Conformations and Metal Ion Affinities of Glutamine Binding with Alkali and Alkaline Earth Metal Cations:an ab initio Study

Rui Pang,Zi-jing Lin

Department of Physics and Collaborative Innovation Center of Suzhou Nano Science and Technology, University of Science and Technology of China,Hefei 230026,China

(Dated:Received on November 11,2013;Accepted on December 20,2013)

Conformations and Metal Ion Affinities of Glutamine Binding with Alkali and Alkaline Earth Metal Cations:an ab initio Study

Rui Pang,Zi-jing Lina∗

Department of Physics and Collaborative Innovation Center of Suzhou Nano Science and Technology, University of Science and Technology of China,Hefei 230026,China

(Dated:Received on November 11,2013;Accepted on December 20,2013)

Conformations and reaction energetics are important for understanding the interactions between biomolecules and metal ions.In this work,we report a systematic ab initio study on the conformations and metal ion affinities of glutamine(Gln)binding with alkali and alkaline earth metal ions.An efficient and reliable method of searching low energy conformations of metalated Gln is proposed and applied to the complexes of Gln·M+/++(M+/++=Li+, Na+,K+,Rb+,Cs+,Be++,Mg++,Ca++,Sr++,and Ba++).In addition to all conformers known in literatures,many new important conformations are located,demonstrating the power of the new method and the necessity of the conformational search performed here. The metal coordination modes,relative energies,dipole moments,and equilibrium distributions of all important conformations of Gln·M+/++are calculated by the methods of B3LYP, BHandHLYP,and MP2.IR spectra and metalation enthalpies and free energies are also presented and compared with the available experiments.The results form an extensive database for systematic examination of the metalation properties of Gln.

Conformational search method,Complexation structure,Conformational distribution,IR spectrum,Binding energy

I.INTRODUCTION

Metal ions are involved in about 40%of all proteins and enzymes in living systems.Amino acids are the basic building blocks of all natural proteins and the knowledge about amino acids interacting with metal ions is of broad interest.Metalated amino acids in gas phase that are free of the complicated solvent ef f ect are relatively easy to study,but are important basic models for understanding their counterparts in living bodies[1-3].In fact,the structures and properties of gaseous biomolecules are often closely related to their counterparts in solution[3-6].Consequently,conformations and thermo-chemical properties of metalated amino acids in gas phase are interesting research subjects and have been intensively investigated both experimentally[3-18]and theoretically[19-23].

Glutamine(Gln)is a critically important amino acid and comprises about 50%of all free amino acids in the whole-body pool.It plays an important role in muscle growth and synthesis of enzymes and serves as fuel for many kinds of cells.The IR spectra of gaseous Gln metalated with Li+,Na+,K+,Cs+and Ba2+have been measured[17,18].Experimental and theoretical determinations of the structures and metal ion affinities(MIA)of Gln binding with Li+,Na+and K+have also been reported[7-9,13,24].Nevertheless,there is no systematic comparative study on the conformations and MIA of Gln metalated with the set of alkali and alkaline earth metal ions.Moreover,the experimental conformation assignments are indirect and rely on the candidate structures determined by theoretical studies.Furthermore,the existing theoretical results are often obtained by conformational search based on Monte Carlo sampling using molecular mechanics force fi elds that are prone to missing important conformers [25].To improve the reliability of the computational results,extensive systematic conformational searches of the potential energy surfaces are required[26].

In this work,a conformational search method for systematically exploring the potential energy surfaces of metalated Gln is described.Conformations of Gln metalated with alkali and alkaline earth metal ions are obtained by conducting the systematic search method. The important conformations and their structural characteristics are discussed.MIAs and IR spectra of the metalated Gln are computed by several quantum chemistry methods and compared with the available experiments.

II.COMPUTATIONAL METHOD

In principle,the full conformational space of a biomolecule may be thoroughly searched by optimizingtrial structures generated by combinations of varying all rotational degrees of freedom.The thorough conformational search method has been successfully applied to many small biomolecules,including Gln[26].Systematic searches of the structures of protonated,hydrated and metalated amino acid complexes have also been performed[4,5,21,22].These studies show that the structures of metalated amino acid may be classi fi ed as mono-,bi-,and tri-dentate coordination between a metal ion and the electronegative atoms of amino acid. All mono-dentate coordination structures have high energies relative to the global minimum and may be ignored.These studies also indicate that it is a reliable and efficient approach to locate the low energy conformations of metalated amino acid with trial structures generated by adding the metal ion to possible binding sites of free amino acid conformations.Binding metal ion with di ff erent amino acid conformations may result in the metalated structures with an energy ordering very di ff erent from that of free amino acid conformations.Normally the change in energy ordering is expected to be limited to a few kcal/mol.To add a suffi cient safety margin to the solution,conformers of free Gln that are within 16 kcal/mol of the global minimum are used here to generate the trial structures of metalated complexes.

However,some metalated amino acids may exhibit so called salt bridge(SB)structures as their low energy conformations[11,18,21,22,27],while SB or zwitterions structures are often absent for free amino acids. Therefore,the aforementioned trial structures are only reliable for determining low energy charge solvated(CS) structures of metalated amino acid.To locate low energy SB structures of metalated Gln,the free Gln conformers within 16 kcal/mol of the global minimum are used to generate zwitterions structures by adding the proton removed from the carboxyl group to the amino group.These zwitterions are then used to construct the trial SB structures of metalated Gln.

The sum of the radius of oxygen atom and the radius of metal ion is used as the trial bond length between the ion and an electronegative atom of Gln.The location of metal ion is easily determined in a tridentate coordination.For a given set of three electronegative atoms, there are only two possible locations satisfying the bond length constrain,as illustrated in Fig.1(a).However, there is a circle at which all points meet the bond length constrain in bidentate coordination,as shown in Fig.1(b).The possible ion locations on the circle may be searched like a bond rotational freedom.Hence,N equally spacing points on the circle may be used to generate trial metalated Gln structures.Substantial tests have been performed and N=2 is found to be sufficient to locate the low energy conformations.

FIG.1 Schematic for trial structure generation of metalated Gln:the distance between the metal ion(M+/++)and an electronegative atom(a,b or c)of Gln is set at a given value. (a)Two possible sites for tridentate coordination of M+/++with a,b and c.(b)For bidentate coordination between M+/++and a and b,M+/++may be at any point on the large circle.Two end points of a randomly selected diameter are used as the binding sites of M+/++in the trial structure generation.

The trail structures were fi rst optimized at the level of HF/3-21G,followed by optimization at the B3LYP/6-311++G∗∗level.The vibrational frequencies,zero point vibrational energy(ZPVE)and thermal corrections to enthalpy(Hcoor)and free energy(Gcoor)are calculated at the B3LYP/6-311++G∗∗level to correspond to the geometry determination method.A scaling factor of 0.985 for the frequencies is chosen to best match the published experimental results[18].Most single point energies were calculated with MP2[28], B3LYP[29-34],and BHandHLYP methods using the 6-311++G(3df,3pd)basis set.For the fourth and f i fth row elements,the basis set of LANL2DZ[35-37]was used to reduce the computational cost.The trial structures were generated using our in-house developed software.All the energy calculations and geometry optimizations were performed with Gaussian 03[38].

The enthalpy of free Gln or metalated Gln,HA(A=Gln or Gln·M+/++),is determined by conformational ensemble averaging:

where xiis the population fraction of conformer i determined by relative conformational free energies and summed to 1 for all conformers.All the gaseous species are assumed to be ideal gases at the pressure P of 1 atm. The enthalpy of metal ion,HM+/++,is calculated as itselectronic energy(EM+/++)plus its translational energy (3/2RT)and PV term(RT):

In addition to the ensemble averaging,the contribution by the entropy of mixing is considered in the calculation of free energy of Gln and metalated Gln:

Considering the term due to the translational partition function[39],the free energy of metal ion is calculated as

where M is the mass of metal ion.MIA is def i ned as the negative of the enthalpy change of the reaction Gln+M+/++→Gln·M+/++at the room temperature,

The negative free energy change of the reaction is also computed as follows,

The basis set superposition error is considered in computing the enthalpy and free energy changes.

III.RESULTS AND DISCUSSION

For the convenience of referencing,we f i rst def i ne some nomenclature here.There are f i ve electronegative atoms in Gln,i.e.,three O atoms and two N atoms.The tridentate M+/++coordination modes may in principle include OOOs,NOOs,NOO,NsOO,NsOOs,NNsO, and NNsOs,where N(O)and Ns(Os)denote the backbond and the side-chain nitrogen(oxygen),respectively. The bidentate M+/++coordination modes may in principle include OO,OOs,NO,NOs,NNs,NsO,and NsOs. Low energy conformers are found only for the modes of OOOs,NOOs,OO,and OOs.The mode of NOOsgenerally corresponds to a canonical Gln structure.However,the modes of OOOs,OO and OOsmay correspond to either canonical or zwitterionic Gln.For easy identif i cation,the suffixes.c and.z are used to denote the canonical and zwitterionic structures,respectively. That is,a low energy conformation of Gln·M+/++may be denoted with one of the seven coordination types, NOOs.c,OOOs.c,OO.c,OOs.c,OOOs.z,OO.z,and OOs.z.Dif f erent conformations of a given coordination type are further dif f erentiated with a numeral suffix to indicate their relative stabilities ordered according to their MP2 energies,e.g.,OOOs.c1 and OOOs.c2.For simplicity,a conformer may also be denoted with the name of the metal element followed by a numeral suffi x indicating its relative stability within all conformers ordered according to its relative MP2 energy,e.g.,Na1 and Na2.

A.Conformations of Gln binding with alkali metal ion

Table I shows the relative electronic energies,dipole moment,ZPVE,Hcoorand Gcoorat the room temperature for all important conformations of Gln·M+.An important conformer here means that the room temperature population of the conformer is over 10%as determined by any one of the B3LYP,BHandHLYP and MP2 methods.In case there is only one dominant conformer for a metal ion,the second most abundant conformer is also shown in Table I.The equilibrium conformational distributions at T=298 and 500 K are shown in Table II. The structures of the important conformers are shown in Fig.2.

As seen from Table I and Table II,the B3LYP, BHandHLYP,and MP2 results agree on the global minimum conformer of Gln·Li+and the global minimum is the only important conformer for Gln·Li+.The three methods also agree on the global minimum of Gln·Na+and there are two important conformers for Gln·Na+.For Gln·K+,Gln·Rb+,and Gln·Cs+,however,the global energy minima are zwitterions as predicted by B3LYP,but are canonical conformers as predicted by BHandHLYP and MP2.Based on B3LYP results,it has been concluded that the size of metal ion is the dominant factor for the global minimum to take the zwitterionic form[26].The present BHandHLYP and MP2 results indicate that the conclusion may be only a B3LYP biased claim.Experiments may be used to verify the results as there are some characteristic different features in the IR spectra of canonical and zwitterionic conformers.However,it is usually difficult to distinguish by normal temperature IR spectrum as it is a superposition of contributing conformers.Some low temperature IR measurement may be required to draw convincing conclusion.Before such measurement is performed,we tend to prefer the BHandHLYP and MP2 results.Nevertheless,the tendency is clear that zwitterionic structures are favorable for large metal ions.

With the change in metal ion size,another trend also emerges.The coordination of NOOs.c is strongly favored over OOOs.c in Gln·Li+and Gln·Na+,but the two conf i gurations are comparable in Gln·K+.For Gln·Rb+and Gln·Cs+,OOOs.c is more favorable than NOOs.c.That is,there is a trend that the OOOs.c conf i guration is increasingly favored with the increased ion size.Therefore,assuming NOOsas the favorable canonical conf i guration may produce def i cient results. For example,the global minimum for Gln·Cs+was predicted to be Cs5 in our notation[18]and is in fact an unimportant conformer of Gln·Cs+.The global mini-mum was missed in the previous study by assuming the NOOsconf i guration.There are also other omissions in former studies.For example,OOOs.c1 for Gln·K+and NOOs.c3 for Gln·Na+are found in this work,but are missed in Refs.[7,18].It should be mentioned that all conformers located in previous studies have been found in this work.Considering the newly located important conformers,it is clear that our method of conformational search is more reliable than the ones used before [7,18].

FIG.2 Important conformations of Gln·M+.

TABLE I Metal ion coordination modes(coordination),relative electronic energies E,relative ZVPEs,relative thermal corrections for enthalpy Hcoorand Gibbs free energy Gcoor,and dipole moments u of important conformers(conf.)of Gln bound with alkali metal ion.B3LYP and BHandHLYP are abbreviated as B3 and BH,respectively.All energies are in kJ/mol.

It is worthy pointing out that the relative conformational energies and the conformational distributions determined by the B3LYP,BHandHLYP and MP2 methods are quite dif f erent for Gln binding with large cation, K+,Rb+,or Cs+,as shown in Tables I and II.As none of the methods is def i nitely superior to the others and there is a lack of strong evidence to favor one or the other,it is prudent to treat the results of dif f erent methods on an equal basis.It should also be noted that the global energy minimum may be dif f erent from the global free energy minimum(the most abundant conformer) due to the entropy ef f ect.For example,K1 is predicted by MP2 as the global energy minimum(Table I),but K9 which is 4.42 kJ/mol less favorable than K1 has a population almost twice as much as K1 at the room temperature(Table II).As the entropy ef f ect increases with temperature,K9 is over three times more abundant than K1 at 500 K.Therefore,conformers within a sufficiently large energy range of the global minimum should be thoroughly searched before one may reliably determine the most important conformers.

TABLE II Equilibrium distributions of important conformers of Gln·M+(M+:alkali metal ion)at T=298 and 500 K as computed by the methods of B3LYP(B3),BHandHLYP(BH),and MP2.

TABLE III Metal ion coordination modes(coordination),relative electronic energies E,relative ZVPEs,relative thermal corrections for enthalpy Hcorand free energy Gcor,and dipole moments u of important conformers(Conf.)of Gln bound with alkaline earth metal ion.B3LYP and BHandHLYP are abbreviated as B3 and BH,respectively.All energies are in kJ/mol.

B.Conformations of Gln binding with alkaline earth metal ion

Table III shows the relative electronic energies,dipole moment,ZPVE,Hcoorand Gcoorat the room temperature for all important conformations of Gln·M++,where M++denotes divalent alkaline earth metal ion.The equilibrium conformational distributions at T=298 and 500 K are shown in Table IV.The structures of the important conformers are shown in Fig.3.

AsshowninTablesIIIandIV,theB3LYP, BHandHLYP,and MP2 results all agree on that the global energy minimum is the only important conformer for Gln·Be++and Gln·Mg++.The structures of Gln·Be++and Gln·Mg++look almost the same,with the ion chelated by O,N and Os(Fig.3).For Gln·Ca++, zwitterions are predicted by B3LYP and MP2 to be the global energy minimum,while the global energy minimum is a canonical conformer as determined by BHandHLYP.As NOOs.c1 has a favorable ZPVE in comparison with OOOs.z1 and OOOs.z2,the conformational distribution of Gln·Ca++is dominated by NOOs.c1 as predicted by BHandHLYP at most temperature.The population of NOOs.c1 is comparable tothat of OOOs.z1 and OOOs.z2 as determined by the B3LYP results,except for very low temperature when only the global minimum conformer is dominant.However,the population of NOOs.c1 predicted by the MP2 method is very low at most temperature.IR spectrum measured at low to medium temperature may be used to verify which method provides the best description of the Gln·Ca++conformations.

TABLE IV Equilibrium distributions of important conformers of Gln·M++(M++:alkaline earth metal ion)at T=298 and 500 K as computed by B3LYP(B3),BHandHLYP(BH)and MP2.

FIG.3 Important conformations of Gln·M++.

The B3LYP and MP2 methods show clearly that the global minimum of Gln·Sr++is a zwitterionic conformer with the OOOscoordination.Considering ZPVE that is ef f ective at any temperature,the global minimum of Gln·Sr++is predicted by BHandHLYP to be a canonical conformer,NOOs.c1.As predicted by BHandHLYP,NOOs.c1 is the most abundant conformer at any temperature.However,the MP2 method shows that OOOs.z1 is basically the only conformer in the equilibrium ensemble,with a population of over 95% for T≤500 K.The B3LYP results are intermediate to the BHandHLYP and MP2 results.Again,some experimental IR spectrum,especially that measured at low temperature,may provide a clear test of the theoretical results.Moreover,the dipole moment of OOOs.z1 and NOOs.c1 is quite dif f erent(Table III).The theoretical results may be unambiguously tested by the dipole moment measurement at low temperature[39].

As shown in Table III,all the B3LYP,BHandHLYP, and MP2 results show that the global minimum of Gln·Ba++is zwitterionic with a metal ion coordination of OOOs,also agreeing with the result of Ref.[18]. Nevertheless,the tendency of favoring canonical conformation by BHandHLYP is still quite obvious.The BHandHLYP method shows that the canonical conformer of NOOs.c1 is an important conformer with a population of over 20%for T＞298 K,while NOOs.c1 is predicted by B3LYP and MP2 to be unimportant at any temperature(Table IV).However,the above difference is relatively small in comparison with that for Gln·Ca++and Gln·Sr++and may not be easily distinguishable by IR or dipole moment measurement.

Overall,the NOOscoordination of metal ion is the most favorable canonical conformation and the OOOscoordination mode is the most favorable zwitterionic form for alkali and alkaline earth metal ions binding with Gln.The NOOs mode is strongly favored by small ions,Li+,Na+,Be++,and Mg++.The OOOsmode is increasingly favored with the increased size and charge of metal ion.All three methods show that the global minimum of Gln·Ba++is a salt bridged conformation with the ion coordination mode of OOOs.

FIG.4 IR spectra for complexes of Gln·M+/++.The dotted lines are the theoretical results based on the MP2 conformational energies.The solid lines are experimental results obtained by re-plotting curves from Refs.[17,18].

In terms of the computational methods,B3LYP is most inclined to predict a SB form for the global minimum,while BHandHLYP tends to favor a CS structure.MP2 is intermediate to B3LYP and BHandHLYP in predicting a SB structure for the global minimum. However,when MP2 does predict a SB form for the global minimum,the energy dif f erence between the SB and CS forms is larger for MP2 than for B3LYP.

C.Infrared spectra

Figure4showsthetheoreticalIRspectraof Gln·M+/++at the room temperature using the conformational distribution determined by the MP2 method. The theoretical IR spectra are compared with the available experimental results that are obtained by replotting the data in Refs.[17,18]and are only approximate.The theoretical spectra shown here are substantially more reliable than that in Ref.[17]as numerous important conformers are missed in Ref.[17],while all conformers in Ref.[17]are located in this work.In fact, some of the discussion in Ref.[17]may be incorrect.For example,the IR spectrum for Gln·Cs+in Ref.[17]is based on its global minimum that corresponds to Cs5 here and accounts for only 5%in the equilibrium population.

As shown in Fig.4,even though there are overall agreement between the theory and the experiment, there are clear dif f erences in the details of the theoretical and experimental vibrational peak positions and intensities.The IR spectra of MP2,B3LYP, and BHandHLYP are the same for Gln·Li+,Gln·Na+, Gln·Be++,and Gln·Mg++due to their essentially mono conformational feature.The B3LYP and BHandHLYP IR spectra for Gln·K+,Gln·Cs+,and Gln·Ba++are somewhat dif f erent from that of MP2,but the dif f erences are relatively small compared to that with the experiments.That is,using the B3LYP or BHandHLYP spectra does not substantially af f ect the agreement with the experiments.

There are several factors that may cause the difference between the theoretical and experimental results.One is about the data treatment.The theoretical spectra are generally obtained by Gaussian or Lorentz broadening of discrete vibrational modes and are af f ected by the broadening parameter.The experimental spectra implicitly involve interpolations and are af f ected by the resolution and response of the measuring instrument.Closely neighboring peaks often appear as a broadening peak and it is difficult to see sharp features in experiment.Another factor is the ef f ective temperature for the conformational distribution.Even though IR spectra are often measured at about the room temperature,the conformational ensemble may correspond to a dif f erent ef f ective temperature as the sample needs to be heated f i rst to evaporate the amino acid.The ef f ective temperature is almost always higher than the room temperature,and may be as high as 600 K[41]. In fact,the conformational ensemble may sometime be not in equilibrium yet[40].Meaningful theoretical and experimental comparison may be made only when the equilibrium is reached and the ef f ective temperature is determined experimentally.There are also additional error sources from the theoretical side,e.g.,an averaged empirical scaling factor for vibrational frequencies due to the neglect of anharmonic ef f ect,inaccurate distribution due to unknown inaccuracy in relative conformational energies,inaccurate IR intensity due to inaccurate charge distributions and the use of pesudopotentials for heavy elements(the fourth and f i fth row metal ions).

Due to the above mentioned reasons,a comprehensive comparison between the theoretical and experimental IR spectra is unrealistic at the present.We also defera detailed comparison of the theoretical IR spectra to a future study.However,the above analysis does point out the best situation for comparing the theory with the experiment.That is,the experimental IR spectrum for single conformer,namely measured at adequately low temperature with the equilibrium carefully assured, may be used to test the qualities of the predictions by B3LYP,BHandHLYP and MP2.Moreover,the IR spectra shown here are the f i rst systematical results to show the trend of spectral variation in Gln metalated with alkali and alkaline earth metal ions.For example,there are feature peaks in the range of 800-1200 cm-1for light elements such as Li+and Be++,but these peaks weaken with the increased size of the metal ion and disappear for Ba++.These feature peaks may be used to determine the type of metal ion involved.

TABLE V Reaction enthalpies∆H and free energies∆G for Gln metalated with alkali and alkaline earth metal ions as calculated by the methods of B3LYP(B3),BHandHLYP(BH),and MP2.

D.Metal ion affinities

Table V shows the reaction enthalpies and free energies of Gln binding with alkali and alkaline earth metal ions as calculated by the methods of B3LYP, BHandHLYP and MP2.The conformations of Gln reported in Ref.[26]are used here.To our knowledge,only the MIAs of Gln with Na+and K+have been published [7,9,10,13].Data in Table V are the f i rst comprehensive compilation on MIAs of Gln with alkali and alkaline earth metal ions.

As shown in Table V,there is a general trend regarding the results computed by the three methods: the MIAs and the negative free energy changes are the largest as predicted by BHandHLYP,followed by the B3LYP results,and the smallest as predicted by MP2. The dif f erences among the results of the three methods are smaller for alkali metal ions than that for alkaline earth metal ions,due mainly to the larger MIAs of the latter.For a given series of monovalent or divalent metal ions,the dif f erence tends to decrease with the increase of the ion size,though the trend may be disturbed by the use of pseudopotential for the fourth and f i fth row elements.

There are experimental reports on MIAs of Gln with Na+and K+,as shown in Table V.The MP2 results are in very good agreement with these available experiments.It is natural for one to favor the MP2 results. However,one should be cautious about this conclusion as all the existing experimental data are obtained by the simple kinetic method(SKM)that ignores the relative entropy changes[41-44].As shown in Table V, the reaction free energy changes are quite dif f erent at T=298 and 500 K,indicating a strong entropic ef f ect. Since the free energy change is probed more directly than the enthalpy change in experiments and the ef f ective experimental temperature is often close to 500 K, deducing the enthalpy change at T=298 K from the experiment may sometime cause signif i cant error.In fact, it has been shown that the SKM result underestimates the protonation energy of Gln by about 30 kJ/mol[26, 41].Due to the similarity in the metalation and protonation processes,some underestimation of SKM on MIA is expected.Therefore,one should not jump into the conclusion that the MP2 results provide the best estimate of MIAs of Gln.To truly resolve the theoretical dispute,measurements of MIAs of Gln by the more rigorous method of the extended kinetic method[41]are highly desirable and urged.

IV.CONCLUSION

A simple and reliable method of systematically searching low energy conformations of metalated Gln is described.The method is used to f i nd the low energy conformations of Gln binding with alkali and alkaline earth metal ions.In addition to all conformersknown in literatures,many new important conformations are found,demonstrating the power of the new search method and the necessity of the conformational search performed here.

The metal coordination modes,relative energies, dipole moments and equilibrium distributions of all important conformations of Gln metalated with alkali and alkaline earth metal ions as calculated by the methods of B3LYP,BHandHLYP and MP2 are presented.Most of the results are reported for the f i rst time and form an extensive database useful for systematic examination of the metalation properties of Gln.

The tridentate charge solvation coordination mode of NOOsis the dominant conformation for Gln·Li+, Gln·Na+,Gln·Be++,and Gln·Mg++.The global minimum of Gln·Ba++adopts a tridentate SB coordination mode of OOOs.The global minima of Gln·K+, Gln·Rb+,Gln·Cs+,Gln·Ca++,and Gln·Sr++are dependent on the computational method and may take either the CS or SB form.Overall,B3LYP tends to favor the SB form,while BHandHLYP tends to favor the CS conformation.

IR spectra at the room temperature for Gln bound with alkali and alkaline earth metal ions are presented and compared with the available experiments. It is noted that the feature peaks in the range of 800-1200 cm-1may be useful for identifying the metal ion bound with Gln.

MIAs and reaction free energies of Gln binding with alkali and alkaline earth metal ions are computed by the methods of B3LYP,BHandHLYP and MP2.The BHandHLYP MIA values are the largest,while the MP2 values are the smallest.The MP2 results are in good agreement with the available SKM experiments.However,large entropic ef f ects on the reaction free energies are observed.As SKM tends to underestimate MIAs when large entropy changes are involved,MIAs measured by the extended kinetic method are urged to provide more def i nite answers.

V.ACKNOWLEDGMENTS

This work was supported by the National Natural Science Foundation of China(No.11074233 and No.11374272)andtheSpecializedResearchFund fortheDoctoralProgramofHigherEducation (No.20113402110038 and No.20123402110064)

[1]T.Marino,N.Russo,and M.Toscano,Inorg.Chem. 40,6439(2001).

[2]T.Marino,N.Russo,and M.Toscano,J.Inorg. Biochem.79,179(2000).

[3]X.Xu,W.Yu,Z.Huang,and Z.Lin,J.Phys.Chem. B 114,1417(2010).

[4]S.Liu,A.Hu,and Z.Lin,Chin.J.Chem.Phys.25, 409(2102).

[5]H.Li,Z.Lin,and Y.Luo,Chin.J.Chem.Phys.25, 681(2012).

[6]A.K.Rai,W.Fei,Z.Lu,and Z.Lin,Theo.Chem.Acc. 124,37(2009).

[7]A.L.Heaton,R.M.Moision,and P.B.Armentrout,J. Phys.Chem.A 112,3319(2008).

[8]A.L.Heaton,S.J.Ye,and P.B.Armentrout,J.Phys. Chem.A 112,3328(2008).

[9]P.Wang,G.Ohanessian,and C.Wesdemiotis,Int.J. Mass spectrom.269,34(2008).

[10]M.M.Kish,G.Ohanessian,and C.Wesdemiotis,Int. J.Mass spectrom.227,509(2003).

[11]A.S.Lemof f,M.F.Bush,C.C.Wu,and E.R. Williams,J.Am.Chem.Soc.127,10276(2005).

[12]W.Y.Feng,S.Gronert,and C.Lebrilla,J.Phys.Chem. A 107,405(2003).

[13]A.L.Heaton and P.B.Armentrout,J.Phys.Chem.B 112,12056(2008).

[14]K.K.Lehmann,G.Scoles,and B.H.Pate,Annu.Rev. Phys.Chem.45,241(1994).

[15]D.P.Little,J.P.Speir,M.W.Senko,P.B.Oconnor, and F.W.McLaf f erty,Anal.Chem.66,2809(1994).

[16]X.Wang and L.Wang,Annu.Rev.Phys.Chem.60, 105(2009).

[17]M.F.Bush,J.Oomens,R.J.Saykally,and E.R. Williams,J.Am.Chem.Soc.130,6463(2008).

[18]M.F.Bush,J.Oomens,R.J.Saykally,and E.R. Williams,J.Phys.Chem.A 112,8578(2008).

[19]J.Wang,Chin.J.Chem.Phys.20,509(2007).

[20]X.Yang,L.Cao,Q.Zhang,J.Chen,G.Chu,Y.Zhao, X.Shan,F.Liu,and L.Sheng,Chin.J.Chem.Phys. 25,379(2012).

[21]W.Fei,A.K.Rai,Z.Lu,and Z.Lin,J.Mol.Struc.: THEOCHEM 895,65(2009).

[22]L.Meng and Z.Lin,Comput.Theor.Chem.976,42 (2011).

[23]C.Kapota,J.Lemaire,P.Maitre,and G.Ohanessian, J.Am.Chem.Soc.126,1836(2004).

[24]A.S.Lemof f,C.C.Wu,M.F.Bush,and E.R. Williams,J.Phys.Chem.A 110,3662(2006).

[25]S.Ling,W.Yu,Z.Huang,Z.Lin,M.Haranczyk,and M.Gutowski,J.Phys.Chem.A 110,12282(2006).

[26]R.Pang,M.Guo,S.Ling,and Z.Lin,Comput.Theor. Chem.1020,14(2013).

[27]S.Jang,M.J.Song,H.Kim,and S.S.Choi,J.Mass Spectrom.46,496(2011).

[28]M.Headgordon,J.A.Pople,and M.J.Frisch,Chem. Phys.Lett.153,503(1988).

[29]A.D.Becke,J.Chem.Phys.98,5648(1993).

[30]C.T.Lee,W.T.Yang,and R.G.Parr,Phys.Rev.B 37,785(1988).

[31]P.J.Stephens,F.J.Devlin,C.F.Chabalowski,and M. J.Frisch,J.Phys.Chem.98,11623(1994).

[32]S.H.Vosko,L.Wilk,and M.Nusair,Can.J.Phys.58, 1200(1980).

[33]X.Song,W.Hua,Y.Ma,C.Wang,and Y.Luo,J. Phys.Chem.C 116,23938(2012).

[34]I.Y.Zhang,J.Wu,Y.Luo,and X.Xu,J.Comput. Chem.32,1824(2011).

[35]W.R.Wadt and P.J.Hay,J.Chem.Phys.82,284 (1985).

[36]P.J.Hay and W.R.Wadt,J.Chem.Phys.82,299 (1985).

[37]P.J.Hay and W.R.Wadt,J.Chem.Phys.82,270 (1985).

[38]M.J.Frisch,G.W.Trucks,H.B.Schlegel,G.E.Scuseria,M.A.Robb,J.R.Cheeseman,J.A.Jr.Montgomery,T.Vreven,K.N.Kudin,J.C.Burant,J.M. Millam,S.S.Iyengar,J.Tomasi,V.Barone,B.Mennucci,M.Cossi,G.Scalmani,N.Rega,G.A.Petersson,H.Nakatsuji,M.Hada,M.Ehara,K.Toyota, R.Fukuda,J.Hasegawa,M.Ishida,T.Nakajima,Y. Honda,O.Kitao,H.Nakai,M.Klene,X.Li,J.E.Knox, H.P.Hratchian,J.B.Cross,C.Adamo,J.Jaramillo,R. Gomperts,R.E.Stratmann,O.Yazyev,A.J.Austin, R.Cammi,C.Pomelli,J.W.Ochterski,P.Y.Ayala, K.Morokuma,G.A.Voth,P.Salvador,J.J.Dannenberg,V.G.Zakrzewski,S.Dapprich,A.D.Daniels,M. C.Strain,¨O.Farkas,D.K.Malick,A.D.Rabuck,K. Raghavachari,J.B.Foresman,J.V.Ortiz,Q.Cui,A. G.Baboul,S.Clif f ord,J.Cioslowski,B.B.Stefanov,G. Liu,A.Liashenko,P.Piskorz,I.Komaromi,R.L.Martin,D.J.Fox,T.Keith,M.A.Al-Laham,C.Y.Peng, A.Nanayakkara,M.Challacombe,P.M.W.Gill,B. Johnson,W.Chen,M.W.Wong,C.Gonzalez,and J. A.Pople,Gaussian 03,Pittsburgh PA:Gaussian,Inc., (2003).

[39]C.J.Cramer,Essentials of Computational Chemistry: Theories and Models,John Wiley and Sons,325(2005).

[40]I.Compagnon,F.C.Hagemeister,R.Antoine,D. Rayane,M.Broyer,P.Dugourd,R.R.Hudgins,and M.F.Jarrold,J.Am.Chem.Soc.123,8440(2001).

[41]G.Bouchoux,D.A.Buisson,C.Colas,and M.Sablier, Eur.J.Mass Spectrom.10,977(2004).

[42]L.Drahos and K.Vekey,J.Mass Spectrom.38,1025 (2003).

[43]K.M.Ervin and P.B.Armentrout,J.Mass Spectrom. 39,1004(2004).

[44]A.G.Harrison,Mass Spectrom.Rev.16,201(1997).

∗Author to whom correspondence should be addressed.E-mail：zjlin@ustc.edu.cn