5740
J. Phys. Chem. B 2007, 111, 5740-5747
Is the Peptide Bond Formation Activated by Cu2+ Interactions? Insights from Density Functional Calculations A. Rimola,† L. Rodrı´guez-Santiago,*,† P. Ugliengo,‡ and M. Sodupe*,† Departament de Quı´mica, UniVersitat Auto` noma de Barcelona, Bellaterra 08193, Spain, and Dipartimento di Chimica IFM, UniVersita` di Torino and NIS, Nanostructured Interfaces and Surfaces, Centre of Excellence, Via P. Giuria, 7, 10125 Torino, Italy ReceiVed: February 7, 2007; In Final Form: March 27, 2007
The catalytic role that Cu2+ cations play in the peptide bond formation has been addressed by means of density functional calculations. First, the Cu2+-(glycine)2 f Cu2+-(glycylglycine) + H2O reaction was investigated since mass spectrometry low collision activated dissociation (CAD) spectra of Cu2+-(glycine)2 led to the elimination of a water molecule, which suggested that an intracomplex peptide bond formation might have occurred. Results show that this intracomplex condensation is associated to a very high free energy barrier (97 kcal mol-1) and reaction free energy (66 kcal mol-1) because of the loss of metal coordination during the reaction. Second, on the basis of the salt-induced peptide formation theory, the condensation reaction between two glycines was studied in aqueous solution using discrete water molecules and the conductor polarized continuum model (CPCM) continuous method. It is found that the synergy between the interaction of glycines with Cu2+ and the presence of water molecules acting as proton-transfer helpers significantly lower the activation barrier (from 55 kcal/mol for the uncatalyzed system to 20 kcal/mol for the Cu2+ solvated system) which largely favors the formation of the peptide bond.
Introduction Amino acids are held together by peptide bonds, whose successive linkages lead to polypeptides and proteins. Peptide bond formation is a process of paramount importance not only in protein chemistry but also in the context of current theories of polymerization of amino acids under prebiotic conditions.1-3 The condensation reaction between two R-amino acids occurs via a nucleophilic attack of the NH2 group of one amino acid to the carbon atom of the carboxylic group belonging to the second amino acid. This step is followed by the elimination of one water molecule with formation of the HN-CdO peptide bond. The uncatalyzed reaction between glycines has been theoretically studied and has been found to follow two possible routes: a concerted or a stepwise mechanism.4 The two processes are close in energy, with free-energy barriers around 50-55 kcal/mol and isoergonic reaction energies. The catalysis of the peptide bond formation has also been addressed in several works both from experimental and computational points of view. The catalysts reported can be mainly grouped in (1) protic solvents,5-10 such as water or ammonia; (2) mineral oxides surfaces,10-30 such as silicas, clays, or aluminosilicates; and (3) metal dications,29,31-35 such as Mg2+, Ca2+, Zn2+, Cu2+, or Ni2+. Among these catalysts, particularly interesting are metal cations since they have been suggested long ago to activate amino acids toward the nucleophilic attack. In particular, the salt-induced peptide formation (SIPF) theory, proposed by Rode and co-workers,33,34,36-38 invokes Cu2+ to play a major role under conditions that resemble those of the primitive earth. The mechanism proposed for this condensation * To whom correspondence should be addressed. E-mail: mariona@ klingon.uab.es (M. S.);
[email protected] (L. R.-S.). † Universitat Auto ` noma de Barcelona. ‡ Universita ` di Torino and NIS.
reaction in aqueous solution is based on the formation of a monochlorocuprate complex with two amino acids that coordinate the metal cation in the presence of high concentrations of NaCl. From such a complex, the reaction leads to the formation of the peptide bond through an intracomplex condensation between the two amino acids. This process is in principle activated and is favored as a consequence of the coordination of the amino acids to the metal dication. Since the presence of water in excess will inevitably favor the hydrolysis rather than the formation of the peptide bond, high concentrations of Na+ ions are needed to act as a dehydrating agent. On the other hand, an experimental study in gas phase using soft electrospray ionization mass spectrometry techniques showed that low-energy collision-activated dissociation spectra of Cu2+-(glycine)2 complex led to the elimination of a water molecule without loss of charge.39 On the basis of these spectra and of the previously proposed SIPF mechanism, the authors suggested that an intracomplex peptide bond formation might have occurred. Efforts toward understanding the role of metal cation interactions with early biomolecules, in particular Cu2+, in the formation of the first biopolymers are, thus, of great importance in the field of prebiotic chemistry. Nonetheless, to our knowledge, there are no recent works that analyze the role of Cu2+ as catalyst on the peptide bond formation from a mechanistic point of view. In this work, we present a quantum chemical study that addresses this point. We have considered both the naked and hydrated metal cation interacting with two glycine amino acids as starting structures to study the mechanism of the intracomplex peptide bond formation. The aim of this work is to explore the feasibility of this process both in gas phase and in solution.
10.1021/jp071071o CCC: $37.00 © 2007 American Chemical Society Published on Web 05/01/2007
Peptide Bond Formation and Cu2+ Interactions
Figure 1. BHLYP free-energy profile [kcal mol-1] of the uncatalyzed concerted peptide bond formation between two isolated glycine (Gly) molecules. Relative free energies refer to the 2 Gly reference state. (Bond lengths in Å).
Computational Details Recently, some theoretical works of open-shell Cu2+-ligand systems have demonstrated that functionals such as BHLYP with larger percentages of exact exchange (50%) than the popular B3LYP (20%) better compare to the highly correlated CCSD(T) method.40-42 This is because that when GGA or hybrid functionals with small percentages of exact exchange are used to study radicals, they overstabilize delocalized situations as a result of a bad cancellation of the self-interaction part by the exchange-correlation functional. In particular, it was shown that B3LYP relative energies between ground and low-lying states of Cu2+-H2O system differed significantly from BHLYP and CCSD(T) results43 because of an overdelocalization of the spin density with B3LYP. Similarly, significant differences were observed between B3LYP and CCSD(T) methods when analyzing the relative stability of Cu2+-(glyoxilic acid oxime) complexes with different coordination modes and, thus, different spin density distribution.42 Furthermore, binding energies computed with the B3LYP hybrid functional were found to be significantly larger than those computed with BHLYP or CCSD(T).42,43 Nevertheless, for the Cu2+-(glycine)2 complex, for which we have also performed calculations at the B3LYP and single points at the CCSD(T) level (correlating all valence electrons), results have shown that reaction energy values are
J. Phys. Chem. B, Vol. 111, No. 20, 2007 5741
Figure 3. BHLYP free-energy profile [kcal mol-1] corresponding to the peptide bond formation reaction in the gas phase with Cu2+. Relative free energies refer to the CuG2-1 reference state. (Bond lengths in Å.)
very similar with all methods, the largest differences being about 3 kcal/mol. For the energy barriers, differences are somewhat larger, BHLYP values being about 5-6 kcal/mol larger than with B3LYP, whereas CCSD(T) values lie in between. Because of that, all geometry, frequency, and energy calculations in the present study have been carried out with the BHLYP functional.44,45 The basis sets used are described as follows. The Cu2+ cation basis is derived from (14s9p5d)46 primitive set of Watchers supplemented with one s, two p’s, and one d diffuse functions47 as well as one f polarization function, the final contracted basis set being (15s11p6d1f)/[10s7p4d1f]. For C, N, O, and H atoms, we have used the 6-31++G(d,p) standard basis set. All calculations have been performed using the Gaussian03 package programs.48 Structures have been characterized by the analytical calculation of the harmonic frequencies as minima or saddle points. In some intriguing cases, we have carried out intrinsic reaction coordinate (IRC) calculations at the same level of theory to corroborate that a given transition state connects the expected reactants and products. Thermochemical corrections to the energy values have been computed using the standard rigid rotor/harmonic oscillator formulas.49 Solvent effects have been introduced using the conductor polarized continuum model (CPCM),50 which is an implementation of the conductor-like screening solvation model (COSMO)51 in Gaussian03.
Figure 2. BHLYP optimized geometries for the Cu2+-(glycine)2 isomers. Relative free energies in kcal mol-1. Distances in Å.
5742 J. Phys. Chem. B, Vol. 111, No. 20, 2007
Rimola et al.
Figure 4. BHLYP free-energy profiles [kcal mol-1] corresponding to the elimination of water in the gas phase: path A starting from CuG2-4 and path B starting from CuG2-1. Relative free energies refer to the CuG2-1 reference state. (Bond lengths in Å.)
Figure 5. BHLYP free-energy profile [kcal mol-1] corresponding to the elimination of water in the gas phase in the presence of one background water molecule. Relative free energies refer to the CuG2-1+ H2O reference state. (Bond lengths in Å.)
Manipulation, visualization, and preparation of structures have been dealt with MOLDRAW program, whereas graphical rendering of the pictures has been made by the POVRAY program using the input files prepared by MOLDRAW.52 Results and Discussion First, the results obtained for the uncatalyzed concerted reaction between two isolated glycines will be shown. Next, inspired by the available experimental data,39 we will present the results obtained for the Cu2+-(glycine)2 gas-phase system. Once its intrinsic reactivity is analyzed, that is, without complicating factors such as solvation, the peptide bond formation between two glycines coordinated to a hydrated Cu2+ cation and also in the presence of water molecules will be examined to explore the reactivity in aqueous solution.
Uncatalyzed Gas-Phase System. The results for the uncatalyzed concerted reaction between two glycines (2NH2CH2COOH f NH2CH2CONHCH2COOH + H2O) are reported in Figure 1. The activation and reaction free energy of this process are 55.4 kcal/mol and -1.1 kcal mol-1, respectively, in good agreement with previous studies.4 The energy barrier is 3 kcal mol-1 larger, a fact which is not surprising since the BHLYP includes more percentage of exact exchange. The transition structure is characterized by a simultaneous N-C bond formation and a proton transfer through a constrained fourthmembered ring, giving rise, consequently, to a significant highenergy barrier. Cu2+ Gas-Phase System. Before tackling the gas-phase reactivity, the structure of the complex generated in the source must be first elucidated by analyzing the different modes of
Peptide Bond Formation and Cu2+ Interactions
Figure 6. BHLYP optimized geometries (bond length in Å) of the structures obtained for the Cu2+-(glycine)2(H2O)2 complex. Italics bond distances for CuG2W2-1 correspond to experimental values of the bis(glycinato)copper(II) complex. Bare energy values refer to the free relative energies of the Cu2+-(glycine)2(H2O)2 complexes. Value in bracket is the free energy of reaction of CuG2W2-1 + H2O f CuG2W3, in kcal mol-1.
coordination. We have considered four possible isomers, which are represented in Figure 2. For all cases, the open shell corresponds to an in-plane d orbital of the metal cation. In the lowest energy isomer (CuG2-1), the glycine ligands are in trans disposition and coordinate to the metal cation through the carbonyl oxygen and the amino nitrogen. The cis isomer (CuG23) is found to lie 3.4 kcal/mol above. Since zwitterionic forms of amino acids are stable in gas phase when interacting with metal cations,43,53,54 their stability increasing the higher the charge of the cation is,55,56 we have also considered an isomer derived from the interaction of the metal cation with the COOgroup belonging to one zwitterionic glycine (CuG2-2), which has been found to lie 1.5 kcal/mol above CuG2-1. This result indicates that the charge of the metal is significantly screened as a consequence of the interaction with the other glycine ligand, and so the interaction of the zwitterionic form with Cu2+ becomes less stable than that of its neutral form, in agreement with previous works.42,57,58 Since this structure is less stable than CuG2-1, the interaction of two zwitterionic glycines would be even more unstable than CuG2-2, and it has not been considered. Finally, CuG2-4 in which Cu2+ is coordinated by the NH2 and the OH groups of one glycine is 21.7 kcal/mol high in energy with respect to CuG2-1. Despite the oxidative character of Cu2+ described in the literature for coordinatively unsaturated Cu2+ complexes,59-61 in the present systems the spin density at the metal cation is always around 0.80, in agreement with what is usually found in four-coordinated square planar systems. The most stable form of Cu2+-(glycine)2, CuG2-1, has been taken as the starting complex for the peptide bond formation in gas phase. As shown in Figure 3, the free-energy barrier of this
J. Phys. Chem. B, Vol. 111, No. 20, 2007 5743 process (97.2 kcal/mol) is much larger than that obtained for the uncatalyzed case (55.4 kcal/mol), whereas the final products Cu2+-(glycylglycine) (CuGG) + H2O lie 65.6 kcal/mol above CuG2-1. This process exhibits a high activation energy, which can be reasonably understood by examining the transition structure. To allow the nucleophilic attack of the amine nitrogen, the NH2 group needs first to decoordinate from the metal cation, which induces instability of TS1-CuG2 with respect to CuG21. Moreover, the transfer of one H atom from the attacking NH2 to OH implies the formation of a highly strained fourmember ring. The loss of coordination to Cu2+, from square planar in the reactant to trigonal in the product, is also responsible for the high value obtained for the reaction free energy (65.6 kcal/mol). As mentioned above, collisionally activated dissociations of doubly charged Cu2+-(glycine)2 complexes, generated in electrospray ionization (ESI) mass spectrometry experiments, led to the formation of a product ion [Cu2+(glycine)2-H2O] in which a water molecule has been lost without charge reduction, which suggested the occurrence of an intracomplex condensation reaction between the two glycines of the ion.39 Overall, our results indicate that the loss of water from low-collisionactivated dissociation spectra of Cu2+-(glycine)2 at room temperature does not arise from an intracomplex condensation reaction. The question is thus which mechanism enables the loss of water in low collision conditions? The elimination of water from transition metal-amino acid systems is generally observed, and it has been traditionally invoked to occur via insertion of the metal cation into the C-C or C-OH backbone bonds of the amino acids.62-69 Because of that, these mechanisms were theoretically explored, and the results are reported in Figure 4. All attempts to find a path involving the insertion of Cu2+ into the C-C bond for the elimination of water failed. In fact, when simulating the insertion of the metal cation by enlarging the C-C bond, the final products corresponded to the loss of +H NdCH ion, which was also observed in the experimental 2 2 collision-activated dissociation (CAD) spectra. Instead, a Cu2+ insertion into the C-OH bond from CuG2-4 (see path A of Figure 4) yielded to the formation of a water molecule through TS2-CuG2 with an overall energy barrier of 64.2 kcal/mol. In addition to the metal bond insertion, this process also involves a hydrogen transfer from the CH2 to the OH, converting glycine in H2O + NH2-CHdCdO. Such a transfer is not surprising since the C-OH cleavage is highly heterolytic, the OH acquiring an important negative charge and the NH2CH2CO fragment acquiring a positive charge, which increases the CH acidity. The direct cleavage of water from I1-CuG2 leads to the Cu(G2-W)1 product ion (66.4 kcal/mol above CuG2-1), in which the amino-ketene coordinates Cu2+ through the N and C atoms. This pathway presents an overall energy barrier that is 33.0 kcal/ mol lower than that computed for the peptide bond formation in the previous scheme (see Figure 3), thus indicating that at least another more favorable process leading to the loss of water exists in the collision cell. The loss of water can also be produced from the most stable isomer CuG2-1 through a mechanism that involves a 1,3 hydrogen transfer from CH2 to the OH group (see path B of Figure 4). However, the activation energy associated to this path is 86.5 kcal/mol, around 20 kcal/ mol higher than that of path A, and thus, it would not take place. Finally, since background water (BW) was observed in the collision cell, in such a way that the Cu2+-(glycine)2(H2O) complex was formed,39 we have also considered another decomposition mechanism that leads to the elimination of H2O
5744 J. Phys. Chem. B, Vol. 111, No. 20, 2007
Rimola et al.
Figure 7. BHLYP Gibbs free energies [kcal mol-1] of the peptide bond formation reaction for two glycine molecules (solid gray line), intracomplex condensation starting from CuG2W3 (dashed green lines), water-assisted intracomplex condensation starting from [CuG2W3]wa (dashed-dotted red lines), and double water-assisted intracomplex condensation starting from [CuG2W3]wa2 (dashed-double-dotted purple lines). In parenthesis, the values are computed using the CPCM solvation model.
in the presence of discrete water molecules. Figure 5 shows the free-energy profile for this elimination after forming the Cu2+-(glycine)2(H2O) adduct ([CuG2]bw), in which the water is in the apical position of the complex. As a first step, this water molecule decoordinates from the metal center and acts as an assistant in the proton transfer from NH2 to the OH group. This transfer takes place through a barrier of 53.7 kcal/mol above CuG2-1 + H2O and leads to the formation of the intermediate [I1-CuG2]bw, in which the H3O+ hydronium ion is formed. Because direct cleavage of water cannot proceed from [I1CuG2]bw, a C-C bond breaking takes place through the [TS2CuG2]bw structure, which has an energy 66.9 kcal/mol higher than CuG2-1. Finally, direct elimination of two water molecules from [I2-CuG2]bw occurs, giving rise to the Cu(G2-W)2 product, which presents a square planar coordination in which the metal cation interacts with glycine in a bidentate manner with CO (through the O atom) and with NHdCH2 (through the N atom). This Cu(G2-W)2 + H2O asymptote lies 46.7 kcal/ mol above CuG2-1. A slightly more stable isomer of the final copper complex, Cu(G2-W)3, in which the CO coordinates through the C atom, has also been found. Although this mechanism leads to a much more stable square-planar product (Cu(G2-W)3), compared to those shown in Figure 3 and Figure 4 (CuGG and Cu(G2-W)1, respectively), the overall energy barrier (66.9 kcal/mol) is very similar to that obtained for path A in Figure 4 (66.4 kcal/mol) and, thus, still would not explain the formation of [Cu2+-(glycine)2-H2O] at low collision activation energies. Accordingly, the loss of water from Cu2+-(glycine)2 is rather puzzling, and other explanations for the experimental observations should probably be considered (see below). Reaction in Aqueous Solution. It is well-known that solvation can largely affect the reactivity of any process, particularly when metal cations are involved, because solvent molecules can participate in the coordination sphere of the cation. In the present section, we will analyze the peptide bond
formation by including discrete water molecules as well as the continuous CPCM solvation model. The starting complex is the Cu2+-(glycine)2(H2O)2 system, for which four isomers have been localized. Optimized structures are reported in Figure 6. The most stable one is CuG2W2-1, whose glycine ligands are in the equatorial plane and in trans disposition, the apical metal-ligand distances being 0.4 Å larger than the equatorial ones, because of the Jahn Teller distortion. This structure is in agreement with an X-ray absorption spectroscopy study for the bis(glycinato)copper(II) complex in aqueous solution,70 the differences in the Cu2+-ligand bond distances being around 0.05 Å at the most. The remaining three isomers CuG2W2-2, CuG2W2-3, and CuG2W2-4 lie 2.4, 5.2, and 6.1 kcal/mol above CuG2W2-1. The glycine ligands of these isomers are not disposed in a coplanar fashion and they differ basically in the relative position of the donor groups: CO in trans and NH2 in cis disposition (CuG2W2-2), CO in cis and NH2 in trans disposition (CuG2W2-3), and both CO and NH2 in cis disposition (CuG2W2-4). The nucleophilic attack of the NH2 group of one glycine to the other one implies changes on the coordination sphere of the metal. As mentioned, in gas phase such changes led to a loss of coordination that gave rise to a high-energy barrier (see Figure 2). Nevertheless, in aqueous solution the vacancy originated upon decoordination of the NH2 and can be occupied by a water molecule arising from the solvent. According to that, CuG2W3 has been found as the prereactant complex for the peptide bond formation. As shown in Figure 6, the NH2 group of one glycine in CuG2W3 is not interacting with the metal whereas three water molecules are coordinated to Cu2+. The free energy of the CuG2W2-1 + H2O f CuG2W3 reaction is 4.5 kcal/mol. Figure 7 shows the energy barriers and reaction energies for the different peptide bond formation processes considered, and Figure 8 summarizes the geometries of the stationary points involved. As previously mentioned, the condensation between
Peptide Bond Formation and Cu2+ Interactions
J. Phys. Chem. B, Vol. 111, No. 20, 2007 5745
Figure 8. BHLYP optimized geometries relative to the reactions involved in Figure 7. Bond distances in Å.
two neutral glycines in gas phase possesses an activation free energy of 55.4 kcal/mol (TS-G2). On the other hand, the peptide bond formation starting from the CuG2W3 complex and occurring via TS-CuG2W3 is associated to an energy barrier of 49.7 kcal/mol, which implies a decrease of around 6 kcal/ mol compared to the uncatalyzed reaction. Thus, the reaction is slightly activated by the interaction of the CO group with Cu2+. This activation toward the nucleophilic attack also manifests itself in the geometry of the transition structure, since the N-C distance of the forming bond in TS-CuG2W3 is shorter (1.51 Å) than in the uncatalyzed reaction (1.56 Å) and the H of the NH2 is much more transferred to the OH group in TS-CuG2W3 (H‚‚‚O distance is 1.17 Å) than in TS-G2 (H‚‚‚O distance is 1.41 Å). Despite that, the decrease in the energy barrier is not as large as one would expect from the enhancement
of the electrophilic character of the carbon atom by the interaction of the CO with the metal cation, as was observed in the presence of other Lewis sites.10 This is mainly because the intracomplex nucleophilic attack implies a lengthening of the Cu-O distance of the attacking glycine from 2.32 Å in CuG2W3 to 2.54 Å in TS-CuG2W3 and, so, there is a loss of stabilizing interaction. In addition, the loss of the hydrogen bond in CuG2W3 during the condensation process may also contribute to this minor decrease on the energy barrier. In TS-CuG2W3, the water molecules are coordinated to Cu2+, just exerting a screening effect on the metal cation. They are simple spectators that do not take part in the reaction. However, it is well-known that protic solvents can act as potential protontransfer helpers.71-73 Accordingly, reactions in which a water molecule assists the hydrogen transfer from NH2 to OH groups
5746 J. Phys. Chem. B, Vol. 111, No. 20, 2007 have also been explored. The water-assisted reaction starting from [CuG2W3]wa and proceeding through [TS-CuG2W3]wa results in an activation energy of 38.3 kcal/mol, which means a significant decrease (around 11 kcal/mol) on the energy barrier with respect to TS-CuG2W3. This is mainly because the presence of the added water allows for a smaller distortion of the CuG2W3 complex upon producing the nucleophilic attack, that is, the Cu-OC distance of the attacking glycine is now 2.43 Å instead of 2.54 Å, and also because the water molecule acts as a proton transport catalyst, the proton from NH2 being transferred to the assistant water molecule in [TS-CuG2W3]wa; that is, the water assistant molecule becomes a hydronium ion in the transition structure. Because such ions are largely stabilized by the interaction with water molecules,74 we have also considered the water-assisted reaction with an additional water molecule that is hydrogen bonded to the helper one. The process occurs now through [TS-CuG2W3]wa2 with an activation barrier of 25.8 kcal/mol with respect to [CuG2W3]wa2. In addition, if the effects of the dielectric medium are accounted for with the continuum CPCM solvation model, at the gas-phase geometries, the energy barrier is reduced to 21 kcal/mol. Nonetheless, since solvent effects can influence the geometry of the stationary points, we have also carried out CPCM optimizations. However, whereas we have obtained reasonably wellconverged geometry parameters for the reactant [CuG2W3]wa2, the calculations on the transition structure [TS-CuG2W3]wa2 oscillated because of small changes on the second more external water molecule, the energy barrier thus ranging from 17 to 20 kcal mol-1. These values are low enough to explain the peptide bond formation catalyzed by copper cations at room temperature and show a significant catalytic effect of the medium. Finally, the values obtained for the reaction free energy indicate that the presence of Cu2+ cation favors the reaction compared to gas-phase system (around -4.0 for Cu2+ complexes versus -1.0 kcal/mol for gas-phase system). This is probably because in the final product, N-glycylglycine coordinates Cu2+ in a bidentate manner through the amino group and the carbonyl oxygen of the amide bond. The interaction of the metal cation with the amide carbonyl oxygen stabilizes the -(Η)+NdC(O-)- resonant form of the peptide bond,75 which leads to a shorter CO-Cu2+ distance, compared to that of CO(carboxylic)-Cu2+ in the reactants, as well as to a strengthening of the peptide bond. The distance corresponding to the peptide bond in the Cu2+ complexes is 1.30-1.31 Å, whereas in Nglycylglycine in gas phase this value is 1.35 Å. These facts stabilize the final complex, making the peptide bond formation a thermodynamically more favorable process. These results are in agreement with the SIPF mechanism suggested by Rode and Suwannachot.34 However, in this theory, Cu2+ is coordinated by a Cl- ion, the main influence of this ligand being probably a screening effect on the metal cation. Since this screening is expected to be similar both in the reactant and in the transition state, we do not expect significant changes on the energy barrier. Overall, these results suggest that the simultaneous presence of both the Cu2+ metal cation and water molecules provides the situation that exhibits the lowest activation barrier to form the peptide bond between two glycine molecules in aqueous solution as a consequence of the synergy between these two factors. That is, Cu2+ strongly coordinates glycine molecules whereas water molecules efficiently catalyze the condensation reaction. Moreover, the peptide product becomes thermodynamically more stable in the presence of Cu2+ cations. According to these results, and considering the mass spectrometry experiments, one may think that the peptide bond
Rimola et al. formation has occurred in the ion source, since Cu2+ cations and glycine molecules are mixed in an aqueous solution and are submitted to a high voltage. In fact, it has been described in the literature that intracomplex rearrangements can take place during the solvent evaporation.76 This means that the condensation would occur in the ion source rather than in the collision cell and that the peak of the ESI mass spectrum observed at m/z ) 106.5 would correspond to the Cu2+-(glycylglycine)(H2O) complex and not to Cu2+-(glycine)2 so that the elimination of water observed in the collision cell would arise from Cu2+-(glycylglycine)(H2O). Because of this, we have optimized the structure of reaction of Cu2+-(glycylglycine)(H2O) and have computed the free energy for the Cu2+-(glycylglycine)(H2O) f Cu2+-(glycylglycine) + H2O reaction at the same level of theory. The reaction free energy obtained is 40 kcal/mol, significantly smaller than any of those previously computed for the loss of water from Cu2+-(glycine)2. Thus, from the present results, the most plausible explanation for the experimental observations is that the peptide bond formation has already occurred at the ion source, the ions being generated with somewhat excess of energy. Conclusions The present calculations suggest that the elimination of water experimentally observed in low-energy collision-activated dissociation spectra does not result from an intracomplex peptide bond formation because of the high-energy barrier (97 kcal/ mol) and reaction free energy (66 kcal/mol) computed for this process. These high values are associated to the loss of coordination of the metal cation in the transition structure and in the product. Other mechanisms yielding to the loss of water and leading to more stable products have been found to present lower energy values. However, these processes still present too large energy barriers considering the low collision activation dissociation spectra, the most plausible explanation for the experimental observations being that the peptide bond formation occurs before generating the ions in gas phase at the ion source. On the other hand, peptide bond formation in aqueous solution has also been studied by including discrete water molecules and by using the CPCM method to include the global solvation effects. It has been found that the important coordination changes observed in gas phase and that lead to so high energy barriers are largely attenuated when considering hydrated cations. Moreover, results show that solvent molecules can act as proton transport catalysts while simultaneously avoiding significant distortions in the coordination sphere. Among the reactions tested, the less energy demanding one, with a freeenergy barrier of around 20 kcal/mol, occurs via an intracomplex water-assisted condensation. Acknowledgment. Financial support from MCYT and DURSI, through the CTQ2005-08797-C02-02 and SGR200500244 projects, and the use of the Catalonia Supercomputer Centre (CESCA) are gratefully acknowledged. A. R. is indebted to the Universitat Auto`noma de Barcelona for a doctoral fellowship. References and Notes (1) Orgel, L. E. Trends Biochem. Sci. 1998, 23, 491. (2) Davies, P. Sci. Prog. 2001, 84, 1. (3) Davies, P. Sci. Prog. 2001, 84, 17. (4) Jensen, J. H.; Baldridge, K. K.; Gordon, M. S. J. Phys. Chem. 1992, 96, 8340. (5) Woon, D. E. Int. J. Quantum Chem. 2002, 88, 226.
Peptide Bond Formation and Cu2+ Interactions (6) Oie, T.; Loew, G. H.; Burt, S. K.; MacElroy, R. D. J. Am. Chem. Soc. 1983, 105, 2221. (7) Antonczak, S.; Ruiz-Lopez, M. F.; Rivail, J. L. J. Am. Chem. Soc. 1994, 116, 3912. (8) Bakowies, D.; Kollman, P. A. J. Am. Chem. Soc. 1999, 121, 5712. (9) Chalmet, S.; Harb, W.; Ruiz-Lopez, M. F. J. Phys. Chem. A 2001, 105, 11574. (10) Rimola, A.; Tosoni, S.; Sodupe, M.; Ugliengo, P. Chem. Phys. Lett. 2005, 408, 295. (11) Bernal, J. D. The Physical Basis of Life; Routledge and Kegan Paul: London, 1951. (12) Rao, M.; Odom, D. G.; Oro, J. J. Mol. EVol. 1980, 15, 317. (13) Lahav, N.; White, D.; Chang, S. Science 1978, 201, 67. (14) Orgel, L. E. Origins Life EVol. Biosphere 1998, 28, 227. (15) Basiuk, V. A.; Sainz-Rojas, J. AdV. Space Res. 2001, 27, 225. (16) Bujda`k, J.; Rode, B. M. J. Mol. EVol. 1997, 45, 457. (17) Bujda`k, J.; Rode, B. M. React. Kinet. Catal. Lett. 1997, 62, 281. (18) Bujda`k, J.; Rode, B. M. J. Mol. Catal. A 1999, 144, 129. (19) Bujda`k, J.; Rode, B. M. Origins Life EVol. Biosphere 1999, 29, 451. (20) Bujda`k, J.; Rode, B. M. Amino Acids 2001, 21. (21) Bujda`k, J.; Rode, B. M. J. Inorg. Biochem. 2002, 90. (22) Bujda`k, J.; Rode, B. M. Catal. Lett. 2003, 91, 149. (23) Plankensteiner, K.; Reiner, H.; Rode, B. M. Chem. BiodiVersity 2004, 1, 1308. (24) Zamaraev, K. I.; Romannikov, V. N.; Salganik, R. I.; Wlassoff, W. A.; Khramtsov, V. V. Origins Life EVol. Biosphere 1997, 27, 325. (25) Collins, J. R.; Loew, G. H.; Luke, B. T.; White, D. H. Origins Life EVol. Biosphere 1988, 18, 107. (26) Aquino, A. J. A.; Tunega, D.; Gerzabek, M. H.; Lischka, H. J. Phys. Chem. B 2004, 108, 10120. (27) Rimola, A.; Tosoni, S.; Sodupe, M.; Ugliengo, P. ChemPhysChem 2006, 7, 157. (28) Porter, T. L.; Eastman, M. P.; Whitehorse, R.; Bain, E.; K. M. Scanning 2000, 22, 1. (29) Porter, T. L.; Eastman, M. P.; Bain, E.; Begay, S. Biophys. Chem. 2001, 91, 115. (30) Boulet, P.; Greenwell, H. C.; Stackhouse, S.; Coveney, P. V. THEOCHEM 2006, 762, 33. (31) Oie, T.; Loew, G. H.; Burt, S. K.; MacElroy, R. D. J. Am. Chem. Soc. 1984, 106, 8007. (32) Yamagata, Y.; Inomata, K. Origins Life EVol. Biosphere 1997, 27, 339. (33) Rode, B. M. Peptides 1999, 20, 773. (34) Rode, B. M.; Suwannachot, Y. Coord. Chem. ReV. 1999, 190192, 1085. (35) Matsubara, T.; Hirao, K. Organometallics 2001, 20, 5056. (36) Liedl, K. R.; Rode, B. M. Chem. Phys. Lett. 1992, 197, 181. (37) Remko, M.; Rode, B. M. Chem. Phys. Lett. 2000, 316, 489. (38) Remko, M.; Rode, B. M. Phys. Chem. Chem. Phys. 2001, 3, 4667. (39) Seto, C.; Stone, J. A. Int. J. Mass Spectrom. 1999, 192, 289. (40) Poater, J.; Sola, M.; Rimola, A.; Rodriguez-Santiago, L.; Sodupe, M. J. Phys. Chem. A 2004, 108, 6072. (41) Georgieva, I.; Trendafilova, N.; Rodriguez-Santiago, L.; Sodupe, M. J. Phys. Chem. A 2005, 109, 5668. (42) Rimola, A.; Rodriguez-Santiago, L.; Sodupe, M. J. Phys. Chem. B 2006, 110, 24189. (43) Bertra´n, J.; Rodriguez-Santiago, L.; Sodupe, M. J. Phys. Chem. B 1999, 103, 2310. (44) Becke, A. D. J. Chem. Phys. 1993, 98, 1372. (45) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (46) Wachters, A. J. H. J. Chem. Phys. 1970, 52, 1033. (47) Hay, P. J. J. Chem. Phys. 1977, 66, 4377. (48) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheesman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin, K. N.;
J. Phys. Chem. B, Vol. 111, No. 20, 2007 5747 Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmanu, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03; Gaussian Inc.: Wallingford, CT, 2004. (49) McQuarrie, D. Statistical Mechanics; Harper and Row: New York, 1986. (50) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem. 2003, 24, 669. (51) Klamt, A.; Schueuermann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799. (52) Ugliengo, P. MOLDRAW, 2003. (53) Jensen, F. J. Am. Chem. Soc. 1992, 114, 9533. (54) Moision, R. M.; Armentrout, P. B. J. Phys. Chem. A 2002, 106, 10350. (55) Strittmatter, E. F.; Lemoff, A. S.; Williams, E. R. J. Phys. Chem. A 2000, 104, 9793. (56) Pulkkinen, S.; Noguera, M.; Rodriguez-Santiago, L.; Sodupe, M.; Bertran, J. Chem. Eur. J. 2000, 6, 4393. (57) Remko, M.; Rode, B. M. J. Phys. Chem. A 2006, 110, 1960. (58) Marino, T.; Toscano, M.; Russo, N.; Grand, A. J. Phys. Chem. B 2006, 110, 24666. (59) Lamsabhi, A. M.; Alcami, M.; Mo, O.; Yanez, M.; Tortajada, J. Chem. Phys. Chem. 2004, 5, 1871. (60) Lamsabhi, A. M.; Alcami, M.; Mo, O.; Yanez, M.; Tortajada, J. J. Phys. Chem. A 2006, 110, 1943. (61) Noguera, M.; Bertran, J.; Sodupe, M. J. Phys. Chem. A 2004, 108, 333. (62) Wen, D.; Yalcin, T.; Harrison, A. G. Rapid Commun Mass Spectrom. 1995, 9, 1155. (63) Yalcin, T.; Wang, J.; Wen, D.; Harrison, A. G. J. Am. Soc. Mass Spectrom. 1997, 8, 749. (64) Lavanant, H.; Hoppilliard, Y. J. Mass Spectrom. 1997, 32, 1037. (65) Talaty, E. R.; Perera, B. A.; Gallardo, A. L.; Barr, J. M.; Van, Stipdonk, M. J. J. Phys. Chem. A 2001, 105, 8059. (66) Rodriguez-Santiago, L.; Sodupe, M.; Tortajada, J. J. Phys. Chem. A 2001, 105, 5340. (67) Shoeib, T.; Siu, K. W. M.; Hopkinson, A. C. J. Phys. Chem. A 2002, 106, 6121. (68) Rimola, A.; Sodupe, M.; Tortajada, J.; Rodriguez-Santiago, L. Int. J. Mass Spectrom. 2006, 257, 60. (69) Hoppilliard, Y.; Ohanessian, G.; Bourcier, S. J. Phys. Chem. A 2004, 108, 9687. (70) D’Angelo, P.; Bottari, E.; Festa, M. R.; Nolting, H. F.; Pavel, N. V. J. Phys. Chem. B 1998, 102, 3114. (71) Gauld, J. W.; Audier, H.; Fossey, J.; Radom, L. J. Am. Chem. Soc. 1996, 118, 6299. (72) Chalk, A. J.; Radom, L. J. Am. Chem. Soc. 1997, 119, 7573. (73) Gauld, J. W.; Radom, L. J. Am. Chem. Soc. 1997, 119, 9831. (74) Yamabe, S.; Minato, T.; Hirao, K. J. Chem. Phys. 1984, 80, 1576. (75) Constantino, E.; Rimola, A.; Rodriguez-Santiago, L.; Sodupe, M. New J. Chem. 2005, 29, 1585. (76) Rogalewicz, F.; Hoppilliard, Y.; Ohanessian, G. Int. J. Mass Spectrom. 2000, 201, 307.
