Snm as a photolabile protecting group for the cysteinyl radical - Direct evidence from time-resolved IR and UV/Vis spectroscopies

Article abstract of DOI:10.1002/ejoc.200390158

The cysteinyl radical and the methyl(phenyl)thiocarbamic acid radical have been generated by laser flash photolysis (266 nm) of BOC-Cys(Snm)-OH in acetonitrile and characterized both by time-resolved (IR and UV/Vis) and standard spectroscopic methods (UV, IR, and NMR). The mode of action of the Snm unit [methyl(phenyl)thiocarbamic acid] is similar to the so-called "caging" technique. Cleavage of the disulfide bridge by irradiation is possible in acceptable quantum yields. Time-resolved IR (TR-IR) spectroscopic experiments using the step-scan FTIR technique allowed the detection of the methyl(phenyl)thiocarbamic acid radical (V_C=O = 1607.7 and 1579.3 cm^-1), which has a lifetime of about 3 ps. Because of the weak absorptions in the IR spectrum and the tendency for fast recombination to BOC-protected cystine, the cysteinyl radical cannot be monitored by TR-IR spectroscopy. Time-resolved UV/Vis spectroscopy revealed the formation of both of these radicals and permitted the determination of the decay rates of the methyl(phenyl)thiocarbamic acid radical in the presence of several quenchers. NMR spectroscopic studies of the photolysis of BOC-Cys(Snm)-OH allowed us to identify N-methylaniline, carbonyl sulfide, and BOC-protected cystine as the major products. This study clearly demonstrates that the thiol protection group Snm is a suitable photolabile protecting group. ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003).

Full text of DOI:10.1002/ejoc.200390158

FULL PAPER
Snm as a Photolabile Protecting Group for the Cysteinyl Radical ؊
Direct Evidence from Time-Resolved IR and UV/Vis Spectroscopies
Christoph Kolano[a] and Wolfram Sander*[a]
Keywords: IR spectroscopy / Cysteine / Disulfide bridge / Radicals / Kinetics / Photochemistry
The cysteinyl radical and the methyl(phenyl)thiocarbamic
acid radical have been generated by laser flash photolysis
trum and the tendency for fast recombination to BOC-pro-
tected cystine, the cysteinyl radical cannot be monitored by
(266 nm) of BOC−Cys(Snm)−OH in acetonitrile and charac- TR-IR spectroscopy. Time-resolved UV/Vis spectroscopy re-
terized both by time-resolved (IR and UV/Vis) and standard
spectroscopic methods (UV, IR, and NMR). The mode of
action of the Snm unit [methyl(phenyl)thiocarbamic acid] is
similar to the so-called ‘‘caging’’ technique. Cleavage of the
disulfide bridge by irradiation is possible in acceptable
quantum yields. Time-resolved IR (TR-IR) spectroscopic
experiments using the step-scan FTIR technique allowed the
detection of the methyl(phenyl)thiocarbamic acid radical
(νC=O = 1607.7 and 1579.3 cm−1), which has a lifetime of
about 3 µs. Because of the weak absorptions in the IR spec-
vealed the formation of both of these radicals and permitted
the determination of the decay rates of the methyl-
(phenyl)thiocarbamic acid radical in the presence of several
quenchers. NMR spectroscopic studies of the photolysis of
BOC−Cys(Snm)−OH allowed us to identify N-methylaniline,
carbonyl sulfide, and BOC-protected cystine as the major
products. This study clearly demonstrates that the thiol pro-
tection group Snm is a suitable photolabile protecting group.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
‘‘reporter’’ position with a strong IR absorption next to the
disulfide bridge. A variety of molecules have been investig-
ated, but the thiol protection group Snm [methyl-
(phenyl)thiocarbamic acid] published by Barany et al.,[13]
which can be cleaved under mild thiolytic conditions, was
found to be most suitable for this application.
Disulfide bridges play an important role in defining the
three-dimensional structure of peptides and proteins. Dur-
ing the last few years, extensive investigations on the photo-
reaction of disulfide bridges in relatively small biomolecules
have been undertaken.[1] The disulfide cleavage can be per-
formed by various techniques. Direct irradiation yields the
resulting cysteinyl radicals only in poor quantum yields, al-
though it is known[2Ϫ5] that aryl disulfide bonds can be
cleaved in less than 100 ps by ultrashort UV laser pulses
(λexc ϭ 270 nm). A major problem is the fate of the cystei-
nyl radical, and efforts have been made to stabilize this rad-
Results and Discussion
Photochemistry of BOC؊Cys(Snm)؊OH
Laser flash photolysis (LFP) (λexc ϭ 266 nm) with step-
ical so as to avoid rapid recombination.[6,7] Several tech- scan detection of 4 in acetonitrile (1.2 mmol, Ar purge) re-
niques have been reported to monitor the radicals.[8Ϫ12]
sulted in the bleaching of the 1678.2 cmϪ1 absorption of 4
Here we report on the photochemical generation and in- and the formation of a transient species with an absorption
direct spectroscopic detection of cysteinyl radicals. Thiyl at 1607.7 cmϪ1. The lifetime of the transient species was
radicals, such as the cysteinyl radical, show only very weak estimated to be 2.5Ϯ0.5 µs. Both signals were detected dur-
absorptions in the IR spectrum, and thus the direct detec- ing the rise time (25 ns) of the step-scan setup. A complete
tion of the cysteinyl radical by the comparatively insensitive measurement covers a time range of 20 µs, within which
step-scan technique is not possible. In this paper, we present no further signals were detected. A partial recovery of the
a convenient method for the generation and detection of bleached signal after 20 µs indicates the formation of a fur-
the cysteinyl radical. We used a ‘‘caging’’ technique, which ther product. Infrared bands in this region are not charac-
has been used in context with the step-scan technique be- teristic of acyl radicals.[15] A possible fragmentation pattern
fore,[14] for the indirect detection of the cysteinyl radical. of 4 is shown in Scheme 1.
This technique allows the detection of radicals through a
Several products and fragmentation pathways can be dis-
regarded immediately. A cleavage of the carboxylic acid
group is energetically unfavourable. The loss of the BOC
protection group would result in the formation of the tert-
butoxycarbonyl radical 8, which undergoes rapid decarb-
[a]
Lehrstuhl für Organische Chemie II, Ruhr-Universität,
44780 Bochum, Germany
Fax: (internat.) ϩ 49-234/3214353
E-mail: wolfram.sander@ruhr-uni-bochum.de
1074
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0306Ϫ1074 $ 20.00ϩ.50/0 Eur. J. Org. Chem. 2003, 1074Ϫ1079

Snm as a Photolabile Protecting Group for the Cysteinyl Radical
FULL PAPER
Scheme 1
oxylation yielding the tert-butyl radical 9. The detection
In order to confirm the assignment, the vibrational spec-
and kinetics of these radicals has been reported by our trum of 1 was calculated using density functional theory
group recently.[16] Since no evidence for them is found in and ab initio methods [UB3LYP/6-311ϩϩG(d,p) and
our spectra, both fragmentation pathways can be disreg- UMP2/6-31ϩG(d,p)] (Figure 2). The calculated infrared
arded. A suitable site for bond breakage is the disulfide spectrum predicts additional bands should exist in the spec-
bridge. The cleavage of the SϪS bond would lead to the tral region of interest. To verify the calculation, the trans-
methyl(phenyl)thiocarbamic acid radical 1 and the cysteinyl parency of the cell was increased by reducing the spacer size
radical 2. Because of the lack of an intense IR chromophore (0.05 mm). In addition to the bands at 1607.7 cmϪ1 and
and the tendency for rapid recombination, radical 2 cannot 1678.2 cmϪ1, a new absorption is observed at 1579.3 cmϪ1
,
be monitored by the step-scan technique, but the methyl- which decays with the same kinetics as the band at 1607.7
(phenyl)thiocarbamic acid radical 1 seems a promising can- cmϪ1. The results obtained are in good agreement with the
didate for it (Figure 1).
Figure 2. Top: calculated vibrational spectrum of 1 [UB3LYP/6-
Figure 1. Time-resolved difference IR spectra (time resolution
500 ns, spectral resolution 6 cmϪ1, spacer size 0.2 mm) showing the
photochemistry of 4 in argon-purged acetonitrile; bands appearing
on irradiation (λ ϭ 266 nm) are pointing upwards, bands disap-
pearing are pointing downwards; black: spectrum 500 ns after irra-
diation; dotted: spectrum after 20 µs
311ϩϩg(d,p), unscaled]; bottom: time-resolved difference IR spec-
tra (time resolution 500 ns, spectral resolution 6 cmϪ1, spacer size
0.05 mm) showing the photochemistry of 4 in argon-purged aceton-
itrile; bands appearing on irradiation (λ ϭ 266 nm) are pointing
upwards, bands disappearing are pointing downwards; black: spec-
trum 500 ns after irradiation; dotted: spectrum after 20 µs
Eur. J. Org. Chem. 2003, 1074Ϫ1079
1075

C. Kolano, W. Sander
FULL PAPER
theoretical predictions [the CϭO stretching vibrations of 1
are calculated at the UB3LYP/6-311ϩϩG(d,p) level of the-
ory to be 1619.4 and 1651.3 cmϪ1].
The IR spectra of the radicals 3 and 5 Ϫ as well as the
trapping products 7, 11, and 12, as potential secondary
products of 1 Ϫ also were calculated using DFT and ab
initio methods [(UB3LYP/6-311ϩϩG(d,p) and UMP2/6-
31ϩg(d,p)]. None of the calculated spectra matched the ex-
perimental results. Radical 3 is expected to decarbonylate
rapidly to the aminyl radical 6, but neither radical 3 nor
carbon monoxide, with a characteristic and strong IR ab-
sorption at 2140 cmϪ1, was found in our spectra. Because
of the lack of strong IR chromophores, the aminyl radical
6 is not expected to be observable by step-scan FTIR. This
radical will be discussed in context with the time-resolved
UV/Vis experiments.
It is known[17,18] that the acetylthiyl radical and the
benzoylthiyl radical undergo β-fragmentations:
·
CH3COS· Ǟ CH3 ϩ COS
PhCOS· Ǟ Ph·ϩ COS
The PhCOS· radical undergoes β-fragmentation to form
the phenyl radical and carbonyl sulfide with a rate constant
of 8.5·103 sϪ1 at room temperature;[17] for CH3COS· this
rate constant is 6.6·104 sϪ1 [18]
Thus, we expect a similar
.
rate constant for the loss of COS from radical 1. Within the
time range of 20 µs, however, no COS signal (2049.5
cmϪ1 [19]
was observed in the IR spectra. This observation
might be explained by the slow hydrolysis of COS to CO2
and H2S (at 25 °C, the pseudo-first-order rate constant is
0.0011 sϪ1).[20]
)
Figure 3. Structures of 1 and the transition state for the loss of
calculated at the UB3LYP/6-311ϩϩG(d,p) level of theory
Because the intial step Ϫ the loss of COS Ϫ takes place radicals have been examined extensively by several research
on a time scale too slow to be monitored by our step-scan groups using LFP.[21,22] These radicals generate transient
measurement, the subsequent step Ϫ very slow hydrolysis absorptions in the region around 400 nm. We conducted
to carbon dioxide Ϫ also can not be observed directly. The LFP experiments with 4 under the same conditions as de-
activation barrier for the loss of COS from 1 was calculated scribed for the TR-IR experiments. The TR-UV/Vis spectra
using DFT and ab initio methods to be 12.5 kcal/mol [(UB- showed a short-lived transient species with absorptions at
3LYP/6-311ϩϩG(d,p)] (Figure 3).
λmax ϭ 630 and 345 nm. Furthermore, a long-lived transi-
Thus, the loss of COS should take place at room temper- ent species is observed that overlaps with the absorption at
ature, but on a much longer time scale than the observed λ ϭ 345 nm (Figure 4). The transient species observed at
decay of 1. The stability of 1 was confirmed by further inde- λ ϭ 370 and 600 nm (Figure 5) decayed with second-order
pendent techniques (LFP, liquid nitrogen Cryostat and kinetics, yielding a half-life of approximately t1/2
ϭ
product studies after preparative irradiation). The forma- 5.3·10Ϫ6 s.
tion of radical 1 reveals clearly that the Snm protection
At 295 nm the long-lived transient decayed with k ϭ
group is photolabile. The cysteinyl radical 2 must be formed 1.8·104 sϪ1 (τ ϭ 55 µs) in a first-order reaction. In agree-
as the second fragment of the cleavage of 4. Radical 2 could ment with the lifetime obtained from the TR-IR experi-
either undergo hydrogen abstraction from the solvent or re- ments, we assigned the short-lived transient species to the
combine with 1 to give the starting compound 4. A further methyl(phenyl)thiocarbamic acid radical 1 and the long-
possibility is the dimerization to yield cystine 10. We per- lived species to the cysteinyl radical 2. In order to confirm
formed a TR-IR spectroscopic experiment under 1 bar of our assignment, the UV/Vis spectra of radical 1 was calcu-
oxygen in order to examine the quenching/oxidation chem- lated using density functional theory [(UB3LYP/6-
istry of 1. The spectra obtained in this experiment showed 311ϩϩG(d,p)] on the previously optimized structure of the
a variety of signals with a much more complex photochem- radical. The calculated absorptions are in good agreement
istry and an inferior S/N ratio than in the argon-purged ex- with the experimental results (Table 1).
periments.
Our time-resolved UV/Vis measurements cover the time
To confirm that a loss of COS from 1 takes place on a ranges up to 75 µs. Within this time range, we did not ob-
relatively long time scale, and that the cysteinyl radical 2 is serve the appearance of an absorption in the region of λ ഠ
formed, we used time-resolved UV/Vis spectroscopy. A loss 400 nm indicating the formation of aminyl radical 6 by the
of COS from radical 1 forms the aminyl radical 6. Aminyl loss of COS, a situation that is in accordance with the con-
1076
Eur. J. Org. Chem. 2003, 1074Ϫ1079

Snm as a Photolabile Protecting Group for the Cysteinyl Radical
FULL PAPER
ent absorption at λ ഠ 380 nm.[8,9] The formation of the
CysSS· radical, however, is energetically not favoured.
Thiyl radicals add quickly to double bonds in an anti-
Markovnikov fashion.[23Ϫ25] We performed quenching ex-
periments and measured the rate constant for the reaction
of 1 with 1,4-cyclohexadiene (1,4-CHD) by determining the
decay rate of 1 in the presence of 1,4-CHD at seven differ-
ent concentrations (Figure 6). The rate constant for this re-
action is 3.5·106 L·molϪ1·sϪ1. Similarly, the rate constant
for the reaction of 1 with methyl methacrylate (MMA) was
determined to be k ϭ 1.95·107 L·molϪ1·sϪ1 by determining
the decay rate of 1 in the presence of five different concen-
trations of MMA (Figure 7).
Figure 4. Time-resolved UV/Vis spectra [traces recorded after 1 µs
(black dots), 5 µs (white dots), 17 µs (black triangles), and 75 µs
(white triangles), spectral resolution 5 nm, cell size 1.0 cm] dis-
playing the photochemistry of 4 in argon-purged acetonitrile
Figure 6. Decay rate of 1 in the presence of seven different concen-
trations of 1,4-cyclohexadiene
Figure 5. Transient decay trace, recorded at λ ϭ 370 nm, upon laser
flash photolysis (λexc ϭ 266 nm) of 4 in acetonitrile at ambient tem-
perature
Table 1. Experimental and calculated UV absorptions of the
methyl(phenyl)thiocarbamic acid radical 1
Experimental λ [nm]
Calculated λ [nm][a]
345.0
465.0
630.0
325.9
478.8
621.6
Figure 7. Decay rate of 1 in the presence of five different concentra-
tions of MMA
[a]
UB3LYP/6-311ϩϩg(d,p).
clusions from the TR-IR measurements. According to pub-
Monochromatic UV irradiation (λexc ϭ 266 nm) of 4 in
lished LFP data on cystine and its derivatives, the cysteinyl degassed acetonitrile at 230 K resulted in slow bleaching of
radical should exhibit an absorption maximum at a shoulder at λ ϭ 284 nm. Simultaneously, a band at λ ϭ
330 nm.[8,9] We assume that this absorption overlaps with 330 nm with the same kinetic behaviour is formed. This new
the absorption of 1 at λ ϭ 345 nm, which has to be taken absorption at 330 nm is in good agreement with the ex-
into account when evaluating the kinetic data. It has been pected absorption for the cysteinyl radical 2.
reported in the literature that the cysteinyl radical can ab-
We performed a product study to confirm the formation
stract sulfur atoms yielding the CysSS· radical with a transi- of the cysteinyl radical 2. A degassed acetonitrile solution
Eur. J. Org. Chem. 2003, 1074Ϫ1079
1077

C. Kolano, W. Sander
FULL PAPER
of 4 in a sealed quartz cell was irradiated for 1 h with the change to slightly brown as was observed during the laser
irradiation of an Nd:YAG laser at 266 nm. After irradi- irradiation. To confirm the ability of Snm to act as a photo-
ation, the previously colourless solution had turned slightly labile protecting group, we examined the irradiated sample
brown and the odour of thiol-containing compounds was by FTIR spectroscopy. After irradiation, the IR spectra
1
sensed. A H NMR spectrum in [D6]DMSO revealed that showed three new bands in the carbonyl region: a strong
clean photochemistry had occurred. Under these condi- absorption at 1753.7 cmϪ1 and two very strong bands at
tions, the loss of COS takes place as indicated by the forma- 1715.6 and 1512.7 cmϪ1. These signals are assigned to the
tion of N-methylaniline. To trap the COS we repeated the three prominent vibrations of the BOC-protected cystine 10
1
experiment in a fused quartz NMR tube. We recorded H (Table 3), and is in good agreement with literature data.[27]
NMR, 13C NMR, and HMQC spectra. All signals in the
Table 3. IR absorptions of photoproduct 10 in solution (CH3CN)
1H NMR (Table 2) and 13C NMR (Figure 8) spectra can be
assigned clearly to COS, N-methylaniline, BOC-protected
Exptl. ν˜ [cmϪ1
]
Exptl. ν˜ [cmϪ1 [27]
]
Assignment
cystine 10, and traces of nonphotolized 4. There is no evid-
ence that 7 is formed since the characteristic shift of the
ϪNMe group at δ ϭ 3.36 ppm[26] is missing.
1753.7
1715.6
1512.7
1743
1718
1509
νCϭO CO2H
νCϭO
Amide II
In conclusion we have shown that Snm can be used as a
photolabile protecting group. Photolysis leads, in good
yields, to the interesting methyl(phenyl)thiocarbamic acid
radical 1 and the well-known cysteinyl radical 2, which un-
dergo various subsequent reactions (e.g., 2 dimerizes to
form cystine 10). Thus, the precursor molecule 4 can be
considered as a ‘‘caged’’ cysteinyl radical 2. In view of the
many biochemical functions of cysteinyl radicals, this pro-
tecting group and these spectroscopic techniques might find
applications in the exploration of biochemical reaction
pathways.
Figure 8. 13C NMR spectra (150 MHz) after preparative photolysis
(1 h, 266 nm, 10 Hz, 0.6 W sϪ1) of 4 in degassed CD3CN in a fused
quartz NMR tube
Experimental Section
Step-Scan FTIR: For time-resolved measurements we have used a
set-up that has been described previously.[16] Excitation of the
sample is carried out by the 4th harmonic (266 nm) of an Nd-YAG
laser (Spectra-Physics Quanta-Ray LAB 100) with a repetition rate
of 10 Hz. To avoid thermal effects and shockwaves, the laser energy
is attenuated to between 15 and 28 mJ/pulse by an external attenu-
The lability of the Snm protection group towards irradi-
ation can be monitored by a simple experiment. Exposure
of a normal NMR sample of 4 (for example in [D6]DMSO)
to daylight for several weeks resulted in the same colour ator. All measurements were carried out with 6 cmϪ1 resolution in
1
Table 2. H NMR spectroscopic data (600 MHz) of the products obtained after preparative photolysis (1 h, 266 nm, 10 Hz, 0.6 W·sϪ1
)
of 4 in degassed CD3CN
Exptl. δ [ppm]
Assignment
Exptl. δ [ppm][a]
Exptl. δ [ppm][27]
1.42
2.74
2.97
3.27
3.33
4.39
4.46
5.73
BOC CH3, of 10 and unchanged 4
CH3, methylaniline
Cys β-CH2, of 10 and unchanged 4
Cys β-CH2, of 10 and unchanged 4
NCH3 of unchanged 4
NH, methylaniline
Cys α-CH, of 10 and unchanged 4
NH of 10
COOH of 10
NH of unchanged 4
Aromatic ring, methylaniline
Aromatic ring, methylaniline
Aromatic ring of unchanged 4
Aromatic ring of unchanged 4
1.47
3.25
2.82
4.30
4.47
5.78
6.28
6.15
6.58
7.12
7.37
7.47
6.69
7.26
[a]
Recorded as a reference (200 MHz, CD3CN).
1078
Eur. J. Org. Chem. 2003, 1074Ϫ1079

Snm as a Photolabile Protecting Group for the Cysteinyl Radical
a spectral range between 0 and 3160 cmϪ1, resulting in an interfero-
FULL PAPER
by the Deutsche Forschungsgemeinschaft (Sonderforschungs-
gram containing 1060 points. The signal is averaged 20 times per bereich 452) and the Fonds der Chemischen Industrie.
sampling position. In all experiments, a time range of 20 µs is re-
[1]
M. Volk, Eur. J. Org. Chem. 2001, 14, 2605Ϫ2621.
T. W. Scott, S. N. Liu, J. Phys. Chem. 1989, 93, 1393Ϫ1396.
N. P. Ernsting, Chem. Phys. Lett. 1990, 166, 221Ϫ226.
N. A. Borisevich, N. A. Lysak, S. V. Melnichuk, S. A. Tikhomi-
rov, G. B. Tolstorozhev, in: Ultrafast Phenomena in Spectro-
scopy, vol. 49 (Eds.: E. Klose, B. Wilhelmi), Springer-Verlag,
Berlin, Heidelberg, 1990, pp. 276Ϫ281.
Y. Hirata, Y. Niga, T. Okada, Chem. Phys. Lett. 1994, 221,
283Ϫ288.
M. Volk, Y. Kholodenko, H. S. M. Lu, E. A. Gooding, W. F.
DeGrado, R. M. Hochstrasser, J. Phys. Chem. B 1997, 101,
8607Ϫ8616.
H. S. M. Lu, M. Volk, Y. Kholodenko, E. Gooding, R. M. H.
Hochstrasser, W. F. DeGrado, J. Am. Chem. Soc. 1997, 119,
7173Ϫ7180.
D. W. Grant, J. H. Stewart, Photochem. Photobiol. 1984, 40,
285Ϫ290.
G. H. Morine, R. R. Kuntz, Photochem. Photobiol. 1981, 33,
1Ϫ5.
S. P. Mezyk, Chem. Phys. Lett. 1995, 235, 89Ϫ93.
M. Z. Hoffman, E. Hayon, J. Phys. Chem. 1973, 77, 990Ϫ996.
S. A. Everett, C. Schöneich, J. H. Stewart, K. D. Asmus, J.
Phys. Chem. 1992, 96, 306Ϫ314.
corded with a resolution of 25 ns. After FT, the average of 20 suc-
cessive spectra is calculated to achieve a better signal-to-noise ratio,
resulting in an effective time resolution of 500 ns. If necessary, this
figure can be improved by averaging a smaller number of spectra.
The sample is dissolved (between 1.0 and 6.0 g·LϪ1 depending on
the layer thickness of the cell) in acetonitrile (Riedel-de Hae¨n, spec-
troscopic grade) and purged with argon for at least 45 min.
[2]
[3]
[4]
[5]
[6]
Laser Flash Photolysis: A standard LFP set-up was used, consisting
of a Spectra-Physics Quanta-Ray LAB 130 Nd-YAG laser, oper-
ated at 1 Hz and 266 nm (50 mJ/pulse, 8 ns pulse duration), a
pulsed Xe arc lamp by Müller, Germany, a TRIAX 180 monoch-
romator coupled to a photoelectron multiplier tube, and a LeCroy
9361 digital oscilloscope. The entire set-up was controlled from a
personal computer using LabView software. Data evaluation was
performed using SigmaPlot software. In order to avoid depletion
of the precursor and product build-up, a flow-cell was used.
Quenching experiments were performed in the same way, except
that the flow-cell was replaced by a static cell. The concentration
of 4 was adjusted so that an optical density of 0.3 was achieved at
the laser wavelength used for excitation.
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
A. L. Schroll, G. Barany, J. Org. Chem. 1989, 54, 244Ϫ247.
C. Allin, K. Gerwert, Biochemistry 2001, 40, 3037Ϫ3046.
C. Chatgilialoglu, D. Crich, M. Komatsu, I. Ryu, Chem. Rev.
1999, 99, 1991Ϫ2069.
G. Bucher, M. Halupka, C. Kolano, O. Schade, W. Sander, Eur.
J. Org. Chem. 2001, 3, 545Ϫ552.
Calculations: Calculations were performed with the Gaussian98[28]
suite of programs. Transition states, energies, vibrational and
UV spectra were calculated using B3LYP or MP2 theory with a
6-311ϩϩG(d,p) or 6-31ϩG(d,p) basis set.
[16]
[17]
[18]
[19]
´
R. Zhao, J. Lind, G. Merenyi, T. E. Eriksen, J. Am. Chem. Soc.
1998, 120, 2811Ϫ2816.
Product Studies: Photolyses were performed using acetonitrile or
[D3]acetonitrile and an Nd-YAG laser (Spectra-Physics Quanta-
Ray LAB 100) operating at λexc ϭ 266 nm with a repetition rate of
10 Hz. The preparative photolysis was carried out with 0.6 W·sϪ1
for 1 h. Samples (0.04 mol/L) were poured in quartz NMR tubes
and degassed by the use of a supersonic bath and the ‘‘freeze-and-
´
R. Zhao, J. Lind, G. Merenyi, T. E. Eriksen, J. Phys. Chem. A
1999, 103, 71Ϫ74.
J. Fournier, C. Lalo, J. Deson, C. Vermeil, J. Chem. Phys. 1977,
66, 2656Ϫ2659.
M. M. Sharma, Trans. Faraday Soc. 1964, 61, 681Ϫ688.
B. D. Wagner, G. Ruel, J. Lusztyk, J. Am. Chem. Soc. 1996,
118, 13Ϫ19.
E. Leyva, M. S. Platz, B. Niu, J. Wirz, J. Phys. Chem. 1987,
91, 2293Ϫ2298.
K. Griesbaum, Angew. Chem. 1970, 82, 276Ϫ290; Angew.
Chem. Int. Ed. Engl. 1970, 9, 273Ϫ287.
W. K. Busfield, K. Heiland, I. D. Jenkins, Tetrahedron Lett.
1995, 36, 1109Ϫ1112.
W. K. Busfield, K. Heiland, I. D. Jenkins, Tetrahedron Lett.
1994, 35, 6541Ϫ6542.
G. Barany, A. L. Schroll, A. W. Mott, D. A. Halsrud, J. Org.
Chem. 1983, 48, 4750Ϫ4761.
[20]
[21]
1
pump’’ technique. The tubes were fused and H, 13C, and HMQC
NMR spectra were taken before and after the photolysis (Bruker,
Avance DPX 200, 200.13 MHz and Bruker, Avance DRX 600,
600.13 MHz).
[22]
[23]
[24]
[25]
[26]
[27]
[28]
Nitrogen Cryostat: Measurements were carried out in acetonitrile
solutions at 230 K in a liquid-nitrogen cryostat (Oxford, DN1714
equipped with Suprasil windows). Samples were poured into self-
made quartz cuvettes (length 60 cm, diameter 1.0 cm), degassed in
the same way as described for the preparative photolysis and sealed
with a stopper. For irradiation, we used a high-pressure mercury
lamp (Mercury Light Source 200, Bausch & Lomb, Rochester, NY,
USA) with a monochromator (Bausch & Lomb, grating 1200
grooves/mm) operating at λexc ϭ 265 nm. UV spectra were recorded
S. Ranganathan, N. Tamilarasu, R. Roy, Tetrahedron 1996,
52, 9823Ϫ9834.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgom-
ery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Mil-
lam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Po-
melli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P.
Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A.
G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challac-
ombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L.
Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle, J. A.
Pople, Gaussian 98, Revision A.6, Gaussian, Inc., Pittsburgh,
PA, USA, 1998.
every minute for
1
h
with
a
diode-array spectrometer
(HewlettϪPackard, HP 8452) in the spectral range 210Ϫ700 nm
with a resolution of 2 nm. To avoid influences on the sample by
the deuterium lamp of the spectrometer, the scan time was set to
0.5 s.
Synthesis: The precursor BOCϪCys(Snm)ϪOH was synthesized
according to a literature procedure described by Barany et al.,[13]
except that the workup was modified slightly.
Acknowledgments
C. K. is indebted to PD Dr. G. Bucher for his introduction and
advice to the LFP technique. This work was financially supported
Received August 14, 2002
[O02468]
Eur. J. Org. Chem. 2003, 1074Ϫ1079
1079