Photochemical generation and reactivity of the major hydroxyl radical adduct of thymidine

Article abstract of DOI:10.1021/ol301109z

5,6-Dihydro-5-hydroxythymidin-6-yl radical (1), the major reactive intermediate resulting from hydroxyl radical addition to C5 of the pyrimidine, is produced via 350 nm photolysis of a 2,5-dimethoxyphenylsulfide precursor (2). Competition between O₂ and thiol for 1 suggests that the radical reacts relatively slowly with β-mercaptoethanol compared to other alkyl radicals. Overall, aryl sulfide 2 should be an effective precursor for the major hydroxyl radical adduct of thymidine in DNA.

Full text of DOI:10.1021/ol301109z

ORGANIC
LETTERS
2012
Vol. 14, No. 11
2866–2869
Photochemical Generation and Reactivity
of the Major Hydroxyl Radical Adduct of
Thymidine
Joanna Maria N. San Pedro and Marc M. Greenberg*
Department of Chemistry, Johns Hopkins University, 3400 North Charles Street,
Baltimore, Maryland 21218, United States
mgreenberg@jhu.edu
Received April 25, 2012
ABSTRACT
5,6-Dihydro-5-hydroxythymidin-6-yl radical (1), the major reactive intermediate resulting from hydroxyl radical addition to C5 of the pyrimidine, is
produced via 350 nm photolysis of a 2,5-dimethoxyphenylsulfide precursor (2). Competition between O2 and thiol for 1 suggests that the radical
reacts relatively slowly with β-mercaptoethanol compared to other alkyl radicals. Overall, aryl sulfide 2 should be an effective precursor for the
major hydroxyl radical adduct of thymidine in DNA.
Hydroxyl radical abstracts hydrogen atoms and adds to
double bonds with rate constants that can approach the
diffusion controlled limit.1 The reactivity of the hydroxyl
radical is particularly important in nucleic acid damage by
γ-radiolysis and radiomimetic systems, such as Fe•EDTA.1,2
Nucleobase radicals resulting from π-bond addition are the
major family of reactive intermediates formed when nucleic
acids are exposed to the hydroxyl radical. Because the
hydroxyl radical reacts approximately equally with each
nucleotide and without any sequence selectivity in nucleic
acids,2,3 methods have been developed for independently
generating the corresponding reactive intermediates in
DNA and RNA.4ꢀ7 We describe herein photochemical
generation of the major hydroxyl radical adduct of thymidine
(5,6-dihydro-5-hydroxythymidin-6-yl, 1) from 2 using
350 nm irradiation and examination of the reactivity of 1
(Scheme 1).
Scheme 1
The purpose of independently generating reactive inter-
mediates is to simplify studying their reactivity. Previously,
nucleobase radicals have largely been generatedvia Norrish
(1) von Sonntag, C. Free-Radical-Induced DNA Damage and Its
Repair; Springer-Verlag: Berlin, 2006.
(2) Pogozelski, W. K.; McNeese, T. J.; Tullius, T. D. J. Am. Chem.
Soc. 1995, 117, 6428–33.
(3) Balasubramanian, B.; Pogozelski, W. K.; Tullius, T. D. Proc.
Natl. Acad. Sci. U.S.A. 1998, 95, 9738–9743.
(4) Greenberg, M. M. Org. Biomol. Chem. 2007, 5, 18–30.
(5) Resendiz, M. J. E.; Pottiboyina, V.; Sevilla, M. D.; Greenberg,
M. M. J. Am. Chem. Soc. 2012, 134, 3917–3924.
(6) Jacobs, A. C.; Resendiz, M. J. E.; Greenberg, M. M. J. Am. Chem.
Soc. 2011, 133, 5152–5159.
(8) Hong, I. S.; Carter, K. N.; Sato, K.; Greenberg, M. M. J. Am.
Chem. Soc. 2007, 129, 4089–4098.
(9) Carter, K. N.; Greenberg, M. M. J. Am. Chem. Soc. 2003, 125,
13376–13378.
(10) Greenberg, M. M.; Barvian, M. R.; Cook, G. P.; Goodman,
B. K.; Matray, T. J.; Tronche, C.; Venkatesan, H. J. Am. Chem. Soc.
1997, 119, 1828–1839.
(7) Giese, B.; Dussy, A.; Meggers, E.; Petretta, M.; Schwitter, U.
J. Am. Chem. Soc. 1997, 119, 11130–11131.
r
10.1021/ol301109z
Published on Web 05/22/2012
2012 American Chemical Society

Scheme 2
Scheme 3
Type I photocleavage of ketones.4ꢀ6,8ꢀ10 Ketones are
potentially limited in their use with respect to sequence
due to potential excited state quenching by dG.11 Mono-
meric 5,6-dihydro-5-hydroxythymidin-6-yl (1) was inde-
pendently generated from a m-trifluoromethyl benzoate
via photoinduced single electron transfer, but this method
was incompatible with forming 1 at defined sites in
oligonucleotides.12 More recently, an aryl sulfide (3) was
used to produce 1, but this molecule required 254 nm
irradiation.13,14 Products attributable to 1 were formed in
low yield, and it is likely that irradiation of DNA contain-
ing this molecule at 254 nm would damage the biopolymer.
Aryl sulfide 3 was previously synthesized from 6 following
treatment of the bromohydrin (5) with ZnO in the presence of
thiophenol (Scheme 2).13 We attempted using this strategy
toprepare 2, butthe sulfurization reaction wasinconsistent
in our hands. Consequently, we pursued a slightly different
approach using the protected thymidine glycol (7, Scheme 3)
in which the C6-hydroxyl was activated for substitution
by benzoylation. Secondary benzoate 8 was obtained from
the major diastereomer of the glycol (7).16 After significant
experimentation, the aryl sulfide was reproducibly ob-
tained by treating 8 with ZnCl2 (2.5 equiv) in the presence
of the appropriate thiol (2.0 equiv) at ꢀ78 °C.15 Similar
yields were obtained when either 4-methoxythiophenol or
2,5-dimethoxythiophenol were used as nucleophiles. In
each instance two diastereomers were obtained, albeit in
approximately a 10:1 ratio. NOE experiments carried out
on the bis-silyl ethers (9a,b) indicated that the cis isomer
(9a) was the major product.17 In order to enhance UV-
detection of the anticipated products from 1 following
HPLC separation, the 50-benzoyl derivative of 2 (10), as
well as 13ꢀ15, were independently synthesized (Figure 1).
Irradiation (30 min) using lamps with maximum output
at 350 nm under degassed conditions in the presence of
0ꢀ350 mM β-mercaptoethanol (BME) consumed more
than 97% of 10. The major product formed was the
thymidine C5-hydrate (14), even in the absence of BME
(Table 1). Significant amounts of thymidine glycol (13, as a
mixture of epimers atC6) werealsoformed, and the overall
mass balance was >85%. The ratio of diastereomeric glycols
(∼1:1), their yield relative to 14 (∼1.5), and their overall
Figure 1. Structures of 50-benzoylated photochemical substrates
(10ꢀ12), anticipated photoproducts (13, 14), and internal stan-
dard (15).
yield showed little dependence on BME concentration
(Table 1). The formation of glycols (13) and 14 in the
absence of O2 and BME, respectively, warranted closer
examination. Glycol formation under anaerobic condi-
tions suggested that the carbocation (16, Scheme 4) was
an intermediate. The observation that the yield of 13 in the
presence of NaN3 (0.6 M) is half as much as in the absence
of azide is consistent with the intermediacy of 16. The
carbocation (16) could form by direct heterolysis upon
excitation and/or electron transfer between the radical pair
(Scheme 4).18 The lack of an effect of BME concentration
on the yield of 13 indicates that the glycols arise from
(11) Adam, W.; Arnold, M. A.; Nau, W. M.; Pischel, U.; Saha-
Moeller, C. R. J. Am. Chem. Soc. 2002, 124, 3893–3904.
(12) Barvian, M. R.; Barkley, R. M.; Greenberg, M. M. J. Am. Chem.
Soc. 1995, 117, 4894–4904.
(13) Zhang, Q.; Wang, Y. J. Am. Chem. Soc. 2004, 126, 13287–13297.
(14) Zhang, Q.; Wang, Y. Chem. Res. Toxicol. 2005, 18, 1897–1906.
(15) Gordon, D. M.; Danishefsky, S. J. Carbohydr. Res. 1990, 206,
361–366.
(16) Barvian, M. R.; Greenberg, M. M. J. Org. Chem. 1993, 58, 6151–
4.
(17) See Supporting Information.
(18) Fleming, S. A.; Jensen, A. W. J. Org. Chem. 1996, 61, 7040–7044.
Org. Lett., Vol. 14, No. 11, 2012
2867

Table 1. Product Yields upon Photolysis of Aryl Sulfide (10)
under Anaerobic Conditions As Function of BME
Concentrationa
% yieldb
[BME]
(mM)
mass
13
14
balance
0
1
36.4 ( 2.0
33.8 ( 5.0
41.6 ( 1.6
42.3 ( 1.3
43.5 ( 2.4
52.3 ( 4.7
51.5 ( 6.2
58.5 ( 3.0
58.3 ( 2.9
57.7 ( 0.9
89.7 ( 6.1
86.0 ( 8.0
100.1 ( 4.4
100.6 ( 4.4
101.1 ( 2.9
20
150
350
a Initial [10] = 30 μM. b Based upon unrecovered starting material.
Yields are the average (std. dev. of three experiments determined by HPLC.
heterolysis and/or electron transfer between the neutral
intermediates within the initially formed radical cage. The
aryl thiol produced in conjunction with 16 (and ultimately
13) was suspected asthe source ofthe hydrogen atom in the
absence of BME. Taking advantage of the relatively low
pKa of the thiol supported this hypothesis. No thymidine
hydrate (14) was observed when 10 was photolyzed in the
presence of NaOH (1 mM) and no BME, presumably due
to deprotonation of any aryl thiol that formed in situ.19
Figure 2. Relative yields of thymidine C5-hydrate (14) and
glycols (13) from photolysis of 10 as a function of thiol (BME)
concentration. (Yields of 13 are corrected for nonradical path-
ways. See text.)
were unable to detect 13 or 14 in
[14]
[13]
kBME[50-Bz-1][BME]
¼
ð1Þ
k
O2[50-Bz-1][O2]
the photolyzates by HPLC. These data suggested that overall
the dimethoxyphenyl sulfide (10) was the best photochemical
radical precursor, and all subsequent experiments were
carried out using it.
Scheme 4
Photolysis of 10 under aerobic conditions resulted in
considerably higher glycol (13) yields. For instance, 13
was formed in 98.9 ( 0.5% under conditions where BME
(1 mM) would not be expected to effectively compete with
O2 for 50-Bz-1. The amount of glycol (13) attributable to O2
trapping of the radical (50-Bz-1) was determined by subtract-
ing the yield of this product detected under anaerobic
conditions from the total yield. A range for the rate constant
describing BME trapping of 50-Bz-1 (kBME) was determined
by subtracting either the low or high yield of 13 detected
under anaerobic conditions (Table 1). The rate constant for
hydrogen atom abstraction from BME by 50-Bz-1 was
estimated by using O2 as a competitor. The ratio of 14:13
varied linearly with respect to the concentration of BME
(Figure 2). An approximate rate constant for BME trapping
of 50-benzoylated 1 (kBME) was obtained from the slope of
For comparison, we analyzed the photolyzates of 11 and
12 in order to gauge their suitability as precursors for the
nucleobase radical (1). Although aryl sulfide 12 was not
susceptible to irradiation at 350 nm under anaerobic condi-
tions, photolysis using lamps that emit maximally at 300 nm
produced 13 and 14. Irradiation of 12 for 30 min resulted in
g70% conversion. The yields of glycols (13) and hydrate (14)
from 12 were independent of BME concentration, but the
ratio of 14:13 was greater (∼2.0) than that obtained from
photolysis of 10 (∼1.5). The 50-benzoyl derivative (11) of the
previously reported phenyl sulfide required irradiation in
quartz tubes using 254 nm lamps. Irradiation of 11 under
these conditions resulted in its rapid consumption and the
formation of a complex mixture of products. However, we
this line by assuming that kO = 2 ꢁ 109 Mꢀ1sꢀ1 and [O2] =
2
0.2 mM (eq 1). The rate constant for BME trapping
(kBME = (4.7 ( 0.1) ꢁ 105 to (5.7 ( 0.1) ꢁ 105 Mꢀ1sꢀ1) is
considerably lower than that for typical alkyl radical reac-
tions with thiols.20 The estimated rate constant (kBME) is
only ∼5-fold slower than the rate constant for the reaction
of the related radical (17) with BME (kBME = (2.6 ( 0.5) ꢁ
106 Mꢀ1 sꢀ1).21 Computational experiments support the rela-
tively slow reactivity of 17 (and by inference 1).5 The even
(20) Newcomb, M. Tetrahedron 1993, 49, 1151–1176.
(21) Newman, C. A.; Resendiz, M. J. E.; Sczepanski, J. T.; Greenberg,
M. M. J. Org. Chem. 2009, 74, 7007–7012.
(19) Bonesi, S. M.; Fagnoni, M.; Dondi, D.; Albini, A. Inorg. Chim.
Acta 2007, 360, 1230–1234.
2868
Org. Lett., Vol. 14, No. 11, 2012

greater recalcitrance of 50-Bz-1 to react with BME may be
due to increased steric hindrance by the C5-substituents.
Moreover, the reactivity difference between 1 and 17 in
hydrogen atom abstraction reactions may carry over to
studies in DNA. However, the higher barriers encountered
when abstracting hydrogen atoms from carbon-hydrogen
bonds should result in an even greater difference in reactivity
between these two nucleobase radicals.
using the 2,5-dimethoxyaryl sulfide (2, 10) as a photo-
chemical precursor in DNA. The experiments de-
scribed above suggest that the sulfide will be useful for
generating 1 under conditions that will not randomly
damage DNA.
Acknowledgment. We are grateful for support of this
research by the National Institute of General Medical
Sciences (GM-054996).
Supporting Information Available. Experimental pro-
cedures for the synthesis of 2. Spectroscopic character-
ization of all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
This possibility and the reactivity of 5,6-dihydro-5-
hydroxythymidin-6-yl (1) in general will be addressed
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 11, 2012
2869