Complexation of DNA with Poly (methacryl oxyethyl

Biomacromolecules 2004, 5, 1853-1861

1853

Complexation of DNA with Poly(methacryl oxyethyl trimethylammonium chloride) and Its Poly(oxyethylene) Grafted Analogue Toni Andersson, Vladimir Aseyev,† and Heikki Tenhu* Laboratory of Polymer Chemistry, University of Helsinki, PB 55, FIN-00014 Helsinki, Finland Received April 5, 2004; Revised Manuscript Received May 25, 2004

Intermolecular complexes of genomic polydisperse DNA with synthetic polycations have been studied. Two cationic polymers have been used, a homopolymer poly(methacryl oxyethyl trimethylammonium chloride) (PMOTAC) and its analogue grafted with poly(oxyethylene). The amount of poly(oxyethylene) grafts in the copolymer was 15 mol % and Mw of the graft was 200 g/mol. Salmon DNA (sodium salt) was used. The average molecular weight (Mw) of DNA was 10.4 × 106 g/mol. Conductivity, pH, and dynamic light scattering studies were used to characterize the complexes. The size and shape of the polyelectrolyte complex particles have been studied as a function of the cation-to-anion ratio in aqueous solutions of varying ionic strengths. The polyelectrolyte complexes have extremely narrow size distributions taking into account the polydispersity of the polyelectrolytes studied. The poly(oxyethylene) grafts on PMOTAC promote the formation of small colloidally stabile complex particles. Addition of salt shifts the macroscopic phase separation toward lower polycation content; that is, complexes partly phase separate with the mixing ratios far from 1:1. Further addition of salt to the turbid, partly phase separated solution results in the dissociation of complexes and the polycation and DNA dissolve as individual chains. Introduction Macromolecules of different chemical structures may bind to each other by various mechanisms leading either to the precipitation of the polymer aggregates or to the formation of colloidally stabile polymer particles.1-4 Thus, oppositely charged polyelectrolytes form stable interpolymer complexes in aqueous media as a result of Coulombic attraction between charged groups of the polymer chain.1,5 At present, noticeable interest to polyelectrolyte complexes (PECs) is focused on the complexation of DNA with various polycations, owing to the potential use of soluble DNA complexes in gene therapy.6-9 The goal in gene delivery is to transfer DNA segments bound to an appropriate polycationic carrier to the cell, through the cell membrane, and finally to release the DNA segment from the complex. Formation of interpolymer complexes is entropically favorable. When two oppositely charged polyions form a complex, low molar mass counterions are released in the outer solution and thus increase the entropy of the solution. The intermolecular complex shrinks due to the compensation of the polyion charges and increasing hydrophobicity of the complex. The latter results in multi molecular association, which stops when a particle of an equilibrium size is formed. Close to the 1:1 cation-to-anion ratio, charges of polyions are neutralized by interpolymer binding, the complexes turned hydrophobic and typically insoluble in aqueous * To whom correspondence should be addressed: Phone: +358-919150335. Fax: +358-9-19150330. E-mail: [email protected]. † Permanent address: Institute of Macromolecular Compounds, Russian Academy of Science, Bolshoi Prospect 31, 199004 St. Petersburg Russia.

solvents. Because of the ionic nature of the cellular membranes, the electroneutrality of the complexes is crucial to avoid clinging due to the opposite charges or repulsion of the charges of a same sign. For this reason, copolymers with nonionic water-soluble grafts or blocks, such as poly(oxyethylene) (POE), are promising candidates for gene delivery vectors in future.3,7 We have earlier studied PECs formed by cationic poly(methacryl oxyethyl trimethylammonium chloride) (PMOTAC) and anionic poly(oxyethylene-block-sodium methacrylate) and found that these objects may turn extremely soluble in water, if either of the polymers contains POE grafts or blocks.3 These particles also show monomodal and surprisingly narrow size distributions. Thus, it is evidently useful to have a look on the complexes of synthetic polycations with another polyanion, such as DNA. Recently, an important contribution on this topic was published by Nisha et al.10 These authors studied complexes formed by calf thymus DNA and a synthetic polycation, where the polymer was build of the same monomers we have used. However, in this case, the content of the POE-containing monomer was very high ranging from 68 to 94 mol %. We have used a different approach and synthesized polymers with a low degree of poly(oxyethylene) grafting. A polydisperse sodium salt of salmon DNA was a polymer of choice for our investigation. The high polydispersity of the sample allows for a better understanding of the formation of the monodisperse complexes. In the present report, the formation of the PEC of genomic polydisperse DNA with PMOTAC and PMOTAC-g-POE is discussed. Colloidal stability of the particles in aqueous

10.1021/bm049799n CCC: $27.50 © 2004 American Chemical Society Published on Web 06/29/2004

1854

Biomacromolecules, Vol. 5, No. 5, 2004

Andersson et al.

solutions with varying ionic strengths and with a varying cation:anion ratio was investigated. The effect of poly(oxyethylene) grafted to the polycation on the solubility of PEC is clearly demonstrated. Also the size and the shape of complexes have been examined. Experimental Section Materials. For the synthesis of the polycation, an aqueous solution (75 wt %) of methacryl oxyethyl trimethylammonium chloride, MOTAC, was purchased from Aldrich and used as received. Poly(oxyethylene) monomethyl ether monomethacrylate with 4-5 oxyethylene repeating units, POE-methacrylate, (Polysciences Inc.), ammonium peroxydisulfate, APS, (Merck), and N,N,N,N-tetramethylethylenediamine, TEMED 99.5%+, (Aldrich) were also used with no further purification. The phosphate buffer used in polymerization was Titrisol pH 7.00 (Merck). Salmon DNA (sodium salt) was purchased from Sigma. DNA was dialyzed for 3 days in water. After dialysis, it was dried under vacuum in a freeze-drier. Synthesis of PMOTAC and PMOTAC-gPOE has been reported earlier.3 Poly(methacryl oxyethyl trimethylammoniumchloride) (PMOTAC) with molar mass Mw ) 3 × 105 g/mol and its analogue with POE grafts (PMOTAC-g-POE) with molar mass Mw ) 2.8 × 105 g/mol were synthesized and purified with dialysis.3 Molar masses were measured by size exclusion chromatography (Waters) using POE as a standard. The amount of the POE-substituted repeating units in the copolymer was 15 mol %. The distributions of the hydrodynamic radius of the pure polymers measured by DLS were not much affected by the ionic strength within the studied measurement range (0.021.0 M NaCl). The size distributions of the PMOTAC, PMOTAC-g-POE, and DNA in 0.05 M aqueous NaCl are shown in Figure 1. To prepare solutions of polyelectrolyte complexes, DNA, PMOTAC, and PMOTAC-g-POE were first dissolved separately in water and in aqueous 0.02 M NaCl. Polycation solutions were added to DNA solutions. The molar ratios of cationic to anionic groups of the studied polyions were varied between 0 and 0.98. In the present investigation, the average value of the molar weight of four deoxyribonucleotide units in DNA has been used to calculate the molar ratios of cationic to anionic repeating units of the polymers in the interpolymer complexes. All the measurements were performed at 20 °C. Instrumentation. ConductiVity. Conductivities of the DNA solutions as a function of PMOTAC and PMOTACg-POE concentrations were measured with a Consort C833 multi-parameter analyzer. pH Studies. pH values of DNA solutions as a function of PMOTAC content were studied with a Radiometer Copenhagen Meterlab PHM210 standard pH meter. Light Scattering. A Brookhaven Instrument BIC-200 SM goniometer and a BIC-9000 AT digital correlator were employed to study light scattering from solutions. An argon laser (LEXEL 85, 1W) attenuated for 15-50 mW output power and λ ) 514.5 nm wavelength was the light source. Distributions of the hydrodynamic size were studied at scattering angles ranging between 30° and 150°. Those

Figure 1. Size distributions of PMOTAC (0.1 mg/mL), PMOTAC-gPOE (0.1 mg/mL), and DNA (0.025 mg/mL) in aqueous 0.05 M NaCl.

investigations, which are not focused on the absolute size of the particles studied and only qualitatively demonstrate changes in the system, were performed at 90° scattering angle. The autocorrelation function of scattered light intensity G2(t) ) 〈I(0)I(t)〉 was collected in the self-beating mode11 and then converted into an autocorrelation function of the scattered electric field g1(t) using the Siegert’s relationship |g1(t)| ) β1/2

x

G2(t) - G2(∞) G2(∞)

(1)

where G2(∞) is the experimentally determined baseline and β is the coherence factor determined by the geometry of the detection (typically 0.5 e β e 0.8). Characteristic decay times of a field correlation function τi and their relative amplitudes Ai(τi) were evaluated via moments of a corresponding distribution function of decay times A(τ) obtained using an inverse Laplace transform programs CONTIN as g1(t) )

∫0∞ A(τ)-t/τ dτ

(2)

5-7 correlation curves with various accumulation times were collected for every sample to check the reliability of the mathematical solution by the CONTIN analysis. Translation diffusion coefficients were calculated as Di ) (1/τi)q-2, where q is the amplitude of the scattering vector determined as q)

4πn0 θ sin λ0 2

λ0 is the wavelength of the incident laser light source, n0 is the refractive index of the solvent, and θ is the scattering

Biomacromolecules, Vol. 5, No. 5, 2004 1855

Complexation of DNA with Synthetic Polycations

Figure 2. Size distributions of complexes of PMOTAC and DNA as a function of salt concentration in aqueous NaCl solutions. The cation-toanion ratio is 0.28.

angle. The hydrodynamic radii were obtained from diffusion coefficients Di via the Stokes-Einstein equation Rhi )

kT 6πη0Di

(3)

where k is the Boltzmann constant, T is the absolute temperature, and η0 is the solvent viscosity. Mean values of the hydrodynamic radius distributions extrapolated to zero scattering angle were used for the estimation of the average radius Rh. Analysis of the correlation functions using CONTIN is known not to be the best choice for polydisperse samples. In such cases, an alternative single stretched exponent analysis of correlation functions was applied to estimate the width of the size distributions g1(t) ) Ae-(t/τ)

R

(4)

The autocorrelation function of the scattered electric field g1(t) decays from 1 to 0, and thus, the amplitude A equals one. Parameter R characterizes the width of a distribution of relaxation times. For a narrow single mode distribution, parameter R equals 1, and eq 4 turns into a trivial singleexponential decay, whereas for a broad monomodal (or polymodal) distribution, R is between zero and one. In the current investigation, R is used as a parameter alternative to the inverse Laplace transform data analysis to express the changes in the width of a size distribution. Results and Discussion Particles formed by polyelectrolytes have been studied in aqueous solutions of varying ionic strengths, as well as by changing the cation:anion ratio. Of a special interest it has

been to find out the effect of the POE-grafts in the polycation on the size and stability of the colloidal particles. Effect of Salt Concentration. In the first stage of this study, the effect of NaCl concentration on the size of the PMOTAC-DNA and PMOTAC-g-POE-DNA complexes was investigated. Aqueous 0.02 M NaCl solutions of DNA (0.025 mg/mL) with PMOTAC (0.1 mg/mL) and with PMOTACg-POE (0.1 mg/ml) were mixed in a way that the ratio of the cationic to the anionic repeating units was 0.28 in both mixtures. DNA concentration after the mixing was 0.021 mg/ mL. The solutions of PECs were kept for 24 h at 4 °C in a refrigerator, and then appropriate amounts of NaCl were added to the solutions to adjust the ionic strength (0.05 M-1.0 M). Finally, the PEC solutions were kept at least for 24 h in a refrigerator to reach equilibrium. The effect of ionic strength on the distributions of the hydrodynamic size of the PECs was studied by dynamic light scattering at 20 °C. The addition of NaCl within the range of 0.05-1.0 M has a minor effect on the size distributions of PMOTAC, PMOTAC-g-POE, and DNA solutions. However, the complexation and the size distributions of the PECs strongly depend on the ionic strength.12,13 Figure 2 presents the effect of salt concentration of the size distribution of the PMOTAC-DNA complex with a cation-to-anion ratio of 0.28. The size distributions are narrow when the salt concentration is below 0.6 M NaCl. The corresponding correlation functions are well fit by a single-exponential function. Narrow size distributions are typical for PECs and have been discussed in the literature.14 With increasing ionic strength, the size distributions broaden and the corresponding average Rh increases. PEC solutions with ionic strength below 0.05 M NaCl are visually clear. However, the solutions with salt concentration ranging from 0.05 to 0.6 M NaCl are opaque and partially precipitated with increasing the ionic

1856


strength. For the light scattering studies, the supernatant of partially precipitated solutions was taken. As discussed below, large and dense particles are formed in this range of ionic strength resulting in strong scattering from solutions. The polymer complexes are hydrophobic, and the addition of salt screens the remaining charges and promotes aggregation and even precipitation of the complexes. With a further increase in salt content above 0.6 M, the precipitated PECs starts to dissolve, and at 1.0 M NaCl, the size distribution is very broad and the solution is completely clear. The observed size distribution is as broad as that of the aqueous DNA solution. At 1.0 M NaCl, the scattering intensity decreased dramatically to a value similar to that of a pure DNA solution. With a 0.28 cation-to-anion ratio, minute amounts of uncomplexed PMOTAC do not contribute significantly to the scattered light intensity and do not affect size distributions. The particles formed by PECs scatter light much stronger than free DNA or PMOTAC molecules. With increasing salt concentration, the uncompensated ions of the polymers in the complex become screened by salt and the complex turns more hydrophobic. The size of the complex aggregates increases due to the secondary aggregation.3 Further addition of NaCl breaks the complexes and PMOTAC and DNA dissolve as individual polymers. In the case of complexes of PMOTAC-g-POE with DNA, the size distributions and the effect of added salt were similar to those of homopolycation up to 0.6 M NaCl. At 0.8 M NaCl, the complexes with POE-grafted polycation were disintegrated and the precipitated part was dissolved completely as individual polymers. This indicates, that POE grafts affect the complexation. Because cations bind to POE, POEgrafted PMOTAC is expected to self-associate. Thus, more uncomplexed ions of polycations exist in the PECs and less salt is needed to break the complexes apart. Due to the hydrophilicity of POE and a higher amount of uncomplexed free ions, PECs of PMOTAC-g-POE with DNA are not as hydrophobic as in the case of PEC formed by the homopolymer PMOTAC. When the cation-to-anion ratio is higher, the effect of ionic strength on the distributions of the hydrodynamic radius changes in a similar fashion as described above. Thus, for PEC solutions having the cation-to-anion ratio 0.50, the size of the PEC particles increased with increasing salt content up to 0.8 M NaCl and at 0.8 M NaCl the PECs start to dissociate. However, the mean Rh in the case of the 0.50 mixing ratio is significantly higher than that for solutions with 0.28, in the whole range of ionic strength. For example, in a solution of PMOTAC homopolymer complexes with DNA in 0.2 M NaCl, Rh was approximately 5-fold that of the Rh measured for solutions with a mixing ratio of 0.28, and the solution was more turbid. This indicates that this salt concentration is high enough to screen free charges in a 0.50 complex. Complexes form secondary aggregates, and the amount of insoluble fraction is larger than that in solutions with a mixing ratio of 0.28 upon increasing ionic strength. Complexes with higher polycation content could not be measured by dynamic light scattering due to the strong precipitation of the complexes in saline solutions with 0.10.8 M NaCl. However, even with a cation-to-anion ratio of

Andersson et al.

Figure 3. Scattering intensity, conductivity, and pH as a function of cation-to-anion ratio in saltless aqueous solution. (0) ) PMOTAC + DNA complexes, (2) ) PMOTAC-g-POE + DNA complexes, (×) ) values for pure DNA solution.

1.0, the precipitated fraction of PECs dissolves as individual polymers above 0.8 M NaCl and solution turns visually clear. Effect of the Mixing Ratio. The effect of the mixing ratio on the complexation was studied in a 0.02 M aqueous NaCl solution as well as saltless in aqueous solutions. Saltless aqueous solutions were used because the PECs are more soluble and a wider range of cation:anion ratios can be studied, approaching the ratio 1:1. In addition, useful information on the complexation may be gained when studying the conductivities of saltless solutions. 0.02 M NaCl was added to PEC solutions to minimize Coulomb interactions affecting light scattering. However, light scattering from saltless solutions was studied as well. PMOTAC or PMOTAC-g-POE solutions with 0.5 mg/mL concentration were added to 0.2 mg/mL DNA solutions in various mixing ratios. The final DNA concentration was kept constant at 0.16 mg/mL in each solution after the addition of the polycation. The solutions of complexes were kept in a refrigerator for 24 h to reach equilibrium. Figure 3 presents the effect of the mixing ratio on the light scattering intensity, conductivity, and pH of scattered light for PMOTAC/DNA and PMOTAC-g-POE/DNA complexes in saltless aqueous solutions. It may be observed, that the scattering intensity increases continuously with increasing



Figure 4. Size distributions of complexes of PMOTAC-g-POE with DNA (on the left column A) and PMOTAC with DNA (on the right column B) in aqueous saltless solution. Polycation concentration increases from top to bottom (cation-to-anion ratios: 1 ) 0.3, 2 ) 0.6, 3 ) 0.9).

PMOTAC concentration. The intensity of the scattered light from the PECs with the highest PMOTAC concentration approaching a 1:1 cation:anion ratio decreases slightly due to the macroscopic phase separation of the complex particles and thus, decreasing PEC concentration. Similar to PMOTAC/DNA, the intensity of light scattered by PMOTAC-g-POE/DNA increases with increasing the PMOTAC-g-POE concentration untill a cation:anion ratio of 0.5 is reached. Above 0.5, the intensity has a constant value, which is much lower than that observed for the PMOTAC/DNA complexes. Solutions in this region are clear, and no precipitation of complexes was observed. Thus, the POE grafts may decrease the density of the complexes formed. As seen in Figure 4, above the cation:anion ratio 0.5 the size distribution of PMOTAC-g-POE/DNA does not change. Formation of particles of a certain equilibrium size stabilized by POE can be suggested. In aqueous 0.02 M NaCl solutions, similar dependencies of light scattering intensity were observed for both complexes. However, a dramatic decrease of intensity in solutions of PMOTAC/DNA was observed at a PMOTAC concentration that is lower than that in saltless solutions. Thus, at a 0.90 cation:anion ratio, the intensity drops to 50 kcps from the value similar to that of the saltless solutions of about 600 kcps owing to the precipitation of the PECs. Addition of 0.02 M NaCl did not radically affect the scattering intensity from PMOTAC-g-POE/DNA solutions and intensity does not change upon increasing polycation content above 0.5. Near the 1:1 mixing ratio, the solution of PMOTACg-POE/DNA turned slightly turbid but without macroscopic phase separation. A trend similar to the intensity vs mixing ratio dependence of saltless PECs solutions can also be seen on the conductiv-

ity vs mixing ratio dependence. The conductivity increases monotonically with increasing PMOTAC concentration up to the mixing ratio 0.90. At this point, the conductivity smoothly starts to level off due to the phase separation. Conductivity of the PMOTAC-g-POE/DNA complexes is slightly lower than that of PECs with the homopolycation over the whole range of mixing ratios studied. The pH decreases monotonically with increasing the degree of neutralization of DNA with PMOTAC. In the case of PMOTAC-g-POE, the pH reaches a constant value at the mixing ratio of 0.6. It is interesting to note that the observed decrease in pH with increasing polycation concentration shows a similar difference between PMOTAC and PMOTAC-g-POE as was observed by light scattering. The effect of the mixing ratio on the size distributions of the complexes is presented in Figure 4 for saltless solutions and in Figure 5 for 0.02 M NaCl solutions. The size distributions of the polymers themselves (PMOTAC, PMOTACg-POE, and DNA) are broad in aqueous solutions. As seen in Figures 4 and 5, the size distributions are bimodal, consisting of two narrow peaks. Obviously, the peak corresponding to large particles represents the polyelectrolyte complexes. Its mean Rh decreases with increasing polycation content. The other peak has a much lower relative intensity. Its position does not significantly vary either as a function of ionic strength (Figure 2) or as a function of polycation content. The origin of this low intensity peak may be the polycation (PMOTAC and PMOTAC-g-POE), which is not bound to DNA. One of the possible reasons why all of the polycation is no bound to DNA could be a difference in the degree of the dissociation of these polyelectrolytes. Decrease of the relative intensity of the peak corresponding to the polycation upon increasing the polycation content owes to

1858


Andersson et al.

Figure 5. Size distributions of complexes of PMOTAC-g-POE with DNA (on the left column A) and PMOTAC with DNA (on the right column B) in 0.02 M NaCl. Polycation concentration increases from top to bottom (cation-to-anion ratios: 1 ) 0.3, 2 ) 0.6, 3 ) 0.9).

the stronger scattering from PECs; the number and solubility of the PECs increases with cation and salt content. The size of the complex particles is smaller and the distributions are narrower for solutions with PMOTAC-gPOE than for solutions with PMOTAC. This is especially pronounced in saltless solutions with low polycation concentrations (Figure 4, A1 and B1). Though the interpretation of light scattering from saltless polyelectrolyte solutions is difficult, one can see from the comparison of Figures 4 and 5 that salt containing solutions are similar to saltless ones. With high cation-to-anion ratios (over 0.5) in saltless solutions, the mean Rh of PMOTAC/DNA complexes starts to decrease upon increasing polycation concentration and reaches a value of the same magnitude as in the copolymer complexes. The size distribution also becomes narrow (Figure 4, A3 and B3). The Rh and the width parameter R calculated using the stretched exponent analysis (see eq 4) are presented in Figure 6. The average values of Rh obtained with CONTIN or stretched exponential analyses were approximately the same. At low cation content, the distributions of sizes obtained with CONTIN are bimodal. The distributions may, in fact, also be broad monomodal ones because of the well-known features of the inverse Laplace transform data analysis, though broad distributions are not typical for PECs.10,14 For saltless solutions, Rh shown in Figure 6 corresponds to the mean Rh value obtained with CONTIN, i.e., a value averaged taken into account the relative weight of both peaks shown in Figure 4. As is seen in Figure 4, the peak representing interpolymer complexes shifts toward smaller sizes with increasing polycation concentration and also dominates scattering from individual molecules. This results in an apparent decrease in Rh determined with the single stretched exponential analysis. When the cation-to-anion ratio exceeds

Figure 6. (A) Width parameter for PMOTAC + DNA complexes (0) and PMOTAC-g-POE+ DNA complexes (2) as a function of anionto-cation ratio in aqueous solution. (B) Rh calculated with stretched exponential for PMOTAC + DNA complexes (0) and PMOTAC-gPOE+ DNA complexes (2) as a function of cation-to-anion ratio.

0.6, size distributions are narrow and the width parameter R approaches unity in the case of PMOTAC-g-POE/DNA complexes and 0.8 for PMOTAC/DNA complexes. For the 0.02 M NaCl solutions studied in the range of mixing ratios from 0.3 to 0.6, the size of the complexes formed by both homo- and copolymer does not change significantly with increasing polycation content, see Figure 5. With a cation-to-anion ratio of 0.3 the size distribution is extremely narrow and the mean value of Rh of the PMOTAC/ DNA complex is 163 nm, which is approximately the same



Table 1. Analysis of Dynamic and Static Light Scattering Data for 0.02 M NaCl Solutions of PECs with the Cation to Anion Ratio of 0.3 conc. (mg/mL) PMOTAC/DNA

PMOTAC-g-POE/DNA

0

Rg (nm) 120

Rh (nm) 165

Rg/Rh

dn/dc (mL/g)

Mw (×106) (g/mol)

Fhc (g/cm3)

0.73

0.166a 0.200b

410 280

0.04 0.02

0.166a 0.200b

410 290

0.04 0.03

0.20 0.15 0.10 0.05 0

119 120 120 121 112

163 165 163 168 158

0.73 0.73 0.74 0.72 0.71

0.20 0.15 0.10 0.05

112 110 111 113

160 157 158 156

0.70 0.70 0.70 0.72

a dn/dc value is taken from ref 15. b dn/dc value is taken as a typical one for globular proteins/particles. c For density (F ) estimation the hydrodynamic h volume was used.

as in the case of copolymer complexes. In saltless solutions, Rh of the copolymer complexes was of the same order of magnitude as well. However, the Rh of the PMOTAC homopolymer complexes with DNA was 340 nm. In the saltless solutions, the size distributions of the homopolymer/ DNA complexes were as narrow as those with POE grafted copolymers with a mixing ratio of 0.9. In a 0.02 M NaCl solution, complexes are more hydrophobic than those in saltless solutions and distributions are broad. Due to the screening of remaining charges of polyions by counterions of added salt, PECs aggregate and precipitate partly with mixing ratios close to unity (Figure 5). This can also be seen as a dramatic drop in scattering intensity, as mentioned above. Addition of 0.02 M NaCl does not affect significantly the size distribution and mean value of Rh of PMOTAC-gPOE/DNA complexes. These observations indicate that POE grafts increase the stability of polycation/DNA complexes in aqueous and in saline solutions with different mixing ratios owing to the hydrophilicity of added POE graft. The shape and density of the complexes was estimated using static and dynamic light scattering. Static light scattering cannot be used straightforward owing to the bimodality of the size distributions. On the other hand, at higher polycation content, it is not possible to determine the real polymer concentration owing to the precipitation of PECs. These problems were minimal for 0.02 M NaCl solutions having total polymer concentrations in the range of 0.050.2 mg/mL, meaning DNA concentrations 0.04-0.17 mg/ mL and a cation-to-anion ratio of 0.3. The absolute intensity of the scattered light was measured at scattering angles varying in the range 30-150°. Simultaneously, intensity weighted distributions of the hydrodynamic radius were recorded. The absolute intensities emitted by the two types of scatterers, i.e., the PECs and the polycations, were calculated based on the total absolute intensity and the relative contribution of those scatterers to the corresponding intensity weighted distributions. Primarily, the intensity of light scattered by the PECs was analyzed to obtain the molar masses (Mw) and the radii of gyration (Rg). As stated above, the low intensity peak (Figure 5) most likely originates from individual chains of the polycation and may not be taken as very informative, due to the poor CONTIN resolution and high experimental error. This peak clearly appears at 90° and its relative contribution

to the total intensity grows with increasing scattering angle. At angles below 75°, only one peak representing the PECs was detected. In the 0.02 M NaCl solutions, PECs are expected to be dense spherical particles of the size comparable with the scattering vector q. Therefore, the Guinier form factor function seems to be the best choice to analyze the angular dependence of scattered light intensity. For that reason, the radius of gyration Rg was calculated using the Guinier method, i.e., from the slope of the ln I vs q2 dependence. Molar masses of PECs were estimated using the values of intensity extrapolated to zero scattering angle and then plotted against the total PEC concentration. Table 1 summarizes the results of the light scattering analysis. As seen in Table 1, both Rh and Rg do not have any noticeable concentration dependence. Rg/Rh values are typical for a hard (not a flowing-through) sphere having the value of 0.77. To estimate molar masses of PECs, two possible increments of the refractive index (dn/dc) were used. One of those (0.166 mL/g)15 is typical for DNA in aqueous salt solutions and motivated by a low cation-to-anion ratio. The other one (0.200 mL/g) is typical for compact particles such as globular proteins. In both cases, estimated molar masses are huge though the content of the polymer material inside the particles is lower than one can expect for a solid particle. It may be suggested that the PECs having a cation:anion ratio of 0.3 are spherical and consist of a dense hydrophobic core of bound polyions and loose corona of DNA. The estimated molar masses of both complexes are the same although the size and Rg/Rh ratio of PMOTAC-g-POE/DNA is slightly smaller than that of PMOTAC/DNA. This may be understood assuming a nonuniform spatial distribution of polymer material within the PECs. Thus, the hydrophobic core of PECs is denser than the corona formed mainly by DNA, which is in access to the polycation. The hydrophobic core of PMOTAC-g-POE/DNA can contain more polymer material and result in a smaller Rg value and Rg/Rh ratio with respect to those of PMOTAC/DNA. For steric reasons as well as due to the difference in the degree of dissociation, uncomplexed cationic and anionic groups may exist in the PECs. Cationic binding of POE to polycation hinders the repulsion between positive groups not bound to DNA. Hydrophilic POE can stabilize the particle, and a denser core can be formed without the precipitation of the particles.

1860


Discussion and Conclusions The effect of the mixing ratio, as well as the effect of the POE grafts in the polycation on the size and stability of the complexes with DNA was studied in aqueous saltless solutions and the results were compared with those obtained in 0.02 M NaCl solutions. Saltless solutions were used because of the better solubility of the complexes with higher polycation/DNA ratios. 0.02 M NaCl was added to prevent the polyelectrolyte effect. It has been observed, first, that the polyelectrolyte complexes have surprisingly narrow size distributions taking into account the polydispersity of the polyelectrolytes studied and, second, that the POE grafts in PMOTAC promote the formation of small colloidally stabile complex particles. The size of PMOTAC/DNA and PMOTAC-g-POE/DNA complexes was observed to decrease with increasing polycation concentration. Approaching the cation-to-anion ratio of 1.0, the complexes turned hydrophobic and the sizes of the complexes increased slightly before the phase separation was obviously observed. In saline solutions, the phase separation of the PECs was observed with lower polycation content. Also the effect of salt concentration on the complexes of PMOTAC and PMOTAC-g-POE with DNA was studied. Increasing salt concentration in PEC solutions with a cationto-anion ratio of 0.28 increases the size of the complexes due to secondary aggregation. Thus, an increase in the ionic strength from 0.0 to 0.3 M increases the size of the PECs from 100 to 200 nm. For comparison, the mean Rh for pure DNA at the same conditions is 5-fold larger and distributions are extremely broad. Free ionic groups of DNA and polycation were screened with an increase in ionic strength and the complex turned more hydrophobic. When the ionic strength was above 0.2 M NaCl, large aggregates were observed even with a cation-to-anion ratio of 0.5. Above 0.8 M NaCl, the turbid complex solution turned completely clear, the scattered light intensity decreased dramatically, and a broad size distribution was observed indicating that PMOTAC and DNA were dissolved as individual polymers. It has been shown that POE grafts improve the solubility of PECs. Thus, PMOTAC-g-POE/DNA complexes do not precipitate as easily as the complexes with the homopolymer upon increasing ionic strength, though the size of complexes was approximately the same as those of the PMOTAC/DNA complexes. It was observed that PMOTAC-g-POE/DNA complexes dissolve as individual chains in 0.8 M aqueous NaCl completely, whereas complexes of the homopolymer disintegrate only partly: the size distributions were still narrow, and the intensity of the scattered light was high due to the strong scattering by PECs. In other words, less salt is needed to break the PECs into individual chains due to binding of the ether oxygen of POE to the polycation in the case of PMOTAC-g-POE/DNA complexes. Conformation and density of the PECs were estimated. Measured Rg/Rh ratios were typical for a hard sphere value of 0.77. However, content of the polymer material inside the particles is low. A model with a non-uniform density distribution was suggested. Thus, a spherical particle consisting of a dense hydrophobic core of bound polyions and a loose corona of DNA fits well the experimental data.

Andersson et al.

The formation of PEC particles of a certain equilibrium size should be emphasized. The existence of a specific surface-to-volume ratio for the PECs is remarkable especially taking into account the original polydispersity of the polyions forming the PECs. Thus, as clearly demonstrated in Table 1, in solutions with the ionic strength of 0.02 M NaCl and the cation:anion ratio of 0.3, neither the size nor the Rg/Rh ratio of the PECs are affected by the polymer concentration. The second virial coefficient of the PEC particles by light scattering was zero within the experimental error. The following mechanism of formation of the PECs can be suggested. First the oppositely charged polyions bind to each other. Low molar mass counterions get released to the outer solution, thus increasing the entropy of the system. Water is a poor solvent for the parts of the polyions bound via Coulombic attraction. The electrostatically bound chains further associate via hydrophobic attraction forming a spherical particle. The aggregation of complexed chains stops when a particle of a certain equilibrium surface-to-volume ratio is reached. The size and shape of the PECs does not depend on the polydispersity of the polyions in the complex. The second virial coefficient for the PECs at equilibrium is zero. This reflects that all of the interactions between particles are in balance, i.e., particles do not further aggregate or grow. The formation of dense particles of equilibrium size has recently been reported for charged16 and uncharged17-18 water-soluble homopolymers in thermodynamically unfavorable conditions. From the analysis of these reports, one may conclude that in a thermodynamically poor solvent polymers with a low molar mass associate and form spherical particles of equilibrium size in a dilute solution. The size of the particles formed by uncharged water-soluble homopolymers increases with increasing the polymer concentration.17,18 A polyelectrolyte of a molar mass below 2 × 106 g/mol also associates in a poor solvent until a particle of optimal size is formed.16 However, if the molar mass exceeds 2 × 106 g/mol, the collapsed state of this molecule consists of a sequence of dense spheres, the size of which is comparable with the optimal size of the aggregates formed by the same polymer of lower molar mass. References and Notes (1) Dautzenberg, H. In Physical Chemistry of Polyelectrolytes; Radeva, T., Ed.; Marcel Dekker Inc.: New York, 2001; pp 743-792. (2) Kriz, J.; Dautzenberg, H.; Dybal, J.; Kurkova, D. Competitive/ cooperative electrostatic interactions in macromolecular complexes: multinuclear NMR study of PDADMAC-PmaNa complexes in the presence of Al3+ ions. Langmuir 2002, 18, 9594-9599. (3) Andersson, T.; Holappa, S.; Aseyev, V.; Tenhu, H. Complexation of linear and poly(ethylene oxide)-grafted poly(methacryl oxyethyl trimethylammonium chloride) with poly(ethylene oxide-block-sodium methacrylate). J. Polym. Sci. Part A 2003, 41, 1904-1914. (4) Dautzenberg, H.; Karibyants, N. Polyelectrolyte complex formation in highly aggregating systems. Effect of salt. Response to subsequent addition of NaCl. Macromol. Chem. Phys. 1999, 200 (1),118-125. (5) Michaels, A. S.; Miekka, R. G. Polycation-polyanion complexes: preparation and properties of poly(vinylbenzyltrimethylammonium styrenesulfonate). J. Phys. Chem. 1961, 65, 1765-1773. (6) Gebhart, C. L.; Kabanov, A. V. Evaluation of polyplexes as gene transfer agents. J. Controlled Release 2001, 73, 401-416. (7) Kabanov, A. V.; Gebhart, C. L.; Bronich, T. K.; Vinogradov, S. V. Polycations for gene delivery: problems and solutions. Polym. Prepr. Am. Chem. Soc. DiV. Polym. Chem. 2002, 43 (2), 669-670.

Complexation of DNA with Synthetic Polycations (8) Oupicky, D.; Konak, C.; Ulbrich, K.; Wolfert, M. A.; Seymour, L. W. DNA delivery systems based on complexes of DNA with synthetic polycations and their copolymers. J. Controlled Release 2000, 65, 149-171. (9) McAllister, K.; Sazani, P.; Adam, M.; Cho, M. J.; Rubinstein, M.; Samulski, R. J.; DeSimone, J. Polymeric nanogels produced via inverse microemulsion polymerization as potential gene and antisense delivery agents. J. Am. Chem. Soc. 2002, 124, 15198-15207. (10) Nisha, C. K.; Manorama, S. V.; Ganguli, M.; Maiti, S.; Kizhakkedathu, J. N. Complexes of Poly(ethylene glycol)-based cationic random copolymer and calf thymus DNA: A complete biophysical characterization. Langmuir 2004, 20, 2386-2396. (11) Chu, B. Laser Light Scattering: Basic Principles and Practice, 2nd edition, Academic Press Inc., San Diego, 1991, p 74. (12) Dautzenberg, H.; Rother, G.; Response of polyelectrolyte complexes to subsequent addition of sodium cloride: Time.dependent static light scattering studies. Macromol. Chem. Phys., 2004, 205, 114-121. (13) Seyrek, E.; Dubin, P. L.; Tribet, C.; Gamble, E. A. Ionic strength dependence of protein-polyelectrolyte interactions. Biomacromolecules, 2003, 4, 273-282.

Biomacromolecules, Vol. 5, No. 5, 2004 1861 (14) Buchhammer, H.-M.; Mende, M.; Oelmann, M. Formation of monosized polyelectrolyte complex dispersions: effects of polymer structure, concrentration and mixing conditions. Colloids Surf. A: Physicochem. Eng. Aspects 2003, 218 (1-3), 151-159. (15) Krasna, A. I. Low-angle light-scattering studies on alkali- and heatdenaturated DNA. Biopolymers 1970, 9, 1029-1038. (16) Aseyev, V.; Klenin, S. I.; Tenhu, H.; Grillo, I.; Geissler, E. Neutron scattering studies of the structure of a polyelectrolyte globule in a water-acetone mixture. Macromolecules 2001, 34, 3706-3709. (17) Dawson, K. A.; Gorelov, A. V.; Timoshenko, E. G.; Kuznetsov, Yu. A.; Du Chesne, A. Formation of Mesoglobules from Phase Separation in Dilute Polymer Solutions: a Study in Experiment, Theory, and Applications. Physica A 1997, 244, 68-80. (18) Laukkanen, A.; Valtola, L.; Winnik, F.; Tenhu, H. Formation of Colloidally Stable Phase Separated Poly(N-vinylcaprolactam) in Water: A Study by Dynamic Light Scattering, Microcalorimetry, and Pressure Perturbation Calorimetry. Macromolecules 2004, 37, 2268-2234.

BM049799N

Complexation of DNA with Poly (methacryl oxyethyl

Recommend Documents