Synthesis of deuterium and C-13-labelled ethyl glycolate and their subsequent use in the synthesis of labelled analogues of the DNA adduct O 6-carboxymethyl-2′-deoxyguanosine

Article abstract of DOI:10.1002/jlcr.1923

The adduct O⁶-carboxymethyl-2′-deoxyguanosine (O ⁶CMdG) is of importance as it has been previously linked to high red meat diet in humans, and as yet, a liquid chromatography-mass spectrometry (LC-MS) method has not been developed due to lack of appropriate standards. The synthesis of the deuterated and C-13 analogues required the use of [ ²H₂]- and [¹³C₂]ethyl glycolate to label the carboxymethyl moiety of O⁶CMdG. [²H ₂]Ethyl glycolate was synthesised via acid hydrolysis of ethyl diazoacetate using deuterated solvents (59% yield), whilst [¹³C ₂]ethyl glycolate was synthesised from [¹³C ₂]glycine in a three-step procedure (35% yield). The labelled ethyl glycolates were then used to synthesise [²H₂]- and [ ¹³C₂]O⁶CMdG for future use as internal standards in the LC-MS analysis of biological samples.

Full text of DOI:10.1002/jlcr.1923

Note
Received 1 July 2011,
Revised 25 July 2011,
Accepted 27 July 2011
Published online 7 September 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1923
Synthesis of deuterium and C-13-labelled
ethyl glycolate and their subsequent use in
the synthesis of labelled analogues of the DNA
6
adduct O -carboxymethyl-2′-deoxyguanosine
a
b
Sharon A. Moore * and David E. G. Shuker
6
6
The adduct O -carboxymethyl-2′-deoxyguanosine (O CMdG) is of importance as it has been previously linked to high red
meat diet in humans, and as yet, a liquid chromatography-mass spectrometry (LC-MS) method has not been developed
2
due to lack of appropriate standards. The synthesis of the deuterated and C-13 analogues required the use of [ H
2
]- and
1
3
6
2
[
C ]ethyl glycolate to label the carboxymethyl moiety of O CMdG. [ H ]Ethyl glycolate was synthesised via acid hydrolysis
2 2
13 13
of ethyl diazoacetate using deuterated solvents (59% yield), whilst [ C
in a three-step procedure (35% yield). The labelled ethyl glycolates were then used to synthesise [ H ]- and [ C ]O CMdG for
2
]ethyl glycolate was synthesised from [ C ]glycine
2
2 13 6
2
2
future use as internal standards in the LC-MS analysis of biological samples.
6
6
Keywords: stable labelled synthesis; C-13; H-2; O -carboxymethyl-2′-deoxyguanosine; O CMdG; ethyl glycolate; LC-MS
Introduction
Results and discussion
Cancer is a disease that afflicts approximately one-third of the The synthesis of 1 was carried out with methyl glycolate and6
global population at some point in their lifetime. Consequently, gave a lower overall yield (6.2%) than previously reported
there is a vast quantity of research into the mechanisms and pre- despite several attempts to improve yields at each stage. We
vention of cancer, and differences have been found according to then investigated the hydrolysis of ethyl diazoacetate to give
geographical regions that have been linked to a number of fac- ethyl glycolate in both aqueous and deuterated solvents
1
tors including diet. Mutagenic and carcinogenic compounds (Scheme 2, step (iii)). As discussed later, the C-13-labelled ethyl
arise from many exogenous and endogenous routes, including glycolate was subsequently required; hence, the procedure was
diet, and may result in chemical modifications to the DNA struc- extended to produce ethyl glycolate from glycine (Scheme 2).
6
ture (i.e. DNA adducts). One such adduct is O -carboxymethyl-2′- Kresge and Popik had shown that the hydrolysis of diazoketones
6
deoxyguanosine (O CMdG), which has been linked to nitrosated took place in both water and deuterium oxide to give the corre-
2
7
amines from red meat. Previous research has examined this sponding alcohol. Whilst they found that the reactions took
adduct using a very sensitive immunoslot blot assay, but only a place at different rates according to the structure of the
single adduct can be analysed at a time and difficulties can occur diazoketone, single products were acquired in high yield.
3,4
because of cross-reactivity of the antibodies. A limited supply We found that the hydrolysis of ethyl diazoacetate proceeded
6
of polyclonal antibody exists for O CMdG, but attempts to pro- well under both conditions and with no substantial difference
duce a monoclonal antibody have been unsuccessful, and a dif- in the yields. The NMR for the unlabelled compound showed
ferent technique will therefore be needed in the future. There a singlet for the CH -OH protons at 4.1 ppm that was not pre-
2
has been considerable progress in the area of DNA adduct ana- sent in 4, confirming that the diazo group was converted to
lysis by liquid chromatography-mass spectrometry (LC-MS), and the alcohol with 100% incorporation of deuterium at the
good sensitivity can be achieved (e.g. Ref. 5). Hence, we pursued
the use of LC-MS as a more selective, and potentially sensitive,
technique to analyse DNA adducts in biological samples. The
6
synthesis of unlabelled O CMdG (1; Scheme 1) has been
a
6
School of Pharmacy and Biomolecular Sciences, Liverpool John Moores Univer-
sity, Liverpool L3 3AF, UK
reported by other researchers, with methyl glycolate used to
6
furnish the carboxymethyl group at the O position of 2′-
b
deoxyguanosine. However, we utilised ethyl glycolate because of
the availability of starting materials to insert the labelled atoms
Department of Chemistry and Analytical Sciences, Open University, Milton
Keynes MK7 6AA, UK
2
13
6
for [ H
focussed on the synthesis of [ H
; Scheme 2) and subsequently that of 2 and 3.
2 2
]- and [ C ]O CMdG (2 and 3). Hence, this research initially
*
Correspondence to: Sharon A. Moore, School of Pharmacy and Biomolecular
2
13
2 2
] - and [ C ]ethyl glycolate (4 and
Sciences, Liverpool John Moores University, Liverpool L3 3AF, UK.
E-mail: s.a.moore@ljmu.ac.uk
5
J. Label Compd. Radiopharm 2011, 54 855–858
Copyright # 2011 John Wiley & Sons, Ltd.

S. A. Moore and D. E. G. Shuker
6
2
6
13
6
2 2
Scheme 1. Synthesis of O CMdG (1), [ H ]O CMdG (2), and [ C ]O CMdG (3).
2
13
2 2
Scheme 2. Synthesis of [ H ]- and [ C ]ethyl glycolate (4 and 5).
a-carbon. Subsequently, 4 was used to give 2 (Scheme 1) in a observed. However, for 3, the limiting factor was the quantity
low overall yield (6.4%), which was comparable with 1. of 5 that was available because of the high cost of the
Deuterium–proton exchange was evident in 2 as the NMR material. Hence, yields for intermediates of Scheme 1 (steps
revealed protons at the CH
carboxymethyl group, which had not been present in 4. This was more difficult due to the high quantity of unreacted
was improved by performing the final step in D O (11%), but precursors. Nevertheless, sufficient pure material was obtained
2
position (17%) of the iv–vi) of 3 were expected to be much lower and the purification
2
it is not known whether the exchange occurred during the (0.52% yield) for use as an internal standard, and the problem of
synthesis or purification of 2. However, the NMR spectra of deuterium–proton exchange had been overcome.
intermediates suggest that some exchange was occurring at
each stage. Furthermore, the use of 2 in LC-MS revealed addi- Experimental
tional deuterium–proton exchange, resulting in peaks in the
General
mass spectrum corresponding to single and di-deuterated
6
compounds. Hence, a C-13-labelled O CMdG standard (3)
All chemicals were purchased from Sigma-Aldrich (Dorset, UK)
and used without further purification. Solvents were purchased
from Fisher (Loughborough, UK). Organic extracts were dried
over MgSO . NMR data were obtained on a Bruker Avance 300
was required for use as a stable LC-MS standard.
13
[
C2]Glycine was converted to the ethyl ester by reaction with
acyl chloride rather than an acid/ethanol reflux, which had given
4
lower yields in initial attempts (data not shown). The conversion
1
13
spectrometer at 300.1 MHz ( H) or 75.5 MHz ( C). Chemical shifts
were determined relative to the residual solvent peak and
reported as parts per million (ppm), and coupling constants were
reported in hertz (Hz). Reactions and purifications were moni-
tored by HPLC on a Waters Alliance system equipped with a
Waters 996 photodiode array detector with a narrow-bore
Hypersil BDS C18 column (3 mm, 100 Â 2.1 mm), flow rate
1
3
to [ C ]ethyl diazoacetate was performed following a previously
2
8
described procedure, and the conversion to 5 was done in the
same manner as for 4. The major difference in the NMR
between the labelled and unlabelled compounds was either
the lack of a peak or the very large coupling constants
observed between protons and the C-13 atoms in the
labelled compounds, which enabled unambiguous assign-
ment of the NMR spectra. The C-13-labelled compounds
had peaks for the a-H that showed splitting due to coupling
0.2 ml/min, MeOH/0.02 M ammonium acetate (pH 5.4) 80:20.
ESI-MS spectra were recorded on VG Quattro mass spectrometer
with narrow-bore Hypersil BDS C18 column (3 mm,
a
1
3
2
to the two C-13 atoms with the exception of [ C ]ethyl
1
00 Â 2.1 mm), flow rate 0.2 ml/min, and 0.1% formic acid/
diazoacetate. This compound had broader peaks than the
other compounds, possibly due to the diazo group charge
distribution, which did not allow observation of the splitting
because of coupling with the carbonyl C-13 as well as the
a-C. The synthesis of 3 was then performed in the same man- Glycine (0.5 g, 6.66 mmol) was suspended in ethanol (15 cm )
ner as for 2, and the double splitting pattern of the a-H was and acetyl chloride (1.5 cm , 21.12 mmol), and the mixture was
methanol gradient.
Glycine ethyl ester hydrochloride
3
3
www.jlcr.org
Copyright # 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 855–858

S. A. Moore and D. E. G. Shuker
ꢀ
heated at 85 C for 30 min under N
2
. The solvents were removed NMR (CDCl
3
) d
H
ppm 1.26 (3H, t, CH
OH, J = 145.56, 5.13), 4.15 (2H, dq, J = 7.14, J = 3.30);
ppm 14.13 (CH ), 53.40 (CH CH ), 60.53 (CH OH, d, J = 58.93),
J = 7.14); d ppm 14.03 (CH ), 41.05 (N-C), 64.16 (O-CH ), 169.00 173.32 (C = O, d, J = 58.93).
3
, J = 7.14), 4.08 (2H, dd,
CH
HH
CH
to give a white powder (1.01 g, >100% yield). NMR (D
.19 (3H, t, CH , J = 7.14), 3.81 (2H, s, N-CH ), 4.20 (2H, q, O-CH
2 H 2
O) d ppm CH
1
3
2
2
,
d
C
3
2
3
2
C
3
2
(C = O).
6
O -[1-Azonia-bicyclo[2.2.2]octane]-3′,5′-bis-O-
(isobutyryloxy)-2′-deoxyguanosine
1
3
[
C2]Glycine ethyl ester hydrochloride
1
3
[
2
C ]Glycine (0.5 g, 6.49 mmol) was reacted as above to give a Step (i): 2′-Deoxyguanosine (1.05 g, 3.74 mmol) was dried twice
3
2 H
white powder (0.93 g, >100% yield). NMR (D O) d ppm 1.17 with toluene (20 cm ) and dissolved in dry tetrohydrofuran
3H, t, CH
CH
3
(
3
, J = 7.14), 3.79 (2H, dd, N-CH
2
, J = 145.64, 6.32), (20 cm ) with 4-(dimethylamino)pyridine3 (DMAP) (0.024 g,
HH
CH
4
4
.19 (2H, dq, O-CH
1.02 (N-C, d, J = 62.40), 64.15 (O-CH
2
, J = 7.14, J 2.94); d ppm 14.02 (CH ), 0.2 mmol) and iso-butyric anhydride (2 cm , 12.13 mmol), and
C
3
2
), 169.00 (C = O, d, J = 62.40). the solution was stirred under N2 for 16 h. NaHCO3 (aq) was
added to give pH 8 and stirred for 2 h. The solvent was removed,
Ethyl diazoacetate
Glycine ethyl ester.HCl (0.99 g, 7.09 mmol) was dissolved in water
2
and the precipitate was collected, washed (H O), and dried to
give a white powder (1.30 g, 85% yield). Step (ii): The product
of (i) (1.00 g, 2.45 mmol) was dissolved in dry dichloromethane
3
3
(
1.8 cm ) and dichloromethane (4 cm ), and the solution was
3
(
30 cm ) with DMAP (0.03 g, 0.25 mmol) and methanesulfonyl
ꢀ
stirred and cooled to À10 C. Aqueous sodium nitrite (4.43 M,
3
chloride (1.00 g, 8.73 mmol), then triethylamine (3 cm ) was
added dropwise, and the solution was stirred for 3 h. Purification
was carried out on silica (chloroform/methanol, 97:3) to give a
yellow oil (1.86 g, 98% yield). Step (iii): The product of (ii) was dis-
solved in dry dichloromethane (30 cm ) with quinuclidine HCl
1.0 g, 6.77 mmol) and dry triethylamine (3 cm ), and the solution
3
1
(
1
.8 cm ) was added to the cooled solution, then 5% H SO
2 4
3
0.65 cm ) was added dropwise, and the solution was stirred for
ꢀ
0 min. The organic layer was poured into 5% NaCO at 0 C.
3
The aqueous layer was extracted with dichloromethane, and
the organic extract was combined with the organic/NaCO3
solution. The combined solution was shaken thoroughly and
the pH was checked to ensure that it was alkaline. The solvents
were removed to give 0.66 g of pale yellow oil (5.78 mmol, 82%
yield). NMR (CDCl ) dH ppm 1.21 (3H, t, CH , J = 7.14 Hz), 4.15
3
3
(
was stirred for 2 h. TLC showed a single blue spot on the baseline
methanol/dichloromethane, 10:90). The solvent was removed,
(
and the product was used in further reactions without
purification.
3
3
(
2H, q, CH , J = 7.14), 4.67 (1H, CH); d ppm 14.43 (CH ), 46.11
2
C
3
(CH), 60.84 (CH
2
), 166.86 (C = O).
6
O -Carboxymethyl-2′-deoxyguanosine.Na (1)
1
3
[
C2]Ethyl diazoacetate
Step (iv): The product from (iii) was dissolved in dry dichloro-
3
3
methane (20 cm ). Methyl glycolate (0.5 cm , 6.48 mmol) and
13
[
C
2
]Glycine ethyl ester.HCl (0.80 g, 5.65 mmol) was reacted as
3
1
,8-diazabicyclo[5.4.0]undec-7-ene (0.5 cm , 3.35 mmol) were
described above to give a pale yellow oil (0.36 g, 55% yield).
NMR (CDCl ) d ppm 1.21 (3H, t, CH , J = 7.14), 4.15 (2H, dq,
CH , J = 7.14, J = 3.30), 4.75 (1H, d, CH, J = 259.05);
added and stirred for 16 h. The solvents were removed, and
the resulting oil was purified on silica (methanol/dichloro-
methane, 5:95) to give a yellow oil (1.60 g). Step (v): The oil was
3
H
3
HH
CH
CH
2
d ppm 14.41 (CH ), 46.12 (CH, d, J = 96.84), 60.85 (CH ), 166.87
3
3
C
3
2
dissolved in methanol (20 cm ) with triethylamine (1 cm ), and
the solution was stirred for 72 h. The solvents were removeTdM
to give a yellow oil (1.67 g), which was purified on a Supelclean
Envi-18 6-ml column (Sigma-Aldrich, Dorset, UK) pre-
(
C = O, d, J = 96.84).
Ethyl glycolate
Ethyl diazoacetate (0.40 g, 3.51 mmol) was dissolved in dichloro-
methane (6 cm ) and 1% aqueous acetic acid (6 cm ) was added,
and the solution was stirred for 72 h in the dark. The aqueous
layer was extracted with dichloromethane, and the organic
extracts were combined with the organic layer; the mixture
was washed with water and then dried, and the solvents were
3
3
2
conditioned with methanol (2cm ) and H O (2cm ). The sample
3
3
3
2
was applied to the column in H O (0.5cm ), eluted, and washed
3
3
with H
2 2
O (2cm ) and 5% methanol//95% H O (2cm ). The pure
fractions were lyophilised to give a slightly yellow powder
(
(
0.075 g). Step (vi): The powder was dissolved in 0.1M NaOH
4.4 cm ) and stirred for 4 h, and the solution was lyophilised to
3
give 1 as a white powder (0.080 g, 6.2% overall yield). NMR
DMSO) d 2.21 (1H, m, 2′-H), 2.59 (1H, m, 2′-H), 3.53 (2H, m,
′-H), 3.83 (1H, m, 4′-H), 4.37 (1H, m, 3′-H), 4.58 (2H, s, CH ), 5.17
1H, br, 5′-OH), 5.44 (1H, br, 3′-OH), 6.19 (1H, t, 1′-H, J = 6.27),
.22 (2H, s, NH ), 8.06 (1H, s, 8-H); d ppm 39.49 (2′-C), 61.64
(5′-C), 64.86 (CH ), 70.69 (3′-C), 82.70 (1′-C), 87.57 (4′-C), 114.37
removed to give a pale yellow oil (0.24 g, 66% yield). NMR (CDCl
ppm 1.23 (3H, t, CH , J = 7.14 Hz), 4.08 (2H, s, CH OH), 4.20 (2H,
q, CH CH , J = 7.14); d ), 53.32 (CH CH ), 61.64
OH), 173.42 (C = O).
3
)
(
5
(
6
H
d
H
3
2
2
2
3
C
ppm 14.21 (CH
3
2
3
(CH
2
2
C
2
2
[
2
H ]Ethyl glycolate (4)
(
5-C), 137.23 (8-C), 153.45 (6-C), 159.60 (2-C), 160.74 (4-C), 170.40
3
À
Ethyl diazoacetate (10 cm , 96.41 mmol) was reacted as above, (COOH); M 324.
substituting D O for H O and d-acetic acid for acetic acid, to give
2
2
2
6
a pale yellow oil (6.05 g, 59% yield). NMR (CDCl ) dH ppm 1.28 [ H ]O -Carboxymethyl-2′-deoxyguanosine.Na (2)
3
2
(
3H, t, CH , J = 5.36), 4.24 (2H, q, CH , J = 5.36).
3
2
3
Steps (iv) to (vi) were performed as for 1, substituting 4 (1 cm ,
10.37 mmol) for methyl glycolate and D O for H O, to give 2 as
a white powder (83.4 mg, 6.4% overall yield). NMR (DMSO) d
1
3
2
2
[
C2]Ethyl glycolate (5)
H
13
[
2
C ]Ethyl diazoacetate (0.40 g, 3.49 mmol) was reacted as for ppm 2.27 (1H, m, 2′-H), 2.65 (1H, m, 2′-H), 3.60 (2H, m, 5′-H),
ethyl glycolate to give a pale yellow oil (0.24 g, 66% yield). 3.89 (1H, m, 4′-H), 4.43 (1H, m, 3′-H), 5.19 (1H, br, 5′-OH), 5.45
J. Label Compd. Radiopharm 2011, 54 855–858
Copyright # 2011 John Wiley & Sons, Ltd.
www.jlcr.org

S. A. Moore and D. E. G. Shuker
6
(
(
(
(
1H, br, 3′-OH), 6.27 (1H, t, 1′-H, J = 6.87), 6.34 (2H, s, NH
2
), 8.14 Thus, 5, the C-13-labelled analogue of O CMdG, was obtained
1H, s, 8-H); d ppm 35.42 (2′-C), 61.99 (5′-C), 68.41 (CD
C
2
), 70.85 in sufficient yield for future analyses of biological samples by
3′-C), 83.40 (1′-C), 87.82 (4′-C), 114.39 (5-C), 133.95 (8-C), 140.22 LC-MS.
À
6-C), 143.54 (2-C), 160.15 (4-C), 175.65 (COOH); M 326.
1
3
6
[
C ]O -Carboxymethyl-2′-deoxyguanosine.Na (3)
Acknowledgement
2
3
Steps (iv) to (vi) were performed as for 1, substituting 5 (0.2cm ,
.07mmol) for methyl glycolate, to give 3 as a white powder
3.5 mg, 0.52% overall yield). NMR (DMSO) d ppm 2.59 (2H, m,
′-H), 3.57 (2H, m, 5′-H), 3.83 (1H, m, 4′-H), 4.37 (1H, m, 3′-H), 4.53
, J = 3.83, 145.49), 5.07 (1H, br, 5′-OH), 5.30 (1H, br, References
Financial support was obtained from World Cancer Research
Fund (2004/24).
2
(
H
2
(1H, dd, CH
2
3
H); d
′-OH), 6.18 (1H, t, 1′-H, J = 6.86), 6.30 (2H, s, NH
2
), 8.04 (1H, s, 8-
[
1] World Cancer Research Fund, American Institute for Cancer Research,
Washington DC, AICR, 2007.
C
ppm 36.90 (2′-C), 61.76 (5′-C), 64.98 (CH , J = 54.19), 70.76
2
(
(
3′-C), 82.75 (1′-C), 87.55 (4′-C), 114.43 (5-C), 137.07 (8-C), 153.42 [2] M. H. Lewin, N. Bailey, T. Bandaletova, R. Bowman, A. J. Cross,
À
J. Pollock, D. E. G. Shuker, S. A. Bingham, Cancer Res. 2006, 66, 1859.
3] B. C. Cupid, Z. T. Zeng, R. Singh, D. E. G. Shuker, Chem. Res. Toxicol.
6-C), 159.57 (2-C), 160.95 (4-C), 169.39 (COOH, J = 54.19); M 326.
[
[
[
2
004, 17, 294.
Conclusion
4] S. A. Moore, O. Xeniou, Z. T. Zeng, E. Humphreys, S. Burr, E. Gottschalg,
S. A. Bingham, D. E. G. Shuker, Anal. Biochem. 2010, 403, 67.
5] R. Singh, P. B. Farmer, Carcinogenesis 2006, 27, 178.
Both of the labelled ethyl glycolates, 4 and 5, were obtained in
high yield. However, when 4 was used to synthesise 2, some [6] K. L. Harrison, N. Fairhurst, B. C. Challis, D. E. G. Shuker, Chem. Res.
deuterium–proton exchange was observed. This was exacer-
bated when LC-MS analysis was attempted. The synthesis of 5
did not pose any major problems, but this compound was a lim-
iting factor in the synthesis of 3. In terms of cost, 2 was the pre-
Toxicol. 1997, 10, 652.
[
7] A. J. Kresge, V. V. Popik, In: Electronic Conference on Heterocyclic
Chemistry ’96 (Eds.: H. S. Rzepa, J. Snyder, C. Leach), Royal Society of
Chemistry, Online http://www.ch.ic.ac.uk/ectoc/echet96/papers/037/,
1
997.
ferred option, but proved to be problematic for LC-MS analysis. [8] N. E. Searle, Org Synth Coll. 1963, 4, 424.
www.jlcr.org
Copyright # 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 855–858