Anion Interactions - The Journal of

19676

J. Phys. Chem. B 2005, 109, 19676-19680

Strength and Character of Peptide/Anion Interactions A. Chavanieu,† J. F. Guichou,† R. Prado-Gotor,‡ P. Perez-Tejeda,‡ R. Jimenez,‡ P. Lopez-Cornejo,‡ and F. Sanchez*,‡ Centre de Biochemie Structurale CNRS UMR 5048 INSERM UMR 554, UniVersite´ de Montpellier 1 29, route de NaVacelles 34090 Montpellier Cedex, France, and Department of Physical Chemistry, Faculty of Chemistry, UniVersity of SeVilla, C/Profesor Garcıá Gonza´ lez s/N 41012 SeVilla, Spain ReceiVed: March 10, 2005; In Final Form: June 17, 2005

The binding free energy of complex [Co(C2O4)3]3- to three peptides H-Lys-Gly-Lys-Gly-Lys-Gly-Lys-NH2 (P-1), H-(Lys-Gly-Lys-Gly-Lys-Gly-Lys)2-NH2 (P-2), H-(Lys-Gly-Lys-Gly-Lys-Gly-Lys)3-NH2 (P-3) and to the monomers (amino acids) forming the peptides has been obtained using the kinetics of the electron-transfer reaction between [Ru(NH3)5py]2+ and [Co(C2O4)3]3- as the probe. The polymerization of the monomers increases the negative free energy of binding and changes its character, noncooperative for the monomers and anticooperative for the peptides. This increase in the negative free energy represents a driving force for the polymerization process. The magnitude of the gain in negative free energy, as a consequence of the anticooperative character of the binding of the cobalt complex to the peptide, depends on the ratio of [complex]/ [monomers].

Introduction

Chart 1

Interactions of ions with peptides are relevant in the field of ion-assisted self-organizing molecular processes, which are of utility in the formation of surpramolecular structures and a promising tool for the design of topologically predetermined peptides and proteins.1 Recently, several groups have observed self-association and/or polymerization of monomers induced by cations.2-4 These ion-induced processes could be important in the promotion of transmembrane ion channels that would not only conduct ions, but also build up only if suitable ions are present, thus presenting a most intriguing ability to perform an ion-selective self-regulation of ion flow. Moreover, fundamental to understanding biological electron transfer is discerning the involvement of heterogeneous polypeptide environments surrounding the redox sites, particularly in electron-transfer processes coupled to conformal changes or proton transfer, such as the electron-transfer processes in photosynthetic reactions.5 As a contribution to this field, in a previous paper,6 using a kinetic approach, we obtained the binding free energy of the anionic complex [Co(C2O4)3]3- (C2O42- ) oxalate anion) to the peptide H-(Lys-Gly)10-Lys-NH2 and to the monomers (amino acids) forming the peptide. It was shown that polymerization of the monomers increases the negative free energy of binding and changes its character: noncooperative for the monomers and anticooperative for the peptide. To gain a deeper insight into this question, the interaction of [Co(C2O4)3]3- with amino acids (monomers) and different peptides containing 7, 14, and 21 monomers, respectively (Chart 1), has been studied. The objective of this work was to investigate the influence of the polymerization degree on the free energy of binding and on the anticooperativity of the anion/peptide union. As in a previous paper, a kinetic approach was employed for monitoring this union. The results of this work are presented in this paper.

H-Lys-Gly-Lys-Gly-Lys-Gly-Lys-NH2

* To whom correspondence should be addressed. Tel: 34-954 557177. Fax: 34-954557174. E-mail: [email protected]. † Universite de Montpellier. ‡ University of Sevilla.

(P-1 in the text);

H-(Lys-Gly-Lys-Gly-Lys-Gly-Lys)2-NH2 (P-2 in the text); H-(Lys-Gly-Lys-Gly-Lys-Gly-Lys)3-NH2 (P-3 in the text); K ) lysine. G ) glycine. Experimental Section The reactants [Co(C2O4)3]3- (as potassium salt) and [Ru(NH3)5py]2+ (py ) pyridine) (as perchlorate salt) were prepared following published procedures.7 The peptides containing Lys and Gly residues in a 4/3 proportion were prepared as described in a previous paper.6 Kinetic runs were carried out in the presence of different concentrations of monomers and peptides at 298.2 ( 0.1 K. The experiments were performed under pseudo-first-order conditions ([Co(C2O4)3]3- ) 3 × 10-4 mol dm-3 and [Ru(NH3)5py]2+ ) 3 × 10-5 mol dm-3). In the case of the experiments with monomers, the proportion of the two amino acids was always maintained the same as in the case of the peptides. The pH of the solutions ranged from pH ∼ 7 (water) to pH ∼ 6 (for the more concentrated peptide solutions), and it was pH ∼ 7 for the solutions containing the monomers. However, we checked in preliminary experiments that these changes in pH do not change the value of the rate constant of the reaction studied here. NMR experiments were recorder on a Bruker AMX 500 spectrometer (500.13 MHz for 1H) in D2O at 298 K. Experiments corresponding to peptide solutions, in deuterated water, with and without [Co(C2O4)3]3- showed a change in the NMR spectra consistent with an interaction between the complex and the peptides.

10.1021/jp051233l CCC: $30.25 © 2005 American Chemical Society Published on Web 10/04/2005

Peptide/Anion Interactions

J. Phys. Chem. B, Vol. 109, No. 42, 2005 19677

TABLE 1: Rate Constants for the Reaction of Ru(NH3)5py2+ + Co(C2O4)33- in the Presence of the Monomers 104 [monomer]/mol dm-3 k/s-1 0 6.3 12.6 21.0 42.0 63.0

104 [monomer]/mol dm-3 k/s-1

292 280 266 243 205 189

84.0 126 168 210 294

169 148 130 112 92.0

TABLE 2: Rate Constants for the Reaction of Ru(NH3)5py2+ + Co(C2O4)33- in the Presence of the Peptide P-1 104 [P-1]/mol dm-3

k/s-1

104 [P-1]/mol dm-3

k/s-1

0 0.45 0.90 1.26 1.80 2.40 2.88

292 244 197 179 158 113 87.0

4.50 7.20 12.6 18.0 21.0 25.2 28.8

59.1 28.3 17.0 12.8 11.7 11.3 10.2


k/s-1

104 [P-2]/mol dm-3

k/s-1

0 0.30 0.45 0.9 1.2 1.5 1.8 2.25

292 238 187 166 135 105 83.0 47.0

3.0 3.6 4.2 6.0 9.0 12 15

25.0 16.0 11.0 5.0 4.1 3.8 4.0


k/s-1

104 [P-3]/mol dm-3

k/s-1

0 0.25 0.5 0.75 1.0 1.25 1.5

292 245 191 140 98.0 52.0 40.0

2.0 2.5 3.0 4.0 6.0 10.0

13.0 5.7 1.9 1.1 0.68 0.62

Figure 1. Plot of the experimental rate constants of the reaction [Ru(NH3)5py]2+ + [Co(C2O4)3]3- vs concentration. Symbols ((O) monomers, (b) P-3, (4) P-2, (0) P-1) are experimental data, and lines are the best fit using eq 2 for the monomers and this equation and K given by eq 4 in the case of the peptides.

The results corresponding to the mixture of monomers will be considered first. These results can be fitted to the equation

k)

kf + Kkb[m] 1 + K[m]

This equation, in fact the Olson-Simonson equation,9 corresponds to the behavior expected for a two-state reactive system. That is, a system in which the reactant or substrate, S, can be found in two states in equilibrium, a free state (f) and a bound state (b), to a receptor, R. These two states react at different rates, kf and kb, respectively K S+R 9 8 S/R Vkb Vkf products products

Results and Discussion Tables 1-4 contain the results of our experiments, corresponding to the pseudo-first-order rate constant, k, of the process8 k

[Ru(NH3)5py]2+ + [Co(C2O4)3]3- 98 [Ru(NH3)5py]3+ + [Co(C2O4)3]4- (1) It is clear that efficiency of the peptides for decreasing the rate of reaction increases as the length of the peptide increases according to the data in the tables. Figure 1 gives the plot of these data. In this figure the concentrations in the abscissa axis are given in such a way that the monomers and the lower peptides can be directly compared with those of the peptide containing 21 residues. Thus, a concentration of 10-3 mol dm-3 of the mixture of the monomers means that this mixture contains 10-3 × 9 ) 9 × 10-3 mol dm-3 of Gly and 10-3 × 12 ) 1.2 × 10-2 mol dm-3 of Lys. Similarly, for a value of 10-3 in the abcisa axis, the concentrations of the peptides containing 7 and 14 amino acids residues were 3 × 10-3 mol dm-3 and (3/2) × 10-3 mol dm-3, respectively.

(2)

(3)

It is worth pointing out that, strictly speaking, eq 2 can be applied only in the case of unimolecular processes. However, as we have shown in a previous paper,10 eq 2 is still valid for a second-order process provided that only one of the reactants (the [Co(C2O4)3]3- in the present case, given the positive charge of the monomers and peptides) is partitioned between the two states. In fact eq 2 fits our data corresponding to the solutions containing the monomers with the following values of the parameters appearing in eq 2: kf ) 293 s-1, kb ) 34 s-1, and K ) 109 mol-1 dm3. It is important to say that these values have been obtained using the natural concentration of the monomers, not the concentrations appearing in Figure 1 (see below). It is also important to realize that the reactivity of the bound state is one order of magnitude lower than the reactivity of the free state. As to the results corresponding to the solutions containing the peptides, these results cannot be fitted by eq 2 (with [P] instead of [m]) unless an allowance was made for a variation of K with the ratio between the concentrations of the [Co(C2O4)3]3- anion and the peptide. However, since the

19678 J. Phys. Chem. B, Vol. 109, No. 42, 2005

Chavanieu et al.

TABLE 5: Values of the Best Fit Parameters for Eq 2 in the Case of Monomers and for Eqs 2 and 4 in the Case of Peptides monomers P-1 P-2 P-3 a

Kmaxa

ha

ja

kfa

kba

109 3.4 × 104 1.3 × 106 4.4 × 106

7.3 × 10-4 9.8 × 10-4 4.5 × 10-4

0 3.5 × 10-4 1.8 × 10-4 6.6 × 10-5

293 282 280 289

34 7.9 3.0 0.6

kf, s-1; kb, s-1; j, mol dm-3; h, mol dm-3; Kmax, mol-1 dm3.

concentration of this reactant is a constant in our experiments, K depends only on the concentration of the peptide. At first, this dependence is unknown. However to have a physical meaning K must show a saturation behavior, that is, it must reach a constant value after a given concentration of the peptide. A dependence of K accomplishing this requirement, frequently found in biological systems, is given by eq 4 which corresponds to a sigmoidal dependence11

K)

Kmaxet 1 + et

(4)

In this equation t ) ([P] - h)/j, Kmax is the maximum (limiting) value of K, h is the value of the concentration of the peptide, [P], for which K ) (1/2)Kmax, and j is an adjustable parameter (see below). The values of Kmax, h, j, kf, and kb for the three peptides used here are given in Table 5. Notice that, according to the value of kb, the reaction of the [Co(C2O4)3]3- bound to the peptides with the ruthenium complex is much slower than that in the bulk, as expected, taking into account that the peptide bears a positive charge, as does the ruthenium complex, in such a way that the encounter between the peptide bound cobalt complex and the ruthenium complex will be difficult. The origin of this can be seen clearly if kb is expressed as

kb ) ko exp(-zRuRFnψ/RT)

Figure 2. Plot of ln kb for the different peptides vs the number of monomers contained in the peptide.

(5)

This equation follows from a decomposition of the activation free energy in an electrostatic and nonelectrostatic part,12 that is, ∆Gq ) ∆Gqne + ∆Gqelec. In eq 5

ko ) exp(-∆Gqne/RT)

(6a)

∆Gqelec ) zRuRFnψ

(6b)

and

In eq 6b, zRu is the charge of the ruthenium complex, n is the number of monomers in the peptide, Ψ is the potential originated by each monomer of the peptide, and R gives the fraction of the potential influencing kb. According to eq 5, a plot of ln kb vs n should be linear. Figure 2 gives this plot, supporting our interpretation. On the other hand, the fact that the values of K, for the three peptides, increase when the [Co(C2O4)3]3-/[P] decreases means that the union of the cobalt complex and the peptides is anticooperative. This fact implies that, in comparison with the monomers, the peptides not only change the strength of binding but also change its character: it is noncooperative in the case of the monomers and anticooperative in the case of the peptides.13 This anticooperative character has also been observed in the case of the binding of small ions to DNA14 and dendrimers.15 The quantification of the anticooperative character of the union peptide/cobalt complex can be done through the values of the parameter j in Table 5. Notice that j measures the

Figure 3. Plot of ln Kmax for the different peptides vs the number of monomers contained in the peptide.

degree of anticooperativity, according to eq 4: for j ) 0, K ) Kmax for all the concentrations, and thus, noncooperativity appears. Thus, anticooperativity increases as j does. Anticooperativity does not increase when the number of monomers in the peptide does according to the values of this parameter appearing in Table 5. In this regard it is interesting to note that ln Kmax, that is, the free energy of binding corresponding to Kmax for the processes in eq 3, changes as represented in Figure 3; that is, there is an asymptotic increase of the free energy. The fact that the plots of ln kb and ln Kmax vs the number of monomers are different implies that in the binding of the cobalt complex and in the repulsion of the ruthenium complex different factors are in play. The electrostatic factor, acting on the ruthenium complex, will do the same in the case of the cobalt

Peptide/Anion Interactions

J. Phys. Chem. B, Vol. 109, No. 42, 2005 19679

complex, but in the latter case other factors appear, adding other components of free energy to the electrostatic free energy. These factors are precisely responsible for the origin of anticooperativity. In fact this sheds some light on the origin of the anticooperativity. Of course, a part of this anticooperativity arises as a consequence of the fact that, when one complex is bound, a second complex would feel repulsion from the first one bound. However, other causes of anticooperativity cannot be ruled out. Thus, the binding of a cobalt complex, with a charge sign opposite the charges on the peptide, would produce a screening between the charges on a given peptide molecule, thus allowing a more compact conformation of the receptor, with different binding properties from the more extended conformation in the absence of the cobalt complex. In other words, anticooperativity would be a consequence of a change in the equilibrium between different conformations of the peptide, in such a way that, after the union with the first cobalt complex, polymer conformations less favorable for the binding would be promoted.16 It is also possible that the anticooperative character of the binding of the cobalt complex was, in some sense, a reflection of the cooperativity of the intramolecular peptide hydrogen bonds which, as is well-known, play an important role in organization, assembly, and molecular recognition processes.17 That is, the union of the cobalt complex to the peptide through the hydrogen bond would imply a reduction in the number of intramolecular hydrogen bonds of the peptide. This would produce a conformational change in the peptide and, thus, a variation of the binding constant. Notice that this explanation of the anticooperative character of the cobalt complex/peptide union, based on different effects that do not act in the same direction, is in agreement with the fact that cooperativity, as measured by the j parameter, does not increase monotonically with the number of units in the peptide. Now, a different aspect of the interactions of the peptides and monomers with the anionic cobalt complex will be considered. This aspect has to do with the field of ion-assisted self-organizing and/or polymerization molecular processes.2-4 To clarify this point we will refer to Figure 4. It is seen in this figure that the free energy of polymerization in the presence and in the absence of cobalt complex can be established as follows: When the cobalt complex is absent, the free energy of polymerization corresponds to the process

n monomers f peptide + (n - 1)H2O,

∆Gpolymerization (7)

n ) 7, 14, 21 in our case and in the presence of cobalt complex

n monomers/[Co(C2O4)3]3- f peptide/[Co(C2O4)3]3- + (n - 1)H2O,

∆G′polymerization (8)

where n monomers/[Co(C2O4)3]3- and peptide/[Co(C2O4)3]3represent the n monomers, with only one of them bound to the cobalt complex, and the peptide bound to this complex, respectively. The difference in free energy for the species on the right-hand side of eqs 7 and 8 is, of course, dependent on the ratio [Co(C2O4)]3-/[P]. The most favorable value for this difference corresponds to the values of Kmax appearing in Table 5. These values of the free energies are -25.9, -34.8, and -37.7 kJ/mol for n ) 7, 14, and 21, respectively. On the other hand, the difference in the free energy of species on the left-hand side of eqs 7 and 8 cannot be obtained using

Figure 4. Schematic diagram showing that polymerization is more favorable in the presence of the [Co(C2O4)3]3- anion. ∆Gm and ∆Gp are, respectively, the free energies corresponding to the binding of the anion to the n monomers (see text) and polymer. ∆Gpolymerization and ∆G′polymerization are the free energies of the polymerization processes in the absence and in the presence of the anion, respectively.

the value of K corresponding to the monomers in Table 5, because the value of K in this table corresponds to process 9

monomer + [Co(C2O4)3]3- f monomer/[Co(C2O4)3]3- (9) which, of course is different from the processes in eq 10

n monomers + [Co(C2O4)3]3- f n monomer/[Co(C2O4)3]3-,

∆Gm (10)

However the free energies corresponding to eq 10, ∆Gm (see Figure 4), can be obtained easily: It suffices to express the monomer concentration in units of 7, 14, and 21 monomers, respectively (in fact, for n ) 21, this unit has been used in Figure 1). Using these values for [m], the values of K for eq 10 are 7.6 × 102, 1.5 × 103, and 2.3 × 103 mol-1 dm3, respectively, and the corresponding free energies, ∆Gm, -16.4, -18.1, and -19.2 kJ mol-1, respectively. Now, according to Figure 4

∆G′polymerization + ∆Gm ) ∆Gpolymerization + ∆Gp

(11)

∆G′polymerization - ∆Gpolymerization ) ∆Gp - ∆Gm

(12)

or

where ∆Gp represents the free energy of interaction peptide/ complex. The differences on the right-hand side of eq 12 are, thus -9.5, -16.7, and -18.5 kJ mol-1 for n ) 7, 14, 21, respectively. Thus, polymerization, in the presence of the cobalt complex, is more favorable, from a thermodynamic point of view, for all three peptides. This explains the effect of ions on the polymerization processes mentioned in the Introduction. It is interesting to note that the values of ∆Gp - ∆Gm and, thus, the values of ∆G′polymerization - ∆Gpolymerization (see eq 12), that is, the thermodynamic efficiency of the ion for promoting polymerization, depend on the relation of concentrations of the cobalt complex and the monomers. This is a consequence of the anticooperative character of the binding of the cobalt complex to the peptides which are the products of the polymerization process. In the limit of zero concentration of monomers these efficiencies decrease, in terms of free energy, to -4.04, -3.23, and -1.76 for n ) 7, 14, and 21, respectively. In fact this trend of the free energy is the opposite of that found when the concentration of

19680 J. Phys. Chem. B, Vol. 109, No. 42, 2005 the monomers is high enough to produce peptides to a concentration corresponding to Kmax. Notice that this would permit the selection of the peptide produced in the polymerization by changing the proportion of the monomers/cobalt complex: When this ratio is low, the short polymer would be favored and the longer polymer would be produced at the higher monomer concentrations. In conclusion, the binding free energy of the complex [Co(C2O4)3]3- to three peptides has been studied. In comparison with the monomers, the peptides not only change the strength of binding but also change its character: It is noncooperative in the case of the monomers and anticooperative in the case of the peptides. The degree of anticooperativity is not proportional to the number of monomers which implies that more than one factor determines noncooperativity. Finally, it is interesting to point out that, in some sense, the induction of polymerization by the presence of the ligand (the cobalt complex in the present case) is reminiscent of some enhancement of aggregation phenomena observed in other supramolecular aggregates. Thus, in the case of the micellization process of ionic surfactants, an analogue phenomenon would be the enhancement of micellization in the presence of ions of opposite sign from that of the surfactant, as indicated by an increase in the aggregation number and a decrease in the critical micellar concentration.18 Acknowledgment. This work was financed by the D.I.G.Y.T. (BQU-2002 01063) and the Consejerıá de Educacioń y Ciencia de la Junta de Andalucıá.

Chavanieu et al. References and Notes (1) Ghadiri, M. R.; Soares, C.; Choi, C. J. Am. Chem. Soc. 1992, 114, 4000. (2) Petijean, A.; Cuccia, L A.; Lehn, J. M.; Nierengarten, H.; Schmutz, M. Angew. Chem., Int. Ed. 2002, 41, 1195. (3) Takeuchi, D.; Anada, K.; Osakada, K. Angew. Chem., Int. Ed. 2004, 43, 1233. (4) Shan, N.; Vickers, S. J.; Adams, H.; Ward, H. D.; Thomas, J. A. Angew. Chem., Int. Ed. 2004, 43, 3938. (5) Utschig, L. M.; Thurnaner, M. C. Acc. Chem. Res. 2004, 37, 439. (6) Ortiz, J.; Guichou, J. F.; Chavanieu, A.; Sańchez, F.; Prado-Gotor, R. Chem. Phys. Lett. 2004, 384, 266. (7) (a) Cannon, R. D.; Stillmann, J. Inorg. Chem. 1975, 14, 2207. (b) Creutz, C.; Taube, H. J. Am. Chem. Soc. 1973, 95, 1086. (8) Equation 1 is the rate determining step for the studied reaction. This step is followed by the decomposition of the cobalt complex: [Co(C2O4)3]4- f 3C2O42- + Co2+. (9) Olson, A. R.; Simonson, J. R. J. Phys. Chem. 1949, 17, 1167. (10) Lopez-Cornejo, P.; Sańchez, F. J. Phys. Chem. B 2001, 105, 10523. (11) Hammes, G. G. Thermodynamics and Kinetics for the Biological Sciences; Wiley-Interscience: New York, 2000; p 124 and ff. (12) Laidler, K. J. Chemical Kinetics; McGraw-Hill: London, 1965. (13) McGhee, J. D.; Von Hippel, P. H. J. Mol. Biol. 1974, 86, 469. (14) Secco, F.; Venturini, M.; Lo´pez, M.; Pe´rez, P.; Prado, R.; Sańchez, F.; Phys. Chem. Chem. Phys. 2001, 3, 4412. (15) Lo´pez-Cornejo, P.; Pe´rez, P.; Garcıá, F.; de la Vega, R.; Sańchez, F. J. Am. Chem. Soc. 2002, 124, 5154 and references therein. (16) Lo¨wik, D. W. P. M.; Garcia-Hartjes, J.; Meijer, J. T.; van Hest, J. C. M. Langmuir 2005, 21, 524. (17) Casarati, A.; Sansone, F.; Ungaro, R. Acc. Chem. Res. 2003, 36, 246. (18) See, for example, Aswal, V. K.; Goyal, P. S. Chem. Phys. Lett. 2002, 364, 44.

Anion Interactions - The Journal of

Recommend Documents