Theoretical study of fragmentation pathways and product distribution of deprotonated aspartic acid

Accepted Manuscript Theoretical study of fragmentation pathways and product distribution of deprotonated aspartic acid Zu-xu Chen, Hua Zhong, Hai-tao Yu PII: DOI: Reference:

S2210-271X(16)30445-5 http://dx.doi.org/10.1016/j.comptc.2016.11.006 COMPTC 2292

To appear in:

Computational & Theoretical Chemistry

Received Date: Revised Date: Accepted Date:

18 September 2016 1 November 2016 4 November 2016

Please cite this article as: Z-x. Chen, H. Zhong, H-t. Yu, Theoretical study of fragmentation pathways and product distribution of deprotonated aspartic acid, Computational & Theoretical Chemistry (2016), doi: http://dx.doi.org/ 10.1016/j.comptc.2016.11.006

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Theoretical study of fragmentation pathways and product distribution of deprotonated aspartic acid Zu-xu Chen, Hua Zhong*, Hai-tao Yu*

Key Laboratory of Functional Inorganic Material Chemistry (Ministry of Education), School of Chemistry and Materials Science, Heilongjiang University, Harbin 150080, China

* Correspondence authors. Tel.: +86–451–86608576 E–mail address: [email protected] (Hua Zhong), [email protected] (Prof. H.–T. Yu)

Abstract In this study, the fragmentation processes of deprotonated aspartic acid to eliminate CO2, NH3, and H2O were investigated by a quantum-mechanism computation at the B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory. By constructing the fragmentation reaction potential energy profile using the Gibbs free energies and enthalpies of the located stationary points, we explored both the preferred dissociation pathways and the product distribution. Furthermore, the thermal energy correction to key stationary points was computed to evaluate the temperature dependences of the dominant reaction channels and product distribution. Additionally, the similarities and differences between the present theoretical investigation and available experiment were discussed in detail.

Keywords Deprotonated aspartic acid; Fragmentation; Theoretical computation; Dissociation; Potential energy profile

1. Introduction Aspartic acid (Asp) is one of the α-amino acids [1,2] and includes D- and L-enantiomorphs [3]. D- and L-Asp are important precursors or intermediates in the syntheses of other amino acids [4] and medicines [5], and polymerized L-Asp has often been used as key reactants in chemistry [6], materials science [7], and biomedicine [8], among other research areas. 1 / 25

Asp can form three crystalline forms: monoclinic (L-Asp, P21 space group, Z = 2) [9–12], orthorhombic (L-Asp-H2O, P212121 space group, Z = 4) [11,13], and monoclinic (DL-Asp, C2/c space group, Z = 8) [14–16] systems. However, in aqueous solution, twenty-two zwitterionic structures have been suggested [17]. For Asp in the gas phase, Chen and co-workers theoretically predicted the existence of 139 isomers, among which only eight structures were believed to be thermodynamically stable [17]. In a pioneering study [1], Sanz identified six of the eight structures based on their distinct rotational

14

N nuclear quadrupolar coupling constants. Furthermore, the

vibrational spectrum of L-Asp is available from a theoretical computational work by Alam and Ahmad [18]. Mass spectrometry (MS) plays an extremely important role in characterizing the structures of amino acids [19], especially when used in combination with the liquid chromatography (LC) and gas chromatography (GC) techniques, i.e., LC-MS [20] and GC-MS [21]. Recently, some newly developed MS methods [22–25] have received increasing attention, such as the atmospheric pressure ionization (API) technique [26]. Common API methods include atmospheric pressure chemical ionization (APCI) [27], electrospray ionization (ESI) [28], and atmospheric pressure photoionization (APPI) [29], and they have been widely used in analyzing and identifying the structures of some amino acids [30] and various biomolecules [31]. The use of API is often followed by MS [32]. Generally, the parent ions generated in an initial collision-induced dissociation (CID) process have a narrow distribution of internal energy and, thus, fewer fragment ions [33,34]. Although initial CID can provide some very important characteristics of compounds [35], more information about the parent fragment ions is still required [36]. Such information can be provided by analyzing the molecular ion fragments resulting from a multilevel CID process [33,34] by tandem MS. In 2013, Choid and co-workers ionized twenty types of amino acids by the APCI technique [37], followed by a further investigation of the resulting deprotonated amino acids in a single quadrupole mass spectrometer. They found three molecular ion products [37], i.e., [Asp–H–CO2]– (m/z 88), [Asp–H–NH3]– (m/z 115), and [Asp–H–H2O]– (m/z 114), whose structures were assigned as 2, 3, and 4 (see Fig. 1), respectively. The three molecular ion fragments were believed to form from the dissociation of deprotonated Asp ([Asp-H]−) via the pathways of losing smaller fragment molecules CO2, NH3, and H2O, as shown in Fig. 1. Another interesting 2 / 25

aspect should be given special attention; i.e., the yield of [Asp–H–NH3]– is higher than that of [Asp–H–CO2]– when the cone voltage is less than 120 V. However, when the cone voltage is larger than 120 V, the yield of the latter becomes higher than the former [37]. It seems that the kinetic processes and product distribution are strongly temperature-dependent when they arise from the voltage change [38,39]. McLafferty gave a reasonable explanation for the fragmentations of a compound in CID; i.e., the formation and distribution of different fragments are strongly related to the rearrangement and isomerization reactions of parent molecular ions and intermediates [34]. As such [37], the aforementioned losses of the fragment molecules CO 2, NH3, and H2O from [Asp-H]−, along with the formation of the molecular ions [Asp–H–CO2]–, [Asp–H–NH3]–, and [Asp–H–H2O]–, imply that some rearrangement and isomerization processes must proceed to reach the prone configurations for such dissociations. Therefore, the isomerization and dissociation mechanism of [Asp-H]− and the corresponding fragment distribution are interesting aspects. In this study, we performed a DFT computation and constructed the isomerization and dissociation potential energy profile of [Asp-H]−, which was further used to analyze its fragmentation mechanism and ion product distribution. The present investigation provides a representative example for the establishment of a universal fragmentation model of deprotonated amino acids.

2 Computational details

All computations were performed using the Becke three-parameter hybrid functional in conjunction with the Lee-Yang-Parr correlation functional (B3LYP) [40,41] and second-order Møller-Plesset (MP2) [42,43] method with the 6-311++G(2df,2pd) and 6-31+G(d,p) basis sets [44,45], respectively, implemented in the Gaussian09 program package [46]. The molecular geometries of stationary points were optimized fully in vacuum, and no symmetry constraint was adopted in the calculations. The located structures were proven to be energy minima (no imaginary frequency) or first-order saddle points (transition states, only one imaginary frequency) by computing their vibrational frequencies within the harmonic approximation [47]. Furthermore, the thermal energy correction at different temperatures to electronic total energies for some stationary points was calculated at the same level of theory to investigate the temperature

dependence

of

the

fragmentation 3 / 25

mechanism.

Because

the

B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels gave the same results, unless otherwise mentioned, the energies and structural data used in the next discussion are that computed at the B3LYP/6-311++G(2df,2pd) level of theory. All geometries in Cartesian coordinates and MOL format are available in the Supporting Information.

3 Results and discussion 3.1 Rotamers and rotational isomerization of [Asp-H]− In the computations, we first searched and optimized the rotamers of [Asp-H]− and further investigated their isomerization processes by locating their interconnecting transition states. The schematic structures of the optimized minimum points are given in the supporting Fig. S1, and the energies of all stationary points are listed in Tables S1 and S2 (see Supporting Information). Figs. 2 and S2 show the constructed rotational isomerization potential energy profiles using the computed Gibbs free energies and enthalpies at 598.15 K at the B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory, respectively. Based on the data listed in Fig. 2 and Table S1, it can be seen that 5 is the lowest-lying rotamer among all rotational structures of [Asp-H]−, followed by low-lying 6 and 7, which have relative enthalpies of 1.93 kcal mol–1 and 4.69 kcal mol–1, respectively. We can rationalize the high thermodynamic stabilities of 5 and 6 based on their structural characteristics, i.e., the existence of intramolecular hydrogen bonds. In 5 and 6, there are apparent −O−H┄O− structures (see Fig. 2) between the carboxylate ion and carboxyl group, with an OHO angle of 175.1° and two OH bond lengths of approximately 1.1 Å and 1.4 Å, which are very similar to the structural features of the suggested intramolecular hydrogen bond by Dixon [46]. With the exception of 5 and 6, the structures with carboxylate ion and carboxyl group in cis mode are 2.3 kcal mol–1 ~ 4.7 kcal mol–1 higher in Gibbs free energy and 0.4 kcal mol–1 ~ 5.3 kcal mol–1 higher in enthalpy than the conformers in trans mode (see the supporting Table S1), which is believed to originate from steric hindrance between carboxylate ion and carboxyl group. According to the relative Gibbs free energies listed in Fig. 2, one can readily see 4 / 25

that the relative Gibbs free energy of the rotational transition states separating the rotamers of [Asp-H]− are lower than 23.80 kcal mol–1 (598.15 K); thus, the concentration distribution of these isomers is determined by their thermodynamic stabilities. 3.2 Rearrangement and dissociation pathways of [Asp-H]−

Amino acids can undergo fragmentation reactions in MS via various modes, mainly involving the losses of CO2 [49], NH3 [50], and H2O [50]. The aforementioned experiment by Choid [37] is a characteristic example (see Fig. 1). In the following sections, we will discuss the detailed fragmentation pathways of [Asp-H]− in losing CO2, NH3, and H2O and the corresponding fragment structures. The energies of all stationary points in the energy surface are relative to isomer 5. In search of the dissociative sites [37], we considered all located isomers of [Asp-H]−. The pathways given below are those possessing high kinetic and/or thermodynamic predominance. 3.2.1 Fragmentation pathways to [Asp–H–CO2]– Figs. 3a and S3a show the dissociation pathways towards the molecular ion products [Asp–H–CO2]– via the loss of CO2 at the B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory, respectively. The energies of the optimized stationary points on those pathways are provided in Tables S3 and S4 (see Supporting Information). The computational results indicate that such dissociations can begin from either 5 or 6 via two types of fragmentation, i.e., with or without activated dissociation transition states. The isomers of the resulting molecular ion [Asp–H–CO2]– involve 19, 20, 21, 22, and 23. The fragmentation reactions losing CO2 via the located dissociation transition states are characterized by coinstantaneous bond cleavage and intramolecular proton shift. As shown in Figs. 3a and S3a, 5 can undergo direct bond cleavage with the release of CO2 accompanied by intramolecular proton transfer to α and β carbons through the interconversion transition states TS17 and TS18 to furnish the molecular ion products 19 (Path I) and 20 (Path II), respectively. Similarly, 6 proceeds via such a fragmentation mechanism to provide the molecular ion fragments 20 (Path III) and 21 (Path IV) with the concomitant elimination of CO2. 5 / 25

For the fragmentation processes without activated dissociation transition states, three pathways, Path V ~ Path VII, were obtained in the present theoretical search. As shown in Figs. 3a and S3a, 5 can undergo a H-shift rearrangement from the carboxyl group linked to β carbon to the carboxylate anion linked to α carbon via the interconnecting transition state TS19 to generate intermediate 24 and then by a barrierless bond-rupture process to furnish the ion product 22 and fragment CO2 (Path V). Similarly, 6 can isomerize into a pair of conformational isomers 25 and 26 by H-shift rearrangements through the transition states TS20 and TS21, followed by barrierless thermodynamic bond-cleavage reactions to molecular ions 22 (Path VI) and 23 (Path VII), respectively, with the removal of CO2. The three barrierless fragmentation processes are theoretically supported by the relaxed dissociation potential energy surface scans for intermediates 24, 25 and 26. The results plotted in Figs. 3b (at the B3LYP/6-311++G(2df,2pd) level)) and S3b (at the MP2/6-31+G(d,p) level) indicate that starting from the equilibrium distance, their relative energies first increase significantly with increasing distance of the breaking CC bond and then tend to constant values as the breaking CC bond becomes longer than 4.5 Å. Thus, no local maximum points appear on the dissociation potential energy curves. 3.2.2 Fragmentation pathways to [Asp–H–NH3]– The present theoretical computation results show three channels for removing the fragment NH3 from [Asp-H]−. Rotamers 15 and 12 are found to be predisposed for such removals, accompanied by the formation of molecular ion products 27 and 28, respectively. Figs. 4 and S4 display the fragmentation reaction pathways computed at the B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory, respectively, and the energies of the relevant stationary points are given in supporting Tables S5 and S6. As shown in Figs. 4 and S4, 15 can undergo a direct NH3-loss reaction via the four-membered ring transition state TS24 to give the molecular ion product 27 (Path VIII). Evidently, the formation of the unsaturated CC double bond in 27 originates from the simultaneous losses of –NH2 and –H at the two neighboring carbon atoms. Alternatively, 15 can undergo a sequential H-shift arrangement and bond rotation isomerization process via the separating transition states TS25 and TS26 to reach the 6 / 25

intermediates 29 and 30. Then, a NH3-loss reaction via the four-membered transition state TS27 can transform 30 into the molecular ion product 27 (Path IX). Furthermore, starting from rotamer 12 and proceeding via the intermediates 31 and 32, the isomerization can reach another intermediate, i.e., 33. In turn, 33 can eliminate a NH3 fragment to generate the molecular ion product 28 (Path X). 3.2.3 Fragmentation pathways to [Asp–H–H2O]–

The hydroxyl of the carboxyl group is a kinetically and thermodynamically less stable site in the fragmentation reactions of amino acids [51,52]. Therefore, it should be the fragment source of the loss of H2O. The optimized results indicate that 14 and 15, a pair of rotamers, are prone to dissociation towards the fragment H 2O and molecular ion

product

34.

The

computed

fragmentation

processes

at

the

B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory are shown in Figs. 5 and S5, respectively, and the energies of the optimized minimum points and transition states are listed in the supporting Tables S7 and S8. As shown in Figs. 5 and S5, to reach the suitable activated structures separating the reactant and products, the –CH2–COOH bond rotations of 14 and 15 can proceed from two orientations to form the bicyclic transition states TS33 and TS34. Then, the H2O molecule can be eliminated from the two transition states to generate the same four-membered ring molecular ion product 34 (Path XI and Path XII, respectively).

3.3 Fragmentation mechanism and comparison with experiment

Based on the data listed in Figs. 3a and S3a, it can be seen that the four reactions losing CO2 via Path I ~ Path IV are thermodynamically spontaneous, although they are endothermic. Furthermore, Path II leading to ion product 20 is kinetically more favorable than the other three dissociation pathways by at least 2 kcal mol–1. For the barrierless thermodynamic dissociation reactions losing CO2 via Path V ~ Path VII, Path V is thermodynamically more favorable with the formation of the molecular ion product 22. To estimate the temperature dependence of product distribution and fragmentation manner, we performed a thermal energy correction computation of the four reaction transition states on Path I ~ Path IV and thermodynamic products 22+CO2 and 23+CO2. The results plotted in Fig. 6a indicate that the direct removal of 7 / 25

CO2 from 5 via TS18 to give the molecular ion product 20 (Path II, see Fig. 3a) is kinetically more favorable than other CO2-loss channels when the temperature is lower than 716 K. In contrast, the formation of 22 with the loss of CO2 via Path V (see Fig. 3a) becomes more favorable as the temperature becomes higher than 716 K. Note that the inversion temperature is 760 K at the MP2/6-31+G(d,p) level of theory, as shown in Fig. 6c. For the reaction channels losing NH3, the transition states of the rate-determining step of Path VIII, Path IX, and Path X are TS24, TS27, and TS28 with relative Gibbs free energy heights of 53.18 kcal mol–1, 36.86 kcal mol–1, and 44.00 kcal mol–1, respectively, as illustrated in Fig. 4. Thus, at 598.15 K, Path IX is kinetically the most favorable among the three pathways, and the resulting molecular ion product should be 27 in the trans mode and not the estimated cis species 28 by Choid and co-workers (see Fig. 1) [37]. The thermal energy correction to the three key transition states gives a similar stability ordering (see Figs. 6b and 6d), implying that Path IX is more favorable than Path VIII and Path X in the considered temperature range of 298.15 K ~ 898.15 K. The direct H2O-loss reactions via Path XI and Path XII give the same molecular ion product 34 (see Figs. 5b and S5b), whose structure agrees with the experimental estimation [37]. It should be noted that Path XI is kinetically more favorable than Path XII based on the listed energies in Figs. 5b and S5b. The thermal energy correction in the temperature range of 298.15 K ~ 898.15 K gives the same ordering of kinetic stabilities, as shown in Figs. 6b and 6d. To better understand the preferred dissociation pathways and product distribution, we list the computed results with thermal energy correction for the kinetically favorable channels losing CO2 (Path II and Path V), NH3 (Path IX), and H2O (Path XI), as shown in Fig. 7. The B3LYP/6-311++G(2df,2pd)-computed results apparently indicate that Path IX leading to the ion product 27 and fragment NH3 is more favorable than other pathways when the temperature is lower than 719 K (see Fig. 7a). However, when the temperature is higher than 719 K, 22 becomes the dominant ion product with the elimination of CO2 via Path V. Note that the inversion temperature is 788 K at the MP2/6-31+G(d,p) level of theory (see Fig. 7b). In the available experiment by Choi and coworkers [37], they observed the molecular ion products [Asp–H–H2O]– (m/z 114), [Asp–H–NH3]– (m/z 115), and [Asp–H–CO2]– (m/z 88). The computed structure (34) of [Asp–H–H2O]– is very close 8 / 25

to that (4) experimentally estimated by Choi and coworkers, whereas the structure (3) of [Asp–H–NH3]– resulting from the experiment does not match the present computational result (27). Furthermore, the experimentally estimated structure (2) of [Asp–H–CO2]– is in good agreement with the present result (22) when the temperature is higher than 716 K; however, when the temperature is lower than 716 K, it is inconsistent with the computed structure (20). Additionally, the distribution of the predominant molecular ion products coincides well with the experimental estimation resulting from the characterization of m/z of the observed molecular ions [37], which implies the existence of an inversion temperature. When the temperature is higher and lower than the inversion temperature, the dominant dissociation processes are the eliminations of CO2 and NH3, respectively. The computed inversion temperature is 719 K and 788 K at the B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory, respectively.

4. Conclusions

The preferred fragmentation channels losing CO2 were either Path II or Path V with inversion temperatures of 716 K and 760 K at the B3LYP/6-311++G(2df,2pd) and MP2/6-31+G(d,p) levels of theory, respectively. In the three optimized fragmentation reactions with the loss of NH3, Path IX was kinetically more favorable. Furthermore, Path XI was the energetically preferred pathway for the loss of H2O. For the four kinetically favorable reactions losing CO2, NH3, and H2O, the B3LYP/6-311++G(2df,2pd)-computed thermal energy corrections gave the dominant product distribution at different temperatures, i.e., Path IX losing NH3 and providing the molecular ion product 27 is more favorable than the other pathways when the temperature is lower than 719 K; however, Path V became kinetically the most favorable channel with the dominant ion product 22 and the lost fragment CO2 when the temperature was higher than 719 K. When the MP2/6-31+G(d,p) level was used, the inversion temperature is 788 K. Although the present theoretical dissociation mechanism is in good agreement with the experimental observation, the inversion temperature is dependent on the used computational levels. Thus, a deeper analysis is required to obtain more accuracy inversion temperature in further work. 9 / 25

Acknowledgements

This work was supported by National Natural Science Foundation of China (Grant No. 21173072).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/

References

[1] M.E. Sanz, J.C. López, J.L. Alonso, Six conformers of neutral aspartic acid identified in the gas phase, Phys. Chem. Chem. Phys. 12 (2010) 3573–3578, http://dx.doi.org/ 10.1039/b926520a. [2] K.G.R. Pachler, Nuclear magnetic resonance study of some α-amino acids-II. Roational isomerism, Spectrochim. Acta 20 (1964) 581–587, http://dx.doi.org/ 10.1016/0371-1951(64)80055-2. [3] C. Viedma, J.E. Ortiz, T. de Torres, T. Izumi, D.G. Blackmond, Evolution of solid phase homochirality for a proteinogenic amino acid, J. Am. Chem. Soc. 130 (2008) 15274–15275, http://dx.doi.org/ 10.1021/ja8074506. [4] J.E. Baldwin, M.G. Moloney, M. North, Non-proteinogenic amino acid synthesis : synthesis of β, γ-unsaturated α-amino acids from aspartic acid, Tetrahedron 45 (1989) 6319–6330, http://dx.doi.org/ 10.1016/S0040-4020(01)85141-3. [5] R. Yoshioka, O. Ohtsuki, M. Senuma, T. Tosa, Efficient preparation of D-aspartic acid β-methyl ester as an aspoxicillin material by optical resolution, epimerization, and asymmetric transformation, Chem. Pharm. Bull. 37 (1989) 883–886, http://dx.doi.org/ 10.1248/cpb.37.883. [6] S.L. Lim, W.N.H. Tang, C.W. Ooi, E.S. Chan, B.T. Tey, Rapid swelling and deswelling of semi-interpenetrating network poly(acrylic acid)/poly(aspartic acid) hydrogels prepared by freezing polymerization, J. Appl. Polym. Sci. 133 (2016) 43515–43523, http://dx.doi.org/ 10.1002/app.43515. [7] J.Q. Lu, Y.H. Li, D. Hu, X.L. Chen, Y.M. Liu, L.P. Wang, Y.S. Zhao, Synthesis and 10 / 25

Properties of pH-, thermo-, and salt-sensitive modified poly (aspartic acid)/poly (vinyl alcohol) IPN hydrogel and its drug controlled release, Biomed. Res. Int. 2015 (2015) 236745–236756, http://dx.doi.org/ 10.1155/2015/236745. [8] H.J. Wang, L.Y. Wang, One-pot syntheses and cell imaging applications of poly (amino acid) coated LaVO4: Eu3+ luminescent nanocrystals, Inorg. Chem. 52 (2013) 2439–2445, http://dx.doi.org/ 10.1021/ic302297u. [9] J.D. Bernal, The crystal structure of the natural amino acids and belated compounds,

Z.

Krist.

78

(1931)

363–369,

http://dx.doi.org/

10.1524/zkri.1931.78.1.363. [10] J.L. Derissen, H.J. Endeman, A.F. Peerdeman, The crystal and molecular structure of L-aspartic acid,

Acta Cryst.

B 24 (1968)

1349–1354,

http://dx.doi.org/ 10.1107/S0567740868004280. [11] A.M. Silva, S.N. Costa, B.P. Silva, V.N. Freire, U.L. Fulco, E.L. Albuquerque, E.W.S. Caetano, F.F. Maia, Assessing the role of water on the electronic structure and vibrational spectra of monohydrated L-aspartic acid crystals, Cryst. Growth Des. 13 (2013) 4844–4851, http://dx.doi.org/ 10.1021/cg401016v. [12] E.E. Bendeif, C. Jelsch, The experimental library multipolar atom model refinement of L-aspartic acid, Acta Cryst. C 63 (2007) o361–o364, http://dx.doi.org/ 10.1107/S0108270107021671. [13] K. Umadevi, K. Anitha, B. Sridhar, N. Srinivasan, R.K. Rajaram, L-Aspartic acid monohydrate, Acta Cryst. E 59 (2003) o1073–o1075, http://dx.doi.org/ 10.1107/S1600536803014041. [14] S.T. Rao, Refinement of DL-aspartic acid, Acta Cryst. B 29 (1973) 1718–1720, http://dx.doi.org/ 10.1107/S0567740873005339. [15] A. Sequeira, H. Rajagopal, M. Ramanadham, A neutron study on the structure of DL-aspartic acid, Acta Cryst. C 45 (1989) 906–908, http://dx.doi.org/ 10.1107/S010827018801412X. [16] G.M. Wang, Z.X. Li, C.S. Duan, H. Li, DL-aspartic acid, Acta Cryst. E 63 (2007) o4003, http://dx.doi.org/ 10.1107/S1600536807043097. [17] M.L. Chen, Z.J. Lin, Ab initio studies of aspartic acid conformers in gas phase and in solution, J. Chem. Phys. 127 (2007) 154314–154324, http://dx.doi.org/ 10.1063/1.2777161. [18] M.J. Alam, S. Ahmad, Anharmonic vibrational studies of L-aspartic acid using HF and DFT calculations, Spectrochim. Acta A 96 (2012) 992–1004, 11 / 25

http://dx.doi.org/ 10.1016/j.saa.2012.07.135. [19] R. Gahoual, A. Beck, Y.N. François, E. Leize-Wagner, Independent highly sensitive characterization of asparagine deamidation and aspartic acid isomerization by sheathless CZE-ESI-MS/MS, J. Mass Spectrom. 51 (2016) 150–158, http://dx.doi.org/ 10.1002/jms.3735. [20] H. Kaspar, K. Dettmer, Q. Chan, S. Daniels, S. Nimkar, M.L. Daviglus, J. Stamler, P. Elliott, P.J. Oefner, Urinary amino acid analysis: A comparison of iTRAQ®–LC–MS/MS, GC–MS, and amino acid analyzer, J. Chromatogr. B 877 (2009) 1838–1846, http://dx.doi.org/ 10.1016/j.jchromb.2009.05.019. [21] J.P. Hardy, S.L. Kerrin, Rapid determination of twenty amino acids by gas chromatography,

Anal.

Chem. 44 (1972) 1497–1499, http://dx.doi.org/

10.1021/ac60316a054. [22] D.B. Robb, T.R. Covey, A.P. Bruins, Atmospheric pressure photoionization: An ionization method for liquid chromatography-mass spectrometry, Anal. Chem. 72 (2000) 3653–3659, http://dx.doi.org/ 10.1021/ac0001636. [23] V.V. Laiko, M.A. Baldwin, A.L. Burlingame, Atmospheric pressure matrix-assisted laser desorption/ionization mass spectrometry, Anal. Chem. 72 (2000) 652–657, http://dx.doi.org/ 10.1021/ac990998k. [24] Z. Takáts, J.M. Wiseman, B. Gologan, R.G. Cooks, Mass spectrometry sampling under ambient conditions with desorption electrospray ionization, Science 306 (2004) 471–473, http://dx.doi.org/ 10.1126/science.1104404. [25] R.B. Cody, J.A. Laramée, H.D. Durst, Versatile new ion source for the analysis of materials in open air under ambient conditions, Anal. Chem. 77 (2005) 2297–2302, http://dx.doi.org/ 10.1021/ac050162j. [26] E.C. Huang, T. Wachs, J.J. Conboy, J.D. Henion, Atmospheric pressure ionization

mass

spectrometry,

Anal.

Chem.

62

(1990)

713A–725A,

http://dx.doi.org/ 10.1021/ac00212a727. [27] W.C. Byrdwell, Atmospheric pressure chemical ionization mass spectrometry for analysis

of

lipids,

Lipids

36

(2001)

327–346,

http://dx.doi.org/

10.1007/s11745-001-0725-5. [28] N.B. Cech, C.G. Enke, Practical implications of some recent studies in electrospray ionization fundamentals, Mass Spectrom. Rev. 20 (2001) 362–387, http://dx.doi.org/ 10.1002/mas.10008. [29] A. Raffaelli, A. Saba, Atmospheric pressure photoionization mass spectrometry, 12 / 25

Mass Spectrom. Rev. 22 (2003) 318–331, http://dx.doi.org/ 10.1002/mas.10060. [30] S Özcan, H.Z. Şenyuva, Improved and simplified liquid chromatography/ atmospheric pressure chemical ionization mass spectrometry method for the analysis of underivatized free amino acids in various foods, J. Chromatogr. A 1135 (2006) 179–185, http://dx.doi.org/ 10.1016/j.chroma.2006.09.039. [31] J.B. Fenn, M. Mann, C.K. Meng, S.F. Wong, C.M. Whitehouse, Electrospray ionization for mass spectrometry of large biomolecules, Science 246 (1989) 64–71, http://dx.doi.org/ 10.1126/science.2675315. [32] E. Brewer, J. Henion, Atmospheric pressure ionization LC/MS/MS techniques for drug disposition studies, J. Pharm. Sci. 87 (1998) 395–402, http://dx.doi.org/ 10.1021/js9701059. [33] P.M. Mayer, C. Poon, The mechanisms of collisional activation of ions in mass spectrometry, Mass Spectrom. Rev. 28 (2009) 608–639, http://dx.doi.org/ 10.1002/mas.20225. [34] F.W. Mclafferty, P.F. Bente, R. Kornfeld, S.C. Tsai, I. Howe, Collisional activation spectra of organic ions, J. Mass Spectrom. 30 (1995) 797–806, http://dx.doi.org/ 10.1002/jms.1190300604. [35] D.K. Han, J. Eng, H.L. Zhou, R. Aebersold, Quantitative profiling of differentiation-induced microsomal proteins using isotope-coded affinity tags and mass spectrometry, Nat. Biotechnol. 19 (2001) 946–951, http://dx.doi.org/ 10.1038/nbt1001-946. [36] N.C. Maragou, N.S. Thomaidis, M.A. Koupparis, Optimization and comparison of ESI and APCI LC-MS/MS methods: A case study of irgarol 1051, diuron, and their degradation products in environmental samples, J. Am. Soc. Mass Spectrom. 22 (2011) 1826–1838, http://dx.doi.org/ 10.1007/s13361-011-0191-z. [37] S.S. Choi, O.B. Kim, Fragmentation of deprotonated amino acids in atmospheric pressure chemical ionization, Int. J. Mass Spectrom. 338 (2013) 17–22, http://dx.doi.org/ 10.1016/j.ijms.2013.01.006. [38] M.A. Van Agthoven, P. Colomby, M. Surugue, C. Beaugrand, F.L. Wind, J.C. Tabet, Ion kinetic energy measurements in two soft ion ejection methods from a quadrupole ion trap, Int. J. Mass Spectrom. 296 (2010) 59–64, http://dx.doi.org/ 10.1016/j.ijms.2010.08.026. [39] R.S. Mason, D.R. Williams, I.P. Mortimer, D.J. Mitchell, K. Newman, Ion formation at the boundary between a fast flow glow discharge ion source and a 13 / 25

quadrupole mass spectrometer, J. Anal. At. Spectrom. 19 (2004) 1177–1185, http://dx.doi.org/ 10.1039/B400563P. [40] A.D. Becke, Density-functional exchange-energy approximation with correct asymptotic behavior, Phys. Rev. A 38 (1988) 3098–3100, http://dx.doi.org/ 10.1103/PhysRevA.38.3098. [41] P.J. Stephens, F.J. Devlin, C.F. Chabalowski, M.J. Frisch, Ab initio calculation of vibrational absorption and circular dichroism spectra using density functional force fields, J. Phys. Chem. 98 (1994) 11623–11627, http://dx.doi.org/ 10.1021/j100096a001. [42] M. Head-Gordon, J.A. Pople, M.J. Frisch, MP2 energy evaluation by direct methods,

Chem.

Phys.

Lett.

153

(1988)

503–506,

http://dx.doi.org/10.1016/0009-2614(88)85250-3. [43] J.J.W. McDouall, K. Peasley, M.A. Robb, A simple MC SCF perturbation theory: Orthogonal valence bond Møller-Plesset 2 (OVB MP2), Chem. Phys. Lett. 148 (1988) 183–189, http://dx.doi.org/10.1016/0009-2614(88)80296-3. [44] A. Lenz, L. Ojamäe, On the stability of dense versus cage-shaped water clusters: Quantum-chemical investigations of zero-point energies, free energies, basis-set effects and IR spectra of (H2O)12 and (H2O)20, Chem. Phys. Lett. 418 (2006) 361–367, http://dx.doi.org/ 10.1016/j.cplett.2005.11.013. [45] G.S. Maciel, A.C.P. Bitencourt, M. Ragni, V. Aquilanti, Studies of the dynamics around the O–O bond: Orthogonal local modes of hydrogen peroxide, Chem. Phys. Lett. 432 (2006) 383–390, http://dx.doi.org/ 10.1016/j.cplett.2006.10.073. [46] S. Pan, M. Solà, P.K. Chattaraj, On the validity of the maximum hardness principle and the minimum electrophilicity principle during chemical reactions, J. Phys. Chem. A 117 (2013) 1843−1852, http://dx.doi.org/ 10.1021/jp312750n. [47] F. Anagnostou, E. Drougas, A.M. Kosmas, Quantum mechanical analysis of decomposition pathways of chloromethyl hypochlorite, J. Mol. Struct. THEOCHEM.

728

(2005)

189–196,

http://dx.doi.org/

10.1016/j.theochem.2005.05.021. [48] Z. Li, M.H. Matus, H.A. Velazquez, D.A. Dixon, C.J. Cassady, Gas-phase acidities of aspartic acid, glutamic acid, and their amino acid amides, Int. J. Mass Spectrom. 265 (2007) 213–223, http://dx.doi.org/ 10.1016/j.ijms.2007.02.009. [49] M. Eckersley, J.H. Bowie, R.N. Hayes, Collision-induced dissociations of deprotonated α-amino acids. The occurrence of specific proton transfers 14 / 25

preceding fragmentation, Int. J. Mass Spectrom. Ion Processes 93 (1989) 199–213, http://dx.doi.org/ 10.1016/0168-1176(89)80097-7. [50] J.H. Bowie, C.S. Brinkworth, S. Dua, Collision-induced fragmentations of the (M-H)– parent anions of underivatized peptides: An aid to structure determination and some unusual negative ion cleavages, Mass Spectrom. Rev. 21 (2002) 87–107, http://dx.doi.org/ 10.1002/mas.10022. [51] F.M. Benoit, A.G. Harrison, Predictive value of proton affinity. Ionization energy correlations involving oxygenated molecules, J. Am. Chem. Soc. 99 (1977) 3980–3984, http://dx.doi.org/ 10.1021/ja00454a015. [52] N.E. Middlemiss, A.G. Harrison, The structure and fragmentation of protonated carboxylic acids in the gas phase, Can. J. Chem. 57 (1979) 2827–2833, http://dx.doi.org/ 10.1139/v79-459.

15 / 25

Figure captions Fig. 1. Proposed structures of [Asp-H]− and its dissociation fragments by Choid and co-workers Fig. 2. Rotational isomerization potential energy profile of [Asp-H]− using the B3LYP/6-311++G(2df,2pd)-computed relative Gibbs free energies and enthalpies (in parenthesis) at 598.15 K Fig. 3. B3LYP/6-311++G(2df,2pd)-computed (a) relative Gibbs free energies and enthalpies (in parenthesis) on the CO2-loss pathways to [Asp–H–CO2]– at 598.15 K and (b) potential energy surface scanning curves of the CO2-elimination reactions Fig. 4. Potential energy profile of the NH3-loss reaction of [Asp-H]− using the B3LYP/6-311++G(2df,2pd)-computed relative Gibbs free energies and enthalpies (in parenthesis) at 598.15 K Fig. 5. Stationary point structures (a) and potential energy profile (b) for the H2O-loss pathways using the B3LYP/6-311++G(2df,2pd)-computed relative Gibbs free energies and enthalpies (in parenthesis) at 598.15 K Fig. 6. Computed temperature dependence of dissociation pathways eliminating CO2 (a), NH3 and H2O (b) at the B3LYP/6-311++G(2df,2pd) level of theory. The corresponding results computed at the MP2/6-31+G(d,p) level of theory are outlined in (c) and (d), respectively Fig. 7. Computed temperature dependence of the kinetically favorable pathways to eliminate CO2, NH3, and H2O at the B3LYP/6-311++G(2df,2pd) (a) and MP2/6-31+G(d,p) (b) levels of theory

16 / 25

17 / 25

18 / 25

19 / 25

20 / 25

21 / 25

22 / 25

23 / 25

Highlights The NH3-loss pathway of [Asp-H]− is dominant as the temperature is lower than 719 K The CO2-loss is the kinetically most favorable as the temperature is higher than 719 K The H2O-loss pathway is less favorable than the NH3- and CO2-loss channels

24 / 25

Graphical Abstract Fragmentation pathways and product distribution of deprotonated aspartic acid were investigated

25 / 25