Spectroscopic evidence for a radical cation as intermediate in a model reaction of the 4′-DNA radical strand cleavage

Full text of DOI:10.1021/ja971677y

8740
J. Am. Chem. Soc. 1997, 119, 8740-8741
Spectroscopic Evidence for a Radical Cation as
Intermediate in a Model Reaction of the 4′-DNA
Radical Strand Cleavage
Andreas Gugger,† Rohit Batra,† Piotr Rzadek,‡
Gu¨nther Rist,‡,* and Bernd Giese†,*
Department of Chemistry, UniVersity of Basel
St. Johanns-Ring 19, CH-4056 Basel, Switzerland
Physics Department, NoVartis SerVices AG
CH-4001 Basel, Switzerland
ReceiVed May 22, 1997
ReVised Manuscript ReceiVed July 23, 1997
Figure 1. Time-resolved CIDNP spectrum (150 ns) of 6a exhibiting
polarized resonances of 11 (upper trace) and reference NMR spectrum
of 11 (lower trace).
Radicals are effective neighboring groups to induce nucleo-
philic substitution reactions.1 This activation effect might play
an important role in the anaerobic scission of 4′-DNA radicals.
Whereas ab initio calculations on primary systems make SN2-
type substitutions feasible,1a studies on the radical-induced DNA
strand cleavage (1 f 3) favor the SN1-type reaction with radical
cation 2 as intermediate (Scheme 1).2,3 CIDNP (chemically
induced dynamic nuclear polarization) experiments have led to
the first spectroscopic proof for the existence of a radical cation
as intermediate in a reaction that models the C,O-bond scission
of 4′-DNA radicals.
Scheme 1
As precursor for these radical reactions, we synthesized
ketone 6a via benzonitrile oxide cycloaddition to furan,4
subsequent hydrogenation (4 f 5),5 methylation, and phospho-
rylation. The diastereoisomers could be separated by flash
chromatography.6
Two different kinds of CIDNP experiments, time-resolved
experiments with a maximum time resolution of approximately
50 ns and multiple laser pulse experiments,7,8 were performed
to study the photochemistry of 6a. In methanol solution, several
spin-polarized products could be observed. The major nuclear
spin polarizations are due to the enol ether 11. The resonances
recorded in a time-resolved experiment with a resolution of 150
ns are reproduced in Figure 1 and were also observable in the
50 ns experiment but with much smaller signal to noise (S/N)
ratio.9 The chemical shifts and the polarizations of the signals
are compiled in Table 1. Chemical analysis proved that 11 is
an important product and not solely formed in a side reaction.
CIDNP intensities, governed by the radical pair theory, for
comparable nuclear spin relaxation times and for hyperfine
couplings smaller in magnitude than the absolute value of the
Table 1. CIDNP Spectrum (Multiple Laser Pulse Experiment) of
Enol Ether 11 Observed by Irradiation of Phenyl Ketone 6a and
Calculated from the Hyperfine Coupling Constants of Table 2
rel intensity
chemical
shift [ppm] polarization expt
relaxation
time [s]
calcda
H
3-H
4-H
5-H
6-H
4.55
2.57
4.30
1.68
emission
absorption
absorption
absorption ≡1.0 ≡1.0 (≡1.0)
-5.0 -3.4 (-6.3)
27.9
14.5
15.2
11.7
10.1
1.4
8.6 (10.1)
2.3 (3.2)
† University of Basel.
a The numbers in parentheses take the spin-lattice relaxation times
into account.
‡ Novartis Services AG.
(1) (a) Zipse, H. Angew. Chem., Int. Ed. Engl. 1994, 33, 1985. (b) Curran,
D. P. Chemtracts: Org. Chem. 1995, 8, 62.
(2) (a) Giese, B.; Dussy, A.; Elie, C.; Erdmann, P.; Schwitter, U. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1861. (b) Giese, B.; Beyrich-Graf, X.;
Erdmann, P.; Petretta, M.; Schwitter, U. Chem. Biol. 1995, 2, 367.
(3) For pioneering work in this area, see: Behrens, G.; Koltzenburg,
G.; Schulte-Frohlinde, D. Z. Naturforsch. C 1982, 37, 1205.
(4) Caramella, P.; Cellerino, G.; Corsico Coda, A.; Gamba Invernizzi,
A.; Gru¨nanger, P.; Houk, K. N.; Marinone Albini, F. J. Org. Chem. 1976,
41, 3349.
difference in Zeeman energies of the two radicals are ap-
proximately proportional to the hyperfine coupling constants.10
The character of the polarization in such a case is determined
by Kaptein’s rules.11 Since the strongest polarization in Figure
1 is observed for the C-4 methylene group, the radical pair after
R-cleavage (7 + 8) cannot be responsible for the polarization.
However, it might be built up in a pair of benzoyl radical (7)
and radical cation (9) after rapid phosphate elimination (Scheme
2). In order to verify this hypothesis, we calculated the ESR
hyperfine coupling constants of the radical cation 9 using three
different density functionals known to give reliable results (Table
2).12
(5) Curran, D. P. J. Am. Chem. Soc. 1983, 105, 5826.
(6) The new compounds were fully characterized by spectroscopic
techniques and elemental analysis (data given in the Supporting Information).
Compounds 6a and 6b (assigned by NOE experiments and cis-γ-effect)
were separated by flash chromatography on silica gel with diethyl ether/
ethyl acetate (100:0 to 80:20) as eluents.
(7) For the multiple laser pulse experiment, the observing radiofrequency
pulse followed a saturation pulse sequence and 10 laser pulses with delay
times of 5 ms.
(8) Phenyl ketone 6a (28 mg) was dissolved in MeOH-d4 (0.6 mL),
degassed for 10 min with N2, and sealed in a NMR quartz tube. The solution
was irradiated at 25 °C with an excimer laser (Questek) coupled with a
dye laser (Lambda Physik) at λ ) 342 nm (11 mJ/pulse) in the magnetic
field of a Bruker AM200 NMR spectrometer (200 MHz wide bore).
(9) The experiment was carried out as in ref 8 using a Nd-Yag laser
(Continuum, Surlite II) at λ ) 355 nm (100 mJ/pulse).
Alkyl aryl ketones in general undergo R-cleavage from the
(10) Goetz, M. Concepts Magn. Reson. 1995, 7, 69-86. Adrian, F. J. J.
Chem. Phys. 1970, 53, 3374. Kaptein, R.; Oosterhoff, J. L. Chem. Phys.
Lett. 1969, 4, 195, 214. Closs, G. L. J. Am. Chem. Soc. 1969, 91, 4552.
Fischer, H. Z. Naturforsch. A 1970, 25, 1957.
(11) Kaptein, R. J. Am. Chem. Soc. 1972, 94, 6251. Kaptein, R. J. Chem.
Soc., Chem. Commun. 1971, 732.
S0002-7863(97)01677-6 CCC: $14.00 © 1997 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 119, No. 37, 1997 8741
Scheme 2
dynamically feasible. This reaction leads to the benzoyl cation
10 which should react with methanol yielding methyl benzoate.
Actually, methyl benzoate is formed if phenyl ketone 6a is
irradiated in methanol (Scheme 2).18 Enol ether 11 could not
be detected by GC because it reacts under these acidic conditions
(diethylphosphorous acid is liberated) to acetal 12. The higher
yield of acetal 12 compared to that of methyl benzoate is an
indication that the trapping of radical cation 9 by methanol
competes successfully with the SET reaction (7 + 9 f 10 +
11).19
As mentioned above, the radical pair between benzoyl radical
7 and radical cation 9 is generated in a cage by scission of the
C,O-bond from the first formed triplet pair between benzoyl
radical 7 and tetrahydrofuranyl radical 8. Thus, this phosphate
cleavage (8 f 9) has to compete (a) with the diffusion of the
radicals into the solvent and (b) with the triplet-singlet
interconversion. Because of this competition, the observed
CIDNP effect demands a very rapid reaction for the cleavage
step (8 f 9). Actually, competition kinetic experiments with
an excess of thiophenol as radical trap showed that the
Table 2. Calculation of ESR Hyperfine Coupling Constants aH
(mT) of Radical Cation 9 by Three Density Functionals
method
3-H
4-H
5-H
6-H
B3LYP/6-31G*
BLYP/6-31G*
BPW91/6-31G*
-1.677
-1.489
-1.662
4.276
4.463
4.434
1.157
1.397
1.437
0.495
0.564
0.495
heterolysis of the C,O-bond (8 f 9) is faster than 3 × 109 s-1 20
.
first excited triplet state.13 For compound 6a this cleavage is
accelerated by the oxygen in the five-membered ring. The
radical pair theory predicts that cage products (recombination
or disproportionation within the pair) carry opposite polarization
with respect to their escape counterparts. The observed escape
polarization is usually weaker than the cage polarization. The
major nuclear spin polarization observed for enol ether 11
therefore strongly points to a formation of this product in a cage
reaction. This is further corroborated by the observation of a
weak and opposite spin polarization for 2-methylfuran (Vide
infra). As a consequence of precursor state, product formation,
and the calculated signs of the hyperfine coupling constants, a
benzoyl radical (7) has to be involved in the cage reaction to
fulfill Kaptein’s rule. Its extremely small g-value leads to the
required positive sign for the difference of the g-values of
radicals 9 and 7.14
Under the assumption that the elimination of phosphate
(Scheme 2) does not affect the electron spin state and that the
enol ether 11 is formed by an electron transfer (SET) from the
benzoyl radical 7, the observed polarizations and intensities are
in good agreement with the calculated values (Table 1).
For a more accurate determination of the CIDNP intensities,
the spin lattice relaxation times T1,15 the external magnetic field,
and the estimated difference in g-values of 2.4 × 10-3 14 were
also used (Table 1, values in parentheses).16 The agreement
between the observed and calculated polarizations and the
intensities of the CIDNP spectrum is a direct proof for the radical
cation 9 as intermediate in this C,O-bond cleavage reaction
which undergoes a SET reaction.
With strongly improved S/N ratios such as those in multiple
laser pulse experiments,8 weak resonances of 2-methylfuran
were detected carrying escape-type polarization, lending further
proof for radical cation 9. No resonances were observed in the
CIDNP experiment for 6a and its diastereomer 6b, indicating
that phosphate elimination is more effective than recombination
to the starting material. However, the aldehyde proton of
benzaldehyde (10.0 ppm), formed by hydrogen abstraction, was
observed in enhanced absorption.
Acknowledgment. This work was supported by the Swiss National
Science Foundation. We acknowledge helpful discussions with Dr. T.
Winkler, Novartis AG.
Supporting Information Available: Spectral data and elemental
analysis for compounds 5, 6a, and 6b, including the NOE experiments
for 6a and 6b (3 pages). See any current masthead page for ordering
and Internet access instructions.
JA971677Y
(17) The ionization potentials were calculated with the compound method
CBS-Q: Ochterski, J. W.; Peterson, G. A.; Montgomery, J. A. J. Chem.
Phys. 1996, 104, 2598. The calculated ionization potential of 8.0 eV for 11
is in agreement with its photoelectron spectrum.
(18) For the analysis of the products, phenyl ketone 6a (35 mg) was
dissolved in MeOH (1 mL), degassed for 10 min with N2, and sealed in a
NMR quartz tube. The solution was irradiated at 25 °C with a Nd-Yag
laser (Surelite) at 355 nm (39 mJ/pulse) for 30 min. The products were
identified and quantified by GC using undecane as an internal standard
and synthesized or commercially available reference substances.
(19) We have not observed CIDNP polarizations for acetal 12, which is
formed by trapping reaction of the radical cation 9 with methanol and
subsequent H-abstraction from methanol. The reason is probably the slow
H-abstraction step, which is in the order of 10-102 M-1 s-1 (see Burchill,
C. E.; Ginns, I. S. Can. J. Chem. 1970, 48, 1232, 2628). Presumably, after
this slow reaction step, the polarizations are diminished to such an extent
that they are no longer observable for acetal 12.
(20) The competition kinetic experiment was carried out at 25°C in
MeOH with phenyl ketone 6a (46 mM) and a 10-fold excess of thiophenol
(460 mM). After photolysis, the product analysis was carried out by gas
chromatography. The product formed by trapping of radical 8 by thiophenol
could not be observed. The limit of detection between the stable cleavage
product 12 and the H-abstraction product of radical 8 was 200. With this
number and an estimated rate of H-abstraction for thiophenol of 3 × 107
M-1 s-1 (see Tronche, C.; Martinez, F. N.; Horner, J. H.; Newcomb, M.;
Senn, M.; Giese, B. Tetrahedron Lett. 1996, 37, 5845), the rate for the
heterolytic cleavage (8 f 9) was calculated to be >3 × 109 s-1. With a
benzoate instead of a diethyl phosphate as leaving group (6a, PhCO instead
of PO(OEt)2) the intermediate neutral radical analogous to 8 could be
trapped, and the C,O-bond cleavage was determined as 8 × 106 s-1 at 25
°C in MeOH (unpublished results of S. Peukert and M. Obkircher, Basel).
Kaptein’s rule and the primary R-cleavage have led to the
benzoyl radical as donor in the SET reaction. The calculated
gas phase redox potentials of the benzoyl radical 7 and the enol
ether 11 are 6.5 and 8.0 eV, respectively.17 Thus, the electron
transfer from benzoyl radical 7 to radical cation 9 is thermo-
(12) Batra, R.; Giese, B.; Spichty, M.; Gescheidt, G.; Houk, K. N. J.
Phys. Chem. 1996, 100, 18371.
(13) Turro, N. J. Modern Molecular Photochemistry; University Science
Books: Sausalito, CA, 1991; p 531.
(14) The following g-values were used: g(7) ≈ 2.0006; g(9) ≈ 2.0030
(see (a) Grossi, L.; Placucci, G. J. Chem. Soc., Chem. Commun. 1985, 943.
(b) Madelung, O.; Fischer, H. Landolt-Bo¨rnstein New Series, II/17h, refs
80-84).
(15) T1 Values were measured by the inversion recovery method on a
Varian Gemini 300 (300 MHz) NMR spectrometer.
(16) Borer, A.; Kirchmayer, R.; Rist, G. HelV. Chim. Acta 1978, 61,
305.