SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

The reactions of ketone/methionine systems are widely used as efficient and selective sources of biorelevant radical species. In this study, we address intramolecular variants of this couple with respect to its photosynthetic utility and as a mechanistic model of underlying elementary reaction steps of biological importance, especially with respect to the study of photoinitiated electron transport in complex peptides. The outcomes of this study are two-fold: (1) steady-state irradiation of sterically constrained benzophenone/methionine dyads afforded stable photocyclization products with high yield and product selectivity. (2) Mechanistic insights into the triplet-triggered product formation were obtained from an analysis of the flash photolysis results and the molecular structure of the stable product formed upon irradiation. Time-resolved experiments identified (net) hydrogen-atom transfer from the methionine as the mechanism of the triplet quenching and the resulting biradicals as the major precursor of the isolated stable product. Both the analyses of triplet quenching and stable-product formation in the diastereomeric pairs point to effects of chiral center configuration, i.e., significant stereoselectivity is observed for all elementary steps. The underlying stereochemical restraints were quantitatively addressed by means of molecular dynamics simulations.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

The thioether side chain of methionine in peptides and proteins is susceptible to oxidation. Indeed, methionine is one of the most easily oxidized amino acid residues by reactive oxygen species [1, 2]. Such oxidation is of great interest because of its relevance to the inactivation of protein pharmaceuticals [3], aging [4, 5], and neurodegenerative diseases, such as Parkinson's [6] and Alzheimer's [7]. The important role of methionine-containing compounds in biological processes has stimulated numerous studies on simpler systems to understand the oxidation of the thioether group in biologically relevant compounds [8-17]. These efforts have involved both the characterization of the reactive species in these oxidations and the detailed reaction mechanisms in which these species participate. Besides radiolytic techniques, particularly photoinduced reactions of ketone/methionine are widely used as efficient and selective sources of biorelevant radical species like sulfur-centered radical cations.

Extended work by Marciniak et al. (transient spectroscopy; [10, 11, 14, 18]), Yurkovskaya and Vieth and Goez et al. (chemically induced dynamic nuclear polarization; [19-21]) has addressed intermolecular one-electron photooxidations of methionine derivatives by 4-carboxybenzophenone (CB) in detail. The primary photochemical step in these reactions was identified to involve an electron transfer from the sulfur atom to the triplet state of the carbonyl. It was shown that the radical-ion complex decays competitively by the following: (1) a back electron-transfer process to form the ground state of the reactants, (2) proton transfer within the complex, with diffusion apart of the CB ketyl radical and an α-thio-alkyl radical, and (3) escape of the radical ions into the bulk solution. The ratio between these three decay channels varied, depending on the quencher and molecular structures. Notably, this chemistry between benzophenone and methionine is the basis of the widely used biochemical method of photoaffinity labeling of proteins, which gives highly selective photochemical coupling reactions [22, 23]. Recently, Moretto et al. have reported data on an intramolecular side-chain to side-chain macrocyclization reaction between BP and Met residues in a helical heptapeptide [24]. This type of photocyclization resembles earlier studies by Griesbeck et al., who explored the preparative potential to some extent [25, 26].

The underlying geometrical restraints of photocyclization reactions are not yet well understood, so that we were attracted by the question whether this type of intramolecular side-chain to side-chain ring formation can occur in the already quite rigid structure of a diketopiperazine (DKP)-based BP-Met. Besides their synthetic potential, these intramolecular variants, with both functionalities in the same molecular structure, can serve as model systems for providing detailed mechanistic insights into the underlying elementary reaction steps of biological relevance. Thus, our study of intramolecular triplet quenching in a pair of benzophenone/methionine (BP-Met) dyads is motivated by its model character for photoinitiated electron transport in complex peptides [27, 28]. Although this approach can serve as a useful tool in understanding the radical processes that take place in vivo, it has been the subject for only a few studies. Thus, in this contribution we address the triplet reactivity of benzophenone in conformationally controlled donor–acceptor (D–A) pairs, with special respect to the effect of chiral center configurations on the diastereoselectivity of triplet-quenching processes and intermediates' dynamics. Structures of the dyads are summarized in Scheme 1.

image

Scheme 1. (a) Structures of the diketopiperazines 12 and notation of atom positions, (b) notation of the side-chain rotamers (R denotes the respective BP or CH2–S–CH3 from methionine residue).

Download figure to PowerPoint

Mechanistic features of the overall UV light-induced intramolecular reactions, i.e., transient lifetimes and quantum yields of the primary reactions were studied by nanosecond laser flash photolysis (LFP). In this manner, intramolecular quenching rate constants were obtained, whereas the efficiencies of irreversible reaction pathways were tested and quantified by steady-state UV photolysis. In addition, theoretical work involved studies of the ground-state structures and conformations, which yielded quantitative information on the populations of the side-chain rotamers in the dyads' ground states and, in particular, furnished distributions of the sulfur to carbonyl oxygen distances. To provide insight into these aspects, molecular dynamics MD simulations were performed. A correlation is established between intramolecular reactivity and the structural preferences that are reflected by the non bonding distances of the interacting functional groups in the side chains.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Synthesis and characterization

Chemicals and solvents (Sigma–Aldrich, Merck and BACHEM) for synthesis were of the highest available analytical grade and used without further purification. Products were identified by means of NMR techniques and high-resolution mass spectrometry. The synthesis of the cyclic dipeptides (DKP) 12 followed the method invented by Nitecki et al. [29]. Full synthetic procedures and the details of product characterizations are given in the supporting information. Solvents for time-resolved spectroscopy and steady-state measurements were of the highest available analytical grade and were used without further purification.

Table 1. Summary of kinetics data and quantum yields observed during the photolysis (steady state and time resolved) of 1 and 2 in acetonitrile (ACN) solutions
 kT [s−1]kBR [s−1]ΦketylΦconsumption
1 3.0 × 1061.9 × 106≤0.180.18
2 48.0 × 1068.3 × 1060.6 ± 0.10.35

Steady-state photolysis

Steady-state irradiation was used to determine the quantum yields of substrate disappearance and, in a further step, to identify the stable products. Irradiation with a low-pressure mercury lamp (λirr = 254 nm) was chosen because it was necessary to work with low concentrations to avoid side reactions such as self-quenching. For quantum yield determinations, irradiations were performed in rectangular quartz cells (1 cm × 1 cm). UV–Vis spectra were measured at room temperature using a Cary 300 Bio Varian spectrophotometer (Varian, Palo Alto, CA). The progress of the reactions was followed by HPLC using a Waters 600E Multisolvent Delivery System Pump (Waters Corporation, Milford, MA). The detection system consisted of a Waters 996 Photodiode Array UV–Vis Detector. Analytical HPLC analyses were carried out on a Waters XTerra RP18 reverse phase column (4.6 × 250 mm, 5 μm particle size). The mobile phases were mixtures of solutions of acetonitrile (ACN) and 20 mm KH2PO4: ACN (95:5 vol/vol), with a flow rate of 1 mL min−1. The quantum yields, Φ were measured using uranyl oxalate actinometry at 254 nm, taking its quantum yield to be 0.602 [30]. In the case of dyad 2, larger quantities of the irradiated solutions were required to identify stable products. Irradiations on a preparative scale, with concentrations of 4.7 × 10−5 m, were carried out in an immersion-lamp photoreactor (low-pressure mercury lamp, λirr = 254 nm). After irradiations, solutions were concentrated by evaporating the solvent under reduced pressure and temperature ≤303 K and subsequently the products were isolated by thin-layer chromatography (TLC). The isolated photoproduct as a mixture of diastereoisomeric alcohols was analyzed by means of NMR studies (1H, 13C, 1H-1H COSY, 1H-13C HMBC) and mass spectrometry.

Nanosecond LFP

The set up for the nanosecond LFP experiments and its data acquisition system have been previously described in detail [31]. LFP experiments employed a pulsed Nd:YAG laser (355 and 266 nm, 5 mJ, 7–9 ns) for excitation. Transient decays were recorded at individual wavelengths by the step-scan method with a step distance of 10 nm in the range of 320–700 nm as the mean of 10 pulses. Samples for LFP were deoxygenated with high-purity argon for 15 min prior to the measurements. Concentrations in the range of 2–4 × 10−5 and 1–3 × 10−3 m were used at 266 and 355 nm, respectively. We used 266 nm excitation to avoid self-quenching processes and because of the limited solubility of compound 2. Experiments were performed in rectangular quartz cells (1 cm × 1 cm). In the case of 266 nm photolysis a flow-through-system procedure without recycling was employed. This procedure was accomplished by putting a volume of 100 mL of the investigated solution inside a reservoir and then deoxygenating it. The flow (by gravity) of the solution (from the reservoir into the cell) was adjusted to one drop for 1–2 s. Experiments were performed with freshly prepared solutions at room temperature (295 ± 1 K).

MD simulations

All MD simulations were performed with the SANDER module of the AMBER 10 suite of programs [32]. MD simulations were performed on the investigated compounds 1–2 using the general AMBER force field (GAFF), which well-describes small organic molecules. Reliable statistical distributions for the conformers of the compounds in this work resulted from free MD simulations with a 2 fs time step and 100 ns propagation times, preceded by 20 ps of heating from 0 to 300 K and 60 ps of equilibration. Cross-checks by running MD simulations also for 200 ns propagation time resulted in the same distributions, which verify the complete sampling of the conformational subspace. The robustness of the recorded distributions was addressed by sampling of each compound from different starting configurations. Starting coordinates for the MD simulations were the optimized geometries of molecules of the ground-state obtained using a density functional theory (DFT) method (PBE1PBE/6-31 + G(d)) [33]).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Steady-state photolysis

Mechanistic information concerning quenching of the triplet states of 1–2 was obtained from steady-state irradiations in ACN solutions and subsequent product analyses. Steady-state irradiations were conducted with 4.7 × 10−5 m deoxygenated ACN solutions of 1–2 with a low-pressure mercury lamp (254 nm). Concentrations in the range of 10−5 m were used due to the low solubility of 2 and also to limit the intermolecular quenching pathway with 1, which was shown to occur in the LFP experiments in ACN (see elsewhere in the text). The course of the reactions was followed by HPLC analyses (inset of Fig. 1). In both cases, as the reaction proceeded, a new peak progressively emerged in the HPLC profile at the expense of the substrate. Quantum yields of substrate disappearance (Φirr), for 2 and 1 were found to be 0.35 and 0.18, respectively (Table 1). The dominance of only one single photoproduct in the HPLC profiles points to a high degree of selectivity in both cases. This is a finding which is in accord with the observation of clean isosbestic points in the reaction UV spectra measured at regular time intervals (Fig. 1).

image

Figure 1. Absorption spectra recorded after irradiation at 254 nm in ACN (a) 2 (4.7 × 10−5 m) and (b) 1 (4.7 × 10−5 m); time of irradiation from (a) 0 to 40 s and (b) 0 to 100 s (following direction of arrows); Insets: HPLC chromatograms recorded after irradiation at 254 nm, monitored at λmonit = 210 nm.

Download figure to PowerPoint

The highly irreversible character of the photoinduced chemistry of both dyads is reflected by the disappearance of the absorption band at 258 nm, which is characteristic of the π [RIGHTWARDS ARROW] π* transition of the benzophenone chromophore. This indicates the absence of this moiety in the products of the irradiations of 1–2.

To aid in the understanding of the mechanism of intramolecular triplet-state quenching by the methionine residue in the cyclic DKP-based BP-Met dyads, the structure of the photoproducts was determined. Preparative scale irradiations of 2 in ACN were carried out in a photoreactor with 4.7 × 10−5 m solutions, having a total volume of 1 L. After irradiation, the solutions were concentrated by evaporating the solvent, and, subsequently, the products were isolated by TLC. As expected, based on the HPLC results, the photoreaction led to one major photoproduct, whereas a second product was formed with a significantly lower quantum yield. Further analyses were performed only on the main photoproduct. This photoproduct was separately analyzed by UV/Vis, mass spectrometry, and NMR techniques.

The UV spectrum of the product showed a dominant absorption band at 200 nm, but only minor spectral response around 260 nm. This disappearance of the typical absorption band at 258 nm is consistent with the reduction in the benzophenone chromophore (Fig. 2). A high-resolution EI–MS analysis of the product showed a molecular ion at m/z = 382.13581, which corresponds to an atomic composition of C21H22N2O3S (calculated m/z = 382.13510). Thus, mass spectrometry data confirmed that the product and the starting compound are isomers. High-resolution EI–MS analysis of the base peak (m/z = 196.2) confirmed its proposed atomic composition of C14H12O (mass obtained m/z = 196.08837, mass calculated: m/z = 196.08882). The MS-features of the base peak fit to the structure presented in red in Fig. 2. In addition, very similar peaks were recognized in the MS spectra of the substrate and the product. This observation can be understood if the fragmentation pathway is connected with the cleavage of the bond, which is formed during the irradiation of 2. Extended one- and two-dimensional NMR studies (1H, 13C, 1H-1H COSY, 1H-13C HMBC) revealed that this photoinduced bond had formed between the carbonyl carbon and the terminal methyl group of the methionine side chain, δ-CH3, adjacent to the sulfur atom. This conclusion is based on the results of a detailed NMR study on the isolated photoproduct. First, the characteristic signal at δ = 2.1 ppm for this methyl group in the spectrum of 2 is missing in the NMR spectra (600 MHz, DMSO-d6) of the product. In the photoproduct, the δ protons are no longer equivalent, but give rise to separate resonances around 3.05 and 3.55 ppm. Their signals are split to doublets by geminal coupling with = 14.4 Hz. This signature is fully in accord with the data reported by Moretto et al. for a related macrocyclization reaction between methionine and benzophenone [24]. Sharp 1H-NMR singlets at 5.50 and 5.57 ppm are assigned as the protons of tertiary alcohol functions that are formed during the C–C coupling. Deshielding of these resonances from the values typical for tertiary diphenyl alcohols [34, 35] may be a consequence of anisotropic effects. This is supported by the DFT-minimized structure of the photoproduct (PBE1PBE/6-31 + G(d) level of theory; see Fig. S1) where the –OH function actually points to the π system of the terminal phenyl ring. An additional source of the down-field shift may be hydrogen bonding of the –OH group with the strongly hydrogen-bond accepting solvent DMSO.

image

Figure 2. Absorption spectra of 2 (- - -) and photoproduct (solid line) of the irradiation in ACN and its chemical structure. Outlined in red is the fragment of the molecule that corresponds to the base peak in the MS spectrum; arrows denote diagnostic couplings from the heteronuclear multiple-bond correlation (HMBC) NMR experiment.

Download figure to PowerPoint

Further evidence for C–C coupling between the former carbonyl group and the terminal δ-carbon atom stems from heteronuclear multiple-bond correlation experiments (HMBC). By means of this experiment, long-range through bond 1H-13C couplings are recorded selectively. Diagnostic couplings support photoinduced macrocyclization via C–C bond formation. In particular, the δ protons of the methionine side chain exhibit coupling with carbon atoms of both aromatic rings of the BP moiety. Additional HMBC signals couple the hydroxy protons of the tertiary alcohols with carbon atoms of the BP moiety as well as with the δ carbon of methionine (see arrows in the structural scheme in Fig. 2).

Significant shielding of the met side-chain resonances (β and γ) with δ < 1.5 ppm points to enhanced anisotropic effects (“ring currents”) of the phenyl ring in the crowded macrocyclic product. This conclusion is in agreement with the results from DFT (see Fig. S1). Due to the close vicinity of phenyl and methionine moieties the rotation of the phenyl ring is slow on the NMR time scale (individual signals for each proton of the nominal AA'BB' system).

On the basis of the collected data, a cyclic tertiary alcohol with an additional stereogenic center is deduced as the structure of this main product from 2 (Fig. 2). The occurrence of a “doubled” signal set both in the 1H and 13C NMR spectra (supporting information) provides clear evidence for the formation of an additional stereogenic center and, thus, of two stereoisomers of the photoproduct. The stereochemistry of the photoproduct was not determined but, as indicated by a “doubled” signal set in the NMR spectra of the product, both configurations of the new stereocenter will be accessed in similar amounts.

Importantly, the nature of the main product signals a (net) hydrogen-atom transfer (HAT) to have occurred during the course of the reaction prior to CC-coupling. As a consequence, the reaction coordinate must reflect this fact by the emergence of intermediate biradicals, made up of BP-derived ketyl radicals and methionine-derived α-thioalkyl radicals. By means of time-resolved spectroscopic techniques, we could verify this hypothesis (see below). Another important implication of the product structure given above deserves attention: Intrinsically, two potential sources of H-atoms are present in the methionine side chain, namely the γ-CH2 and the δ-CH3 group. In the past, both sources have been identified to be active [36]. The structure of the photoproduct of 2 in the current study, in turn, implies that the δ-CH3 group acts as the source with a striking selectivity.

LFP. Spectral and kinetic analysis

Nanosecond laser photolysis was carried out to characterize the excited-state dynamics of the dyads 1–2 and to establish the structural conditions that maximize the efficiency of the intramolecular reactivity of the triplet states. This work addressed, in particular, the effect of chiral center configurations on the diastereoselectivity of the triplet-state quenching by the methionine residue. It was accomplished by a comparison of the reactivities of the pair of cyclic diastereoisomers.

The effect of a change in the chirality of one stereocenter on the reactivity is illustrated by the comparison of the transient absorption spectra and decay profiles observed for solutions of 1–2 in ACN (Figs. 3a and 5a). The early spectra obtained for the cis-isomer 2 could not be assigned to the triplet-state unambiguously. Even at short times after the laser pulse, e.g. after 22 ns, the transient absorption spectra contained already a significant contribution from ketyl radicals (Fig. 3a). The presence of the ketyl radical, in turn, was unambiguously identified through its characteristic spectrum with maxima at 340 and 540 nm.

image

Figure 3. (a) Summary of the results obtained during laser flash photolysis at 266 nm of a deoxygenated solution of 2 (2 × 10−5 m) in dry ACN, (a) transient absorption spectra: time delays after flash (from top to bottom): 22, 40, 50 and 80 ns, (b) decay profiles of the transient absorption monitored at 520 nm, the solid line represents a biexponential fit to the decay curve; the numbers represent the values obtained from the fits.

Download figure to PowerPoint

Spectral resolution techniques were used to obtain concentrations of the intermediates in the excited-state processes of 2. Although the presence of the triplet state was not obvious from the transient absorption spectra, resolution into spectral components revealed the presence of the triplet state at the short delay times. To simulate the experimental data quantitatively, three components were needed: the triplet state 3BP, the ketyl radical BPH and the benzophenone radical anion BP•− (Fig. 4a). Contributions from the methionine-derived radical cation were not detected in the transient absorption spectra because its weak absorption at λ < 400 nm was masked by the strong absorption of the BP-derived transients.

image

Figure 4. (a) Resolutions of transient absorption spectra taken 28 ns after 266 nm laser pulsing (5 mJ) of a solution of 2 in ACN, (b) concentration profiles for the triplet state and ketyl radical obtained from the resolution of the transient absorption spectra of 2 in ACN. The symbols represent: □, triplet state 3BP; •, ketyl radical BPH; ◊, BP radical anion BP•−; and *, experimental data; solid curves in (a) and (b) are the resulting fits from the regression analyses, the number represents the value obtained from the fit to the triplet concentration profile.

Download figure to PowerPoint

It is worth noting that the concentration of BP•− was found to be very small ≤0.4 μm, hence not any unambiguous conclusion concerning the presence and the mechanism of the formation of BP•− could be made. The initial triplet concentration, actinometrically determined, was 7.9 μm. The maximum concentration of the ketyl radical was determined to be 4.7 μm. This value was obtained from fitting the BPH concentration profile to a function which takes into account the growth and the decay of the species (Fig. 4b). The quantum yield of the ketyl radical formation was calculated to be 0.6 (Table 1). The uncertainty of this figure, obtained from the concentration profile, is quite large due its rapid decay and is estimated to be ±0.1.

Although the transient spectrum obtained from 2 in pure ACN contained significant contributions from a ketyl radical even after 22 ns, the spectrum obtained from 1 after 500 ns still largely resembled the original triplet spectrum with its characteristic absorption maxima at 325 and 525 nm, in addition to a long-wavelength absorption (Fig. 5a). In the case of dyad 1, hardly any further evolution of the shape of the transient absorption spectra was noticeable aside from the triplet absorption decay (Fig. 5a). However, the normalized spectra from 1 obtained 70 and 800 ns after the flash did not overlap (inset of Fig. 5a), which would be expected if no additional intermediates were formed.

image

Figure 5. Summary of the results obtained during laser flash photolysis (LFP) at 266 nm of a deoxygenated solution of 1 (2 × 10−5 m) in dry ACN, (a) transient absorption spectra: time delays after the flash (from top to bottom): 70, 150, 500 and 800 ns, (b) normalized decay profiles of the transient absorption obtained in ACN monitored at 520 (○), 340 () and 520 nm (Δ) obtained during LFP in 2,2,2-trifluoroethanol. Inset to 5a: normalized transient absorption spectra obtained 70 ns and 800 ns after the flash.

Download figure to PowerPoint

Further evidence for the formation of intermediates in the quenching process of 1 in ACN came from a kinetic treatment of the transient decays. The kinetic traces at 340 and 520 nm did not coincide with each other (Fig. 5b). The kinetic trace at 520 nm decays faster than that at 340 nm, but the respective decay components of these traces differ only by a factor of 1.6. This explains why, at first sight, it was difficult to see spectral evolution in the transient absorption spectra. Comparison of both kinetic traces (340 and 520 nm) and the normalized transient absorption spectra speaks; however, clearly in favor of the formation of an intermediate after the triplet decay, which due to its maximum absorption at 340 nm can be assigned to the ketyl radical. For dyad 1, the resolutions technique failed to provide a realistic determination of the transients' concentrations because the difference in the decay rate constants of the BP triplet state and ketyl radical was too small. Thus, to obtain the quantum yield of the ketyl radical formation, the kinetic trace at 340 nm was analyzed. The maximum absorbance of the ketyl radical was determined by fitting the 340 nm kinetic traces to a function which takes into account the growth and the decay of the species and by subtracting the 340 nm absorbance of the BP triplet state (Fig. 5b). Afterwards, knowing the molar absorption coefficient of BPH at 340 nm (ε = 20500 m−1 cm−1), which was obtained by rescaling the values given by Hayon et al. [37] to ε = 3400 m−1 cm−1 at 540 nm obtained by us, the concentration of the ketyl radical was determined to be 1.4 μm. With the triplet concentration of the 7.9 μm, the upper limit of the quantum yield of the ketyl radical formation was calculated to be 0.18 (Table 1).

The chiral effects on the intramolecular triplet quenching of the pair of diastereoisomers are strongly reflected by the triplet lifetimes, which exhibited remarkable differences. The triplet lifetime of dyad 2 in ACN, obtained from a biexponential fit to the transient decay at 520 nm, under the assumption that the decay of the fast component can be attributed to the triplet state (Fig. 3b), was found to be only 21 ns. This value was cross-checked by comparison with the rate constant obtained from a monoexponential fit to the triplet concentration-time profile resulting from the spectral resolutions (Fig. 4b). It was, within experimental error, identical to the triplet-decay rate obtained from the kinetic decays at 520 nm. The very short triplet lifetime is taken as evidence for the intramolecular nature of the quenching process.

On the other hand, triplet lifetimes obtained from mono- or biexponential fits to the transient decays at 630 and 520 nm, respectively, for dyad 1 in ACN showed concentration-dependent triplet decay. The triplet lifetimes were found to be 230 ns and 330 ns for the concentrations of 1 × 10−3 and 2 × 10−5 m, respectively. This behavior is due to a bimolecular self-quenching process involving unexcited dyads. From the two sets of data points that were collected (from 355 and 266 nm excitation) in ACN, the self-quenching rate constant for dyad 1 was estimated to be kSQ = 1.3 × 109 m−1 s−1. This value is in good agreement with the rate constant for the bimolecular quenching of the triplet state of CB by methionine [14]. On the basis of this value, self-quenching processes can be neglected at a concentration of 2 × 10−5 m, despite the relatively low-intramolecular reactivity of 1.

It is worth noting that the triplet lifetimes of the two diastereoisomers 12 measured for the same concentration (2 × 10−5 m), differed by a factor of 15 in ACN. The change in the configuration at a single chiral center in the cyclic BP-Met dyads produced a marked stereoselectivity, indicating that there are special structural constraints on the triplet-state quenching by the methionine residue. Such significant stereoselectivity has been reported recently also for the HAT quenching of triplet excited BP in rigid benzophenone-tyrosine dyads [38]. It was recently shown that the reactivity of the excited molecule can be increased by the changes in the nature of the solvent [39]. To probe that possibility in the rigid BP-Met dyads, solvent effects on triplet quenching were checked in 2,2,2-trifluoroethanol (TFE) solution for dyad 1, which has a fairly low intramolecular reactivity in ACN. The LFP results on 1 in TFE as the solvent were totally different than those observed in ACN. First of all, the triplet lifetime was reduced by a factor of 10 (Fig. 5b). In addition, the triplet decays of 1 in TFE were connected with the fairly efficient formation of ketyl radicals (Φ = 0.47). The possibility of the HAT from the solvent or from the α-carbon of the DKP ring was excluded by measuring, in TFE, the triplet lifetime and transient absorption spectra of a reference DKP system that lacked of the methionine side chain; no evidence for H-atom transfer was observed. This negative result showed that the triplet state of dyad 1 was being efficiently quenched intramolecularly via (net) HAT from the methionine side chain even though dyad 1 has its benzophenone and methionine moieties on opposite sides of its DKP ring. This dyad is another example showing that the reactivity of an electronically excited molecule can be greatly enhanced by changes in the nature of the solvent [38, 39]. The absence of significant amounts of BP•−, in ACN as well as in TFE, implies that rapid proton transfer can occur within the radical-ion complex via intramolecular proton transfer, involving the δ-CH3 or γ-CH2 group, both adjacent to the sulfur. In TFE solution, another possibility is that the protonation step can occur from the solvent, which is known to be a good proton donor [39].

Conformational analysis

In the light of the results from the time-resolved and steady-state studies on the cyclic BP-Met dyads, it remains clear that the origin of the observed differences in the triplet lifetimes between the diastereoisomers should be sought from the conformational preferences and differences in the carbonyl oxygen-to-sulfur distance distributions. To gain an understanding of the geometrical requirements underpinning the observed differences, MD simulations on 1–2 were performed. The goals of the MD simulations of the DKP-based BP-Met dyads were to establish the distribution of the sulfur to carbonyl oxygen distances and to see how the change in a single chiral center changes the populations of the side-chains' rotamers. Within cyclopeptide molecules, the conformations at the β-carbon atom of the side chains within cyclopeptide molecules can be characterized by three different orientations of the side-chain substituents (R) in relation to the DKP ring. The acronyms EN, EO and F denote side-chain rotamers wherein the substituent R is extended to the adjacent nitrogen atom, extended to the adjacent carbonyl oxygen atom, and folded back across the DKP ring, respectively (Scheme 1). This rotamer notation is based on the orientation of the side-chain substituent R rather than on the dihedral angles, to simplify the comparison between the epimeric compounds 1–2. In the rotamer notation, the first symbol always correspond to the methionine side.

The conformational space of 1–2 was sampled by means of long-time MD simulations in an ACN implicit-solvent model. As quenching of the triplet state of the BP residue by methionine should depend on the distance between the sulfur atom and the BP carbonyl oxygen atom, d(S–O), a quantitative analysis of this parameter was performed.

MD simulation results show that mainly three rotamers (EO, EN and F) of the methionine and only two rotamers (EN and F) of the benzophenone side chain are populated. In particular, the conformational preferences of the BP side chain of both dyads are well defined, meaning that the probabilities for the most likely conformations are large (Table 2). The total probabilities for the BP side chain of 2 and 1 to be in the F structure were found to be 0.85 and 0.79, respectively. From the analysis of the result of the MD simulations, it can be concluded that there are two most favorable configurations for both dyads, namely EΝ–F and F–F. The probabilities of the dyads 2 and 1 to be in one of these two conformations were 78% and 72%, respectively. It is important to note that the FF structure of 2 corresponds to a conformation of the side chains wherein both moieties (BP, Met) are back folded across the DKP ring at the same time (Fig. 6a).

image

Figure 6. Two types of the most populated conformations found in long-time MD simulations (implicit-solvent model for ACN) for (a) 2 and (b) 1. The numbers denote the S–O distance in Å; Yellow, sulfur; red, oxygen; dark blue, nitrogen.

Download figure to PowerPoint

Table 2. Summary of the side-chain rotamers' joint probabilities of 1–2 computed from the MD simulations (implicit ACN model)
Conformation (Met-BP)Joint probability*
2 1
  1. *Joint probabilities—the probability of both events occurring together at the same time.

FF0.310.23
ENF0.470.49
FEN0.070.07
ENEN0.020.12
EOF0.070.07

As shown in Fig. 6a, the F–F structure of 2 allows for close contact of the sulfur and carbonyl oxygen. On the other hand, that same type of conformation, F–F, for 1 corresponds to a structure, with a long interside-chain S–O distance (Fig. 6b). It is worth noting that another type of conformation, namely EΝ–F also gives different S–O distances for both dyads 12 (Fig. 6a,b).

Based on the MD simulations, pair-distribution functions were calculated for the carbonyl/sulfur distance d(S–O) (Fig. 7). Although the populations of the particular conformations were found to be quite similar (Table 2), the pair-distribution functions revealed very important differences between this pair of diastereoisomers. The sampling of 2 in an implicit model of acetonitrile solution gave a quite narrow pair-distribution function with a maximum at a distance d(S–O) = 5 Å. The situation is different for dyad 1 for which the pair-distribution function of the carbonyl/sulfur distance d(S–O) not only is much wider compared to the one for 2 but also exhibits even another noticeable maximum at about 10.6 Å.

image

Figure 7. Pair-distribution functions of the carbonyl/sulfur group distance d(S–O), obtained with MD simulation in ACN for (a) the cis-isomer 2 and (b) the trans isomer 1. Inset: pair-distribution functions of the carbonyl/hydrogen d(O–H) hydrogen from δ-CH3 [solid line] and γ-CH2 [dashed line] distance for 2.

Download figure to PowerPoint

The most important observation based on the pair-distribution functions is that, although both dyads were capable of forming rotamers with relatively close contact between the sulfur and carbonyl oxygen, the probability of d(S–O) to be less than 5.5 Ǻ was calculated to be only 0.09 for 1, whereas the analogous value for 2 was 0.42. This provides a rationalization for the observed stereodifferentiation in the triplet quenching of 1 and 2.

As in the methionine side chain there are two places that may act as proton donors in the quenching mechanism, namely the γ-CH2 and the δ-CH3 group, the pair-distribution function for the distance between the carbonyl oxygen and the hydrogen atoms from the γ-CH2 and the δ-CH3 were quantified (inset to Fig. 7a). In fact, the pair-distribution function of the carbonyl/hydrogen d(O–H; γ-CH2) distance is significantly shifted to longer distances compared to the pair-distribution function for carbonyl/hydrogen d(O–H; δ-CH3).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Intramolecular excited triplet-state quenching in diastereomeric compounds composed of a benzophenone chromophore and a methionine moiety were investigated by LFP and steady-state irradiations to gather information about chiral discrimination in the primary step, as well as in the overall photoprocess. As expected, introduction of a rigid linker between the reactive moieties manifested in a significant chiral discrimination of the triplet-state quenching between the 1 and 2 pair of diastereoisomers. It is noted that stereoselectivity, expressed as the ratio of the triplet lifetimes for the diastereoisomers measured in ACN, was found to be 15 whereas triplet lifetimes of the flexible open-chain analog of rigid BP-Met dyads differed only by a factor 2.5 (A. Lewandowska-Andralojc, G. Hörner, G.L. Hug, B. Marciniak, unpublished data). As the benzophenone moiety is covalently linked to the methionine, the stereodifferentiation must be the result of the steric hindrance introduced by the rigid spacer with respect to a close approach between the two active moieties. The reactivity trend was rationalized by MD simulations that showed a marked difference between these two stereoisomers with regard to their probabilities of forming rotamers with relatively close contact between the sulfur and carbonyl oxygen. The much higher probability of forming a close contact distance d(S–O) found for 2 coincides, indeed, with a faster triplet-state quenching.

The mechanism of the photoinitiated intramolecular reaction can be proposed based on the analogy to the intermolecular reactions of triplet CB with methionine-containing compounds [10, 11, 14, 18]. It was assumed that intramolecular electron transfer from the sulfur atom to the triplet state of the BP is the primary photochemical step in the triplet-state quenching of both diastereoisomers. Decay of the initial charge-transfer complex may involve [1] back electron transfer and [2] proton transfer within the complex. Reaction of the BP ketyl radical and an α-thio-alkyl radical may subsequently lead to stable products, or the system may revert to the starting materials. The contributions of different decay pathways were studied by LFP and steady-state irradiations. Importantly, the spectrum obtained for 2 in pure ACN carries significant contributions from a ketyl radical. Based on the spectral resolution technique, it was shown that the quantum yield of ketyl radical formation is Φ ≈ 0.6 (Scheme 2). Quenching of the triplet state of 2 can thus be attributed to a fairly efficient intramolecular H-atom transfer reaction. In the absence of any sizeable amounts of ketyl radical anions as the nominal primary ET products, a very efficient and rapid intramolecular proton transfer seems to prevail. On the other hand, the transient absorption spectra of 1 showed basically just the benzophenone triplet-state absorption and only a weak absorption that could be assigned to ketyl radicals. It was estimated that the quantum yield of ketyl radical formation was Φ ≈ 0.18, which points to an efficient back electron transfer.

image

Scheme 2. Overview of mechanistic pathways upon excitation of 2.

Download figure to PowerPoint

Further mechanistic information concerning the triplet-state quenching for 1–2 was obtained from steady-state irradiations and product analyses. The significant diastereodifferentiation in the quenching of the excited triplet benzophenone chromophore by methionine raises the important question whether diastereodifferentiation is conserved in the overall photoreaction quantum yields. A higher quantum yield of substrate disappearance (Φ = 0.35) was measured for 2, for which also a faster triplet quenching was observed by LFP. For the second diastereoisomer, 1, the quantum yield of substrate disappearance was found to be Φ = 0.18. Interestingly, although the triplet-state quenching of the compounds 2 and 1 showed high chiral discrimination, the quantum yield of the substrate disappearance differed only by a factor of 2. In addition, for the diastereoisomer 1 the quantum yield of substrate disappearance was comparable with the quantum yield of ketyl radical formation, whereas for the dyad 2 only about half of the ketyl radical formed a stable product. Interestingly, the triplet states of 2 and its open-chain congener were found to be quenched with the same rate constants in ACN (A. Lewandowska-Andralojc, G. Hörner, G.L. Hug, B. Marciniak, unpublished data), but the respective quantum yields of substrate disappearance for these compounds differed by a factor of 7 (0.35 vs 0.05, respectively). This behavior with regard to the efficiency of the substrate consumption between the studied dyads indicates that there are not only special structural constraints influencing the triplet-state quenching rate constants but also additional geometric factors that control the efficiency of the secondary reactions that led to stable products. To gain further insight into the mechanism of the triplet-state quenching of BP by methionine, the photoproduct from the irradiation of 2 was analyzed. Combining the results of the analysis of the LFP experiments with the molecular structure of the stable product from the irradiation of 2 helped in establishing the mechanism of the reaction (Scheme 2). The suggested mechanism for the formation of this product is as follows. First, there is an electron-transfer step, resulting in a radical-ion pair complex. Second, within this complex, an intramolecular proton transfer occurs, involving exclusively the δ-CH3 group, adjacent to the sulfur. The result of this two-step process is a biradical: a ketyl radical and an α-thio-alkyl radical. Subsequently, these biradicals undergo intramolecular recombination to form the macrocyclic ring system (efficiency, φ ≈ 0.6). This efficiency was computed from the quantum yield (Φ ≈ 0.6) of ketyl radical formation and the quantum yield (Φ = 0.35) of overall disappearance of the substrate. The generated carbon-centered biradicals can also revert to the starting materials via back hydrogen transfer (efficiency, φ ≈ 0.42), as shown in Scheme 2.

Interestingly, the investigated intramolecular photoreaction of 2 was found to be strictly regioselective, as formation of the stable product involved only the original Met side-chain δ-CH3 carbon. This exclusiveness is likely to be a consequence of the higher strain than is expected if the neighboring Met side-chain γ-CH2 group was involved. The regioselectivity trend for the product formation was quantitatively rationalized by DFT PBE1PBE/6-31 + G(d) by optimizing the structures of the two possible macrocyclic products (optimized structures in Fig. S1). These calculations showed that a product that would involve the Met side-chain γ-CH2 carbon would, indeed, give rise to a highly strained product, with an energy higher by 20 kcal mol−1 than the energy of the identified product. It is worth noting although that the observation of the product, that involved deprotonation from the Met side-chain δ-CH3 carbon, does not necessarily exclude the possibility of a deprotonation from the Met side-chain γ-CH2 carbon followed by back HAT to reform the starting compound. However, based on the pair-distribution functions of the carbonyl/hydrogen (γ-CH2, δ-CH3), the distances which reflect the accessibility of the proton deprototonation from the δ-CH3 are presumably privileged for dyad 2.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

In this study, we addressed chiral effects on unimolecular photoinduced reactions in two benzophenone/methionine dyads. The results presented in this study very clearly demonstrate the complexity pertaining to intramolecular reactions even in relatively simple model compounds. Combined results from time-resolved and preparative steady-state experiments showed that quenching of the triplet state of the dyad 2 led to a biradical that recombined to form a macrocyclic photoproduct with one additional asymmetric carbon center. The high selectivity and efficiency observed for this intramolecular carbon–carbon bond formation makes it a useful tool in synthetic organic photochemistry for the preparation of small rings and large macrocycles.

The experimental and theoretical study shows that two stereoisomers of a rigid cyclic dipeptide can undergo intramolecular reactions, even though one of these isomers has its two reactive moieties on opposite sides of its DKP ring. However, the difference in the geometrical restraints between the two stereoisomers was reflected in significant chiral discriminations that were observed in the rate constants and in the stable-product formation. These remarkable variations in the reaction rates of the isomeric BP-Met dyads are attributed to the difference between the distributions of the carbonyl/sulfur group distances in the two stereoisomers. A forthcoming article will discuss the photolysis of compound 1 and its photoproduct formation in more detail.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

This work was supported by the Ministry of Science and Higher Education (Poland; grant no. NN204 143138) and COST Chemistry CM0603. This is Document No. NDRL 4923 from the Notre Dame Radiation Laboratory which is supported by the Division of Chemical Sciences, Geosciences and Biosciences, Basic Energy Sciences, Office of Science, United States Department of Energy through grant number DE-FC02-04ER15533. We thank Bozena Wyrzykiewicz for help with NMR experiments. AL-A thanks Professor Ian Carmichael, Director of the Radiation Laboratory, for support of this work and Carrie Miller who introduced AL-A to the field of Molecular Dynamics.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information
  • 1
    Asmus, K.-D. and M. Bonifačić (1999) Sulfur-centered reactive intermediates as studied by radiation chemical and complementary techniques. In S-Centered Radicals (Edited by Z. B. Alfassi), pp. 141191. John Wiley and Sons, New York.
  • 2
    Vogt, W. (1995) Oxidation of methionyl residues in proteins: Tools, targets, and reversal. Free Radic. Biol. Med. 18, 93105.
  • 3
    Meyer, J. D., B. Ho and M. C. Manning (2002) Effects of conformation on the chemical stability of pharmaceutically relevant polypeptides. In Rational Design of Stable Protein Formulations (Edited by J. F. Carpenter and M. C. Manning), pp. 85108. Springer, New York.
  • 4
    Stadtman, E. R. (2006) Protein oxidation and aging. Free Radic. Res. 40, 12501258.
  • 5
    Stadtman, E. R., H. Van Remmen, A. Richardson, N. B. Wehr and R. L. Levine (2005) Methionine oxidation and aging. Biochim. Biophys. Acta 1703, 135140.
  • 6
    Hokenson, M. J., V. N. Uversky, J. Goers, G. Yamin, L. A. Munishkina and A. L. Fink (2004) Role of individual methionines in the fibrillation of methionine-oxidized a-synuclein. Biochemistry 43, 46214633.
  • 7
    Ciccotosto, G., K. Barnham, R. Cherny, C. Masters, A. Bush, C. Curtain, R. Cappai and D. Tew (2003) Methionine oxidation: Implications for the mechanism of toxicity of the β-amyloid peptide from Alzheimer's disease. Lett. Pept. Sci. 10, 413417.
  • 8
    Bobrowski, K., G. L. Hug, D. Pogocki, B. Marciniak and C. Schöneich (2007) Stabilization of sulfide radical cations through complexation with the peptide bond: Mechanisms relevant to oxidation of proteins containing multiple methionine residues. J. Phys. Chem. B 111, 96089620.
  • 9
    Glass, R. S., G. L. Hug, C. Schöneich, G. S. Wilson, L. Kuznetsova, T.-M. Lee, M. Ammam, E. Lorance, T. Nauser, G. S. Nichol and T. Yamamoto (2009) Neighboring amide participation in thioether oxidation: Relevance to biological oxidation. J. Am. Chem. Soc. 131, 1379113805.
  • 10
    Hug, G. L., K. Bobrowski, H. Kozubek and B. Marciniak (1998) Photooxidation of methionine derivatives by the 4-carboxybenzophenone triplet state in aqueous solution. Intracomplex proton transfer involving the amino group. Photochem. Photobiol. 68, 785796.
  • 11
    Hug, G. L., K. Bobrowski, H. Kozubek and B. Marciniak (2000) Photo-oxidation of methionine-containing peptides by the 4-carboxybenzophenone triplet state in aqueous solution. Competition between intramolecular two-centered three-electron bonded (SS) and (SN) formation. Photochem. Photobiol. 72, 19.
  • 12
    Hug, G. L., K. Bobrowski, D. Pogocki, G. Hörner and B. Marciniak (2007) Conformational influence on the type of stabilization of sulfur radical cations in cyclic peptides. ChemPhysChem. 8, 22022210.
  • 13
    Hug, G. L., B. Marciniak and K. Bobrowski (1996) Sensitized photo-oxidation of sulfur containing amino-acids and peptides in aqueous solution. J. Photochem. Photobiol., A 95, 8188.
  • 14
    Marciniak, B., G. L. Hug, K. Bobrowski and H. Kozubek (1995) Mechanism of 4-carboxybenzophenone-sensitized photooxidation of methionine-containing dipeptides and tripeptides in aqueous solution. J. Phys. Chem. 99, 1356013568.
  • 15
    Marciniak, B., G. L. Hug, J. Rozwadowski and K. Bobrowski (1995) Excited triplet state of N-(9-methylpurin-6-yl)pyridinium cation as an efficient photosensitizer in the oxidation of sulfur-containing amino acids. Laser flash and steady-state photolysis studies. J. Am. Chem. Soc. 117, 127134.
  • 16
    Wojcik, A., A. Lukaszewicz, O. Brede and B. Marciniak (2008) Competitive photosensitized oxidation of tyrosine and methionine residues in enkephalins and their model peptides. J. Photochem. Photobiol., A 198, 111118.
  • 17
    Glass, R. S., C. Schöneich, G. S. Wilson, T. Nauser, T. Yamamoto, E. Lorance, G. S. Nichol and M. Amman (2011) Neighboring pyrrolidine amide participation in thioether oxidation. Methionine as a “Hopping” site. Org. Lett. 13, 28372839.
  • 18
    Bobrowski, K., G. L Hug and B. Marciniak (1992) 4-carboxybenzophenone-sensitized photooxidations of sulphur-containing amino acids. Nanosecond laser flash photolysis and pulse radiolysis studies. J. Am. Chem. Soc. 114, 1027910288.
  • 19
    Goez, M. and J. Rozwadowski (1998) Reversible pair substitution in CIDNP: The radical cation of methionine. J. Phys. Chem. A 102, 79457953.
  • 20
    Goez, M., J. Rozwadowski and B. Marciniak (1998) CIDNP spectroscopic observation of SN radical cations with a two-center three-electron bond during the photooxidation of methionine. Angew. Chem. Int. Ed. 37, 628630.
  • 21
    Morozova, O. B., S. E. Korchak, H.-M. Vieth and A. V. Yurkovskaya (2009) Photo-CIDNP study of transient radicals of Met-Gly and Gly-Met peptides in aqueous solution at variable pH. J. Phys. Chem B. 113, 73987406.
  • 22
    Wittelsberger, A., B. E. Thomas, D. F. Mierke and M. Rosenblatt (2006) Methionine acts as “magnet” in photoaffinity crosslinking experiments. FEBS Lett. 580, 18721876.
  • 23
    Rihakova, L., M. Deraet, M. Auger-Messier, J. Perodin, A. A. Boucard, G. Guillemette, R. Leduc, P. Lavigne and E. Escher (2002) Methionine proximity assay, a novel method for exploring peptide ligand–receptor interaction. J. Recept. Signal Transduct. Res. 22, 297313.
  • 24
    Moretto, A., M. Crisma, F. Formaggio, L. A. Huck, D. Mangion, W. J. Leigh and C. Toniolo (2009) Photoinduced intramolecular macrocyclization reaction between a Bpa and a Met residue in a helical peptide: 3D structures of the diastereomeric products. Chem. Eur. J. 15, 6770.
  • 25
    Görner, H., A. G. Griesbeck, T. Heinrich, W. Kramer and M. Oelgemoller (2001) Time-resolved spectroscopy of sulfur- and carboxy-substituted N-alkylphthalimides. Chem. Eur. J. 7, 15301538.
  • 26
    Griesbeck, A. G., H. Mauder and I. Muller (1993) Photochemistry of N-phthaloyl derivatives of methionine. Tetrahedron Lett. 34, 453456.
  • 27
    Giese, B., M. Wang, J. Gao, M. Stoltz, P. Müller and M. Graber (2009) Electron relay race in peptides. J. Org. Chem. 74, 36213625.
  • 28
    Wang, M., J. Gao, P. Müller and B. Giese (2009) Electron transfer in peptides with cysteine and methionine as relay amino acids. Angew. Chem. Int. Ed. 48, 42324234.
  • 29
    Nitecki, D. E., B. Halpern and J. N. Wesley (1968) Simple route to sterically pure dioxopiperazines. J. Org. Chem. 33, 864866.
  • 30
    Murov, S. L., I. Carmichael and G. L. Hug (1993) Handbook of Photochemistry. Marcel Dekker, New York.
  • 31
    Pedzinski, T., A. Markiewicz and B. Marciniak (2009) Photosensitized oxidation of methionine derivatives. Laser flash photolysis studies. Res. Chem. Intermed. 35, 497506.
  • 32
    Case, D. A., T. A. Darden, T. E. Cheatham III, C. L. Simmerling, J. Wang, R. E. Duke, R. Luo, M. Crowley, R. C. Walker, W. Zhang, K. M. Merz, B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry, K. F. Wong, F. Paesani, J. Vanicek, X. Wu, S. R. Brozell, T. Steinbrecher, H. Gohlke, L. Yang, C. Tan, J. Mongan, V. Hornak, G. Cui, D. H. Mathews, M. G. Seetin, C. Sagui, V. Babin and P. A. Kollman (2008) AMBER 10. University of California, San Francisco.
  • 33
    Frisch, M. J., G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, 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, V. Bakken, 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. C. Ortiz, A. G. Baboul, S. Clifford, 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 (2003) Gaussian 03, Revision C02. Gaussian, Inc., Pittsburgh, PA.
  • 34
    Berini, C. and O. Navarro (2012) Ni-catalysed, domino synthesis of tertiary alcohols from secondary alcohols. Chem. Commun. 48, 15381540.
  • 35
    Rung, M. B., M. T. Mwangi, A. L. Miller II, M. Perring and N. B. Bowden (2008) Cascade reactions using LiAlH4 and grignard reagents in the presence of water. Angew. Chem. Int. Ed. 47, 935939.
  • 36
    Deseke, E., Y. Nakatami and G. Ourisson (1998) Intrinsic reactivities of amino acids towards photoalkylation with benzophenone—A study preliminary to photolabelling of the transmembrane protein glycophorin A. Eur. J. Org. Chem. 1, 243251.
  • 37
    Hayon, E., T. Ibata, N. N. Lichtin and M. Simic (1972) Electron and hydrogen atom attachment to aromatic carbonyl compounds in aqueous solution. Absorption spectra and dissociation constants of ketyl radicals. J. Phys. Chem. 76, 20722078.
  • 38
    Lewandowska, A., G. L. Hug, G. Hörner, D. Pogocki, F. Kazmierczak and B. Marciniak (2011) Intramolecular H-atom transfer reactions in rigid peptides. Correlated solvent and structural effects. Can. J. Chem. 89, 266279.
  • 39
    Lewandowska, A., G. L. Hug, G. Hörner, T. Pedzinski, P. Filipiak and B. Marciniak (2010) Photochemical oxidation of anisole in protic solvents. Electron transfer driven by specific solvent–solute interactions. ChemPhysChem. 11, 21082117.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
php1210-sup-0001-FigureS1.docWord document294KData S1. Supplementary Materials

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.