Gel particles sedimented overnight and were repeatedly washed with methanol followed by equilibration with purified water. Gel particles were then sieved using a Retsch 5657 test sieve (Germany). Fractions below the mesh size of 300 µm were collected and stored in water. (24) Okuzaki, H.; Osada, Y. Macromolecules 1995, 28, 380-382. (25) Wang, G.; Li, M.; Chen, X. J. Appl. Polym. Sci. 1997, 65, 789-794.

Langmuir, Vol. 22, No. 12, 2006 5477

Figure 1. Degree of charge as a function of pH for pLys (9) and pAA ([) at ionic strengths of 20 mM (filled symbols) and 220 mM (unfilled symbols) calculated by the Poisson-Boltzmann cell model. Spherical, 50-150 µm sized gel particles were confirmed by viewing in an Olympus Bx-51 light microscope (Olympus, Japan) with 20× magnification. The dry mass of microgels in solution was determined from freeze-drying experiments to be 1.7 mg of gel/g of gel solution using a Flexidry µP freeze-dryer (Kinetics Thermal Systems, Stone Ridge, NY). Calculating the Degree of Charge. The degree of charge of pLys and pAA was theoretically estimated by using the PoissonBoltzmann cell model.15,26 Calculations were based on isolated polymer chains, and a cylindrical cell model describing a polymer chain as a cylindrical rod with uniform charge density centered in a cylindrical cell was used. These calculations were performed to get a simplified picture of how charge varies with pH for pLys and pAA. The reality of the situation is of course more complicated, and results from these calculations should therefore be seen as approximate, although comparisons with experimental data from the literature show similarities.27-29 The radius and length of the approximated cylindrical polymer chain were set to 2 and 2.5 Å15 for pAA and to 11 and 3.5 Å26 for pLys, assuming a random coil conformation for the latter. The radius of the cell was given from the volume per polyion charge in the fully swollen state18 for pAA microgels and from swelling ratios at different pH values. The radius of the cell for pLys was determined from the peptide monomer concentration, assuming very dilute conditions. The pKa values used were 4.6 and 10.5 for pAA and pLys, respectively (i.e., pKa for the isolated monomers). The concentration and distribution of ions at the surface of the cylinder was calculated by solving the PoissonBoltzmann equation using the computer program PBCell, developed by Prof. Bengt Jo¨nsson, Lund University.26 To determine how the charge degree varies with pH, an iterative procedure including the law of mass action and Poisson-Boltzmann equation was performed.26 Figure 1 shows the variation in the degree of charge with pH for pLys and pAA at ionic strengths of 20 and 220 mM, respectively. Deswelling of Microgels. The pLys-induced deswelling of microgel particles was monitored by micromanipulator-assisted light microscopy, using an Olympus Bx-51 light microscope (Olympus, Japan) equipped with a Narishige ONM-1 manipulator (Narishige, Japan) and an Olympus DP 50 digital camera (Olympus, Japan) with the software Olympus DP-soft (Olympus, Japan). Micropipets (10-20 µm in diameter) were pulled using a Narishige PC-10 puller and an MF-9 forger (Narishige, Japan). Gel particles were captured by suction with micropipets using an IM-5A injector (Narishige, Japan) and placed inside a 2 mm diameter flow pipet. Captured gel particles were flushed with pLys solution for 30 min using a Peristaltic pump P-1 (Pharmacia, Sweden) at a flow rate of 1.8 mL/min. Using this experimental setup, single gel particles may be exposed to a constant concentration of pLys without any concentration decrease due to uptake of pLys in the gels. Gel particles were photographed and their diameters measured with the software Olympus DP-soft. (26) Nilsson, S.; Zhang, W. Macromolecules 1990, 23, 5234-5239. (27) Cui, C.; Schwendeman, S. P. Macromolecules 2001, 34, 8426-8433. (28) Makowska, J.; Baginska, K.; Kasprzykowski, F.; Vila, J. A.; Jagielska, A.; Liwo, A.; Chmurzynski, L.; Scheraga, H. A. Biopolymers 2005, 80, 214-224. (29) Philippova, O. E.; Hourdet, D.; Audebert, R.; Khokhlov, A. R. Macromolecules 1997, 30, 8278-8285.

5478 Langmuir, Vol. 22, No. 12, 2006 The deswelling ratios are expressed as V/Vmax, where V is the final volume of a gel particle after exposure to pLys and Vmax is the volume of the gel particle in a reference state (taken to be pH 9.5, ionic strength 20 mM). Deswelling of single gel particles with incorporation of pLys of different concentrations (0.01, 0.1, 1, 10, and 100 mg/L) was studied with respect to pH (4.5, 7, and 9.5), ionic strength (20 and 220 mM), and pLys molecular weight (10000 and 170000). pLys 1000 was also studied at pH 7 and ionic strengths of 20 and 220 mM. Five gel particles (diameter 80-150 µm at high pH and low ionic strength) were studied at each condition, and the results are presented as the average and standard deviation of deswelling ratios V/Vmax. Uptake of pLys in Microgels. The uptake of pLys into microgel particles was studied by equilibrating gel solutions with different concentrations of pLys. Samples were left on a shaking board for 48 h, whereafter gel particles were separated from the pLys solution by centrifugation at 5000 rpm for 10 min. The pLys concentration in the supernatant was determined by complexation with bisinchoninic acid30 (duplicate measurements were performed). Absorbance readings were performed on a Labsystems Multiscan MCC/340 plate reader (Labsystems, Finland) at 540 nm. The uptake (mg of pLys/ mg of gel) of pLys in the microgel particles was calculated by comparing the concentration of pLys in the supernatant with that of a control pLys solution which had not been exposed to microgels. Three independent binding experiments were performed (duplicate samples for each concentration), and the results are presented as the average values and standard deviation of the equilibrium uptake. In addition, exemplifying binding isotherms from individual binding experiments are presented. Confocal Laser Scanning Microscopy. FITC-Labeling. pLys was labeled with FITC as described earlier.31 A 2 µg of FITC/mg of pLys concentration was used, and the reaction was carried out for 1 h. Unreacted FITC was separated by repeated size exclusion using PD-10 colons (GE Health Care, Sweden). The concentration of pLys was measured spectrophotometrically after complexation with bisinchoninic acid,30 while the FITC concentration was determined with a Heλiosγ v 4.60 spectrophotometer (Thermospectronic, United Kingdom) at 495 nm. Deswelling ratios of microgels were not affected by this degree of FITC-labeling at the labeling density used in the present investigation (the FITC/pLys molar ratio was less then unity) as confirmed by micromanipulatorassisted light microscopy (results not shown). Distribution. The distribution of FITC-pLys in microgel particles was analyzed with a confocal Leica DM IRE2 laser scanning microscope (Leica Microsystems, Germany) equipped with an Ar/ He laser using the software Leica TCS SL after equilibration of a 1:1 mixture of gel solution with pLys for 48 h. The excitation wavelength was 488 nm. Throughout, a pLys concentration of 600 mg/L was used to ensure a good signal-to-noise ratio and fast kinetics. Note that, due to the pH sensitivity of the fluorescent marker used, FITC confocal laser scanning microscopy (CLSM) is used only semiquantitatively and no quantitative comparisons between different pH conditions are made. As no qualitative effect of ionic strength was found regarding the distribution of pLys in microgels at any pH value investigated, and the pLys distribution was investigated for concentrations corresponding to maximal uptake in pAA gels, the effects of pLys deficiency on surface layer formation could be excluded. For conditions of surface layer formation, the average thickness of the layer was estimated from intensity profiles of 10 gel particles. Surface Layer Characterization. The surface layer formed by high molecular weight pLys in gel particles was further investigated regarding the steric and electrostatic properties of the layer. To study whether the surface layer creates a steric barrier, a sequential binding study was performed. A 1:1 mixture of microgel solution and unlabeled pLys 170000 was equilibrated for 24 h on a shaking board. (30) Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K.; Gartner, F. H.; Provenzano, M. D.; Fujimoto, E. K.; Goeke, N. M.; Olson, B. J.; Klenk, D. C. Anal. Biochem. 1985, 150, 76-85. (31) Lassen, B.; Malmsten, M. J. Colloid Interface Sci. 1996, 179, 470-477.

Bysell and Malmsten Gel particles were then separated by centrifugation and the supernatant removed and replaced by fresh buffer solution. The latter procedure was repeated three times followed by addition of FITC-labeled pLys of various molecular weights. These samples were then equilibrated for 24 h, and the ability of differently sized FITC-labeled pLys to penetrate the surface layer was analyzed by CLSM as described above. To investigate possible charge reversal of the gel surface after binding of high molecular weight pLys, an oppositely charged polypeptide, pAsp (MW 11000), was labeled with a different fluorescent marker, Alexa 633. In short, the succinimidyl ester fluorescent dye was added to a solution of pAsp and incubated overnight to allow complete reaction with primary amines. Unreacted Alexa 633 was separated by repeated size exclusion using PD-10 colons (GE Health Care, Sweden). Alexa 633-labeled pAsp was added to preloaded pLys 170000 microgels using the same method as described above. This was performed both for FITC-labeled pLys 170000 and for unlabeled pLys 170000 to ensure that the fluorescent markers did not influence the results. The binding of pAsp was analyzed by CLSM in sequential scanning mode to ensure that overbleeding between the two fluorescent markers did not occur. The charge reversal and sequential adsorption study was conducted at pH 7 and ionic strengths of 20 and 220 mM, again at a pLys concentration of 600 mg/L. Conformation of pLys. The R-helix content of pLys in buffer and in microgels was evaluated by circular dichroism using a Jasco J-810 spectropolarimeter (Jasco, Japan). Ten scans of each sample were collected in a quartz cuvette of 1 cm path length. Reference spectra for 100% R-helix and 100% random coil were obtained using pLys 58000 in 0.1 M HCl and 0.1 M NaOH, respectively, as previously described.32 The R-helix content was calculated using the CD signal recorded at 225 nm as reported earlier.33 To ensure the absence of other secondary conformations, the CD spectra were scanned between 210 and 250 nm. The R-helix content of pLys 10000 and pLys 170000 was analyzed using an initial peptide concentration of 100 mg/L to ensure that no free pLys was left in solution. Duplicate measurements were performed at 20 °C.

Results and Discussion Deswelling Experiments. Gel beads consisting of cross-linked pAA show pH-dependent swelling behavior due to ionization of carboxylic acid (-COOH) groups in the polymer network. The pH-dependent swelling is influenced by ionic strength, and a high electrolyte concentration reduces the extent of swelling. For example, when the pH was raised from 4.5 (∼pKa for an isolated acrylic acid group) to 7 for ionic strengths of 20 and 220 mM, a volume increase of 2 and 1.6 times, respectively, was observed for the microgel particles investigated in this study. The volume did not change notably when the pH was increased further, which is in agreement with the theoretical prediction of the degree of charge versus pH shown in Figure 1. When the ionic strength was decreased from 220 to 20 mM, a roughly 2-fold increase in volume was observed for all pH values investigated. A theoretical effective mesh size of the gel network was approximated to range between 50 and 80 Å on the basis of the gel composition and swelling degrees for the different experimental conditions used. Note, however, that these values were obtained assuming a perfectly homogeneous distribution of cross-links in the network, as well as complete consumption of the cross-linker in the gel synthesis. This estimate of mesh size also suffers from other limitations including polymer chains not being fully extended at maximum deswelling, entanglements, and so on, so the values should be taken only to provide an order of magnitude estimate. pLys is a weak electrolyte with fully protonated amino acid (-NH3) groups at low pH (pH 4.5) (Figure 1). At high pH (pH (32) Greenfield, N.; Fasman, G. D. Biochemistry 1969, 8, 4108-4116. (33) Sjo¨gren, H.; Ulvenlund, S. Biophys. Chem. 2005, 116, 11-21.

Interaction between pLys and pAA Microgels

Langmuir, Vol. 22, No. 12, 2006 5479

> 10-11) the peptide is weakly charged, with 7 than at pH 5. They also reported, different from our results, a dramatically increased partition of (45) Eckenrode, H. M.; Dai, H.-L. Langmuir 2004, 20, 9202-9209.

Interaction between pLys and pAA Microgels

Figure 4. (a) Confocal images (selected top to middle sections) and intensity profiles of single gel particles showing the distribution of pLys in microgels at pH 4.5 and an ionic strength of 220 mM for pLys 1000, pLys 10000, and pLys 170000. (b) Thickness of the surface layer created by pLys 10000 and pLys 170000 at pH 4.5 and ionic strengths of 20 mM (gray bars) and 220 mM (black bars).

charged solutes into oppositely charged gels upon a decrease in ionic strength. Distribution of pLys in Microgels. In Figure 4a, the distribution of pLys in pAA gel beads at pH 4.5 is shown for pLys 1000, 10000, and 170000. A striking feature of this figure is that both pLys 10000 and pLys 170000 were unable to penetrate the polymer network mesh at this low pH and were therefore concentrated only within the surface layer of gel beads, in comparison to higher pH values where pLys 10000 distributed evenly throughout the microgel particles (see Figure 5 and the discussion below). pLys 1000, on the other hand, distributed evenly in the microgel particles at all conditions investigated. This could be explained by the fact that, at pH 4.5, gel beads have a low degree of swelling due to the small fraction of ionized -COOH groups. The mesh size of such a network is therefore small, not allowing penetration of peptides with a molecular weight larger than 10000. It has been reported that the effective pore size distribution of spherical microgel particles is nonuniform

Langmuir, Vol. 22, No. 12, 2006 5481

throughout the particles, with a larger pore size in the surface decreasing toward the core.46 This could be one explanation for the ability of pLys to penetrate a limited distance of the surface of gel particles (cf. the finite width of the surface layer). Another possible explanation to the surface phase, or shell, formation is the fast and substantial contraction of the polymer network upon binding of oppositely charged pLys, resulting in a composite shell condensed enough not to allow further polypeptide diffusion toward the microgel center. Note the shell folding observed for pLys 170000 but not for pLys 10000, obvious from the top to middle sections of confocal images shown in Figure 4a. This could reflect osmotic pressure differences in the gel core due to differences in stress relaxation of the network for differently sized peptides. The distribution of pLys of different molecular weights in microgel particles at pH 7 is shown in Figure 5. Here, the molecular weight cutoff for free penetration of pLys throughout the microgel particles was increased to 28000, compared to 10000 at pH 4.5, indicating a less dense network and/or composite shell at the higher pH. This is expected, since at pH 7 the fraction of dissociated carboxyl groups is larger, causing swelling of the network. Again, an increased folding of the gel particle surface with pLys molecular weight was observed. The thickness of the surface layer was somewhat higher for pLys 170000 than for pLys of lower molecular weights, forming shells, an effect more obvious at low ionic strengths (Figure 5b). This may suggest a higher degree of connectivity in the surface phase and larger difficulties in stress relaxation in the case of pLys 170000, such that redistribution of pLys occurs during the deswelling process for the low molecular weight polypeptides but not for pLys 170000. The thicker surface layer for pLys 170000 may also be related to the higher helix content of this peptide when bound to the microgel particles (see below). At pH 9.5, finally, the distribution of pLys in pAA gel particles was identical to that at pH 7 for the conditions investigated (results not shown). Vasheghani-Farahani et al.43 found that the partition coefficient of PEG in NIPAM-based gels decreased with increasing molecular weight. Specifically, PEG of molecular weight 18500 was excluded from the gels investigated, whereas PEG 800 was completely absorbed in the gels. They also investigated the partitioning of different proteins into ionic gels at various pH values and reported that the partition was dependent not only on the net charge but also on hydrophobic effects and surface adsorption in some cases. Although indirect, these findings are in line with those of the present investigation. Furthermore, Bromberg et al.38 investigated the absorption of hydrophobic anticancer drugs in polyether-modified poly(acrylic acid) microgels and estimated the effective pore size of the network by inclusion and exclusion of differently sized proteins, a method previously adopted by Eichenbaum et al.36 Properties of the Surface Layer. The dense surface layer created from pLys 170000 was further investigated regarding whether this surface layer constitutes a steric and/or an electrostatic barrier. In doing so, a sequential binding study was first performed to investigate the steric barrier properties of the shell by monitoring the ability of differently sized pLys to penetrate the surface layer formed by preloading pAA microgels with pLys 170000 at pH 7 and ionic strengths of 20 and 220 mM. Results are shown as selected confocal images in Figure 6. At low ionic strength, pLys of all molecular weights investigated was unable to penetrate through the surface layer, indicating that the surface layer creates a steric and/or net positively charged electrostatic barrier at these conditions. At high ionic strength (Figure 6b), on the other hand, (46) Bradley, M.; Bruno, N.; Vincent, B. Langmuir 2005, 21, 2750-2753.

5482 Langmuir, Vol. 22, No. 12, 2006

Bysell and Malmsten

Figure 5. (a) Confocal images (middle sections) and intensity profiles showing the distribution of pLys 1000, 10000, 28000, 58000, 84000, and 170000 in microgel particles at pH 7 and an ionic strength of 220 mM. (b) Thickness of the surface layer created by pLys 28000, pLys 58000, pLys 84000, and pLys 170000 at pH 7 and ionic strengths of 20 mM (gray bars) and 220 mM (black bars).

Figure 6. Confocal images showing the binding of FITC-labeled pLys of various molecular weights to microgels preloaded with unlabeled pLys 170000 after 24 h of equilibration at pH 7 for (a) an ionic strength of 20 mM and (b) an ionic strength of 220 mM. (c) Deswelling ratios of microgel particles before and after interaction with pLys 170000 and finally after interaction with Lys 1000 (gray bars) and 10000 (black bars) at an ionic strength of 220 mM obtained from micromanipulator-assisted light microscopy. As shown, the deswelling ratios did not decrease further upon addition of lower molecular weight pLys even though these peptides were able to penetrate into the gel core as shown in (b).

the pLys distribution within the microgel particles was essentially the same as that in the absence of a preloaded surface layer (Figure 5a). The latter finding implies that the surface layer at this condition creates no additional barrier to penetration of

microgel beads for pLys 1000 and 10000. (It should be noted that in some gel particles (less than 10%) pLys 10000 was not able to penetrate into the gel core, indicating some heterogeneity between microgel particles, but overall the findings are as described above.) These findings are in line with strong electrostatic interactions between pLys and pAA at low ionic strength, resulting in a dense barrier excluding diffusion off additional pLys into the gel core. In addition, micromanipulatorassisted light microscopy was performed to determine whether the volume of preloaded pLys 170000 microgel particles was further decreased upon addition of pLys 1000 and pLys 10000 at high ionic strength. As can be seen in Figure 6c, this was not the case even though these peptides were able to penetrate the surface layer and interact with the polymer network in the gel core. The latter finding suggests that there are both free and bound pLys molecules in the composite microgels, and/or that the binding of the sequentially added low molecular weight pLys is unable to cause further deswelling due to the compact shell resisting such deswelling. A similar experiment, where an oppositely charged peptide was added to microgels preloaded with pLys 170000 as described above, was also conducted to investigate a possible charge reversal of the gel particle surface. In these experiments, either unlabeled or FITC-labeled pLys 170000 was first allowed to form the surface layer at pH 7 and ionic strengths of 20 and 220 mM, respectively, for at least 24 h. Binding of Alexa 633-labeled pAsp to the surface layer was then monitored, and the results are shown in Figure 7. As binding of the labeled pAsp did not occur in the absence of pLys preloading (results not shown), it may be taken to suggest a surface charge reversal and shells of a net positive charge. However, although nonelectrostatic binding mechanisms are unlikely due to the high water content of the gel particles and

Interaction between pLys and pAA Microgels

Figure 7. Confocal images and intensity profiles for Alexa 633labeled pAsp (black line) on microgels preadsorbed with FITClabeled pLys 170000 (gray line) at pH 7 and ionic strengths of (a) 20 mM and (b) 220 mM. Adsorption of negatively charged pAsp within the surface layer of gel particles preloaded with unlabeled pLys 170000 indicates a surface charge reversal in the shell formed for these particles. (pAsp binding to bare microgels not preexposed to pLys was nondetectable.) In contrast, positively charged pLys of the same molecular weight (10000) is able to penetrate this surface layer at equal conditions as can be seen in Figure 6b.

hence minor van der Waals and hydrophobic interactions, we have no direct evidence in the form of electrokinetic data to support this effect. Therefore, we cannot at this point with absolute certainty exclude the possibility of an essentially neutral shell after pLys binding. In contrast to pAsp, positively charged pLys of the same molecular weight (10000) was able to penetrate this surface layer at equal conditions (cf. Figure 6b), suggesting that the barrier to penetration of higher molecular weight pLys in Figure 6b to a substantial extent is given by steric interactions, although electrostatic interactions may still contribute as well. On the basis of the cutoff molecular weights and the estimated radius of gyration of pLys,39 we infer that the “pore size” of the shells is in the range