Polyplexes of Poly(methylaminophosphazene): Energetics of DNA

Aug 10, 2011 - Physics Department, M. V. Lomonosov Moscow State University, .... Using the Mark–Houwink relation for DNA(14), the molecular weight o...
0 downloads 0 Views 1MB Size
ARTICLE pubs.acs.org/Langmuir

Polyplexes of Poly(methylaminophosphazene): Energetics of DNA Melting Tatiana V. Burova,*,† Natalia V. Grinberg,† Dzidra R. Tur,† Vladimir S. Papkov,† Alexander S. Dubovik,‡ Valerij Y. Grinberg,‡ and Alexei R. Khokhlov§ †

A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, Vavilov St. 28, 119991 Moscow, Russian Federation ‡ N. M. Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, Kosygin St. 4, 119334 Moscow, Russian Federation § Physics Department, M. V. Lomonosov Moscow State University, Vorobyevy Gory, 119992 Moscow, Russian Federation ABSTRACT: The interaction of DNA with a synthetic biocompatible and biodegradable cationic polymer, poly(methylaminophosphazene) hydrochloride (PMAP 3 HCl), was investigated by high-sensitivity differential scanning calorimetry under conditions of strong and weak electrostatic interactions of the macroions. Thermodynamic parameters of the DNA doublehelix melting were determined as a function of pH and the PMAP 3 HCl/DNA weight ratio. PMAP 3 HCL was shown to reveal two functions with respect to DNA: the polyelectrolyte function and the donoracceptor one. The first function stabilizes the helical conformation of DNA, and the second one destabilizes it. The stabilizing effect of PMAP 3 HCl is of entropic origin, related to a displacement of mobile counterions from the DNA’s nearest surroundings by the poly(methylaminophosphazene) charged groups. The donoracceptor function of poly(methylaminophosphazene) dominates when its electrostatic interaction with DNA is either saturated (in the complex coacervate phase at high poly(methylaminophosphazene) concentrations) or completely suppressed (in a salt medium when the polycation carries a small charge). Under these conditions, poly(methylaminophosphazene) destabilizes DNA. It preferentially binds to the DNA coil form likely via the formation of multiple labile hydrogen bonds with the donoracceptor groups of DNA.

’ INTRODUCTION In recent years, there has been an increasing interest in cationic polyorganophosphazenes as components of DNA-based nonviral transport systems.14 The fundamental advantage of the polyphosphazenes compared to other synthetic polycations is determined by their biodegradability.5,6 The biodegradability of polyphosphazenes allows one to avoid polymer accumulation in the organism during long-term treatment. In addition, the polyphosphazenes exhibit lower cytotoxicity when compared to other synthetic polycations.7 The polyphosphazenes containing amino groups have been shown to form nanosized polyplexes with DNA,2 and they possess a high transfection efficiency.8,9 However, the mechanism of DNA complexation with polyphosphazenes has not been investigated in detail. Until now, no data could be found on the conformational changes of DNA in the interpolyelectrolyte complexes with these polycations. Poly(alkylaminophosphazenes) represent a particular class of polyphosphazenes. These are linear polymers that can be obtained as a result of the complete aminolysis of poly(dichlorophosphazene).10 In aqueous solutions, some of them can carry positive charges in their backbone because of protonation in accordance with the scheme11 ½  PðNHRÞ2  NdPðNHRÞ2  NdPðNHRÞ2  NdPðNHRÞ2  n þ 2nHþ a ½  PðNHRÞ2  NHþ dPðNHRÞ2  NdPðNHRÞ2  NHþ dPðNHRÞ2  n

where R = CH3, C2H5. r 2011 American Chemical Society

Previously, we reported on the conformational energetics of the interaction of a double helical biopolymer, ι-carrageenan, with poly(ethylaminophosphazene) hydrochloride (R = C2H5)12 and poly(methylaminophosphazene) hydrochloride (R = CH3).13 Using high-sensitivity differential scanning calorimetry (HSDSC), we showed that both cationic polymers reveal an unwinding effect on the ι-carrageenan helix at pH 3.8 where their degree of ionization is at its maximum (f = 0.5). Alternatively, poly(methylaminophosphazene) stabilizes the helical state of the polysaccharide at pH 7.4 when it is less ionized (f ≈ 0.2). In the present work, we report the results of a thermodynamic study of polyplexes of poly(methylaminophosphazene) (i.e., its interpolyelectrolyte complexes with DNA). By means of HS-DSC, we found that upon complexation poly(methylaminophosphazene) is able to stabilize the DNA double helix as well as to unwind it. We present the thermodynamic parameters of DNA melting as a function of the polycation content in the system and discuss parameters affecting the electrostatic interaction of the macroions (pH and ionic strength). The mechanisms of the stabilizing and destabilizing effects of poly(methylaminophosphazene) on the helical conformation of DNA are discussed. Received: June 21, 2011 Revised: August 9, 2011 Published: August 10, 2011 11582

dx.doi.org/10.1021/la202321t | Langmuir 2011, 27, 11582–11590

Langmuir

ARTICLE

’ EXPERIMENTAL SECTION Materials. Poly(methylaminophosphazene) (PMAP) was obtained by the complete aminolysis of a linear poly(dichlorophosphazene) (Mη = 13  106). To prepare poly(methylaminophosphazene) hydrochloride (PMAP 3 HCl), PMAP was dissolved in 0.01 M HCl in an equimolar PMAP monomole/HCl ratio and lyophilized. The chlorine content in the PMAP 3 HCl sample was 14.9%. An aqueous solution of PMAP 3 HCl at a concentration of 1.0 mg 3 mL1 had a pH of 3.16 ( 0.01. These data revealed that the maximum degree of ionization of PMAP 3 HCl is approximately 0.5. The molecular weight of PMAP 3 HCl was determined from measurements of its sedimentation and diffusion coefficients in an aqueous solution (40 mM glycine buffer, pH 3.5, 0.15 M KCl), MsD =2.2  105.13 A DNA sample isolated as a sodium salt from salmon testis was a gift from Professor S. N. Potekhin (Institute of Protein Research, Pushchino, Russia). The D260/D280 ratio for the sample was 1.75. The amount of the protein impurity was less than 0.5%. The molecular weight of DNA was determined via viscosity measurements on DNA solutions at concentrations from 0.1 to 0.8 mg 3 mL1 in 0.2 M NaCl. Measurements were performed with an AV-1 automatic Zimm-type viscosimeter (Biopribor, Russia) at a temperature of 20.0 ( 0.1 °C and an average shear gradient of 1.2 s1. The intrinsic viscosity of DNA, [η] = 5.32 ( 0.17 dL 3 g1, was obtained. Using the MarkHouwink relation for DNA14, the molecular weight of DNA was estimated to be 7.3  105. Interpolyelectrolyte Reactions. DNAPMAP polyplexes were prepared by mixing stock solutions of DNA and PMAP 3 HCl in a buffer solution at the desired pH. Ten millimolar sodium citrate at pH 4.2 and 10 mM potassium phosphate at pH 7.0 were used as buffer solutions with an ionic strength of 0.02. DNA stock solutions were prepared by dialysis against the corresponding buffer solution at 4 °C for 20 h. The DNA concentration after dialysis was spectrophotometrically ascertained by assuming DNA extinction E1260mgnm3 mL1 = 20.3.15 PMAP 3 HCl stock solutions were obtained by the dissolution of the polymer sample in the buffer solutions. The DNAPMAP 3 HCl mixture composition was characterized by a ratio of the weight concentrations of PMAP 3 HCl and DNA, qw. After being mixed, the systems were incubated at 4 °C for 20 h prior to measurements. The phase analysis of the DNAPMAP 3 HCl mixtures was performed according to the following protocol. Mixtures of various compositions were subjected to ultracentrifugation at 3  104 rpm (∼105 g) for 30 min. As a result, the system was separated into two phases: a supernatant (transparent solution) and the complex coacervate phase. The DNA concentration in the coexisting phases and the masses of both phases were directly determined. The concentrations of the polycation in the coacervate phase and the supernatant were calculated using these data, the lever rule, and the PMAP 3 HCl mass balance. The equivalent ratio of the polyelectrolytes, qew, was determined as the ratio of the equivalent weights of PMAP 3 HCl and DNA. The equivalent weight of the polymer was assumed to be the polymer weight per charge. The degree of ionization of PMAP 3 HCl at a given pH was calculated from the acidity indices (pKa1 7.95 and pKa2 5.2) for octamethylaminocyclotetraphosphazene.11 The linear charge density, ξ, of PMAP 3 HCl was calculated from its degree of ionization and the crystallographic data for poly(dichlorphosphazene).16 The charge of DNA was determined from the potentiometric data of Cavalieri and Rosenberg.17 The estimated values of the equivalent PMAP 3 HCl/DNA weight ratio, qew, were 0.7 and 1.5 at pH 4.2 and 7.0, respectively. Calorimetric measurements were carried out with a DASM-4 differential scanning microcalorimeter (Biopribor, Russia) within the temperature range of 10130 °C at a heating rate of 1 K 3 min1 and an excess pressure of 0.25 MPa. The primary data processing and transformation of the partial heat capacity of DNA into the excess heat

Figure 1. Excess heat capacity functions of DNA melting at different pH values: 7.0 (1), 6.0 (2), 5.5 (3), 5.0 (4), 4.5 (5), 4.1 (6), and 3.8 (7) with 10 mM phosphate or citric buffer with an ionic strength of 0.02 ( 0.01 and a DNA concentration of 0.35 mg 3 mL1. The curves are arbitrarily shifted along the ordinate axis. capacity function of the helixcoil transition were performed using NAIRTA 2.0 software (Institute of Organoelement Compounds, Moscow, Russia). The baseline of the transition was obtained by a spline interpolation. The temperature of the maximum in the excess heat capacity curve was taken as the melting temperature of DNA, Tm. The melting enthalpy, Δmh, was determined by integrating the excess heat capacity function. Spectrophotometric measurements were performed with a Genesys 2 UVvis spectrophotometer (ThermoSpectronic).

’ RESULTS AND DISCUSSION Effects of pH on DNA Conformational Stability. One of the key factors affecting the type and energetics of the DNApolycation interaction is pH.18 For this reason, it was necessary to take into account the effects of pH on the conformational stability of DNA when choosing conditions for the interpolyelectrolyte complex formation. Figure 1 shows the excess heat capacity profiles of DNA melting at different pH values. A single, rather symmetric heat capacity peak is observed over the whole pH range studied. The peak is attributed to the transition of DNA from the helix to the coil state. Upon the pH shift from physiological to acidic values, the transition peak significantly moves to lower temperatures and decreases gradually in height and area. The pH dependences of the thermodynamic parameters of DNA melting (transition temperature, Tm; width, ΔmT; and enthalpy, Δmh) are shown in Figure 2A,B. The melting transition and enthalpy vary in parallel: they reach a maximum at about pH 6.5 and decrease notably over the acidic pH range. These features point to a significant decrease in the DNA conformational stability in the acidic medium. This effect is accompanied by a narrowing of the melting transition, which suggests an increase in its cooperativity. The correlation between values of the DNA melting enthalpy and temperature determined at different pH values is given in Figure 2C. The slope of the correlation line is equal to 0.46 ( 0.02 J 3 g1 3 K1. According to Kirchhoff’s law,19 this value represents the heat capacity increment of the conformational transition, Δmcp. The estimated value of the DNA melting increment, Δmcp, 11583

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

ARTICLE

Figure 3. Turbidity of DNAPMAP 3 HCl mixtures measured at 340 nm vs the PMAP 3 HCl/DNA weight ratio (pH 4.2, ionic strength 0.02, DNA concentration 0.007 mg 3 mL1).

Figure 2. pH dependences of the (A, 1) temperature, (A, 2) width, and (B) enthalpy of DNA melting and (C) the Kirchhoff correlation between the melting enthalpies and temperatures of DNA determined at various pH values. The slope of the correlation line represents the heat capacity increment of the DNA melting, Δmcp = 0.46 ( 0.02 J 3 g1 3 K1. The thin and dotted lines in panel A represent theoretical function Tm(pH) calculated using the algorithms of Sukhorukov et al.23 and Lando et al.,24 respectively. The melting width is determined as the ratio Δmh/cEp (Tm).

is in agreement with the literature data.20,21 The heat capacity increment of DNA is positive and comparable in magnitude to the denaturation heat capacity increments of globular proteins.22 This fact allows us to suggest that hydrophobic interactions make a definite contribution to the stability of the DNA helix. A reduction of the conformational stability of DNA and a narrowing of the melting transition in the acidic pH range were observed by Sukhorukov et al.23 and Lando et al.24 They explained these effects by the protonation of some nucleotide units that leads to the breakdown of hydrogen bonds supporting the double-helix structure and suggested different theoretical approaches to their analysis. The ionization constants of nucleotide units are the key parameters of both approaches. Sukhorukov et al.23 used a single value of the ionization constant for all types of nucleotide units capable of protonation in the coil state (pKa 4.0), and Lando et al.24 considered experimental values of the ionization constants for each type of nucleotide in the helix and coil states. We used both approaches to calculate the Tm(pH) function by taking into account the following expression for the free energy of DNA melting (per base pair) in a reference state (pH 7.0) Δm Go ¼ Δm H o þ Δm Cp ðT  Tmo Þ " !# Δm H o T þ Δm Cp ln o T Tm Tmo

ð1Þ

where ΔmH° = 25.2 kJ 3 mol1 and Tm° = 341.2 K are the melting enthalpy and temperature at pH 7.0 and ΔmCp = 0.30 kJ 3 mol1 3 K1 is the melting heat capacity increment. The results of these calculations are presented in Figure 2A by the thin and dotted

lines. There is qualitative agreement between both theories and our experimental data. However, in contrast to the theoretical curves, the experimental dependence of the melting temperature on pH passes through a maximum at about pH 6.5 and declines on average more rapidly over the pH range from 6.0 to 3.8. A possible source of disagreement seems to be the fact that the theories apparently do not take into account electrostatic factors, which could be very important at a low ionic strength (∼0.02). Effects of PMAP 3 HCl on DNA Melting at pH 4.2 and 0.02 Ionic Strength. In the acidic medium, PMAP 3 HCl carries a large positive charge (the degree of ionization is f ≈ 0.5, and the linear charge density is ξ ≈ 1.5 at pH e4.2); therefore, a strong electrostatic attraction between the PMAP polycation and DNA takes place and results in the formation of interpolyelectrolyte complexes. Figure 3 shows the turbidity of DNAPMAP 3 HCl mixtures at pH 4.2 as a function of the qw ratio. At low qw values (low polycation content), the turbidity is insignificant, thus the DNAPMAP complexes are soluble. Beginning from a certain qw value, the turbidity increases sharply and then passes through a maximum and decreases with further increases in the PMAP 3 HCl content. The substantial growth in turbidity implies the formation of insoluble interpolyelectrolyte complexes. The turbidity maximum is observed at qw ≈ 2. This composition corresponds to the maximum yield of the insoluble complex. The decreasing turbidity at large qw values seems to reflect an overcharge of the complexes25 and, as a consequence, an increase in their solubility. Let us consider changes in the stability of the DNA helical conformation upon formation of the soluble and insoluble DNAPMAP complexes. Figures 4 and 5 show the excess heat capacity curves of the DNA melting in a reference solution (without PMAP 3 HCl) and in mixtures with PMAP 3 HCl of different compositions at pH 4.2. The thermograms of the DNAPMAP 3 HCl system exhibit a complex evolution with increasing content of PMAP 3 HCl. At low qw values, a single peak is seen on the thermogram. Its position is close to that of the free DNA peak (Figure 4). With increasing qw, this peak shifts slightly to the right and decreases in height. Simultaneously, a second peak appears at a significantly higher temperature. This peak grows, becomes dominating at qw g 0.75 and, at last, remains the only peak at qw g 1.0. A further increase in the polycation content leads to a notable decrease of the high-temperature heat capacity peak (Figure 5). The peak continues to shift slightly to high temperatures. Both transition temperatures and the total melting enthalpy for the DNAPMAP 3 HCl system are shown in Figure 6 as a function of the ratio qw. The dashed-dotted line separates the regions of existence of soluble and insoluble complexes (complex 11584

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

Figure 4. Excess heat capacity functions of DNAPMAP 3 HCl mixtures of different compositions at pH 4.2 and an ionic strength of 0.02 {qw = 0 (1), 0.125 (2), 0.25 (3), 0.5 (4), 0.75 (5), 1.25 (6)}. The curves are arbitrarily shifted along the ordinate axis.

Figure 5. Excess heat capacity functions of DNAPMAP 3 HCl mixtures of different compositions at pH 4.2 and an ionic strength of 0.02 (continuation){qw =1.25 (1), 1.5 (2), 1.6 (3), 2.0 (4)}. The curves are arbitrarily shifted along the ordinate axis.

coacervate phase). In the area of soluble complexes, two transitions are observed, the temperatures of which increase with increasing qw. The total enthalpy of these transitions does not depend on the mixture composition. In the region of complex precipitation, one transition is observed, the temperature of which increases slightly with qw, extending the linear dependence Tm(qw) for the high-temperature transition of the soluble complexes. The DNA melting enthalpy in the region of the coacervate phase formation apparently does not change at 1.0 < qw < 1.5 and then decreases rapidly, tending to zero. We can understand the tendencies of changes in the DNA melting parameters in the presence of PMAP 3 HCl, assuming that the system contains complexes of two types: stoichiometric (saturated) and nonstoichiometric complexes of variable composition. This situation is known to be typical of the interpolyelectrolyte reactions between highly charged polyelectrolytes at low salt concentration.2628 It is believed that an uneven distribution of polyelectrolyte ligands between polyelectrolyte

ARTICLE

Figure 6. Thermodynamic melting parameters of DNA in DNA PMAP 3 HCl mixtures of different compositions at pH 4.2 and an ionic strength of 0.02. (A) 1, first transition; 2, second transition. (B) Total enthalpy of both transitions. At qw g 1, phase separation takes place.

matrices of an opposite charge is a consequence of the cooperativity of the ligandmatrix interaction.29,30 This effect was named disproportionation of the interpolyelectrolyte complexes. In particular, such a disproportionation was described for complexes of DNA with PEG-block-poly(N-methyl-4-vinylpyridinium sulfate) and PEG-graft-poly(ethyleneimine).31 In our case, the DNA stability is maximal in the stoichiometric complex where all mobile counterions of the biopolymer are replaced by the charged units of PMAP 3 HCl. In the nonstoichiometric complexes, some of these counterions are retained. They are released into the environment upon DNA melting. This is the reason for the lower conformational stability of DNA in the nonstoichiometric complex as compared to that in the stoichiometric complex. To understand the mechanism of the change in the conformational stability of DNA in the region of stoichiometric complex precipitation, a phase analysis of the DNAPMAP 3 HCl system was carried out. The results of the phase analysis are given in Figure 7. The precipitation of the DNAPMAP interpolyelectrolyte complexes is observed at qw g1. The degree of precipitation of both polyelectrolytes increases with an increase in qw. The degree of precipitation of DNA reaches almost 100% whereas that of PMAP 3 HCl is about 50%. The PMAP 3 HCl/DNA weight ratio in the complex coacervate phase remains constant, qPw = 0.7 ( 0.1 (Figure 7B). It is close to the equivalent ratio of these polyelectrolytes (qew = 0.7). This confirms our hypothesis that PMAP 3 HCl and DNA precipitate in the form of the stoichiometric interpolyelectrolyte complex. Figure 7C shows the concentration of the coacervate phase as a function of qw. The increase in the PMAP 3 HCl content in the initial mixture leads to a significant increase in the total polymer concentration of the coacervate phase. We further investigated the conformational stability of DNA in the complex coacervate phase. For this purpose, the coacervate phase was separated by ultracentrifugation and analyzed calorimetrically. The thermograms of the coacervate phase obtained for qw g 1.25 showed a single DNA transition, the temperature of which coincided with that one of the high-temperature transition in Figure 6A. The DNA melting enthalpy in the coacervate phase decreased with increasing qw. The average ratio of the DNA 11585

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

ARTICLE

Figure 7. Results of the phase analysis of DNAPMAP 3 HCl mixtures of different compositions at pH 4.2 and an ionic strength of 0.02. The phase separation threshold (qw* =1.0) is marked by a vertical dasheddotted line. (A) β, the precipitation yield of DNA or PMAP 3 HCl. (B) qPw, the PMAP 3 HCl/DNA weight ratio in the complex coacervate phase (the average value of qPw is 0.7 ( 0.1; the PMAP 3 HCl/DNA equivalent weight ratio qew = 0.7 is indicated by a dashed line). (C) cPtotal, the total weight of the polymer concentration in the system at qPw < qw* and in the complex coacervate phase at qw g qw*.

enthalpies in the system before ultracentrifugation (Figure 6B) and in the separated coacervate phase was 1.0 ( 0.2, which consequently implies that the loss in the DNA melting enthalpy at qw g 1.25 (Figure 6B) is related to the changes in DNA stability in the coacervate phase. It seems plausible that these changes are caused by a considerable increase in the total polymer concentration of the coacervate phase at the large qw (Figure 7C). This increase is probably a consequence of the osmotic effect of excess PMAP 3 HCl in the supernatant. In summary, our data revealed two types of effects of the DNAPMAP complex formation on the conformational stability of DNA. The first type is the stabilization of the helical structure of DNA in the polyelectrolyte complexes, and the second type is its destabilization dominating the coacervate phase. The interpolyelectrolyte reaction between PMAP 3 HCl and the double helix of DNA is a reaction of ion exchange between the charged units of the polyelectrolytes ½A 3 3 3 H  B2 Na2 þ1 þ 2ðdNH  Þþ1 Cl1 K1

ss f f½A 3 3 3 H  B2 ðdNH  Þ2 þ1 g0 þ 2Naþ1 Cl1 r ð2Þ where the dotted lines are the hydrogen bonds. A and B are the nucleotide residues in the DNA molecule, for example, adenosine and thymine, respectively. (dNH)+1 is the protonated fragment of the PMAP 3 HCl backbone carrying a positive charge.

The fragments of DNA and the interpolyelectrolyte complex are marked by square brackets and braces, respectively. As a result of this reaction, the territorially bound DNA counterions, which could be released into the environment upon the double helix melting, are replaced by PMAP macrocounterions. The macrocounterions remain to be bound to the DNA chains after the double helix melts. Thus, the melting of DNA involved in the polyelectrolyte complex is not accompanied by the release of counterions. Accordingly, the entropy gain of the melting decreases.3234 This explains the increase in the doublehelix stability of DNA immobilized in the DNAPMAP interpolyelectrolyte complexes. The entropic origin of DNA stabilization upon its reaction with PMAP 3 HCl is directly confirmed by the calorimetric data. In fact, the reaction leads to an increase in the melting temperature of the DNA helix without any change in the melting enthalpy. This fact points to the decrease in the DNA unfolding entropy as a result of the complex formation that makes DNA melting less favorable in the complex than in the free state. The effect of unwinding of the DNA helix in the coacervate phase is related to the specific chemical structure of PMAP 3 HCl, which is capable of secondary interactions that are much weaker than the electrostatic interactions. These may be hydrogen bonds between PMAP 3 HCl and DNA. In the coacervate phase at high PMAP 3 HCl concentrations (of about 0.6 monomol 3 L1), the following type of side reactions could take place 0 f½A 3 3 3 H  B2 ðdNH  Þþ1 2 g þ ðdN  Þ K2

ss f f½ðdN  Þ 3 3 3 H  B1 ðdNH  Þþ1 g0 r

þ f½AÞ1 ðdNH  Þþ1 g0

ð3Þ

f½A 3 3 3 H  B2 ðdNH  Þ2 þ1 g0 þ ðH  NR  Þ K3

ss f f½A 3 3 3 ðH  NR  Þ1 ðdNH  Þþ1 g0 r þ f½H  B1 ðdNH  Þþ1 g0

ð4Þ

where (dN) is the uncharged fragment of the PMAP 3 HCl backbone possessing a proton-acceptor function, (HNR) is the side group of PMAP 3 HCl possessing a proton-donor function. Equations 3 and 4 roughly represent the formation of H bonds between DNA and donoracceptor groups of PMAP 3 HCl. The probability of such H bonds is particularly supported by the fact that some organoaminocyclophosphazenes in a concentrated state (e.g., in crystals) form a large variety of ordered structures via hydrogen bonds between donor and acceptor nitrogen atoms.35,36 The penetration of these groups into the DNA structure leads to local “melting” (unwinding) of some turns of the double helix even at room temperature in the course of the formation of the coacervate phase. This becomes apparent as the experimental melting enthalpy of the DNA at high PMAP 3 HCl contents (qw > 1.25) progressively decreases. At the same time, the melting temperature changes only slightly, which implies that the DNA melting events are local. The blocks of DNA in the coil conformation coexist with rather extended sequences of DNA in the helical conformation. It is probable that the equilibrium constants of the reactions represented by schemes (3) and (4) are much smaller than the equilibrium constant of the interpolyelectrolyte reaction represented by scheme (2) because the reactions represented by schemes (3) and (4) occur only at a sufficiently high concentration of PMAP 3 HCl (∼510%). Interestingly, the formation of hydrogen bonds between the polycation and base pairs of DNA was directly detected for polyplexes of poly(glycoamidoamines).37 11586

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

Figure 8. (A) Excess heat capacity functions of DNA melting at pH 4.2 and an ionic strength of 0.02: 1, a reference DNA solution (without PMAP 3 HCl); 2, the DNAPMAP complex coacervate phase at qw = 2. (B) Excess heat capacity functions of DNA melting in 0.5 M NaCl at pH 7.0: 1, a reference DNA solution (without PMAP 3 HCl); 2, the DNAPMAP 3 HCl mixture (qw = 2) obtained as a result of the dissolution of the complex coacervate phase in 0.5 M NaCl at pH 7.0.

The principal feature of interpolyelectrolyte reactions of the biopolymers is the reversibility of their conformation perturbations in the complexes. To clarify whether a case exists for the DNAPMAP complexes, we designed experiments to find the appropriate conditions for complex dissociation. The dissolution of the complex coacervate phase was initiated by the addition of NaCl up to a concentration of 0.5 M at pH 4.2 with the subsequent pH adjusting to pH 7.0. The dissolution was indicated by changes in the turbidity of the system. Figure 8 shows thermograms of DNA in the coacervate phase (qw = 2) and in solution after its dissolution. The thermograms of DNA without PMAP 3 HCl (serving as a control) are given for comparison. Bound DNA in the coacervate phase melts at a much higher temperature but with a notably lower enthalpy than does free DNA in solution (Figure 8A, curve 2). After the dissolution of the coacervate phase (0.5 M NaCl, pH 7.0), the thermogram of the renatured DNA coincides with that of the free DNA (Figure 8B). Consequently, the denaturation of DNA in the complex coacervate phase is completely reversible. Effects of PMAP 3 HCl on DNA Melting at pH 7.0 and 0.02 Ionic Strength. It was of interest to study the interaction of DNA with PMAP 3 HCl in the physiological pH range. Upon shifting pH from an acidic medium to a neutral one, the degree of ionization of PMAP 3 HCl decreases. At pH 7.0, the charge density of PMAP 3 HCl is approximately 1 charge per 4 chain units (ξ ≈ 0.6). Consequently, the interaction between DNA and PMAP 3 HCl at physiological pH is significantly weakens. Nevertheless, the calorimetric data show that PMAP 3 HCl notably affects the stability of the DNA helical conformation at pH 7.0 (Figure 9). The melting thermograms of DNA gradually change their position and profile with increasing PMAP 3 HCl content in the system. Figure 10 shows the transition parameters as a function of the ratio qw/qew, where qew is the PMAP 3 HCl/DNA equivalent weight ratio. While the transition temperature increases from one level to another one, the transition enthalpy is invariable (43.0 ( 3.7 J 3 g1) and the transition width passes through a maximum and returns to the initial value. Note that the changes in the transition temperature and enthalpy vanish above

ARTICLE

Figure 9. Excess heat capacity functions of DNA melting at pH 7.0 and an ionic strength of 0.02 {qw = 0 (1), 1 (2), 2 (3), 13.8 (4)}. The curves are arbitrarily shifted along the ordinate axis.

Figure 10. (A) Temperature, (B) enthalpy, and (C) width of DNA melting in the DNAPMAP 3 HCl mixtures of different compositions vs the ratio of total equivalent concentrations of PMAP 3 HCl and DNA at pH 7.0 and an ionic strength of 0.02 M. The vertical dashed line indicates the equivalence point of the DNAPMAP complexation. The average value of the melting enthalpy is 43.0 ( 3.7 J 3 g1. Solid lines in panels A and C are the results of the fitting of eqs 6 and 7 to the experimental data.

the equivalence point of the polyelectrolytes (qw/qew = 1; marked by a dashed arrow). In a simple approximation, we can describe these features by assuming that the complexation between DNA and PMAP 3 HCl is an exchange reaction between their equivalent weights: K

ss ½DNA 3 Na þ ½PMAP 3 HClr f ½DNA 3 PMAP þ NaCl

ð5Þ

11587

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

ARTICLE

According to this scheme, DNA is represented in the reaction mixture in a free form, [DNA 3 Na], and in the form of a stoichiometric (equivalent) complex, [DNA 3 PMAP]. In this case, the excess heat capacity of melting can be written as follows CEp ðTÞ ¼ xF CEp, F ðTÞ þ ð1  xF ÞCEp, B ðTÞ

ð6Þ

where xF = (C[DNA 3 Na])/(C[DNA 3 PMAP] + C[DNA 3 Na]) is the free DNA fraction and C[DNA 3 Na], C[DNA 3 PMAP], CEp,F(T), and CEp,B(T) are the concentrations and excess heat capacities of the melting of DNA in the free and bound states, respectively. Functions CEp,F(T) and CEp,B(T) can be determined experimentally in the absence of PMAP 3 HCl (qw = 0) and in excess PMAP 3 HCl (qw > qew), when the transition temperature and width do not already depend on the PMAP 3 HCl content in the system. The experimental curves of the excess heat capacity at qw = 0 and qw > qew are satisfactorily approximated by Gaussian functions with an equal area (Δmh = 43 J 3 g1) with parameters of σF = 3.5, Tm,F = 68 °C, σB = 3.817, and Tm,B = 74 °C. From the law of mass action for the reaction represented by scheme (5), it follows that the free DNA fraction, xF, can be expressed via the equilibrium constant K and the ratio of the total equivalent concentrations of PMAP 3 HCl and DNA, CT[PMAP 3 HCl]/CT[DNA] = qw/qew:

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ! v ! #) !" u ( u qw qw qw t K e þ 1  K K þ K e 2 þ 4 qw qew qw

ð7Þ

Figure 11. (A) Temperature, (B) enthalpy, and (C) width of DNA melting in DNAPMAP 3 HCl mixtures with different compositions vs the PMAP 3 HCl/DNA weight ratio at pH 7.0 and an ionic strength of 0.12 M. The average value of the melting enthalpy is 49.6 ( 1.1 J g1.

Equations 6 and 7 were applied to the approximation of the experimental dependence Tm(qw/qew) by using the “Maximize” and “Minerr” built-in functions of the Mathcad 14 software when assuming the equilibrium constant K to be a fit parameter. The optimal fit (the fit standard error (0.6 °C) was achieved at K = 70. Calculated curves Tm and ΔmT versus qw/qew are given in Figure 10A,C by the solid lines. It is obvious that they adequately reproduce the corresponding experimental dependences: Tm increases from one constant value to another, and ΔmT passes through a maximum. It is important that according to the proposed model for K = 70 the yield of the complex at the equivalence point (qw/qew = 1) reaches almost 90%. This fact indicates a rather high affinity of PMAP 3 HCl for DNA at pH 7.0 and an ionic strength of 0.02 despite the relatively small effects of complexation on the DNA stability. We should note that the increase in the melting temperature of DNA under these conditions is not accompanied by changes in the melting enthalpy. Consequently, the observed increase in the DNA stability due to the complexation is caused by the entropic factor, similar to that observed at pH 4.2. Thus, the contribution of the release of counterions into the DNA melting energetics is diminished by the DNAPMAP complexation. Effects of PMAP 3 HCl on DNA Melting at pH 7.0 and 0.12 Ionic Strength. At pH 7.0 and an ionic strength of 0.12, the DNA melting temperature decreases and the transition width increases linearly with increasing polycation content in the system while the melting enthalpy remains constant (Figure 11). The slopes of the dependences Tm(qw) and ΔmT(qw) are KTm = 0.19 ( 0.02 °C and KΔmT = 0.16 ( 0.02 °C, respectively. Therefore, under these conditions PMAP 3 HCl behaves as a destabilizer of the DNA native structure. Urea is known to be a DNA destabilizers38 that also causes a linear decrease in the melting temperature of the DNA double helix. The slope of the dependence of the DNA melting temperature on the molar urea concentration

is 3 K 3 L 3 mol1, which means that the melting temperature decreases by 3 °C at a urea concentration of 1 mol 3 L1. In comparable units, the slope KTm for PMAP 3 HCl is about 57 K 3 L 3 monomol1. Thus, at a PMAP 3 HCl concentration of about 1 monomol 3 L1 (∼10%39) the melting temperature of DNA may decrease to 30 °C. Thus, we observed that under certain conditions PMAP 3 HCl may be a much stronger DNA destabilizer than urea. The effect of urea and other destabilizers on the double helix is usually considered to be a consequence of the preferential binding of these compounds by the unfolded DNA form40 as a result of the formation of multiple hydrogen bonds with the donoracceptor groups of DNA.38 It is highly probable that the effect of PMAP 3 HCl on the DNA conformational stability at pH 7.0 and an ionic strength of 0.12 is of a similar origin. As already mentioned, PMAP 3 HCl has two opposite functions with respect to DNA: polyelectrolyte (cationic) and donoracceptor functions. The first one stabilizes the double helix because of the inhibition of the counterions released upon melting, and the second function, in contrast, destabilizes the helix because of the replacement of the intramolecular hydrogen bonds in the native DNA structure by the intermolecular hydrogen bonds between DNA and PMAP 3 HCl. At high ionic strength and a rather low charge density of PMAP 3 HCl (∼1 charge per 4 chain units and ξ ≈ 0.6 at pH 7.0), the polyelectrolyte function seems to be suppressed completely, and thus we observe only the manifestation of the donoracceptor function of PMAP 3 HCl— the monotonic decrease in the conformational stability of DNA with increasing PMAP 3 HCl concentration. According to Lazurkin et al.,40 the initial slopes of the dependences of the temperature and width of DNA melting on the ligand concentration are proportional to 2(p  1)/(p + 1) and

xF ¼ 1 

2ðK  1Þ

11588

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

ARTICLE

Table 1. Excess Free Energy of DNA Melting, ΔmGE(T°m), at the PMAP 3 HCl/DNA Equivalent Weight Ratio, qew pH

ionic strength

qew

T°m, °Ca

ΔmGE(Tm°), J 3 mol1

4.2

0.02

0.7

37.1

859

7.0 7.0

0.02 0.12

1.5 1.5

68.7 85.1

230 14

b

a

The melting temperature of DNA in a reference solution (without PMAP 3 HCl). b The excess free energy of melting is normalized per equivalent weight of DNA (387 and 350 g 3 mol1 at pH 4.2 and 7.0, respectively).

4[(p  1)/(p + 1)]2, respectively, where p = Kh/Kc with Kh and Kc being the binding constants of the ligand to the DNA helix and coil forms. It follows that p = (2 + (KΔmT/KTm))/(2  (KΔmT/KTm)). Using the experimental values of KΔmT and KTm, we can roughly estimate that Kc ≈ 2.5Kh for PMAP 3 HCl. This result underlines a preference of PMAP 3 HCl binding to the DNA coil form rather than to the helix form. The obtained calorimetric data on the effects of PMAP 3 HCl on the conformational stability of DNA allowed us to estimate a thermodynamic measure of these effects under different conditions. Such a measure is the excess free melting energy assigned to the interaction of DNA with PMAP 3 HCl Δm GE ðT, qw Þ ¼ Δm GðT, qw Þ  Δm GðT, 0Þ

ð8Þ

where ΔmG(T, qw) and ΔmG(T, 0) are the free energies of melting at a given PMAP 3 HCl content and without it, respectively. By definition,19   T Δm GðTÞ ¼ Δm H 1  ð9Þ Tm where T is the temperature and ΔmH and Tm are the DNA melting enthalpy and temperature under the given conditions. If we assume that T = Tm°, where Tm° is the DNA melting temperature in a reference solution (without PMAP 3 HCl, qw = 0), then the excess free energy of melting can be expressed as Δm GE ðTmo , qw Þ ¼ Δm GðTmo , qw Þ

ð10Þ

because ΔmG(T°m, 0)  0. Table 1 lists the calculated values of the excess free energy of DNA melting, normalized per equivalent of DNA in interpolyelectrolyte reactions, at the equivalent values of the PMAP 3 HCl/ DNA ratio under different conditions. Positive or negative values of the excess free energy indicate a stabilizing or destabilizing effect of PMAP 3 HCl on the DNA helical conformation, respectively. At a low ionic strength (I = 0.02), increasing pH from 4.2 (∼1 charge per 2 units of PMAP 3 HCl) to 7.0 (∼1 charge per 4 units of PMAP 3 HCl) leads to a nearly 4-fold decrease in the excess free energy of melting. Thus, the stabilizing effect of PMAP 3 HCl depends strongly on its charge density. It emphasizes the polyelectrolyte nature of this effect. The suppression of the polyelectrolyte contribution to DNAPMAP interactions at pH 7.0 and the high ionic strength (I = 0.12) cause a sign inversion of the excess free energy of melting. It becomes negative. In this case, the stability of the DNA native conformation decreases under the action of PMAP 3 HCl. This destabilization could be explained by the donoracceptor function of PMAP 3 HCl.

’ CONCLUSIONS The interpolyelectrolyte reaction between poly(methylaminophosphazene) hydrochloride and DNA results in the formation of stoichiometric and nonstoichiometric complexes. We demonstrated a strong influence of pH and ionic strength on the reaction mechanism and the DNA conformational energetics in the complexes. Depending on the magnitudes of these parameters, PMAP 3 HCl either stabilizes the DNA double helix or destabilizes it until a partial unwinding of the helix occurs. The stabilizing effect of PMAP 3 HCl dominates under the conditions of strong electrostatic interaction of the polyelectrolytes (the acidic pH values and the low ionic strength). In this case, we observed a cooperative formation of the stoichiometric interpolyelectrolyte complexes in which the helical DNA segments bound to PMAP positively charged units are entropically stabilized because of the reduced release of the territorially bound counterions of DNA upon its melting. Above the equivalence point of the interpolyelectrolyte reaction (i.e., when the reaction is complete), the destabilizing effect of PMAP 3 HCl on the DNA helix becomes apparent. With a gradual weakening of the electrostatic interaction of the polyelectrolytes (increasing pH and ionic strength), both the mechanism of the interpolyelectrolyte reaction and the DNA conformational energetics are changed. The complex formation loses its cooperativity and acquires the features of a consecutive binding reaction with the formation of complexes with variable composition. The DNA conformational stability in these complexes is controlled by the ionic strength. It decreases with increasing ionic strength. Apparently, the origin of the destabilizing effect of PMAP 3 HCl on DNA is determined by its donoracceptor properties, namely, the formation of hydrogen bonds preferably with the coil form of DNA. ’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT Financial support from the Russian Foundation for Basic Research (project 07-03-00476-a) and BGTC RAS Program “Design and investigation of new-generation macromolecules and macromolecular structures” is appreciated. We thank S. A. Potekhin for the DNA sample and T. V. Laptinskaya for assistance with the light scattering experiments. ’ REFERENCES (1) Chaubal, M. V.; Gupta, A. S.; Lopina, S. T.; Bruley, D. F. Crit. Rev. Ther. Drug Carrier Syst. 2003, 20, 295–315. (2) Luten, J.; van Steenis, J. H.; van Someren, R.; Kemmink, J.; Schuurmans-Nieuwenbroek, N. M. E.; Koning, G. A.; Crommelin, D. J. A.; van Nostrum, C. F.; Hennink, W. E. J. Controlled Release 2003, 89, 483–497. (3) Luten, J.; van Steenis, J. H.; Schuurmans-Nieuwenbroek, N. M. E.; van Nostrum, C. F.; Hennink, W. E. J. Controlled Release 2003, 87, 277–279. (4) Luten, J.; van Nostruin, C. F.; De Smedt, S. C.; Hennink, W. E. J. Controlled Release 2008, 126, 97–110. (5) Lakshmi, S.; Katti, D. S.; Laurencin, C. T. Adv. Drug Delivery Rev. 2003, 55, 467–482. (6) Kumbar, S. G.; Bhattacharyya, S.; Nukavarapu, S. P.; Khan, Y. M.; Nair, L. S.; Laurencin, C. T. J. Inorg. Organomet. Polym. Mater. 2006, 16, 365–385. 11589

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590

Langmuir

ARTICLE

(7) de Wolf, H. K.; de Raad, M.; Snel, C.; van Steenbergen, M. J.; Fens, M. H. A. M.; Storm, C.; Hennink, W. E. Pharm. Res. 2007, 24, 1572–1580. (8) Yang, Y. X.; Xu, Z. H.; Jiang, J. U.; Gao, Y.; Gu, W. W.; Chen, L. L.; Tang, X. Z.; Li, Y. P. J. Controlled Release 2008, 127, 273–279. (9) Yang, Y. X.; Xul, Z. H.; Chen, S. W.; Gao, Y.; Gu, W. W.; Chen, L. L.; Pei, L. Y.; Li, Y. P. Int. J. Pharm. 2008, 353, 277–282. (10) Tur, D. R.; Pergushov, D. V.; Babin, I. A.; Papkov, V. S.; Zezin, A. B. Polym. Sci., Part B 2009, 51, 360–366. (11) Feakins, D.; Last, W. A.; Shaw, R. A. J. Chem. Soc. 1964, 4464–4471. (12) Grinberg, V. Y.; Burova, T. V.; Grinberg, N. V.; Usov, A. I.; Tur, D. R.; Papkov, V. S. Polym. Sci., Part A 2009, 51, 390–395. (13) Grinberg, V. Y.; Burova, T. V.; Grinberg, N. V.; Dubovik, A. S.; Tur, D. R.; Usov, A. I.; Papkov, V. S.; Khokhlov, A. R. Langmuir 2011, 27, 7714–7721. (14) Doty, P.; McGill, B. B.; Rice, S. A. Biochemistry 1958, 44, 432–438. (15) Freifelder, D. Physical Biochemistry; W. H. Freeman and Company: San Francisco, 1982. (16) Giglo, E.; Pompa, F.; Ripamonti, A. J. Polym. Sci. 1962, 59, 293–300. (17) Cavalieri, L. F.; Rosenberg, B. H. J. Am. Chem. Soc. 1957, 79, 5352–5357. (18) Rungsardthong, U.; Ehtezazi, T.; Bailey, L.; Armes, S. P.; Garnett, M. C.; Stolnik, S. Biomacromolecules 2003, 4, 683–690. (19) Prigogine, I. R.; Defay, R. Chemical Thermodynamics; Longmans Green & Co.: London, 1954. (20) Chalikian, T. V.; Volker, J.; Plum, G. E.; Breslauer, K. J. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 7853–7858. (21) Rouzina, L.; Bloomfeld, V. A. Biophys. J. 1999, 77, 3242–3251. (22) Pfeil, W. Protein Stability and Folding: A Collection of Thermodynamic Data; Springer: Berlin, 1998. (23) Sukhorukov, B. I.; Moshkovskii, Y. S.; Birstein, T. M.; Lystov, V. N. Biofizika 1963, 8, 294–300. (24) Lando, D. Y.; Haroutiunian, S. G.; Kul’ba, A. M.; Dalian, E. B.; Orioli, P.; Mangani, S.; Akhrem, A. A. J. Biomol. Struct. Dyn. 1994, 12, 355–366. (25) Grosberg, A.Yu.; Nguyen, T. T.; Shklovskii, B. I. Rev. Mod. Phys. 2002, 74, 297–328. (26) Kabanov, V. A. Vysokomolek. Soed. Ser. A & B 1994, 36, 183–197. (27) Kabanov, A. V.; Kabanov, V. A. Bioconjugate Chem. 1995, 6, 7–20. (28) Izumrudov, V. A. Usp. Khim. 2008, 77, 401–415. (29) Tsuchida, E.; Abe, K. Adv. Polym. Sci. 1982, 45, 1–119. (30) Reiter, J.; Epstein, I. R. J. Phys. Chem. 1987, 91, 4813–4820. (31) Bronich, T. K.; Nguyen, H. K.; Eisenberg, A.; Kabanov, A. V. J. Am. Chem. Soc. 2000, 122, 8339–8343. (32) Anderson, C. F.; Record, M. T., Jr. Annu. Rev. Phys. Chem. 1982, 33, 191–222. (33) Record, M. T., Jr.; Anderson, C. F.; Lohman, T. M. Q. Rev. Biophys. 1978, 11, 103–178. (34) Record, M. T., Jr.; Zhang, W.; Anderson, C. F. Adv. Protein Chem. 1998, 51, 281–351. (35) Bickley, J. F.; Bonar-Law, R.; Lawson, G. T.; Richards, P. I.; Rivals, F.; Steiner, A.; Zacchini, S. Dalton Trans. 2003, 1235–1244. (36) Vorontsov, I. I.; Tur, D. R.; Papkov, V. S.; Antipin, M. Y. J. Mol. Struct. 2009, 928, 1–11. (37) Prevette, L. E.; Kodger, T. E.; Reineke, T. M.; Lynch, M. L. Langmuir 2007, 23, 9773–9784. (38) Hong, J.; Capp, M. W.; Anderson, C. F.; Saecker, R. M.; Felitsky, D. J.; Anderson, M. W.; Record, M. T. Biochemistry 2004, 43, 14744–14758. (39) The unimole weight of PMAP 3 HCl at pH 7.0 is equal to 114. (40) Lazurkin, Y. S.; Frank-Kamenetskii, M. D.; Trifonov, E. N. Biopolymers 1970, 9, 1253–1306.

11590

dx.doi.org/10.1021/la202321t |Langmuir 2011, 27, 11582–11590