Conversion of amino-acids by electrons at subexcitation energies

Abstract

Electron attachment to gas phase cysteine is investigated in a crossed electron/molecular beam experiment. The yield functions of the detected anions exhibit signatures of dissociative electron attachment (DEA), initiated by shape (<3 eV) and core-excited (>5 eV) resonances. These parent anion precursor states decompose into a variety of negatively charged fragments with one of the most dominant channels being dehydrogenation, i.e., abstraction of the H radical with the excess electron remaining on the cysteine-like moiety. While this reaction is operative only within the low energy range, S⁻ and SH⁻ appear from both resonance regions. From energy considerations it is shown that formation of S⁻ below 3 eV must be associated with a conversion of cysteine into alanine. The present results demonstrate the capacity of electrons at subexcitation energies to effectively change the nature of amino acids.

---

Introduction

Cysteine (Cys, NH₂–CH–R–COOH where R = CH₂SH) is an amino-acid implicated in various biological functions such as enzyme activities¹ and the building up of protein structures.^2,3 Here we study the interaction of low-energy (<10 eV) electrons (LEEs) with isolated cysteine targets. Such investigations can be directly related to radiation damage in biological systems.

Exposure of living cells to ionizing radiation induces genomic instabilities leading to diseases, such as cancers^4,5 or mutations.⁶ At the early time after depositing energy into the biological medium (fs–ps), ballistic secondary electrons are found to be the most predominant of the radiolytic species produced.^7–9 These electrons, with energies typically below 20 eV, can thereafter undergo reactions with the surrounding medium, altering then its properties.^10,11 Indeed, the ability of such slow electrons to irreversibly damage DNA has been demonstrated in terms of strand breaks,⁸ and fragmentation of the nucleic acids sub-units: DNA nucleobases,^12–14 sugar¹⁵ and also environing water.¹⁶ Since proteins are closely bound to nucleic acids (e.g., protein histones), these slow electrons may also alter proteins in the same manner, namely by damaging their building-blocks (amino-acids).^17–20

Previous investigations on electron stimulated desorption (ESD) of anions from physisorbed cysteine films have shown that low-energy (<30 eV) electrons are able to efficiently decompose Cys via dissociative electron attachment (DEA).²⁰ In this process, the incoming electron is captured by a Cys molecule to form a transitory negative ion (TNI). This precursor parent anion may then dissociate into a negative fragment ion and one or more neutral radical counterpart(s). In fact from ESD experiments, only light anionic species with mass lower than 35 atomic mass units (amu) have been reported.²⁰ Heavier fragments are not expected to desorb since they might not possess sufficient kinetic energy to overcome the attractive forces (polarization) induced by the solid,²¹ and may thus remain at the surface. On the contrary in the gas phase, the situation is different since any long-lived (i.e., on the mass spectrometric time scale) negative ion can be detected. For instance from low-energy electron impact on gaseous glycine (Gly, NH₂–CH–R′–COOH where R′ = H), Gohlke et al. have reported the production of larger mass anionic fragments, such as (Gly–H)⁻, COOH⁻ and H₂NC₂O⁻,¹⁷ which were not detected in ESD experiments.²⁰ Moreover, the presence of neighboring molecules changes the structure of many of the amino-acids. In the gas phase, cysteine^22,23 adopts a “neutral” conformation: NH₂–CH–R; while in the solid phase, Cys rather appropriates the zwitterionic arrangement NH₃⁺–CH–R⁻.^24,25 Therefore molecular orbitals (MOs) are likely to differ between the “neutral” cysteine and its zwitterionic form, as being observed for glycine.^26–29 Additionally the proton transfer to the amino group produces an amino-acid with a very large dipole moment,^28,29 that could substantially stabilize the lowest unoccupied molecular orbital, and thus giving rise to a bound valence anion state of the zwitterion.^30,31

In the present work we investigate gas phase Cys in order to reveal the intrinsic properties with respect to low energy electron interaction.

Experimental

The crossed-beam experimental apparatus has been described previously.³² Briefly, an incident electron beam of well-defined energy (∼10 nA, FWHM ∼ 0.15 eV) is generated from a trochoidal electron monochromator.³³ The electron beam intersects at a right angle with an effusive molecular beam. Cysteine molecules emanate from a resistively heated oven containing approximately 50 mg of 99% purity powder (Aldrich Ltd.). The temperature, measured by a platinum resistance to be approximately 390 K, is below the molecular decomposition temperature (490 K). Therefore, the original structure of the investigated amino-acid is not likely to be altered (see below). The experimental chamber, having a base pressure of 3 × 10⁻⁸ mbar, is maintained at about oven temperatures by two in vacuo halogen lamps during the experiments to prevent powder condensation on the surfaces (plates, chamber wall). This may otherwise lead to undesirable changes in contact potentials. The negative ions are extracted from the reaction volume by a small electric field (<0.5 V cm⁻¹) towards a quadrupole mass analyzer, and are detected by single pulse counting techniques. The electron energy scale is calibrated by measuring the formation of SF₆⁻ ions, which exhibits a sharp peak at 0 eV.

Results and discussion

Low-energy (<10 eV) electrons induce dissociation of Cysteine, producing the negative fragments shown in Figs. 1 and 2. The pronounced resonance profiles are indicative of DEA. The products O⁻ and/or NH₂⁻ (16 amu), S⁻ (32 amu) and SH⁻ (33 amu) (Fig. 1) have been previously observed from ESD experiments,²⁰ however, this has been restricted to energies above 4 eV. In fact, the fragments detected at 32 amu and 33 amu can in principle also be ascribed to O₂⁻ and HO₂⁻ anions, respectively, arising from the dissociation of the carboxyl group. However, these species have not been detected in previous experiments on electron impact on various amino-acids.^17,34 Therefore, O₂⁻ and HO₂⁻ fragments are not likely to be produced here, although we cannot completely rule out this possibility. In addition, the present gas phase study generates larger anion fragments: (Cys–H)⁻ (120 amu), 101 amu, 71 amu and 60 amu fragment (Fig. 2). Here, (Cys–H) represents the cysteine molecule which has lost an H atom. We do not observe other negative species, such as H⁻. If formed, those signals would be below our detection limits.

Fig. 1 Incident electron energy dependence of (a) SH₂⁻, (b) S⁻ and (c) O⁻ anions produced from electron impact on gaseous cysteine.

Fig. 2 Incident electron energy dependence of (a) (Cys–H)⁻ (120 amu), (b) 101 amu, (c) 71 amu, and (d) 60 amu anion produced from electron impact on gaseous cysteine.

It is noteworthy that light fragments (Fig. 1) are produced within the investigated energy range, while the formation of larger species (Fig. 2) is exclusively restricted to incident electron energies below 3 eV. This observation reflects the possibility of excess energy distribution among the fragments. Indeed, detection of (Cys–H)⁻ below 3 eV is limited to the reaction consisting of only two fragments: (Cys–H)⁻ + H. A part from translational energy of both fragments, the available excess energy must be deposited as internal energy of the ion, which above 3–4 eV becomes unstable with respect to dissociation and autodetachment. On the contrary, observation of a light negative fragment such as S⁻ does not preclude dissociation into more than two neutral species and may hence be also observed at higher energy (see below). This mode of behavior is a conspicuous feature of the decomposition of polyatomic negative ions.³²

Before discussing the observed DEA reactions in more detail we shall consider the problem of possible thermal decomposition of Cys upon heating. Amino-acids may undergo thermal decomposition producing CO₂, via degradation of the carboxyl-group COOH.³⁵ In the electron transmission spectra (ETS) of cysteine, Aflatooni et al. reported a signature at 3.59 eV, which they assigned to the presence of CO₂.²⁹ If CO₂ is formed by thermal decomposition, its signature (O⁻ fragment) should be seen on the recorded 16 amu negative fragment yield function (Fig. 1c). In the energy range below 10 eV, DEA to CO₂ produces two resonances located at 4.4 eV and 8.2 eV (peak positions) with the 8.2 eV feature three times more intense than that at 4.4 eV. However, Fig. 1c does not show the signatures of DEA to CO₂. The 16 amu ion yield exhibits a small feature near 4.5 eV, but not a contribution at 8.2 eV indicating that the presence of CO₂ in the reaction chamber can be excluded. From ETS studies on CH₃SH³⁶ (which may model the sulfur group of Cys), a broad feature around 2.9 eV was reported ascribed to electron accommodation into a σ*(C–S–H) MO. Therefore the resonance observed at 3.59 eV in ETS on Cys²⁹ might be associated to the σ*(C–S–H) MO in Cys.

As mentioned above, the 16 amu fragment can be due to O⁻ and/or NH₂⁻. The peak located at 6.1 eV in the O⁻ and/or NH₂⁻ yield function (Fig. 1c) is indicative of the formation of a core-excited resonance, prior to Cys^#− dissociation.³⁷ The incoming electron generates an electronically excited state cysteine (π* and/or σ*), concomitantly being trapped for some time. Since the electron affinity of O (1.46 eV) is nearly twice that of NH₂ (0.78 eV),³⁸ energy arguments suggest the formation of O⁻ rather than NH₂⁻. However, we cannot completely rule out some contributions of NH₂⁻ to the 16 amu ion yield.

The O⁻ fragment can result from the cleavage of the C=O double bond or a concerted reaction as represented in Fig. 3 by reactions C and A, respectively. The negative species NH₂⁻ arises from a single C–N bond cleavage (reaction C). Based on the electron affinity of O (1.46 eV) and NH₂ (0.78 eV),³⁷ and on the corresponding average bond dissociation energies,³⁹ the thermodynamic threshold for reactions A, B and C (Fig. 3) are estimated to be 2.5 eV, 2.2 eV and 6.0 eV, respectively. The experimental appearance potential of the 16 amu ion is observed near 4.5 eV and thus considerably above the calculated threshold of reaction A and B. For the concerted reaction A one expects an appreciable energy barrier.

Fig. 3 Possible dissociation pathways for (A and C) O⁻ and (B) NH₂⁻ production after low-energy electron attachment to cysteine. Reactions D and E transform cysteine into alanyl and alanine, respectively.

The yield functions of S⁻, SH⁻ (Fig. 1) and dehydrogenated cysteine anion, (Cys–H)⁻ (Fig. 2) exhibits two structures in the energy range below 3 eV. At this energy shape resonances are usually involved in the formation of Cys^#−: the incoming electron is captured by the shape of the electron–molecule potential into an empty molecular orbital of cysteine. In electron transmission, the resonance observed at 1.98 eV is assigned to the accommodation of the excess electron into an empty π* (–COOH) MO.²⁹ Bearing in mind that ET mirrors the energy of the transient negative ion state accessed by electron capture (the initial transition), DEA is a convolution of the capture process with the (energy dependent) decay probability into the particular channel (including auto-detachment).⁴⁰ This results in a shift of the DEA peak to lower energy (survival probability shift) with respect to that observed in ETS.⁴⁰ Therefore, we attribute the SH⁻ (1.6 eV), S⁻ (1.1 eV) and (Cys–H)⁻ (1.1 eV) fragment to arise from the π* (–COOH) precursor parent anion with particular shifts of the DEA resonance position. We also mention that in glycine a resonance at 1.1 eV was observed on the (Gly–H)⁻ channel and assigned to the π* (–COOH) MO.²⁹

The three negative fragments show an additional low-energy peak located at 0.4 eV (S⁻, SH⁻) and 0.6 eV ((Cys–H)⁻). The associated reactions may be due to the decomposition of Cys^#− with the extra electron localized on the sulfur group. Alternatively, these peaks can also arise from DEA to vibrationally excited target molecules (hot band transition).⁴¹ For instance, if at room temperature a compound possesses a DEA resonance at low energy but clearly off zero eV (e.g., DEA to CF₃Cl, which produces Cl⁻ fragment at 1.5 eV), moderate thermal excitation can lead to the appearance of an additional sharp peak near 0 eV.⁴¹

At low energy, formation of SH⁻ arises from a single S–C bond rupture, generating the alanyl radical as the neutral counterpart (Fig. 3D) according to

(NH₂–CH(CH₂SH)–COOH)^#− → SH⁻ + (NH₂–CH(CH₂)–COOH)˙

Taking the electron affinity of SH (2.32 eV³⁷) and the C–S bond dissociation energy (2.4 eV),⁴² this reaction becomes approximately thermo-neutral.

The production of S⁻ may arise from the cleavage of the C–S and the S–H bond:

(NH₂–CH(CH₂SH)–COOH)^#− → S⁻ + (NH₂–CH(CH₂)–COOH)˙ + H˙

With the available thermodynamic values (EA(S) = 2.08 eV, D(C–S) = 2.4 eV and D(S–H) = 3.9 eV^37,38,40), the threshold for this dissociative reaction becomes 4.2 eV indicating that it is only accessible from the high energy resonance near 6 eV.

Alternatively, the formation of S⁻ can be driven by a concerted reaction. During the lifetime of the TNI expulsion of S⁻ is accompanied by the transfer of H to the CH₂ site (Fig. 3E) according to

(NH₂–CH(CH₂SH)–COOH)^#− → S⁻ + NH₂–CH(CH₃)–COOH

thereby forming alanine. With respect to the previous reaction the thermodynamic threshold is lowered by D(C–H) = 4.4 eV.³⁸ Alanine formation then becomes exothermic and is accessible from the low energy resonances. Concerted reaction with substantial displacement of hydrogen has also been reported below 10 eV for DEA to ethylene carbonate.⁴³

SH⁻ and S⁻ anions formed from resonant features above 5 eV (Fig. 1a,b) involve core-excited resonances. The location of these DEA features is in reasonable agreement to those obtained in ESD from films of cystein.²⁰ Slight differences in the position of the peaks arise from the effect of the solid and the particular energy constraints for ion desorption.^21,44

Finally as shown in Fig. 2, various heavy negative species (i.e., 101 amu, 71 amu and 60 amu) are detected with a comparatively high intensity. Since the assessment of the observed negative fragments is not unambigous in every case, we refrain from a closer discussion of the associated reactions at this present stage. We nevertheless can tentatively assign by stoichiometry the 101 amu ion to (C₃H₃NOS)⁻ anion. This negative fragment is produced via the lost part of a 20 amu neutral counterpart, e.g., H₂O + H₂. The formation of the 71 amu ion, (C₃H₃O₂)⁻, can probably be accompanied by that of the NH₃ molecule and SH radical. Although we can not definitely assign each decomposition channel, the present findings show the ability of slow electrons to initiate complex reactions in cysteine.

Conclusion

Dissociative electron attachment to gaseous cysteine leads to various fragments attributed to (Cys–H)⁻, O⁻ and/or NH₂⁻, S⁻ and SH⁻ as well as their respective neutral counterparts. Molecular fragmentation arises from simple bond cleavage ((Cys–H)⁻, SH⁻) or more complex reactions involving substantial reorganization of the nuclei within the lifetime of the transitory parent anion.

At subexcitation energies, formation of SH⁻ and S⁻ is accompanied by the formation of the neutral alanyl radical and alanine molecule, respectively. At higher energies multiple fragmentation reactions may occur. The present results demonstrate the ability of low-energy electrons to convert amino-acids. Furthermore, since the sulfur site in cysteine also controls the structure of proteins, C–S bond rupture can modify the structure of proteins leading to a loss of the activities of proteins and enzymes.^45,46

Acknowledgements

This work has been supported by the European Union, The Fonds der Chemischen Industrie and the Freie Universität Berlin. HAC is a fellow of the European Union EU-EPIC (Electron and Positron Induced Chemistry) Network.

References

1. M. Liu, Y. Huang, J. Wu, E. Wang and Y. Wang, Biochemistry, 1999, 38, 11 006–11 011.
2. A. Rodriguez-Casado and G. J. Thomas, Jr., Biochemistry, 2003, 42, 3437–3445.
3. J. Qi, J. Wu, G. A. Somkuti and J. T. Watson, Biochemistry, 2001, 40, 4531–4538.
4. M. Gundestrup and H. H. Storm, Lancet, 1999, 354, 2029–2031.
5. K. H. Kraemer, Proc. Nat. Acad. Sci. USA, 1997, 94, 11–14.
6. A. Kronenberg, Adv. Space Res., 1994, 14, 339–346.
7. ICRU Report, International Commission on Radiation Units and Measurements, DC, Washington, 1979, vol. 31.
8. S. M. Pimblott, J. A. LaVerne, A. Mozumder and N. Green, J. Phys. Chem., 1990, 94, 488–495.
9. V. Cobut, Y. Fongillo, J. P. Patau, T. Goulet, M. J. Fraser and J. P. Jay-Gerin, Radiat. Phys. Chem., 1998, 51, 229–243.
10. L. Sanche, J. Chim. Phys., 1994, 94, 216–225.
11. B. Boudaïffa, P. Cloutier, D. Hunting, M. A. Huels and L. Sanche, Science, 2000, 287, 1658–1660.
12. H. Abdoul-Carime, P. Cloutier and L. Sanche, Radiat. Res., 2001, 155, 625–633.
13. H. Abdoul-Carime, S. Gohlke and E. Illenberger, 2003, submitted.
14. H. Abdoul-Carime, M. A. Huels, E. Illenberger and L. Sanche, 2003, in preparation.
15. D. Antic, L. Parenteau, M. Lepage and L. Sanche, J. Phys. Chem., 1999, 103, 6611–6619.
16. W. C. Simpson, T. M. Orlando, L. Parenteau, K. Nagesha and L. Sanche, J. Chem. Phys., 1998, 108, 5027–5034.
17. S. Gohlke, A. Rosa, E. Illenberger, F. Brüning and M. A. Huels, J. Chem. Phys., 2000, 116, 10 164–10 169.
18. H. Abdoul-Carime, S. Cecchini and L. Sanche, Radiat. Res., 2002, 158, 23–31.
19. H. Abdoul-Carime and L. Sanche, Radiat. Res., 2003, 160, 86–94.
20. H. Abdoul-Carime and L. Sanche, J. Phys. Chem., 2003, to be published.
21. H. Sambe, D. E. Ramaker, L. Parenteau and L. Sanche, Phys. Rev. Lett., 1987, 59, 236–239.
22. G. Junk and H. Svec, J. Am. Chem. Soc., 1963, 85, 839.
23. J. S. Gaffney, R. C. Pierce and L. Friedman, J. Am. Chem. Soc., 1977, 99, 4293.
24. K. Uvdal, P. Bodö and B. Liedberg, J. Colloid Interface Sci., 1992, 149, 162–173.
25. G. J. Leggett, M. C. Davies, D. E. Jackson and S. J. B. Tendler, J. Phys. Chem., 1993, 97, 5348–5344.
26. M. Nyberg, J. Hasselström, O. Karis, N. Wassdhal, M. Weinelt, A. Nielson and L. G. M. Petterson, J. Chem. Phys., 2000, 112, 5420–5427.
27. R. Destro, P. Roversi, M. Barzhaghi and R. E. Marsh, J. Phys. Chem. A, 2000, 104, 1047–1054.
28. M. Gutowski, P. Skurski and J. Simons, J. Am. Chem. Soc., 2000, 122, 10 159–10 162.
29. K. Aflatooni, B. Hitt, G. A. Gallup and P. D. Burrow, J. Chem. Phys., 2002, 16, 6489–6492.
30. I. Powis, J. Chem. Phys., 2000, 104, 878–882.
31. L. Klasine, J. Electron. Spectrosc. Relat. Phenom., 1976, 8, 161–164.
32. E. Illenberger, in Gaseous Molecular Ions, Topics in Physical Chemistry, ed. H.Baumgärtel, E. U. Frank and W. Grünbein, Steinkopff, Darmstadt, Springer, New York, 1992, vol. 2, part III and references cited herein.
33. A. Stamatovic and G. J. Schulz, Rev. Sci. Instrum., 1968, 39, 1752–1753.
34. H. Abdoul-Carime and L. Sanche, Radiat. Res., 2003, 160, 86–94.
35. P. H. Cannington and S. Ham, J. Electron. Spectrosc. Relat. Phenom., 1979, 15, 79–82.
36. C. Dezarnaud, M. Tronc and A. Modelli, Chem. Phys., 1991, 156, 129–138.
37. Electron-Molecule Interactions and their Applications, ed. L. G. Christophourou, Academic Press, Orlando, 1984.
38. B. K. Janousek and J. I. Brauman, in Gas Phase Ion Chemistry, ed. M. T. Bowers, Academic Press, 1979.
39. T. H. Lowry and K. S. Richardson, in Mechanism and Theory in Organic Chemistry, Harper Collins Publishers, 3rd edn., 1987.
40. E. Illenberger and J. Momigny, in Gaseous Molecular Ions. An Introduction to Elementary Processes Induced by Ionization, Topics in Physical Chemistry, Steinkopff, Springer, Weinheim, New York, 1992.
41. I. Hahndorf and E. Illenberger, Int. J. Mass Spectrom., 1997, 167/168, 87.
42. C. W. Bookwalter, D. L. Zoller, P. L. Ross and M. V. Johnston, J. Am. Soc. Mass Spectrom., 1995, 6, 872–876.
43. M. Stepanovic, Y. Pariat and M. Allan, J. Chem. Phys., 1999, 110, 11 376–11 382.
44. L. Sanche, Scanning Microsc., 1995, 9, 619–656.
45. P. H. Kussie, S. Gorina, N. Pavlevitch, V. Marechal, B. Elenbaas, J. Moreau and A. J. Levine, Science, 1996, 274, 948–953.
46. G. C. Telling, P. Parchi, S. J. DeArmond, P. Cortelli, P. Montagna, R. Gabizon, J. Mastrianni, B. Lugaresi, P. Gambetti and S. B. Prusiner, Science, 1996, 274, 2079–2082.

---