Products of the reaction between a diazoate derivative of 2′-deoxycytidine and L-Lysine and its implication for DNA - Nucleoprotein cross-linking by NO or HNO2

Article abstract of DOI:10.1021/tx0001528

Recently, we have reported that a stable diazoate intermediate (dCyd-diazoate) is produced upon the reaction of dCyd with nitrous acid and nitric oxide [Suzuki, T., Nakamura, T., Yamada, M., Ide, H., Kanaori, K., Tajima, K., Morii, T., and Makino, K. (1999) Biochemistry 38, 7151-7158]. In this work, the reaction of dCyd-diazoate with L-Lys was investigated. When 0.4 mM dCyd-diazoate was incubated with 10 mM L-Lys in sodium phosphate buffer (pH 7.4) at 37 °C, two unknown products were formed in addition to dUrd. By spectrometric measurements, the products were identified as dCyd-Lys adducts with C4(dCyd)-N^α(Lys) and C4(dCyd)-N^∈(Lys) linkages (abbreviated as dCyd-αLys and dCyd-∈Lys, respectively). The yields at the reaction time of 72 h were 28.0% dCyd-αLys, 13.4% dCyd-∈Lys, and 11.1% dUrd with 33.9% unreacted dCyd-diazoate. When 0.4 mM dCyd-diazoate was incubated with 22 mg/mL poly(L-Lys) at pH 7.4 and 37 °C for 24 h, 82% of the free dCyd-diazoate disappeared, indicating adduct formation with the polymer. At pH 7.4 and 37 °C, dCyd-αLys and dCyd-∈Lys were fairly stable and gave rise to no product after incubation for 7 days. At pH 4.0 and 70 °C, both adducts disappeared with the same first-order rate constant of 1.7 x 10^-6 s^-1 (t_1/2 = 110 h), which was ～1/3 of that of dCyd. These results suggest that if dCyd-diazoate is formed in DNA in vivo, it may react with free L-Lys and the side chain of L-Lys in nucleoproteins, resulting in stable adducts and DNA-protein cross-links, respectively.

Full text of DOI:10.1021/tx0001528

Chem. Res. Toxicol. 2000, 13, 1223-1227
1223
P r od u cts of th e Rea ction betw een a Dia zoa te Der iva tive
of 2′-Deoxycytid in e a n d L-Lysin e a n d Its Im p lica tion for
DNA-Nu cleop r otein Cr oss-Lin k in g by NO or HNO2
Toshinori Suzuki,†,‡ Masaki Yamada,† Takanori Nakamura,§ Hiroshi Ide,|
Kenji Kanaori,§ Kunihiko Tajima,§ Takashi Morii,† and Keisuke Makino*,†
Institute of Advanced Energy, Kyoto University, Gokasho, Uji 611-0011, J apan, Department of
Polymer Science and Engineering, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku,
Kyoto 606-8585, J apan, and Department of Mathematical and Life Sciences, Graduate School of
Science, Hiroshima University, Kagamiyama, Higashi-Hiroshima 739-8526, J apan
Received J uly 17, 2000
Recently, we have reported that a stable diazoate intermediate (dCyd-diazoate) is produced
upon the reaction of dCyd with nitrous acid and nitric oxide [Suzuki, T., Nakamura, T., Yamada,
M., Ide, H., Kanaori, K., Tajima, K., Morii, T., and Makino, K. (1999) Biochemistry 38, 7151-
7158]. In this work, the reaction of dCyd-diazoate with L-Lys was investigated. When 0.4 mM
dCyd-diazoate was incubated with 10 mM L-Lys in sodium phosphate buffer (pH 7.4) at 37 °C,
two unknown products were formed in addition to dUrd. By spectrometric measurements, the
products were identified as dCyd-Lys adducts with C4(dCyd)-NR(Lys) and C4(dCyd)-Nꢀ(Lys)
linkages (abbreviated as dCyd-RLys and dCyd-ꢀLys, respectively). The yields at the reaction
time of 72 h were 28.0% dCyd-RLys, 13.4% dCyd-ꢀLys, and 11.1% dUrd with 33.9% unreacted
dCyd-diazoate. When 0.4 mM dCyd-diazoate was incubated with 22 mg/mL poly(L-Lys) at pH
7.4 and 37 °C for 24 h, 82% of the free dCyd-diazoate disappeared, indicating adduct formation
with the polymer. At pH 7.4 and 37 °C, dCyd-RLys and dCyd-ꢀLys were fairly stable and
gave rise to no product after incubation for 7 days. At pH 4.0 and 70 °C, both adducts
disappeared with the same first-order rate constant of 1.7 × 10-6 s-1 (t1/2 ) 110 h), which was
1
∼ /3 of that of dCyd. These results suggest that if dCyd-diazoate is formed in DNA in vivo, it
may react with free L-Lys and the side chain of L-Lys in nucleoproteins, resulting in stable
adducts and DNA-protein cross-links, respectively.
under the same conditions. The results indicate that
HNO2 induces a particular class of DNA lesions that are
not recognized by uracil-DNA glycosylase but are by
UvrABC nuclease. We have recently shown that 2′-
deoxyoxanosine (dOxo)1 which is produced from dGuo by
the reaction with HNO2 and NO in addition to 2′-
deoxyxanthosine (dXao), a deamination product, reacts
with Gly, resulting in a stable adduct (10, 11). This result
suggests that if dOxo formed in vivo in DNA it will react
with Gly and other free amino acids in the vicinity,
resulting in adducts which may be recognized by UvrABC
nuclease.
In tr od u ction
Nitrous acid (HNO2) and nitric oxide (NO) under
aerobic conditions react with the amino group of dCyd,
resulting in dUrd, a deamination compound, in vitro (1,
2). It has been thought that the formation of dUrd is one
of the major causes of mutations induced by HNO2 and
NO in vivo, since dUrd can adopt the same base pairing
scheme as dThd in DNA (3, 4). Because the spontaneous
hydrolytic deamination of dCyd also occurs at biologically
significant rates in cells (5), a specific repair enzyme,
uracil-DNA glycosylase, exists abundantly in cells and
excises Ura from DNA by cleaving the N-glycosidic bond
(6, 7). dUrd produced by HNO2 or NO should be also
excised by the same enzyme. However, a recent study
employing several mutant strains of Escherichia coli has
shown that a defect in uracil-DNA glycosylase has little
impact on cell killing and mutagenesis by HNO2 (8). On
the other hand, strains defective in Uvr nuclease, which
excises pyrimidine dimers and various bulky adducts (9),
exhibited pronounced sensitivity to killing and signifi-
cantly stimulated induction of rifampicin resistance
Recently, we have also isolated and characterized the
intermediate produced from dCyd by HNO2 and NO
treatments (12). By spectrometric measurements, the
product was identified as a diazoate derivative of dCyd,
1-(â-D-2′-deoxyribofuranosyl)-2-oxopyrimidine-4-diazo-
ate (dCyd-diazoate) (Figure 1). The time course of the
concentration change of dCyd-diazoate exhibited a profile
characteristic of a reaction intermediate, and the dCyd-
diazoate concentration was greater than that of dUrd at
the initial stage of the reaction. dCyd-diazoate was also
* To whom correspondence should be addressed. Phone: +81-774-
38-3517. Fax: +81-774-38-3524. E-mail: kmak@iae.kyoto-u.ac.jp.
† Kyoto University.
1
Abbreviations: dCyd-diazoate, diazoate derivative of dCyd, 1-(â-
D-2′-deoxyribofuranosyl)-2-oxopyrimidine-4-diazoate; dCyd-RLys, dCyd-
Lys adduct with a C4(dCyd)-NR(Lys) linkage; dCyd-ꢀLys, dCyd-Lys
adduct with a C4(dCyd)-Nꢀ(Lys) linkage; dXao, 2′-deoxyxanthosine;
dOxo, 2′-deoxyoxanosine; RP-HPLC, reversed phase HPLC; tR, reten-
tion time; APCI, atmospheric pressure chemical ionization.
‡ Present address: Unit of Endogenous Cancer Risk Factors,
International Agency for Research on Cancer, Lyon, France.
§ Kyoto Institute of Technology.
| Hiroshima University.
10.1021/tx0001528 CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/16/2000

1224 Chem. Res. Toxicol., Vol. 13, No. 12, 2000
Suzuki et al.
F igu r e 1. Structures of dCyd-diazoate [1-(â-D-2′-deoxyribo-
furanosyl)-2-oxopyrimidine-4-diazoate] and L-Lys.
F igu r e 2. RP-HPLC chromatogram of dCyd-diazoate incubated
with L-Lys. dCyd-diazoate (0.4 mM) was incubated with 10 mM
L-Lys in 100 mM sodium phosphate buffer (pH 7.4) at 37 °C for
72 h. The sample was separated with an ODS column (4.6 mm
× 150 mm). The eluent was 100 mM triethylammonium acetate
buffer (pH 7.0) containing CH3CN, and the flow rate was 1.0
mL/min. The CH3CN concentration was increased from 0 to 20%
for 20 min in a linear gradient mode. Insets are on-line-detected
UV spectra of 1 and 2.
generated from dCyd in neutral solution by NO treat-
ment under aerobic conditions. At pH 7.4 and 37 °C,
dCyd-diazoate was fairly stable (t1/2 ) 330 h). Uracil-DNA
glycosylase did not excise the diazoate residue from an
oligodeoxynucleotide containing this damage. These re-
sults show that the diazoate is generated as a stable
intermediate in the reactions of dCyd with HNO2 and
NO and that the major product is not dUrd but the
diazoate in the initial stage of the reactions. Thus, once
formed in vivo, the diazoate persists for a long time in
DNA and may directly exert major cytotoxic and/or
genotoxic effects. In addition, dCyd-diazoate is likely to
react with various intracellular nucleophiles under physi-
ological conditions since it is an intermediate of diazoti-
zation. The resulting adducts of dCyd in DNA may also
lead to mutagenic or lethal events of cells and be
recognized by UvrABC nuclease. Until now, however,
there has been no information about the reaction of dCyd-
diazoate with cellular molecules.
H-4′), 3.79 (ABX, 2H, H-5′,5′′), 2.98 (m, 2H, H-ꢀ), 2.38 (ddd, 1H,
H-2′′), 2.31 (ddd, 1H, H-2′), 1.88 (m, 1H, H-â or -â′), 1.78 (m,
1H, H-â or -â′), 1.69 (m, 2H, H-δ), 1.46 (m, 2H, H-γ). UV: λmax
) 273 nm (pH 7.0).
Sp ectr om etr ic Da ta of 2 (d Cyd -ELys). 1H NMR (500
MHz, DMSO-d6 at 27 °C): δ 7.81 (dd, J NHCH ) 5.2 and 5.2 Hz,
2
1H, ꢀNH), 7.72 (d, J 56 ) 7.5 Hz, 1H, H-6), 6.15 (dd, 1H, H-1′),
5.76 (d, 1H, H-5), 5.35 (br, 1H, 3′- or 5′-OH), 5.04 (br, 1H, 3′- or
5′-OH), 4.19 (ddd, 1H, H-3′), 3.74 (ddd, 1H, H-4′), 3.54 (ABX,
2H, H-5′,5′′), 3.21 (m, 2H, H-ꢀ), 3.14 (m, 1H, H-R), 2.09 (ddd,
1H, H-2′′), 1.92 (ddd, 1H, H-2′), 1.72 (m, 1H, H-â or -â′), 1.69
(m, 1H, H-â or -â′), 1.46 (m, 2H, H-δ), 1.37 (m, 2H, H-γ). UV:
λmax ) 273 nm (pH 7.0).
We report herein identification and characterization
of the reaction products of dCyd-diazoate with L-Lys at
physiological pH and temperature.
Qu a n tita tive P r oced u r es. The concentrations of products
were evaluated from integrated peak areas of HPLC chromato-
grams and molar extinction coefficients at 260 nm. The values
of ꢀ260 for dCyd-RLys and dCyd-ꢀLys were estimated to be 1.26
× 104 and 1.24 × 104 M-1 cm-1, respectively, from integration
of the H-5 and H-6 proton signals and RP-HPLC peak area
detected at 260 nm relative to those of dCyd. The ꢀ260 values of
7.3 × 103, 7.4 × 103, and 1.01 × 104 M-1 cm-1 were used for
dCyd-diazoate, dCyd, and dUrd, respectively (12, 13). In the
quantitative analysis of the reaction products, the initial RP-
HPLC peak area was used as a standard.
Ma ter ia ls a n d Meth od s
Ma ter ia ls. The dCyd-diazoate was prepared as described
previously (12). dCyd, dUrd, NR-BOC-L-Lys, and poly(L-Lys)
hydrobromide (average molecular weight of 35 000) were ob-
tained from Sigma (St. Louis, MO). All other reagent grade
chemicals were purchased from Wako Pure Chemicals (Osaka,
J apan) or Nacalai Tesque (Osaka, J apan) and used without
further purification. Water was purified with a Millipore Milli-
QII deionizer.
Resu lts
HP LC Con d ition s. The HPLC system consisted of Hitachi
L-6000 and L-6200 pumping systems. On-line UV spectra were
obtained with a Hitachi L-3000 UV/vis photodiode array detec-
tor. For analysis and separation of nucleosides by RP-HPLC, a
Cosmosil 5C18-MS octadecylsilane column (4.6 mm × 150 mm,
particle size of 5 µm, Nacalai Tesque) was used. The eluent was
100 mM triethylammonium acetate buffer (pH 7.0) containing
acetonitrile. The acetonitrile concentration was increased from
0 to 20% for 20 min with a linear gradient mode. RP-HPLC
analyses were performed at room temperature and a flow rate
of 1.0 mL/min.
Sp ectr om etr ic Mea su r em en ts. NMR spectra were meas-
ured using a Brucker ARX-500 spectrometer in H2O/D2O (9:1,
v/v) at 10 °C for 1 or DMSO-d6 at 27 °C for 2. The chemical
shifts (parts per million) were referenced to TSP-d4 for H2O/
D2O or TMS for DMSO-d6 as the internal standard. The signal
assignments were determined by DQF-COSY or TOCSY (40 ms).
Sp ectr om etr ic Da ta of 1 (d Cyd -rLys). 1H NMR (500
MHz, 9:1 H2O/D2O at 10 °C): δ 7.76 (d, J 56 ) 7.6 Hz, 1H, H-6),
7.20 (d, J NHCH ) 7.4 Hz, 1H, NH), 6.28 (dd, 1H, H-1′), 6.06 (d,
1H, H-5), 4.42 (m, 1H, H-3′), 4.35 (m, 1H, H-R), 4.04 (ddd, 1H,
Detection a n d Id en tifica tion of Rea ction P r od -
u cts of d Cyd -Dia zoa te w ith L-Lys. When 0.4 mM
dCyd-diazoate was incubated with 10 mM L-Lys in 100
mM sodium phosphate buffer (pH 7.4) for 72 h at 37 °C,
two unknown products were eluted as major peaks
designated as 1 (tR ) 7.1 min) and 2 (tR ) 9.0 min) in
addition to dUrd, a deamination product (tR ) 8.2 min),
and the starting dCyd-diazoate (tR ) 10.7 min) in the RP-
HPLC chromatogram (Figure 2). These products exhib-
ited similar UV spectra with λmax values of 273 nm
(Figure 2 insets). The products were isolated by prepara-
tive RP-HPLC and subjected to structural assignment.
1
The H NMR spectrum of 2 was recorded in DMSO-d6,
and on the other hand, that of 1 was recorded in H2O/
D2O (9:1) since 1 precipitated in DMSO-d6. Both of the
1H NMR spectra of 1 and 2 exhibited two aromatic
protons, H-5 and H-6, and a set of signals of the intact
2′-deoxyribose moiety originated from dCyd in addition
to four sets of methylene protons (H-â, H-γ, H-δ, and H-ꢀ)

Formation of Adducts between dCyd-Diazoate and L-Lys
Chem. Res. Toxicol., Vol. 13, No. 12, 2000 1225
F igu r e 4. Time course of concentration changes obtained for
dCyd-RLys (0), dCyd-ꢀLys (O), and dUrd (2) together with
that for dCyd-diazoate (b). dCyd-diazoate (0.4 mM) was incu-
bated with 10 mM L-Lys in 100 mM sodium phosphate buffer
(pH 7.4) at 37 °C. The concentration was determined by RP-
HPLC analysis.
F igu r e 3. Structures of 1 (dCyd-RLys) and 2 (dCyd-ꢀLys).
and one methine proton (H-R) from L-Lys. Additionally,
a signal of imino proton was observed in both the spectra
of 1 and 2. The imino proton signal of 1 was a doublet (J
) 7.4 Hz) and correlated with H-R in the DQF-COSY
spectrum, whereas that of 2 was a doublet of a doublet
(J ) 5.2 and 5.2 Hz) and correlated with H-ꢀ in the
TOCSY spectrum. Moreover, when NR-BOC-L-Lys whose
R-amino group is protected by a butoxycarbonyl group
was incubated with dCyd-diazoate under the same condi-
tions, only two products appeared in the RP-HPLC
chromatogram, one of which was dUrd. The HPLC
fraction containing the other product was collected and
further analyzed. The treatment of the collected product
with 50% trifluoroacetic acid at 37 °C for 15 min afforded
a single product peak of 2 in RP-HPLC analysis (data
not shown). Combining these data, we have concluded
that 1 and 2 are products of dCyd with C4 substitution
ate R- and ꢀ-amino groups of L-Lys, respectively (Figure
3). Compounds 1 and 2 are denoted as dCyd-RLys and
dCyd-ꢀLys, respectively.
Rea ction Kin etics for Ad d u ct F or m a tion . Figure
4 shows the time courses of concentration changes in
dCyd-RLys, dCyd-ꢀLys, and dUrd along with conversion
of dCyd-diazoate when 0.4 mM dCyd-diazoate was incu-
bated with 10 mM L-Lys in 100 mM sodium phosphate
buffer (pH 7.4) at 37 °C. The concentrations of the
products were quantified by the combination of RP-HPLC
peak areas and molar extinction coefficients. The values
were calibrated using the initial dCyd-diazoate concen-
tration as a standard. Throughout the reaction, the ratio
between dCyd-RLys and dCyd-ꢀLys was ca. 2:1, and the
yield of dUrd was comparable to that of dCyd-ꢀLys. The
F igu r e 5. Exponential plots of the concentration changes
obtained for dCyd-RLys (0), dCyd-ꢀLys (O), dCyd (9), and
dUrd (2) as a function of incubation time. The solution of dCyd-
RLys, dCyd-ꢀLys, dCyd, or dUrd (0.1 mM each) was incubated
in 100 mM sodium acetate buffer (pH 4.0) at 70 °C. The
concentration was determined by RP-HPLC analysis.
yields at the incubation time of 72 h were 28.0% dCyd-
RLys, 13.4% dCyd-ꢀLys, and 11.1% dUrd with 33.9%
unreacted dCyd-diazoate.
Sta bility of d Cyd -rLys a n d d Cyd -ELys. To evalu-
ate the stability of the products, the isolated dCyd-RLys,
dCyd-ꢀLys, dCyd, and dUrd (0.1 mM each) were incu-
bated in neutral solution (pH 7.4, 37 °C, 100 mM sodium
phosphate buffer) and mild acidic solution (pH 4.0, 70
°C, 100 mM sodium acetate buffer). The reactions were
monitored by RP-HPLC. At pH 7.4 and 37 °C, dCyd-
RLys and dCyd-ꢀLys gave rise to no product after
incubation for 7 days (data not shown). At pH 4.0 and
70 °C, dCyd-RLys and dCyd-ꢀLys disappeared with
incubation time. The plot of concentrations of the adducts
against the incubation time followed first-order kinetics
(Figure 5). The rate constant was determined to be 1.7
× 10-6 s-1 (t1/2 ) 110 h) for both adducts. On the other

1226 Chem. Res. Toxicol., Vol. 13, No. 12, 2000
Suzuki et al.
linked to the R- or ꢀ-amino group of L-Lys. An additional
product of this reaction was dUrd formed by hydrolysis
of the diazoate group. The reaction leading to the dCyd-
Lys adducts is likely to be initiated by nucleophilic attack
of the amino groups of L-Lys on the cation or the
precursor diazo species of dCyd.
Under the conditions described here, incubation of 10
mM L-Lys with dCyd-diazoate resulted in the dCyd-RLys
adduct, and its yield was 2-fold greater than that of dUrd.
Since the intracellular concentration of L-Lys is 1.4 mM
in healthy human muscle (16), it is possible that a
biologically significant amount of the adduct can be
produced by the reaction of dCyd-diazoate in DNA with
intracellular L-Lys. In addition to L-Lys, cells contain
various free amino acids. The total concentration of the
naturally occurring 20 free amino acids is 34 mM in
human muscle cells (16). The average of the pKa values
for the R-amino groups of the 20 amino acids is 9.5 which
is essentially comparable to that of the R-amino group
of L-Lys (pKa ) 9.1) (17). In the first approximation, the
rate of the reaction between the amino acids and dCyd-
diazoate should primarily depend on the pKa value of the
R-amino group since the reaction is initiated by nucleo-
philic attack of the R-amino group. Thus, the rate of
adduct formation with the naturally occurring amino
acids will be comparable to that for L-Lys (R-amino
group). Accordingly, dCyd-diazoate in a free form or in
DNA may preferentially react with amino acids yield-
ing the adducts rather than with water yielding dUrd in
cells.
F igu r e 6. Concentration changes obtained for free dCyd-
diazoate (b) and dUrd (2) in the presence of poly(L-Lys). dCyd-
diazoate (0.4 mM) was incubated with 22 mg/mL poly(L-Lys) in
100 mM sodium phosphate buffer (pH 7.4) at 37 °C for the
indicated time. The sample was filtered with
a Millipore
Cosmonice Filter W to remove poly(L-Lys) and dCyd-poly(L-
Lys) adducts. The concentration of dCyd-diazoate and dUrd in
the filtrate was determined by RP-HPLC analysis.
hand, dCyd disappeared with a first-order rate constant
of 5.0 × 10-6 s-1 (t1/2 ) 39 h), and dUrd was stable.
Rea ction of d Cyd -Dia zoa te w ith P oly(L-Lys). To
elucidate whether dCyd-diazoate reacts with the amino
group of the L-Lys side chain in protein, dCyd-diazoate
was incubated with poly(L-Lys). dCyd-diazoate (0.4 mM)
was incubated with 22 mg/mL poly(L-Lys) in 100 mM
sodium phosphate buffer (pH 7.4) at 37 °C for e24 h.
Aliquots of the solution were filtered by a Millipore
Cosmonice Filter W (pore size, 0.45 µm) to remove the
polymer and analyzed by RP-HPLC. Before chromatog-
raphy, it was confirmed that the filter did not retain
dCyd-diazoate, dCyd, and dUrd. The concentration of
dCyd-diazoate in the filtrate decreased with an increase
in incubation time, and a small amount of dUrd appeared
(Figure 6). At the incubation time of 24 h, 82% of the
dCyd-diazoate disappeared but the yield of dUrd in the
filtrate was 4%. When dCyd and dUrd were incubated
with poly(L-Lys) for 24 h under the same conditions, the
concentration of the compounds did not change before
and after filtration.
In this study, the reaction with the ꢀ-amino group of
L-Lys resulted in another adduct with a yield similar to
that of dUrd. The pKa value of the ꢀ-amino group of L-Lys
is 10.4 (17) and about 1 unit higher than that of the
1
R-amino group of L-Lys. The yield of dCyd-ꢀLys was ∼ /2
of that of dCyd-RLys and was higher than that simply
expected from the difference in pKa values of the R- and
ꢀ-amino groups. The R-amino group of L-Lys has the
bulky carboxyl group in the vicinity, whereas the ꢀ-amino
group is free of steric hindrance. The difference in
accessibility of the amino group to the electrophilic
carbon, i.e., C4 of the cation (or precursor diazo species)
derived from dCyd-diazoate, may affect the reactivity.
After incubation of dCyd-diazoate with poly(L-Lys) and
subsequent removal of poly(L-Lys) by filtration, most of
dCyd-diazoate (82%) disappeared from the filtrate, al-
though the concentration of dCyd and dUrd treated in
the same manner did not change. This result suggests
that dCyd-diazoate reacted with the ꢀ-amino groups of
the side chain in poly(L-Lys) and was trapped by the
polymer. Therefore, it is possible that dCyd-diazoate
formed in DNA may react with the amino group of the
L-Lys side chain in nucleoproteins, resulting in DNA-
protein cross-links, although the rate of cross-link forma-
tion with nucleoproteins would be lower than that with
poly(L-Lys) because of the relatively low abundance of
L-Lys residues on the surface of nucleoproteins and the
possible pKa variation of the ꢀ-amino group of L-Lys.
Among nucleoproteins, histone should be one of the
proteins most likely forming DNA-protein cross-links
since it is in direct contact with DNA and has L-Lys rich
domains. Consistent with this notion, it has been re-
ported that HNO2 treatment of nucleohistone gives rise
to DNA-histone cross-links (18). There are two possible
mechanisms for the cross-link formation caused by nitro-
sation of DNA. The first one involves formation of a
Discu ssion
-
Protonated NO2 (i.e., HNO2) and NO under aerobic
conditions generate several reactive nitrogen species,
including N2O3 that reacts with amines resulting in
nitrosamines (14, 15). The nitrosated primary amines are
spontaneously converted to several intermediates such
as diazohydroxide (R-NdN-OH), diazoate (R-NdN-
O-), diazonium (R-NtN+), and alkyl cation (R+) in
neutral solutions (14). The alkyl cation or its precursor
(diazo species) results in the corresponding hydroxylate
compound as a result of nucleophilic attack of a water
molecule. In the nitrosation of dCyd, the most stable
intermediate is the diazoate derivative, i.e., dCyd-diazo-
ate (12). dCyd-diazoate is converted exclusively to dUrd
by attack of a water molecule in neutral solution. In this
study, we found that the incubation of dCyd-diazoate
with L-Lys resulted in adducts, in which C4 of dCyd was

Formation of Adducts between dCyd-Diazoate and L-Lys
Chem. Res. Toxicol., Vol. 13, No. 12, 2000 1227
covalent bond between the L-Lys side chain on histone
and the reactive aldehyde group in the DNA abasic site.
Abasic sites are known to be generated by spontaneous
depurination of 2′-deoxyxanthosine (dXao), a deamination
product of dGuo caused by nitrosation, since the N-
glycosidic bond of dXao is very labile (19, 20). The second
mechanism involves cross-links between the L-Lys side
chain and nitrosation products themselves. 2′-Deoxy-
oxanosine (dOxo), a byproduct of dGuo nitrosation (10),
reacts with the R-amino group of Gly, resulting in a stable
adduct (11). Therefore, it is possible that dOxo formed
in DNA may react with the ꢀ-amino group of L-Lys on
histone, resulting in DNA-histone cross-links, although
there is currently no direct evidence for the reaction. This
study has suggested a new pathway in the second
mechanism where dCyd-diazoate plays an essential role.
During the course of the reaction, the nitrosated dCyd
exists as the diazoate intermediate for a certain period
of time before being converted to dUrd. When the fact
that isolated dCyd-diazoate is fairly stable in water (t1/2
) 330 h at pH 7.4 and 37 °C) (12) is considered, the
lifetime of the diazoate in cellular DNA would also be
rather long. Combining this and the present data for the
reactivity of the diazoate with the ꢀ-amino group of L-Lys,
we found that covalent bond formation between dCyd-
diazoate and an L-Lys residue would be a feasible
pathway causing DNA-histone cross-links.
Program of J apan Society for the Promotion of Science
(J SPS-RFTF97I00301).
Refer en ces
(1) Schuster, H., and Schramm, G. (1958) Bestimmung der biologish
wirksamen einheit in der ribosenucleinsa¨ure des tabakmosaik-
virus auf chemischem wege. Z. Naturforsch. 13b, 697-704.
(2) Wink, D. A., Kasprzak, K. S., Maragos, C. M., Elespuru, R. K.,
Misra, M., Dunams, T. M., Cebula, T. A., Koch, W. H., Andrews,
A. W., Allen, J . S., and Keefer, L. K. (1991) DNA deaminating
ability and genotoxicity of nitric oxide and its progenitors. Science
254, 1001-1003.
(3) Takahashi, I., and Marmur, J . (1963) Replacement of thymidylic
acid by deoxyuridylic acid in the deoxyribonucleic acid of a
transducing phage for Bacillus subtilis. Nature 197, 794-795.
(4) Warner, H. R., and Duncan, B. K. (1978) In vivo synthesis and
properties of uracil-containing DNA. Nature 272, 32-43.
(5) Frederico, L. A., Kunkel, T. A., and Shaw, B. R. (1990) A sensitive
genetic assay for the detection of cytosine deamination: Deter-
mination of rate constants and activation energy. Biochemistry
29, 2532-2537.
(6) Lindahl, T. (1974) An N-glycosidase from Escherichia coli that
releases free uracil from DNA containing deaminated cytosine
residues. Proc. Natl. Acad. Sci. U.S.A. 71, 3649-3653.
(7) Krokan, H., and Wittwer, C. U. (1981) Uracil-DNA glycosylase
from HeLa-cells: general properties, substrate specificity and
effects of uracil analogs. Nucleic Acids Res. 9, 2599-2613.
(8) Hartman, Z., Henrikson, E. N., Hartman, P. E., and Cebula, T.
A. (1994) Molecular models that may account for nitrous acid
mutagenesis in organisms containing double-stranded DNA.
Environ. Mol. Mutagen. 24, 168-175.
(9) Van Houten, B. (1990) Nucleotide excision repair in Escherichia
coli. Microbiol. Rev. 54, 18-51.
At physiological pH and temperature (pH 7.4 and 37
°C, respectively), dCyd-RLys and dCyd-ꢀLys were fairly
stable and no changes in their concentrations were
observed after incubation for 7 days. Thus, if the adducts
and DNA-protein cross-links are once formed in vivo,
they may persist for a long period, eventually exerting
mutagenic and lethal effects. A recent study employing
several E. coli mutants has shown that defects in
UvrABC nuclease lead to a pronounced sensitivity to cell
killing and an increased frequency of mutation via HNO2
treatment (8). The UvrABC nuclease acts on a broad
spectrum of types of DNA damage, including pyrimidine
dimers, abasic sites, various adducts, and even DNA
interstrand cross-links (9). In view of these results and
the previous study (8), UvrABC nuclease may repair the
amino acid adducts of DNA or DNA-protein cross-links
produced by the reaction between dCyd-diazoate in DNA
and intracellular free amino acids or nucleoproteins.
In conclusion, it has been shown that dCyd-diazoate
reacts with L-Lys, resulting in two stable adducts at
physiological pH and temperature. This result also
implicates a novel mechanism of the formation of DNA
adducts and DNA-protein cross-links generated by
HNO2 and NO.
(10) Suzuki, T., Yamaoka, R., Nishi, M., Ide, H., and Makino, K. (1996)
Isolation and characterization of a novel product, 2′-deoxyox-
anosine, from 2′-deoxyguanosine, oligodeoxynucleotide, and calf
thymus DNA treated by nitrous acid and nitric oxide. J . Am.
Chem. Soc. 118, 2515-2516.
(11) Suzuki, T., Yamada, M., Ide, H., Kanaori, K., Tajima, K., Morii,
T., and Makino, K. (2000) Identification and characterization of
a reaction product of 2′-deoxyoxanosine with glycine. Chem. Res.
Toxicol. 13, 227-230.
(12) Suzuki, T., Nakamura, T., Yamada, M., Ide, H., Kanaori, K.,
Tajima, K., Morii, T., and Makino, K. (1999) Isolation and
characterization of diazoate intermediate upon nitrous acid and
nitric oxide treatment of 2′-deoxycytidine. Biochemistry 38, 7151-
7158.
(13) Sambrook, J ., Fritsch, E. F., and Maniatis, T. (1989) Molecular
Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory
Press, Plainview, NY.
(14) Zollinger, H. (1994) Diazo Chemistry I: Aromatic and Hetero-
aromatic Compounds, pp 39-64, VCH, Weinheim, Germany.
(15) Williams, D. L. H. (1988) Nitrosation, pp 10-16, Cambridge
University Press, Cambridge, U.K.
(16) Roth, E., Mu¨hlbacher, F., Karner, J ., Hamilton, G., and Funovics,
J . (1987) Free amino acid levels in muscle and liver of a patient
with glucagonoma syndrome. Metabolism 36, 7-13.
(17) Brown, T. A. (1991) Molecular Biology Labfax, p 29, BIOS
Scientific Publishers, Oxford, U.K.
(18) Potti, N. D., and Bello, J . (1971) Cross-linking of nucleohistone
by nitrous acid. Mutat. Res. 12, 113-119.
(19) Burnotte, J ., and Verly, W. G. (1971) A kinetic approach to the
mechanism of deoxyribonucleic acid cross-linking by HNO2. J .
Biol. Chem. 246, 5914-5918.
Ack n ow led gm en t. This work was partly supported
by Grants-in-Aid for Scientific Research from the Min-
istry of Education, Science and Culture, J apan [to K.M.
(11101001, 10151219, and 10878092), K.K. (09780545),
and H.I.], and by a Grant from “Research for the Future”
(20) Suzuki, T., Matsumura, Y., Ide, H., Kanaori, K., Tajima, K., and
Makino, K. (1997) Deglycosylation susceptibility and base-pairing
stability of 2′-deoxyoxanosine in oligodeoxynucleotide. Biochem-
istry 36, 8013-8019.
TX0001528