Poly(allylamine hydrochloride) Microcapsules

Biomacromolecules 2005, 6, 1495-1502

1495

Multilayer DNA/Poly(allylamine hydrochloride) Microcapsules: Assembly and Mechanical Properties Olga I. Vinogradova,*,†,‡ Olga V. Lebedeva,†,§ Krasimir Vasilev,† Haofei Gong,† Javier Garcia-Turiel,†,| and Byoung-Suhk Kim† Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany; Laboratory of Physical Chemistry of Modified Surfaces, Institute of Physical Chemistry, Russian Academy of Sciences, 31 Leninsky Prospect, 119991 Moscow, Russia; N. S. Enikolopov Institute of Synthetic Polymer Materials, Russian Academy of Sciences, 70 Profsoyuznaya Street, 117393 Moscow, Russia; and van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Nieuwe Achterracht 166, 1018 WV Amsterdam, The Netherlands Received November 24, 2004; Revised Manuscript Received January 14, 2005

We report the preparation, characterization, and mechanical properties of DNA/poly(allylamine hydrochloride) (PAH) multilayer microcapsules. The DNA/PAH multilayers were first constructed on a planar support to examine their layer-by-layer buildup. Surface plasmon resonance spectroscopy (SPR) showed a nonlinear growth of the assembly upon each bilayer deposited independently on a concentration of salt. A weak increase in the film thickness with the DNA concentration was, however, detected. A post-treatment of the multilayers in the salt solutions has shown a thinning of the film. The optimal conditions of the planar film growth were used to deposit the same multilayers on the surface of colloidal templates and to study their roughness and morphology with the atomic force microscope (AFM) imaging. When an outer layer is formed by DNA, we observe large domains of oriented parallel DNA loops, while an outer layer formed by PAH shows highly porous morphology. The dissolution of colloidal templates led to a formation of highly porous DNA/PAH microcapsules. We probe their mechanical properties by measuring force-deformation curves with the AFM-related setup. The experiment suggests that the DNA/PAH capsules are softer than capsules made from the flexible polyelectrolytes studied before. The softening is due to both higher permeability and smaller Young’s modulus of the shell material. The Young’s modulus of the DNA/PAH shells increases after post-treatment in salt solutions of relatively low concentration. I. Introduction In viruses and cells, DNA is organized in tightly packed structures.1 Much research has been carried out in order to get insight into the mechanism of condensation and aggregation of DNA.2,3 This can be induced by a variety of positive ions, which could interact electrostatically with the oppositely charged phosphate groups on the DNA backbone. DNA molecules form complexes with multivalent cations,4 linear polycations,5,6 cationic dendrimers,7-10 proteins,11,12 colloidal particles,13,14 and lipids.15,16 Most of the cationic agents forming these complexes and aggregates are potentially useful as DNA vectors in novel gene therapies. The past few years have seen a rapid development of effective nonviral drug delivery systems,17 and several DNA systems have been employed for gene delivery applications.18 In this context, DNA-containing multilayer films on colloidal particles19 or microcapsules filled with DNA20 might also be considered as promising * Corresponding author. E-mail: [email protected]. † Max Planck Institute for Polymer Research. ‡ Institute of Physical Chemistry, Russian Academy of Sciences. § N. S. Enikolopov Institute of Synthetic Polymer Materials, Russian Academy of Sciences. | University of Amsterdam.

gene therapy carriers. Another strategy might take advantage of using DNA-containing multilayer microshells or so-called “hollow” polyelectrolyte multilayer microcapsules. Herein, we report an approach to prepare DNA/synthetic polyelectrolyte multilayer microcapsules, by exploiting electrostatic interactions. To date, multilayer microcapsules were comprised of oppositely charged synthetic linear polyelectrolytes,21 polyelectrolyte/inorganic nanoparticles,22 polyelectrolyte/dye,23 and polyelectrolyte/dendrimer24-26 pairs. Multilayer microcapsules containing DNA as a building blocks represent a new system of special interest. In this paper we are primarily interested in the buildup and exploitation of ultrathin multilayer films composed of layers of alternating DNA and poly(allylamine) hydrochloride (PAH), both supported (planar and spherical) and freestanding. Additionally, we examine the mechanical properties of assembled microcapsules. We use the AFM approach in combination with confocal microscopy to measure force vs deformation curves and to scan the capsule shape at different stages of compression.27 We show that in contrast to microcapsules with the shells formed by flexible polyelectrolytes, the DNA-based multilayer microcapsules are highly permeable to water even in the short time scale of the AFM compression experiment. We demonstrate that the enhanced

10.1021/bm049254t CCC: $30.25 © 2005 American Chemical Society Published on Web 03/03/2005

1496

Biomacromolecules, Vol. 6, No. 3, 2005

Vinogradova et al.

permeability of the multilayer is not due to enhanced fragility observed before for aged capsules made from flexible linear polyelectrolytes28 or freshly prepared dendrimer-based capsules26 but rather due to highly porous morphology of DNA/ PAH multilayers. II. Experimental Section A. Materials. The double-stranded deoxyribonucleic acid sodium salt (DNA; salt-free, lyophilized), purchased from Sigma (Fluka) is a highly polymerized natural product originating from calf thymus in the form of a white fibrous substance containing less than 5% protein. A molecular weight (Mw) for this product is reported to be ca. 6.0 × 106 g/mol, which yields an average size per molecule of 9091 base pairs, equal to a contour length L of 3.09 µm. This sample was used without further purification. Here, we use DNA as a polyanion for a multilayer assembly, since its sugar-phosphate backbones are negative charged. The fluorescent dye fluorescein isothiocyanate (FITC) and shell-forming polycationic polyelectrolyte poly(allylamine hydrochloride) (PAH; Mw ∼ 70 000 g/mol) were purchased from Sigma-Aldrich Chemie GmbH. Sodium chloride (NaCl) was purchased from Riedel-de Hae¨n. All chemicals were of analytical purity or higher quality and were used without further purification. Suspensions of monodispersed weakly cross-linked melamine formaldehyde particles (MF particles) with a radius of 2.0 ( 0.1 µm were purchased from Microparticles GmbH (Berlin, Germany). Glass bottom dishes with optical quality surfaces were obtained from World Precision Instruments Inc. Glass spheres (radius 20 ( 1 µm) were purchased from Duke Sci. Co. The glass prism and LaSFN9 substrates for surface plasmon experiments were purchased from Hellma Optik. Water used for all experiments was purified by a commercial Milli-Q Gradient A10 system containing ion exchange and charcoal stages and had a resistivity of 18.2 MΩ/cm. B. Methods. 1. Surface Plasmon Resonance Spectroscopy. A home-built surface plasmon resonance spectrometer (SPR) was used for the thickness measurements of the multilayer films deposited on a planar substrate. The sample was attached to the base of a glass prism optically matched by index oil. The beam of a He-Ne Laser (632.8 nm) with transverse-magnetic polarization was reflected off the base of the prism while the prism was rotated, thus scanning the angle θex of the incident beam relative to the sample surface. Collection of the reflected light with a photodiode (PD) mounted on a second, goniometer stage allowed for angle-resolved reflectivity measurements (see ref 29 for more details). To prepare substrates for SPR measurements, first a gold layer with a thickness of 50 nm was thermally evaporated on high refractive index (n ) 1.8) LaSFN9 substrates. The gold film was then functionalized with a self-assembled monolayer of 3-mercaptopropionic acid (Aldrich) (3MPA), by immersing the substrates in a 0.03 mol/L solution of 3MPA for 24 h. Such a modified surface is negatively charged in water. These substrates were used for multilayer film buildup. In each deposition step enough time was given to reach the adsorption equilibrium and followed by a

Figure 1. Schematic of the capsule deformation experiment.

thoroughout rinsing step with Milli-Q water (for at least 20 min) in order to remove all excess of polyelectrolytes and salt. A Teflon flow cell, allowing injection of polyelectrolytes and rinsing, was used. This made it feasible for the in situ observations of the multilayer growth as well as the detection of the kinetics of deposition. The optical thickness of the samples was analyzed by “Winspall” software based on the transfer matrix evaluation method. The refractive index of n ) 1.54 was used both for PAH and DNA.30 2. Coated Colloids and Capsule Preparation. The positively charged MF particles (50 µL of 10 wt dispersion) as a template were incubated with 1 mL of the negatively charged DNA solution (0.1-0.5 mg/mL containing 0.5 mol/L NaCl, pH 6) at room temperature for 50 min, followed by three centrifugation/rinsing cycles, and finally dispersed in water. A 1 mL portion of a PAH solution (1 mg/mL containing 0.5 mol/L NaCl, pH 6) was then added to the particle dispersion. After 20-45 min given for adsorption, three centrifugation/wash cycles were performed (as above). The DNA and PAH adsorption steps were repeated four times each to build multilayers on the MF particles. Washing out excess polymer and salt was followed by each adsorption. The microcapsules were obtained by dissolving the MF template in HCl at pH 1.2-1.6 and washing with water three times as described before.31 3. Atomic Force and Confocal Microscopy. The AFM images of the coated MF particles were performed with the Dimension 3100 Scanning Probe Microscope (Nanoscope IV, Digital Instrumentals) by using a tapping mode with a Z-limit of 6.60 µm. The positively charged particles (bare and with PAH as an outer layer) easily adhere to the negatively charged glass surface from their aqueous suspension. To provide adhesion of negatively charged (DNA as an outer layer) particles, the surface of glass was coated by a polycation. With this method, the particles remain strongly attached to the glass after rinsing with water and consequent drying. Before measurements, the cantilever was carefully approached to the apex of the particle with the use of a CCD camera. The images were analyzed by Nanoscope 5.30r1 and FemtoScan. The experimental setup used for an AFM compression experiment (Figure 1) was described in details before.27,32,33

DNA/Poly(allylamine hydrochloride) Microcapsules

Briefly, load (force) vs deformation curves were measured with the Molecular Force Probe device (MFP) 1D (Asylum Co., Santa Barbara, CA), which has a nanopositioning sensor. This sensor can correct for the piezo ceramic hysteresis and the creep of the AFM piezotranslator. For the force measurements we used V-shaped cantilevers (Micromash, spring constants k ) 3.0 N/m). The spring constant of the cantilever was estimated from the resonance frequency calibration plot (cantilevers catalog, Micromash). Glass spheres were glued onto the apex of cantilevers with epoxy glue (UHU Plus). The capsule deformation experiment has been described before.32 Here we perform the dynamic measurements at intervals of piezotranslator speed from 0.2 to 20 µm/s. To take confocal images of compressed capsules during force measurements, we have performed some of them in so-called stepped regime (see ref 34 for more details). The intervals between consequent steps of the piesotranslator were 2 s. The result of the measurement represents the deflection vs the position of the piezo translator at a single approach. The load force F was determined from the cantilever deflection, F ) k∆. As before, we assume that zero separation is at the point of the first measurable force.35 Then the deformation is calculated as the difference between the position of the piezo translator and the cantilever deflection. The diameter of the capsule was determined optically with an accuracy of 0.2 µm and from the AFM load vs deformation curves (like in ref 35). The relative deformation of the capsule was than calculated as ) 1 - H/(2r0), where r0 is the radius of the undeformed capsule and H is the distance between glass surfaces.32,35 To get reliable results we have performed several series of force measurements. Every series included at least 10 experiments. Then the average of all force vs deformation curves was calculated. Confocal laser scanning microscopy images were taken with a commercial confocal microscope unit FV300 (Olympus) used in combination with an inverted fluorescence microscope Olympus IX70. A high-resolution (60 ×) bright (NA ) 1.45) immersion oil objective was used. High resolution and contrast of the confocal images was achieved by the use of the fluorescent dye FITC at a concentration of 10-6 mol/L. The excitation wavelength was 488 nm. The confocal images of capsules at different stage of deformation were made together with the force measurements performed at the stepper regime, namely during the intervals between consequent steps of piezotranslator motion. The images were taken at the equatorial plane (dashed line in Figure 1). In other words, they correspond to the distance ∼H/2 from the glass sphere. 4. Scanning Electron Microscopy. For SEM analysis, a drop of each sample solution was applied to a silicon wafer with sequential drying at room temperature for 2-3 h.28 Then measurements were performed using a Gemini Leo (Zeiss) 1530 instrument operating at a working distance of 2 mm and an acceleration voltage of 0.5 kV. Since the samples were not covered with a gold layer before inspection, this low acceleration voltage was applied in order to avoid charging of the sample. The images were recorded using an InLens detector.

Biomacromolecules, Vol. 6, No. 3, 2005 1497

Figure 2. Equilibrium optical thickness of a DNA/PAH film assembled from solutions of different DNA concentrations as a function of deposited layers. Odd layers are from PAH. Even layers are from DNA. The multilayer buildup was performed with the 0.5 mol/L NaCl solution. From top to bottom, the concentration of DNA solutions was 0.7, 0.5, and 0.3 mg/mL. The data for 0.1 mg/mL solutions (not shown) roughly coincide with the results obtained for 0.3 mg/mL solutions. Solid curve represents the best polynomial fit.

III. Results and Discussion A. Assembly and Properties of Supported DNA/PAH Multilayer Films. 1. Planar Support. The DNA/PAH multilayers were first constructed on a planar support in order to characterize their layer-by-layer growth. The effect of both salt and DNA concentration on the film formation was studied. Figure 2 shows the equilibrium optical thickness of the film as a function of the number of the deposition cycles. A variable parameter is the concentration of the DNA solutions. The majority of previous experiments have shown the regular linear growth of the film.36 A clear nonlinearity is, however, seen for DNA/PAH combination. The curve appears linear only at higher layer numbers. This might be attributed to an insufficiently charged bare surface.37 One can also suggest that the first two layers of DNA lay flat on the surface, maximizing contact with the substrate,38 while the next layers are prone to make (vertical) loops and tails (see below). The data presented in Figure 2 suggest a weak increase in thickness with the concentration of DNA in the interval chosen. Higher concentrations of DNA were not explored due to a very high viscosity of more concentrated DNA solutions. Since the use of a 0.5 mg/mL solution provides nearly maximal film thickness and since this is a low viscosity solution, we have chosen this concentration of DNA for further studies. Thus, all results reported below correspond to a DNA concentration of 0.5 mg/mL. Figure 3 shows the results of investigation of the effect of salt (in the DNA solution) on the equilibrium thickness of the assembled multilayers. In contrast to all previous systems studied before,36,39,40 here we found that the thickness of the DNA/PAH multilayer films immersed in water does not depend on the amount of salt in the original DNA solutions. Below, for a multilayer and capsule assembly we use 0.5 mol/L NaCl solutions, as commonly accepted for other polyelectrolyte solutions.

1498


Figure 3. Equilibrium optical thickness of a DNA/PAH film assembled from solutions of different NaCl concentrations as a function of deposited layers. Odd layers are from PAH. Even layers are from DNA. The multilayer buildup was performed with 0.5 mg/mL DNA solution. The concentration of NaCl solutions was 0.05 mol/L (1), 0.1 mol/L (2), 0.5 mol/L (3), and 1.0 mol/L (4). Solid curve represents the best polynomial fit.

Figure 4. Schematic picture of the adsorbed DNA layer when the persistence length is larger than the layer thickness, λp > δ. The deflection length λ scales as λ ∼ δ2/3λp1/3.

In the DNA and NaCl concentration conditions chosen, the bilayer thickness in a quasi-linear regime is found to be approximately equal to ∼8-10 nm. Here, the thickness of a DNA monolayer δ is ∼6-8 nm and of a PAH monolayer is ∼2-3 nm. Thus, in contrast to data for poly(styrene sulfonate) PSS/PAH self-assembled films (where the steps of growth for even and odd layers were roughly the same41), here the step of growth for DNA is much bigger than the one for PAH. The measurements of thickness of DNA/PAH multilayers were made before,42,43 but the monolayer thicknesses were found to be smaller than we observe here. This is likely due to a use of more diluted solutions of DNA and NaCl.42,43 Another reason could be that in refs 42 and 43 the thickness of dried films was studied, but we are measuring the multilayers immersed in water. The cross section of the DNA double helix is ∼2 nm,44 so the measured thickness of the DNA layer in the film corresponds to 3-4 diameters of the DNA, but it is much smaller than the persistence length of DNA (λp ∼ 50 nm). So, the measured thickness δ of a DNA monolayer almost likely indicates the formation of surface loops and tails (Figure 4). We note that L . λp, so there is intramolecular loop formation (ring closure) in the bulk DNA solutions. However, the radius of bulk DNA coils RDNA ∼ λp1/5κ-1/5L3/5,45 where κ-1, the Debye screening length, is much larger than δ. Indeed, the measurements of δ were performed in pure water, so that κ-1 is roughly equal to 300-400 nm.46 This gives RDNA ∼ 800 nm and suggests unfolding/rearrangement of bulk DNA coils during/after adsorption. The surface loops likely do not have intramolecular intersections. The distance between two closest contacts of DNA molecules with the

Vinogradova et al.

Figure 5. The thickness of four DNA/PAH bilayers in NaCl solutions. The solid curve is a guide to the eye.

support, the so-called deflection length λ, can be evaluated as λ ∼ δ2/3λp1/3,47 which gives ∼15 nm. It has to be also noted that the kinetic data (not shown) indicate some peculiarities in adsorption of PAH at a DNAbased film. In contrast to previous data on adsorption of PAH at a PSS/PAH film, where the indicating adsorption abruptly increase in reflectivity followed by a horizontal plateau, here the true plateau regime was established often after 30 min or more, although the main jump in reflectivity was as fast as before. This might indicate that the adsorbed layers of PAH cause some slow conformational changes in the preceding DNA layer. Figure 5 shows the change in thickness of four bilayer DNA/PAH films after contact with salt-containing solutions. A decrease in the multilayer thickness (roughly from 6.5 nm down to 4 nm for one bilayer) was observed until a concentration of ∼2 mol/L. A more significant, abrupt decrease in thickness in more concentrated NaCl solutions indicates a dissociation of the multilayers. The decrease in thickness with the added salt at relatively low concentration is an entirely unexpected new observation. Indeed, the previous studies performed with flexible linear polyelectrolytes (PSS/PAH) of high molecular weight found a moderate swelling of the multilayers in salt solutions,48,49 which was entirely consistent with the theoretical predictions. There have been some observations of loss of polyelectrolytes due to their desorption at low salt, but this concerned only the low molecular weight fraction (Mw below 104 g/mol)50 and was not confirmed for higher molecular weight polyelectrolytes (Mw above 105 g/mol).51 Since the molecular weight of our DNA is even higher, and since the expected polydispersity is (20%, we should rule out any hypothesis of the loss of DNA at low NaCl concentration. A primitive theory52 predicts that for a semiflexible polyelectrolyte, such as DNA, no changes in the film thickness should be observed. So, the reasons for salt thinning observed here are not clear yet. One can speculate that this might be connected with the decrease in the DNA persistence length caused by salt.53 One can also surmise that it could be due to some decrease in the DNA



Figure 7. A typical aggregate of DNA/PAH capsules.

Figure 6. Typical AFM images of bare (top) and coated MF spheres: (middle) (DNA/PAH) DNA multilayer, where the last layer is DNA; (bottom) (DNA/PAH)2 multilayer, where the last layer is PAH.

contour length caused by salt. We have, however, not found any literature data that could support or disprove this hypothesis. 2. Colloidal Templates. The typical images of bare MF sphere and MF spheres coated by DNA/PAH multilayers are shown in Figure 6. These particular examples correspond to (DNA/PAH) DNA and (DNA/PAH)2 multilayers. The rootmean-square (rms) roughness of bare sphere surface was found to be about 2 nm, and the maximum peak-to-valley height difference was less than 10 nm. Therefore, although the spheres (Figure 6, top) might be treated as rough, they are smoother than other (polystyrene) latex particles studied before.54,55 The consequent adsorption of the first DNA/PAH bilayer, corresponding to the nonlinear regime of film growth, leads to the roughening of the surfaces. Thus, the rms was roughly 2 nm for a first deposited bilayer and 4 nm for a third bilayers. Both the rms value and the surface morphology were different depending on which polyelectrolyte (DNA or PAH) formed the outer layer. An outer DNA layer (Figure

6, middle) roughened the surface, which contained large domains of parallel oriented elongated “hills”. This likely indicates the formation of ordered DNA loops and is consistent with the SPR data. Similar ordering, a lamellar phase, was observed before for DNA molecules condensed on cationic lipid membranes56 and was predicted theoretically for semiflexible polyelectrolytes adsorbed on a charged surface.52 The analysis of images has given the lamellar spacing of the order of persistence length λp of DNA. In contrast, in the case of an outer PAH layer, the morphology looks porous, indicating that PAH adsorbs into the “valleys” between “hills”. The diameter of the pores is on the order of 10-20 nm. We remark that the deposition of PAH reduced the rms value as compared with the previous DNA layer. A slight decrease in roughness (smoother morphology) for further deposited layers (corresponding to a linear regime of the film growth) was observed. The rms value was about 3 nm. Moreover, no discernible difference in rms value between PAH and DNA layers was detected within the accuracy of experiment. B. Multilayer Microcapsules. Since some special kinetic behavior was observed for a formation of supported multilayers, we have assembled two types of capsules. In the first case, the adsorption time for PAH was 20 min, as before for different systems.26,31 In the second case, we allowed 45 min for adsorption. The confocal images of DNA/PAH microcapsules are presented in Figure 7. One can see that they show unusual aggregation properties in contrast to all capsule systems studied before. The reasons for this aggregation are not entirely clear. It is likely that this is caused by the fact that the DNA molecules are too long, so that one molecule could cover several template particles during the preparation. One can, therefore, suggest that some DNA/PAH capsules are connected by elastic tails formed by DNA molecules. We have also taken and analyzed SEM images of dried capsules (see Figure 8). It is seen that they collapse upon drying by forming structures with folds and creases. The morphology of the surfaces of dried DNA/PAH capsules seems to be very different from capsules studied before.26,28 SEM images of the surfaces of DNA/PAH capsules clearly show a highly porous structure. The area 600 × 600 nm2 contained approximately 25 nanopores of an average diameter of roughly 8 nm. So, we observe many more nanopores and of much larger size as for PSS/PAH capsules.28 The

1500


Vinogradova et al.

Figure 10. Typical force vs relative deformation, , profiles measured at low compression (symbols). Open symbols correspond to the first type of capsule, and filled symbols represent the results for the second type of capsule. The dashed curve shows the force vs deformation profile expected for impermeable capsules (eq 1) with E ) 100 MPa, h ) 27 nm, and r0 ) 2 µm. The solid curve shows the profile expected for permeable capsules (eq 2) with the same Young’s modulus, shell thickness, and radius.

Figure 8. SEM images of DNA/PAH microcapsules. Scale bar corresponds to 1 µm (top) and 200 nm (bottom)

the results obtained for one sample. By analyzing these profiles and confocal images (Figure 9), we have concluded that the deformation curves indicate high permeability. It is also seen that DNA/PAH capsules are much softer (smaller force at the same relative deformation, ) than PSS/PAH capsules studied before.32,57 To examine the significance of permeability of compressed capsules more closely and to evaluate Young’s modulus of the DNA/PAH multilayers, the small deformation profiles for two types of capsules are given in Figure 10. If we can neglect water drainage through the shell, the dependence of force on relative deformation (for rs . r0 and ν ) 1/2) is given by32 F∼

Figure 9. The average force vs relative deformation, , curve for DNA/PAH capsules of the first (circles) and second (squares) types. Insets show confocal images (equatorial cross section) of deformed capsules at given . From left to right ) 0.1, 0.4, 0.6, 0.7, and 0.9.

results of the SEM image analysis of dried capsules are consistent with the data from AFM imaging of supported multilayers. This likely indicates that the HCl solutions used to dissolve the MF templates do not modify the multilayer structure. No discernible difference in the aggregation properties or in surface morphology between the capsules of the first and the second type was detected. C. Force vs Deformation Profiles. Figure 9 shows load vs deformation profiles typical for capsules made according to procedure one (shorter adsorption time) and two (longer adsorption time). The difference between the results obtained for the two types of capsules is within the variability between

π Eh21/2 + 4πEhr03 2x2

(1)

where E is Young’s modulus of the multilayer and h is its thickness. According to our SPR data, the value of h (for a four-bilayer shell) should be taken as 27 nm. An alternative model, which assumes a rapid (compared to the time scale of the AFM compression experiment) drainage of the inner solution can be obtained if we simply omit the second, stretching, term in the model of impermeable capsules:27 F∼

π Eh21/2 x 2 2

(2)

The data for DNA/PAH capsules at e 0.1 are well-fitted to eq 2 with E ) 100 MPa. This value is two times smaller than Young’s modulus of a PSS/PAH multilayer.32,57,58 A theoretical curve calculated with eq 1 and E ) 100 MPa is included in Figure 10. One can see that this model, previously used for analysis of impermeable (at the time scale of AFM compression experiment) capsules, is not suitable for a description of DNA/PAH microcapsules. We note, however, that the value of Young’s modulus of DNA/PSS multilayers



the multilayer is at the maximum of its stiffness (E ∼ 170 MPa) at 0.7-1 mol/L NaCl solutions. The consequent softening likely indicates that, at the range of concentrations from 1 to 2 mol/L, the multilayer (“tethered”48) structure is still retained, but the number of broken ionic cross-links starts to increase. The higher concentration of salt corresponds to a “melted” state of the multilayer. We remark that, for PSS/ PAH capsules, the transition to a “melted” state was observed at concentrations above 3 mol/L.48 Therefore, lower salt concentration is required to dissociate/deconstruct DNA/PAH multilayer shells. Conclusions

Figure 11. Force at fixed relative deformation, ) 0.1, as a function of salt concentration.

should still be treated as very large. It is confined between the values typical for a highly cross-linked stiff rubber (∼10 MPa) and soft plasticized glass (∼1000 MPa). Since chemical cross-links are absent in the multilayer structure, such a value of E is due to physical, almost likely ionic, crosslinks.27,59 So, the multilayer structure can be treated as “tethered”.48 One can suggest that the drainage of inner water under AFM compression is accelerated due to a great amount of relatively large nanopores in the shell of DNA/PAH capsules. Indeed, as a rough estimate one can assume that water flow through the multilayer shell is πF4∆P 8µh

Q∝N

where N is the number of nanopores, F is their radius, µ is the dynamic viscosity of water, and ∆P is the excess pressure inside the capsule. The latter can be estimated as a force divided by the area of contact, which for ∼ 0.1 would be approximately F/(2πr02) ∼ 104 Pa. With our experimental parameters (Figure 8) for a capsule of r0 ) 2µm, the number of nanopores N ∼ π × 200. From here and assuming F ) 8 nm, we get Q ∼ 10-16 m3/s. Since with our driving speeds it takes 0.2 s to reach ∼ 0.1, the loss of volume due to drainage can be estimated as ∆V ∼ 10-17 m3, which roughly corresponds to a deformation at constant radius of a free area of the capsule. This explains high permeability of DNA/ PAH capsules observed in a force experiment. We leave the matter by discussing the influence of salt post-treatment on the mechanical properties of DNA/PAH capsules. The average force at a fixed relative deformation ) 0.1 measured at different NaCl concentration is shown in Figure 11. With the added salt the capsules first get slightly stiffer and start to soften at concentrations higher than 1 mol/ L. After a concentration of 2 mol/L is reached, the stiffness of the capsules becomes quasiconstant. The stiffening of the capsules at low salt (especially in combination with the decrease in the shell thickness, see Figure 5) indicates the formation of a more compact multilayer structure. The estimates of Young’s modulus made with eq 2 suggests that

We have demonstrated the successful DNA/PAH multilayer film formation on planar substrate as well as on colloid particles and have studied the morphology of these films. The effects of salt and concentration of DNA on deposition, as well as influence of salt postmodification, were explored in detail. We also have assembled the DNA/PAH multilayer microcapsules and studied their mechanical properties in water and water/electrolyte solutions. Both the conditions of multilayer growth and the capsule properties were found to be dramatically different from what was previously observed in the case of flexible polyelectrolytes as building blocks for multilayer formation. This suggested that the unique properties of DNA-based supported and free-standing multilayers reflect the semiflexible character of the DNA molecules. Acknowledgment. H.G. and B.S.K. acknowledge the receipt of an Alexander von Humboldt fellowship. This work was partly funded by the Russian Academy of Sciences within the priority program “Macromolecules and macromolecular structures of new generations”. We thank G. Glasser for taking SEM images. We have benefited from valuable discussions with V. Ball, M. Deserno, I. Kulic, V. V. Lulevich, R. R. Netz, and H. Schiessel. References and Notes (1) Alberts, B.; Bray, D.; Lewis, J.; Raff, M.; Roberts, K.; Watson, J. D. Molecular Biology of the Cell; Garland: New York, 1994. (2) Bloomfield, V. A. Biopolymers 1991, 31, 1471. (3) Bloomfield, V. A. Curr. Opin. Struct. Biol. 1996, 6, 334. (4) Zinchenko, A. A.; Sergeyev, V. G.; Yamabe, K.; Murata, S.; Yoshikava, K. Chem. Bio. Chem. 2004, 5, 360. (5) Howard, K. A.; Dash, P. R.; Read, M. L.; Ward, K.; Tomkins, L. M.; Nazarova, O.; Ulbrich, K.; Seymour, L. W. Biochim. Biophys. Acta 2000, 1475, 245. (6) Gebhart, C. L.; Kabanov, A. V. J. Bio. Compat. Pol. 2003, 18, 147. (7) Loup, C.; Zanta, M. A.; Caminade, A. M.; Majoral, J. P.; Meunier, B. Chem. Eur. J. 1999, 5, 3644. (8) Go¨ssl, I.; Shu, L.; Schlu¨ter, D.; Rabe, J. P. J. Am. Chem. Soc. 2002, 124, 6860. (9) Luo, D.; Haverstick, K.; Belcheva, N.; Han, E.; Saltzman, W. M. Macromolecules 2002, 35, 3456. (10) Chen, W.; Turro, N. J.; Tomalia, D. A. Langmuir 2000, 16, 15. (11) Schiessel, H. J. Phys.: Condens. Matter 2003, 15, R699. (12) Nadassy, K.; Wodak, S. J.; Janin, J. Biochemistry 1999, 38, 1999. (13) Kneuer, C. Biocojugate Chem. 2000, 11, 926. (14) Jeon, S.; Granick, S. Colloids Surf. A 2004, 238, 109. (15) Ra¨dler, J.; Koltover, I.; Salditt, T.; Safinya, C. R. Science 1997, 275, 810. (16) Koltover, I.; Salditt, T.; Ra¨dler, J.; Safinya, C. R. Science 1998, 281, 78. (17) Langer, R. Science 2001, 293, 58.

1502


(18) Lambert, G.; Fattal, E.; Couvreur, P. AdV. Drug. DeliVery ReV. 2001, 47, 99. (19) Dudnik, V.; Sukhorukov, G. B.; Radtchenko, I. L.; Mo¨hwald, H. Macromolecules 2001, 34, 2329. (20) Shchukin, D. G.; Patel, A. A.; Sukhorukov, G. B.; Lvov, Y. M. J. Am. Chem. Soc. 2004, 126, 3374. (21) Donath, E.; Sukhorukov, G. B.; Caruso, F.; Davis, S. A.; Mo¨hwald, H. Angew. Chem.-Int. Edit. 1998, 37, 2202. (22) Caruso, F.; Caruso, R. A.; Mo¨hwald, H. Science 1998, 282, 1111. (23) Dai, Z.; Voight, A.; Leporatti, S.; Donath, E.; Da¨hne, L.; Mo¨hwald, H. AdV. Mater. 2001, 13, 1339. (24) Khopade, A.; Caruso, F. Biomacromolecules 2002, 3, 1154. (25) Khopade, A.; Caruso, F. Nano Lett. 2002, 2, 415. (26) Kim, B. S.; Lebedeva, O. V.; Kim, D. H.; Caminade, A. M.; Majoral, J. P.; Knoll, W.; Vinogradova, O. I. Langmuir, submitted. (27) Lulevich, V. V.; Vinogradova, O. I. Langmuir 2004, 20, 2874. (28) Lulevich, V. V.; Nordschild, S.; Vinogradova, O. I. Macromolecules 2004, 37, 7736. (29) Knoll, W. Annu. ReV. Phys. Chem. 1998, 49, 569. (30) Vasilev, K.; Kreiter, M.; Knoll, W. J. Chem. Phys. 2004, 120, 3439. (31) Sukhorukov, G. B.; Donath, E.; Lichtenfeld, H.; Knippel, E.; Knippel, M.; Budde, A.; Mo¨hwald, H. Colloids Surf. A 1998, 137, 253. (32) Lulevich, V. V.; Andrienko, D.; Vinogradova, O. I. J. Chem. Phys. 2004, 120, 3822. (33) Lulevich, V. V.; Radtchenko, I. L.; Sukhorukov, G. B.; Vinogradova, O. I. Macromolecules 2003, 36, 2832. (34) Lebedeva, O. V.; Kim, B. S.; Vinogradova, O. I. Langmuir 2004, 20, 10685. (35) Lulevich, V. V.; Radtchenko, I. L.; Sukhorukov, G. B.; Vinogradova, O. I. J. Phys. Chem. B 2003, 107, 2735. (36) Bertrand, P.; Jonas, A.; Laschewsky, A.; Legras, R. Macromol. Rapid Commun. 2000, 21, 319. (37) Advincula, R.; Aust, E.; Meyer, W.; Knoll, W. Langmuir 1996, 12, 3536.

Vinogradova et al. (38) Fang, Y.; Hoh, J. H. J. Am. Chem. Soc. 1998, 120, 8903. (39) Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153. (40) Izumrudov, V.; Kharlampieva, E.; Sukhishvili, S. A. Macromolecules 2004, 37, 8400. (41) Lvov, Y.; Decher, G.; Mo¨hwald, H. Langmuir 1993, 9, 481. (42) Lvov, Y.; Decher, G.; Sukhorukov, G. Macromolecules 1993, 26, 5396. (43) Sukhorukov, G. B.; Mo¨hwald, H.; Decher, G.; Lvov, Y. M. Thin Solid Films 1996, 284-285, 220. (44) Lodish, H.; Darnell, J.; Baltimore, D. Molecular Cell Biology; Scientific American Books: New York, 1990. (45) Odijk, T.; Houwaart. A. C. J. Polym. Sci. 1978, 16, 627. (46) Yakubov, G. E.; Butt, H. J.; Vinogradova, O. I. J. Phys. Chem. B 2000, 104, 3407. (47) Odijk, T. Macromolecules 1983, 16, 1340. (48) Lebedeva, O. V.; Kim, B. S.; Vasilev, K.; Vinogradova, O. I. J. Colloid Interface Sci. 2005, 284, 455. (49) Sukhorukov, G. B.; Schmitt, J.; Decher, G. Ber. Bunsen-Ges. Phys. Chem. 1996, 100, 948. (50) Sui, Z.; Salloum, D.; Schlenoff, J. B. Langmuir 2003, 19, 2491. (51) Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626. (52) Netz, R. R.; Joanny, J. F. Macromolecules 1999, 32, 9013. (53) Barrat, J. L.; Joanny, J. F. AdV. Chem. Phys. 1996, 94, 1. (54) Vinogradova, O. I.; Yakubov, G. E.; Butt, H. J. J. Chem. Phys. 2001, 114, 8124. (55) Yakubov, G. E.; Vinogradova, O. I.; Butt, H. J. Colloid J. 2001, 63, 518. (56) Fang, Y.; Yang, J. J. Phys. Chem. 1997, 101, 441. (57) Vinogradova, O. I. J. Phys.: Condens. Matter 2004, 16, R1105. (58) Vinogradova, O. I.; Andrienko, D.; Lulevich, V. V.; Nordschild, S.; Sukhorukov, G. B. Macromolecules 2004, 37, 1113. (59) Kim, B. S.; Vinogradova, O. I. J. Phys. Chem. B 2004, 108, 8161.

BM049254T

Poly(allylamine hydrochloride) Microcapsules

Recommend Documents