Base sequence selectivity in the alkylation of DNA by 1,3-dialkyl-3- acyltriazenes

Article abstract of DOI:10.1021/tx9500742

The base sequence selectivity of DNA alkylation for a series of structurally related 1,3-dialkyl-3-acyltriazenes was examined with calf thymus DNA or polymers containing the sequences GGG, CGC, TGT, and AGA. The reaction products at the N7 and the O⁶ positions of guanine were identified, quantitated, and then correlated with the decomposition rates of the triazenes, 1-(2-chloroethyl)-3-methyl-3-carbethoxy- (CMC), 1-(2-chloroethyl)- 3-methyl-3-acetyl- (CMA), 1-(2-hydroxyethyl)-3-methyl-3-carbethoxy- (HMC), 1- (2-hydroxyethyl)-3-methyl-3-acetyl-(HMA), and 1,3-dimethyl-3-acetyl- (DMA). The results of these studies revealed that DNA sequences with runs of purines were more reactive toward alkylation by all of the triazenes tested, irrespective of whether the alkylation was measured by N7, O⁶, or total guanine adducts. Within this generalization, the (hydroxyethyl)triazenes showed a preference for the AGA sequence, while the (chloroethyl)triazenes favored the GGG sequence. The structure of the 3-acyl group of the triazene also played a role in the extent of alkylation of a particular sequence of DNA. Both the (chloroethyl)- and the (hydroxyethyl)triazenes produced higher alkylation product yields for the 3-carbethoxytriazenes as compared with the 3-acetyl derivatives for most of the sequences examined. These overall patterns correlated well with the order of decomposition of the triazenes at 37 °C: HMC > DMA > HMA > CMC > CMA. This study has demonstrated how varying the structure of 1,3-dialkyl-3-acyltriazenes can modulate DNA alkylation, a finding which may be important in the design of new triazene antitumor agents.

Full text of DOI:10.1021/tx9500742

Chem. Res. Toxicol. 1996, 9, 341-348
341
Ba se Sequ en ce Selectivity in th e Alk yla tion of DNA by
1,3-Dia lk yl-3-a cyltr ia zen es
Marilyn B. Kroeger Smith,*,† Lisa A. Taneyhill,†,‡ Christopher J . Michejda,† and
Richard H. Smith, J r.†,‡
Molecular Aspects of Drug Design Section, Macromolecular Structure Laboratory,
ABL-Basic Research Program, NCI-Frederick Cancer Research and Development Center,
Frederick, Maryland 21702, and Department of Chemistry, Western Maryland College,
Westminster, Maryland 21157
Received April 28, 1995X
The base sequence selectivity of DNA alkylation for a series of structurally related 1,3-dialkyl-
3-acyltriazenes was examined with calf thymus DNA or polymers containing the sequences
GGG, CGC, TGT, and AGA. The reaction products at the N7 and the O6 positions of guanine
were identified, quantitated, and then correlated with the decomposition rates of the triazenes,
1-(2-chloroethyl)-3-methyl-3-carbethoxy- (CMC), 1-(2-chloroethyl)-3-methyl-3-acetyl- (CMA),
1-(2-hydroxyethyl)-3-methyl-3-carbethoxy- (HMC), 1-(2-hydroxyethyl)-3-methyl-3-acetyl- (HMA),
and 1,3-dimethyl-3-acetyl- (DMA). The results of these studies revealed that DNA sequences
with runs of purines were more reactive toward alkylation by all of the triazenes tested,
irrespective of whether the alkylation was measured by N7, O6, or total guanine adducts.
Within this generalization, the (hydroxyethyl)triazenes showed a preference for the AGA
sequence, while the (chloroethyl)triazenes favored the GGG sequence. The structure of the
3-acyl group of the triazene also played a role in the extent of alkylation of a particular sequence
of DNA. Both the (chloroethyl)- and the (hydroxyethyl)triazenes produced higher alkylation
product yields for the 3-carbethoxytriazenes as compared with the 3-acetyl derivatives for most
of the sequences examined. These overall patterns correlated well with the order of
decomposition of the triazenes at 37 °C: HMC > DMA > HMA > CMC > CMA. This study
has demonstrated how varying the structure of 1,3-dialkyl-3-acyltriazenes can modulate DNA
alkylation, a finding which may be important in the design of new triazene antitumor agents.
for chemotherapeutic activity. Such cross-links are
thought to occur by initial attack on the O6 position of
guanine, followed by intramolecular rearrangement in-
volving the N1 position of guanine. This then allows a
-CH2CH2- bridge to form between the N1 of guanine and
the N3 position of a cytosine on the contralateral strand
(7). It is possible that variations in the chemotherapeutic
activity observed with a series of compounds may be
linked to differences in this cross-linking ability.
In tr od u ction
Several 1-(2-chloroethyl)-3-methyl-3-acyltriazenes have
been shown to possess chemotherapeutic activity against
a variety of transplanted tumors in mice (1). The
postulated mechanism of action of these compounds at
acidic or neutral pH (Figure 1) involves N2-N3 heteroly-
sis, leading to production of an alkanediazonium ion. This
intermediate can, in turn, either chloroethylate or hy-
droxyethylate DNA (2), depending on the structure of the
acyltriazene. At basic pH, or in the presence of a
hydrolytic enzyme such as esterase, deacylation becomes
the predominant pathway. Acyl group removal produces
two tautomeric dialkyltriazenes, which can then decom-
pose to give either a chloroethylating or hydroxyethylat-
ing species, or a methylating agent (2). Depending on
the triazene structure, varying ratios of methylation to
chloroethylation or hydroxyethylation were observed in
the reaction of the triazene with calf thymus DNA in the
presence of esterase (2).
In an effort to understand the mechanism of chemo-
therapeutic action of acyltriazenes, we examined the
reaction of several compounds, including 1-(2-chloroeth-
yl)-3-methyl-3-carbethoxytriazene (CMC)1 and 1-(2-chlo-
roethyl)-3-acetyl-3-methyltriazene (CMA), as well as their
2-hydroxyethyl analogs HMC and HMA and 1,3-di-
methyl-3-acetyltriazene (DMA), with calf thymus DNA
and with synthetic polymers of defined sequence. Ex-
periments were undertaken to investigate whether dif-
ferences in triazene structure would alter the observed
initial DNA alkylation patterns. This type of information
would be important in the design of chloroethylating
agents that could modify a particular position on the
various DNA bases, select for a particular sequence in
the DNA, or lead to interstrand cross-links.
While the (hydroxyethyl)guanine adducts are thought
to have little importance for antitumor activity, they may
contribute to the carcinogenic potential of the title
triazenes, as has been postulated for the 1-(2-chloroeth-
yl)-1-nitrosoureas (3, 4). Chloroethylation of DNA, on the
other hand, probably leads to the production of lethal
interstrand cross-links (5, 6), which may be responsible
1 Abbreviations: CMC, 1-(2-chloroethyl)-3-methyl-3-carbethoxytria-
zene; CMA, 1-(2-chloroethyl)-3-methyl-3-acetyltriazene; HMC, 1-(2-
hydroxyethyl)-3-methyl-3-carbethoxytriazene; HMA, 1-(2-hydroxyethyl)-
3-methyl-3-acetyltriazene; DMA, 1,3-dimethyl-3-acetyltriazene; HPLC,
high-pressure liquid chromatography; 7-MeG, 7-methylguanine; 7-HO-
EtG, N7-(hydroxyethyl)guanine; 7-ClEtG, N7-(chloroethyl)guanine; O6-
MeG, O6-methylguanine; O6-HOEtG, O6-(hydroxyethyl)guanine.
† ABL-Basic Research Program, NCI-FCRDC.
‡ Western Maryland College.
X Abstract published in Advance ACS Abstracts, December 15, 1995.
0893-228x/96/2709-0341$12.00/0 © 1996 American Chemical Society
+
+

342 Chem. Res. Toxicol., Vol. 9, No. 1, 1996
Kroeger Smith et al.
F igu r e 1. Decomposition pathways of 1-(2-chloroethyl)-3-acyl-3-methyltriazenes.
Ma ter ia ls a n d Meth od s
Ca u tion ! Alkyltriazenes are potent biological alkylating
agents, mutagens, and carcinogens. All experiments should be
carried out in filter-equipped hoods, and proper clothing,
especially gloves, should be worn in order to minimize exposure.
P olyd eoxyr ibon u cleotid es a n d Oth er Ma ter ia ls. Calf
thymus DNA was purchased from Sigma Chemical Co. (St.
Louis, MO). The synthetic DNA homopolymers poly(dG)‚poly-
(dC) containing the triplet sequence GGG, the alternating
polymers poly(dG-dC)‚poly(dG-dC) containing the sequence
CGC, poly(dA-dC)‚poly(dG-dT) containing the sequence TGT,
and poly(dA-dG)‚poly(dC-dT) containing the sequence AGA were
purchased from LBK/Pharmacia (Piscataway, NJ ). The acyl-
triazenes HMC, HMA, CMC, CMA, and DMA were prepared,
purified, and characterized according to the procedure of Smith
et al. (8, 9) and were >99% pure (NMR). The triazenes were
dissolved in Burdick and J ackson dimethyl sulfoxide (VWR
Scientific, West Chester, PA) immediately prior to use. Stan-
dard alkylated bases were either purchased (7-methylguanine,
Sigma Chemical Co.) or synthesized in our laboratory (10),
except for 7-(chloroethyl)guanine, which was a generous gift of
Dr. David Ludlum, University of Massachusetts Medical School,
Worcester, MA.
NMR Decom p osition a n d P r od u ct Stu d ies. The extent
of decomposition of the various acyltriazenes (Figure 2) after
48 h at 37 °C was measured in 0.05 M sodium phosphate buffer
with 0.25 M NaCl plus 0.0025 M fumaric acid in D2O adjusted
to pH 7.5 with NaOD. Buffer was added to a weighed amount
of triazene, sealed in glass ampules, and incubated for 0, 24, or
48 h. The initial triazene concentration was 0.0100 M. At the
end of the reaction time, an aliquot of the reaction solution was
removed and analyzed by 1H NMR on an XL-200 Varian
spectrometer. Assignment of the NMR peaks arising from the
decomposition of the triazenes was made by comparison with
authentic samples of starting materials and products and
confirmed by coincidence of peaks upon addition of authentic
F igu r e 2. Structures and abbreviations of 1,3-dialkyl-2-acyl-
triazenes.
material. Quantities were determined by comparative integra-
tion of the product peaks and confirmed by use of the vinyl
signal from the fumaric acid as an internal standard. Rates of
decomposition were calculated from the amount of starting
material remaining after 24 h (HMC) or 48 h (DMA, HMA,
CMC, and CMA) according to the equation: ln(A
0/A) ) kT.
The conversion of the acyltriazenes CMC, CMA, HMC, and
HMA to methanol was measured in 0.02 M sodium phosphate
buffer with 0.01 M NaCl plus 0.05 M maleic acid in D2O adjusted
to pH 7.4 with NaOD (nominal pH 6.96). Buffer was added to
a weighed amount of triazene, sealed in ampules, and incubated
for 90 h at 70 °C. The initial triazene concentration in each
reaction was 0.02 M. In addition, the production of methanol
was also assayed from the decomposition of CMA in the presence
of calf thymus DNA (final triazene concentration 0.005 M).
Assignment of the NMR peaks arising from the decomposition
of the triazenes was made as described above, using maleic acid
as the internal standard.
Alk yla tion Assa y. The alkylation of calf thymus DNA or
the synthetic polymers (0.52 mg/mL in 20 mM sodium phos-
+
+

Selectivity in DNA Alkylation
Chem. Res. Toxicol., Vol. 9, No. 1, 1996 343
Ta ble 1. Ra tes of Decom p osition of
1,3-Dia lk yl-3-a cyltr ia zen es in P h osp h a te Bu ffer a
a t p H 7.5
phate buffer, 10 mM NaCl, pH 7.0) by the acyltriazenes was
accomplished at 37 °C. The reaction was initiated by the
addition of 10 µL of a 0.5 M solution of the triazene in DMSO
(5 mM final concentration) to 990 µL of the calf thymus DNA
or polymer solution in buffer. The reaction was mixed on a
Vortex mixer and incubated for 48 h. The reactions were
stopped by the addition of 3 mL of cold absolute ethanol, and
the DNA was pelleted by centrifugation (15 min at 5000g). The
pellet was washed first with cold 2% sodium acetate in ethanol
and allowed to sit in this solution for 15 min. The solvent was
then decanted off and replaced by ethanol/ether (1:1) for 5 min
and finally by ether. After the ether was decanted, the DNA
pellets were dried in a vacuum desiccator and stored at -20 °C
until hydrolysis.
b
acyltriazene
kobs (s-1) at 70 °C
kobsc (s-1) at 37 °C
HMC
DMA
HMA
CMC
CMA
2.22 × 10-3
7.16 × 10-4
4.53 × 10-4
2.46 × 10-4
4.46 × 10-5
3.3 × 10-5
1.3 × 10-5
7.8 × 10-6
3.2 × 10-6
-7
2.5 × 10
a
Buffer concentration 0.05 M; ionic strength of 0.25 M held
b
constant with added NaClO4. The rate constants are an average
of at least two independent runs varying by no more than (3%;
triazene initial concentration 3.0 × 10-5 M; data taken from ref
11. c The rate constants were calculated from the amount of
acyltriazene decomposition in 48 h (see Materials and Methods);
triazene initial concentration 0.0100 M.
The stability of the 7-(chloroethyl)- and 7-(hydroxyethyl)-
guanine adducts to depurination was measured by reacting a
50 µL aliquot of CMC or CMA with 4950 µL of calf thymus DNA
or poly(dG)‚poly(dC) (0.52 mg/mL). The alkylation of the
guanine was allowed to proceed at 37 °C for 48 (CMC) or 90 h
(CMA). At the end of this time, the DNA was isolated by
precipation with 15 mL of ethanol and was then washed and
dried as described above. The dried DNA was redissolved in 5
mL of buffer, and 1 mL aliquots were removed at time points
between 0 and 48 h. The aliquots were precipitated, dried, and
stored as described above until hydrolysis.
The appearance of methanol as a product following
incubation of various triazenes other than DMA for 90 h
at 70 °C in sealed ampules was reinvestigated by NMR.
The spectra showed that when all of the triazene had
decomposed, the maximum amount of methanol produced
was 3.4% from HMA and 1.5% from CMA, while both
HMC and CMC showed no detectable peak for methanol.
Incubation experiments with CMA in the presence of
DNA at 37 °C showed no increase in the levels of
methanol, showing that the presence of DNA did not
change the spectrum of decomposition products.
Alk yla tion Stu d ies. The analysis of synthetic DNA
polymer or calf thymus DNA alkylation products by
HPLC coupled with fluorescence detection, following
alkylation by CMC, CMA, HMC, HMA, or DMA, revealed
the formation of the adducts 7-MeG, 7-HOEtG, 7-ClEtG,
O6-MeG, and/or O6-HOEtG, depending on the triazene
investigated. The results of these studies are sum-
marized in Table 2. Since poly(dG)‚poly(dC) and poly-
(dG-dC)‚poly(dG-dC) contain 50% guanine, while poly(dA-
dC)‚poly(dG-dT) and poly(dA-dG)‚poly(dC-dT) contain
25% guanine and calf thymus DNA contains 21.7%
guanine (12), the yields have been “normalized” by
dividing the actual millimoles of alkylated guanine
detected by the mole percentage of guanine present in
the polymer sequence. The data reported in this manner
allow for direct comparison of the susceptibility of any
particular sequence in the polymer to preferential alky-
lation of DNA guanine upon exposure to a given triazene
for 48 h at 37 °C.
An a lysis of DNA Ad d u cts. Hydrolysis and chromatography
of the DNA adducts (7-MeG, 7-HOEtG, 7-ClEtG, O6-MeG, and
O6-HOEtG) were carried out as previously described (10).
Quantitation of the adducts was accomplished by comparison
of the areas of the HPLC chromatographic peaks to a standard
curve, which was remeasured each day. Results are reported
as mmol of alkylated base/mol of guanine on at least triplicate
experiments.
P olym er Decom p osition Kin etics. The kinetics of triaz-
ene disappearance in the presence and absence of the poly(dG-
dC)‚poly(dG-dC) and the poly(dG)‚poly(dC) polymers were mea-
sured in the following manner. Reactions consisted of 990 µL
of buffer (pH 7.0) or synthetic polymer (0.52 mg/mL) dissolved
in the buffer and 10 µL of a 0.5 M DMSO solution of either CMC
or CMA (final concentration 5 mM). In place of some of the
buffer, EDTA was added to various reaction mixtures to give a
final concentration of 2.3 mM. The reactions were allowed to
run for times up to 119 h, with 50 µL aliquots withdrawn at
times of 0, 3, 7, 24, 30, 48, 57, 72, 95, and 119 h. The aliquots
were assayed immediately by HPLC using a reverse phase C18
column at a flow rate of 1 mL/min with an isocratic solvent
mixture of water/methanol (3:1). The triazene peak was moni-
tored at 234 nm and was integrated by a computer program.
The rate constants reported are the average of at least two
determinations.
As expected, the composition of the polymer sequences
was found to be important in the alkylating pattern of
the triazenes. For any or all of the 7-alkyl guanine
adducts from each triazene, it can be seen that DNA
sequences which have runs of purines were alkylated to
a greater extent than are sequences which alternate
purines and pyrimidines. However, within the frame-
work of this generalization, the structure of the triazene
was also found to be a determinant in the observed
sequence selectivities. Both of the (hydroxyethyl)triaz-
enes examined hydroxyethylated calf thymus DNA to a
greater extent than the other triazenes. For the 7-HO-
EtG adduct from HMC, the sequence of reactivity for the
polymers was in the order: AGA > GGG > CGC . TGT,
while for HMA the order was AGA > GGG . CGC >
TGT. In addition, both (hydroxyethyl)triazenes formed
an unknown 7-alkylguanine product (data not shown),
as has been reported previously both by us (2) and others
(13). Both of the (chloroethyl)triazenes chloroethylated
and/or hydroxyethylated the GGG polymer best, followed
by calf thymus DNA . AGA > TGT > CGC, depending
Resu lts
NMR Stu d ies. The approximate rates of triazene
decomposition, as determined from the amount of tria-
zene remaining after 48 h at 37 °C, are shown in Table
1. The relative order of decomposition of the acyltriaz-
enes at 37 °C was found to be the same as that seen at
70 °C (HMC > DMA > HMA > CMC > CMA) (11). The
ratios of the rates at each temperature were comparable
except for CMA, whose rate of decomposition at 37 °C
was roughly half of that predicted by assuming roughly
parallel temperature dependence of the rates.
The products from the decomposition of the various
triazenes (see Figure 1), measured by NMR, were the
same as determined previously (11). The N1 products
were either methanol (DMA), ethylene glycol and acetal-
dehyde (HMC, HMA), or 2-chloroethanol, ethylene glycol,
and acetaldehyde (CMC, CMA). The N3 product was
N-methylacetamide from the acetyltriazenes and ethyl
N-methylcarbamate from the carbethoxytriazenes.
+
+

344 Chem. Res. Toxicol., Vol. 9, No. 1, 1996
Kroeger Smith et al.
Ta ble 2. Alk yla tion of DNA P olym er s by Va r iou s Acyltr ia zen es
mmol of alkylated base/mol of guanine
agent
HMC
polymer
N
7-MeG
7-HOEtG
7-ClEtG
O6-MeG
O6-HOEtG
poly(dG)‚poly(dC)
GGG
CGC
TGT
AGA
NGN
GGG
CGC
TGT
AGA
NGN
GGG
CGC
TGT
AGA
NGN
GGG
CGC
TGT
AGA
NGN
GGG
CGC
TGT
AGA
NGN
0.08
nda
nd
9.92
8.99
1.53
13.08
12.52
18.30
0.41
0.53
1.57
5.27
7.50
3.83
3.22
11.52
22.95
3.81
0.10
0.21
0.38
1.01
0.16
0.12
0.05
0.32
0.26
0.12
0.04
nd
5.45
5.89
2.19
9.69
8.05
0.10
0.30
0.61
0.90
0.44
4.56
3.30
0.56
4.99
4.60
0.05
0.11
0.54
0.68
0.22
poly(dG-dC)‚poly(dG-dC)
poly(dA-dC)‚poly(dG-dT)
poly(dA-dG)‚poly(dC-dT)
ct DNA
0.67
4.27
nd
0.18
0.47
0.72
2.09
1.32
0.39
nd
1.24
1.81
1.08
0.56
1.12
1.53
2.56
87.40
33.41
34.52
177.89
167.16
CMC
HMA
CMA
DMA
poly(dG)‚poly(dC)
9.99
2.42
0.94
7.71
9.29
poly(dG-dC)‚poly(dG-dC)
poly(dA-dC)‚poly(dG-dT)
poly(dA-dG)‚poly(dC-dT)
ct DNA
nd
nd
poly(dG)‚poly(dC)
1.09
0.79
0.43
1.54
1.92
0.30
0.28
0.11
0.31
0.42
30.92
17.31
16.52
63.23
77.68
poly(dG-dC)‚poly(dG-dC)
poly(dA-dC)‚poly(dG-dT)
poly(dA-dG)‚poly(dC-dT)
ct DNA
poly(dG)‚poly(dC)
2.56
0.14
1.19
nd
poly(dG-dC)‚poly(dG-dC)
poly(dA-dC)‚poly(dG-dT)
poly(dA-dG)‚poly(dC-dT)
ct DNA
2.07
poly(dG)‚poly(dC)
poly(dG-dC)‚poly(dG-dC)
poly(dA-dC)‚poly(dG-dT)
poly(dA-dG)‚poly(dC-dT)
ct DNA
a
nd ) adduct not detected.
Ta ble 3. DNA Alk yla tion Ra tios
7-HOEtG/7-MeG
7-HOEtG +
7-ClEtG/7MeG 7-HOEtG/7-ClEtG
O6/7-HOEtG
O6/7-MeG
DNA polymer
CMC CMA
CMC
CMA CMC CMA HMC HMA CMC CMA HMC HMA CMC CMA HMC HMA DMA
poly(dG)‚poly(dC)
poly(dG-dC)‚
poly(dG-dC)
poly(dA-dC)‚
poly(dG-dT)
poly(dA-dG)‚
poly(dC-dT)
ct DNA
nd
15.3
5.9
0.4
1.8
0.17
1.5
0.7
25.92 3.5 123
5.7 0.01 0.01 0.55 0.61 0.17 0.28 2.0
0.82 0.35
2.2 0.18
nd
9.7 0.73 1.1
0.66 0.86 0.2
0.5 nd
0.1 nd
0.2 0.5
2
0.52
0.48
0.36
0.46
3.1
13
7
1.3
0.56
0.2
0.18
nd
1.1 0.24
nd
nd 1.2 2.5
1.4
0.7
0.17 nd
0.43 nd
nd
1.2
0.25
1.2
2.2 0.25 19.5 9.3 0.57 1.8
2.5 0.4
0.57
0.49
2.9 12.7 0.08 0.22 0.64 0.2 nd
0.17 0.06 1.1
on the triazene. However, they all methylated calf
thymus DNA preferentially to the other sequences, with
the exception of DMA, which methylated the AGA
polymer equally well. Thus, the one triazene which is a
pure methylating agent did not follow the pattern of the
other acetyltriazenes.
HMA > HMC g CMC > CMA, while for poly(dG)‚poly-
(dC), the order was CMC g HMC > HMA g CMA. For
poly(dA-dC)‚poly(dG-dT) the pattern was again different
with HMA > CMA > CMC > HMC. The rates of
decomposition of the triazenes were also reflected in the
alkylation by all of the triazenes on the O6-position of
guanine. For calf thymus DNA and the polymers, the
order of alkylation followed the order of triazene decom-
position.
The ratios of the adducts (Table 3) varied with both
the sequence of the polymer as well as the structure of
the triazene. The (hydroxyethyl)triazenes showed higher
7-HOEt/MeG ratios than do the (chloroethyl)triazenes,
indicative of the fact that the former compounds can only
hydroxyethylate DNA. The ratio of 7-ClEt/7-MeG or
7-HOEt/7-MeG for CMC was always greater or equal to
that for CMA in all the sequences examined. These
ratios reflect the experimental alkylation levels for CMA
relative to CMC (Table 2), which showed an elevation in
the amount of 7-MeG in all sequences examined. This
slight shift in adduct preference (i.e., elevation in the
methylation level at the expense of chloroethylation or
hydroxylation by CMA) is not attributable to the 3-acetyl
group since HMA and HMC clearly do not show the same
shift in the 7-HOEt/MeG ratio. For the CMC and CMA,
the ratio of 7-HOEt/ClEtG was either 9 or 2 times higher
in poly(dG)‚poly(dC) as compared to poly(dG-dC)‚poly-
(dG-dC). However, within the context of each GC se-
Polymers with runs of purines are also preferentially
alkylated in the O6-position of guanine, as compared with
those containing runs of alternating pyrimidines and
purines. For all of the triazenes, the O6-HOEtG adduct
was formed to the greatest extent in the AGA sequences;
however, the second best DNA for the formation of this
product from the (hydroxyethyl)triazenes was calf thy-
mus DNA. The (chloroethyl)triazenes surprisingly alky-
lated the TGT polymer second, followed by calf thymus
DNA. For O6-MeG, calf thymus DNA is alkylated to the
greatest extent by all of the compounds except for HMC,
where AGA was methylated slightly better than calf
thymus DNA. Thus, for O6-methylation, the three acetyl-
triazenes followed the same pattern, while the carbe-
thoxytriazenes showed a different selectivity.
As can be seen from the Table 2, the acyltriazene which
was the best overall alkylating agent varied with the
sequence examined. DMA alkylated DNA to a greater
extent than any other triazene in all sequences. The
order of the rate of decomposition of the triazenes (see
above) paralleled total 7-alkylation only in the GCG or
AGA polymers. For calf thymus DNA, the order was
+
+

Selectivity in DNA Alkylation
Chem. Res. Toxicol., Vol. 9, No. 1, 1996 345
Ta ble 4. Ra tes of Dep u r in a tion of 7-Alk ylgu a n in e
Ad d u cts
determined from the NMR study. The results also
revealed that the rates of decomposition of the two
triazenes, in either the presence or absence of the
chelating agent EDTA, were not significantly different
irrespective of the polymer. Therefore, the polymers
themselves are not acting as catalysts in triazene de-
composition.
kobs (s-1
)
7-ClEtG
7-HOEtG
CMC/ct DNA
CMC/GGG polymer
3.6 × 10-6
3.2 × 10-6
6.3 × 10-7
CMA/ct DNA
CMA/GGG polymer
1.8 × 10-6
2.5 × 10-6
6.4 × 10-6
Discu ssion
quence, both triazenes gave similar numbers for this
ratio, showing that the enhanced apparent reactivity
toward runs of guanines was not due to a preference for
chloroethylation over hydroxyethylation.
The fact that DNA base sequence can affect patterns
of alkylation by various agents; including nitrosoureas
and 1-aryltriazenes, has been previously documented. For
example, even the simple ethyl- and methylnitrosoureas
show such selectivity, with the methyl analog exhibiting
more reactivity toward sequences which are composed
exclusively of A-T base pairs, while the ethyl analogs
demonstrate a preference for consecutive G-C bases (14).
Various (chloroethyl)nitrosoureas have been shown to
favor alkylation at runs of guanines rather than at
random base sequences of DNA, although the degree of
selectivity was found to vary with both the position
alkylated and the nature of the electrophilic species
(chloroethyl or hydroxyethyl) (12, 15, 16). Several 1-aryl-
3-alkyltriazenes and (alkyltriazenyl)imidazoles were also
examined for their ability to selectively alkylate a frag-
ment of pBR322 DNA (17). The monomethyl and
monochloroethyl triazenes alkylated guanines extensively
at the 7-position of guanine with a preference for runs
of contiguous guanines, similar to but not as striking as
that observed for the (chloroethyl)nitrosoureas.
The results of the present study reveal that, like the
nitrosoureas, acyltriazenes exhibit definite preferences
in the alkylation of various DNA polymer sequences, with
the selectvity varying with the structure of the triazene.
The selective alkylation effects are not likely to be due
to DNA conformational changes during the course of the
reaction, since it has been previously reported that the
double-stranded polymers used in our study prefer the
B-DNA conformation under the reaction conditions em-
ployed (16). All of our reactions were carried out for the
same length of time in an attempt to eliminate any
variables except for the differential rates of triazene
decomposition. Further, the rate of disappearance of the
triazene was shown to be independent of the presence of
DNA. Since one can eliminate the above possibilities as
contributing factors in the observed sequence selectivity,
we must then consider the structural differences between
the triazenes as the pertinent factors in their observed
selective alkylation of DNA.
The most important result of the present study in that
regard was that DNA sequences containing runs of
purines (this includes calf thymus DNA) are the most
reactive toward alkylation by all of the triazenes tested,
whether measured by N7, O6, or total guanine adducts.
Within that generalization, (hydroxyethyl)triazenes
showed a preference for the AGA sequence, while (chlo-
roethyl)triazenes favored GGG. The high levels of alky-
lation observed in calf thymus DNA for all of the adducts,
as compared with the levels of the comparable adducts
in the other polymers, had also been observed by Briscoe
et al. (11, 15). They noted that, following alkylation by
1,3-bis(2-chloroethyl)-1-nitrosourea, the 7-HOEtG level
resulting from averaging data obtained separately using
four synthetic polymers was the same as the experimen-
tal value detected in calf thymus DNA. This finding was
also obtained in the DNA alkylation data by choloreth-
Comparison of the O6/7-MeG or O6/7-HOEtG ratios
(Table 3) between different triazenes reveals some in-
teresting trends. The (hydroxyethyl)triazenes, in gen-
eral, and HMA, in particular, had higher O6/7-MeG
ratios, i.e., larger amounts of O6 methylation, than did
DMA, which, in turn, was the same or higher than those
of the (chloroethyl)triazenes. It therefore appears that
the N1-alkyl group on the triazene is the important
determinant in the partitioning of methylation between
the O6- and 7-positions on guanine. The conclusions that
can be drawn from the O6/7-HOEtG ratios are less clear.
In general, the major differences in this ratio for the
(chloroethyl)triazenes are seen when comparing the calf
thymus DNA or the poly(dG)‚poly(dC) sequences to the
others; in the former two cases, the ratio is much lower.
This ratio in the CGC, TGT, and AGA sequences for CMA
was approximately double that for CMC. Except for the
poly(dG)‚poly(dC) sequence, the (chloroethyl)triazenes
had ratios which were the same or higher than the
(hydroxyethyl)triazenes. The presence of a carbethoxy
group in the triazene changed the ratio only for the poly-
(dG)‚poly(dG) and the calf thymus DNA sequences (where
the HMC ratio was quite a bit higher than it was for
CMC); in the other sequences the ratio was identical for
both triazenes.
The stability of the 7-HOEtG and 7-ClEtG adducts was
measured after incubation of the DNA polymers in
phosphate buffer following previous reaction with either
CMC or CMA. As can be seen from the rates of
decomposition of the adducts in Table 4, the 7-HOEtG
and the 7-ClEtG adducts produced from both triazenes
in the calf thymus DNA reactions decomposed, presum-
ably by depurination. However, only 7-ClEtG depuri-
nated from reactions with the GGG polymer. For CMC,
the rate of 7-ClEtG depurination was faster if this adduct
was produced from calf thymus DNA reactions than if it
arose from reactions with GGG; however, the reverse was
true with the 7-ClEtG adduct from CMA. The 7-HOEtG
adduct from calf thymus DNA reactions with CMA
depurinated faster than the same adduct from CMC.
Tr ia zen e Decom p osition Stu d ies in th e P r esen ce
of P olym er . The rates of decomposition of CMC or CMA
in phosphate buffer alone or in the presence of synthetic
DNA polymer in buffer at 37 °C were analyzed by HPLC
measurement of the disappearance of substrate (data not
shown). The rates of decomposition in buffer alone are
essentially the same as those determined by NMR
analysis, making some allowance for the differences in
the methods. The presence of nucleic acid polymer did
not substantially alter these rates within experimental
error. For example, the rate of decomposition of CMC
in the presence of poly(dG)‚poly(dC) was 2.04 × 10-6 s-1
-6
by HPLC analysis, as compared with 3.15 × 10
s-1
+
+

346 Chem. Res. Toxicol., Vol. 9, No. 1, 1996
Kroeger Smith et al.
yltriazenes in the present study. In the case of the
hydoxyethyltriazenes, however, calf thymus DNA was
alkylated to a greater extent than any of the synthetic
polymers; the reason for this finding is unknown, but is
clearly reflective of the structural difference in the N1-
alkyl group.
alkylation products possible from DMA are the result of
methylation.
A second variable examined was whether the alkylat-
ing triazene bears an N1 hydroxyethyl or chloroethyl
group. Inspection of the yields of N7-HOEtG and O6
-
HOEtG in Table 2 reveals that, with almost all of the
polymers, the levels for the (hydroxyethyl)triazenes are
higher than those for the (chloroethyl)triazenes. The
observed enhancement is not a simple rate effect; HMC
decomposed about 10 times faster in buffer at neutral
pH than did CMC. However, the level of O6-HOEtG
produced by HMC in the GGG homopolymer was about
54-fold higher than that produced by CMC. The HMC/
CMC ratio for this same adduct in calf thymus DNA was
about 19. The same type of comparison holds for HMA
and CMA.
Figure 1 shows the mechanisms by which (hydroxy-
ethyl)- and (chloroethyl)- triazenes decompose to alky-
lating agents, resulting in the formation of O6- and N7-
HOEtG. Given that a portion of the alkylating agents
produced by (chloroethyl)triazenes lead to (chloroethyl)-
guanines, it is reasonable to expect that (hydroxyethyl)-
triazenes, purely hydroxyethylating agents, would give
higher overall yields of (hydroxyethyl)guanines.
What is surprising is the range of variation in the O6-
HOEtG/7-HOEtG ratio seen with different triazenes and
polymers. Clearly, this is a result of a complex interac-
tion of several factors. As has already been mentioned,
the varying stability of the same adduct (e.g., N7-HOEtG)
in different polymers will skew the apparent yields and
ratios of the products. Further, it can be seen from
Figure 1 that (chloroethyl)triazenes lead to (hydroxy-
ethane)guanines by two different pathways, C and D.
One of these, C, leads to the same species, the hydroxy-
ethanediazonium ion, which is presumably the sole
alkylating agent derived from (hydroxyethyl)triazenes.
This would lead to the expectation that the O6-HOEtG/
N7-HOEtG ratio from both types of triazenes would be
identical. However, the presence of alternative pathways
by which (chloroethyl)triazenes can produce (hydroxy-
ethyl)guanines through different alkylating agents (e.g.,
chloroethanol, ethylene oxide, and oxadiazoline) (6) would
be expected to give different O6-HOEtG/N7-HOEtG ra-
tios. Because some of these alternative pathways have
been reported to give exclusively N7-HOEtG, it is even
conceivable that in some cases a (chloroethyl)triazene
could give a higher yield of this adduct than does the
corresponding (hydroxyethyl)triazene.
A third factor important in the extent of alkylation of
a given type of DNA was the structure of the triazene
acyl group. For most DNA sequences, 3-carbethoxytria-
zenes gave higher yields of all types of adducts when
compared with 3-acetyltriazenes. This holds true for both
the (hydroxyethyl)- and (chloroethyl)triazenes. These
overall patterns tend to mirror and may be explained by
the rates of decompostion of the triazenes at 37 °C in
phosphate buffer: HMC > HMA > CMC > CMA. The
differences in yields are roughly of the same order as the
differences in the rates of triazene decomposition.
Comparison of the ratios of 7-HOEtG + 7-ClEtG/7-
MeG for CMA vs CMC (see Table 3) shows that an
elevation of methylation occurred from CMA at the
expense of hydroxyethylation and chloroethylation (i.e.,
lower ratios for CMA in all polymer sequences). On an
absolute percent yield basis, a 0.0018% yield of 7-MeG
was produced from CMC, while this number was 0.01%
for CMA (5.5 times greater). Under the conditions of the
The basis for the preferential alkylation in purine runs
is most likely the result of a complex mixture of compet-
ing factors. It has been suggested (12) that one reason
for this may be that guanines in runs of guanines are
the most electronegative of the sequences tested (GGG
. AGA > TGT > CGC) (17). In their study of (chloro-
ethyl)nitrosoureas, which showed the same high selectiv-
ity for alkylation in runs of guanines as we observed,
Briscoe et al. (12) found that this electronegativity
hypothesis correlated well for the 7-HOEt adduct from
various (chloroethyl)nitrosoureas and moderately well for
the 7-ClEtG adduct. Examination of the levels of the
7-(hydroxyethyl)guanine adduct from our studies reveals
that the (chloroethyl)triazenes alkylate between 10-12
times more in the GGG sequence than they do in the AGA
sequence. However, the 7-ClEtG levels are only slightly
elevated in the GGG sequence, as compared to AGA.
Therefore, the electronegativity of a sequence of guanines
is most likely only one contributing factor in total
7-alkylation.
A second important factor may involve steric effects
by neighboring bases. This, along with the decreased
nucleophilicity, probably accounts for the lower alkylation
of TGT by the triazenes, where the 5-methyl group of
thymine may hinder the accessibility of the N7 of the
guanine (18). However, steric considerations should not
be involved in the lower amount of hydroxyethylation
observed in the AGA sequence as compared to the GGG
sequence, since both the N7 and O6 positions in GGG and
AGA should be equally open to attack. The reason for
the preference for the AGA sequences by the (hydroxy-
ethyl)triazenes is therefore unknown.
A third factor which may influence the observed
elevated alkylation yields in runs of purines may involve
the stability of the various adducts themselves. The
(chloroethyl)triazenes yielded more 7-HOEtG in the GGG
polymer than in calf thymus DNA, yet approximately
equivalent amounts of 7-ClEtG are produced in both
polymers. A clue to understanding this result is provided
by the data on depurination of the 7-ClEt and 7-HOEtG
adducts formed from the triazenes in the presence of calf
thymus DNA or poly(dG)‚poly(dC) (see Table 4). While
both adducts were slowly lost through depurination from
the calf thymus DNA reaction, only the the 7-ClEtG
adduct was lost following reaction with the GGG polymer.
The 7-HOEtG adduct from the GGG reaction, by com-
parison, is relatively stable. Thus the observed higher
yield of 7-HOEtG from GGG, as compared with that from
calf thymus DNA, may simply be the result of the partial
decomposition of this adduct from calf thymus DNA
during the 48 h reaction period.
Comparing the two class of triazenes, structural dif-
ferences that affect the pathway of decomposition were
also seen to play a significant role in the noted patterns
of alkylation. The results of the studies clearly showed
that DMA, a substance which produces exclusively meth-
anediazonium ion, was the most effective alkylating
agent of all of the acyltriazenes tested. This presumably
is simply a result of the high susceptibility of DNA
towards methylating agents, as compared with other
types of alkylating agents and the fact that the only
+
+

Selectivity in DNA Alkylation
Chem. Res. Toxicol., Vol. 9, No. 1, 1996 347
methyl-3-(N-methylcarbamoyl)triazene (25). In addition,
the acyl group also appears to influence DNA alkylation
by a more subtle mechanism, which is as yet undefined.
In any case, the modulation of alkylation by the presence
of specific groups in the acyl position of the triazene has
important implications in the design of new, more potent
chemotherapeutic alkylating agents.
experiment, increased deacylation of CMA (pathway B,
Figure 1) would be necessary to account for this ad-
ditional methylation. Deacylation of CMA would produce
2-(chloroethyl)-3-methyltriazene as the first product.
This, in turn, has been shown preferentially to give the
methanediazonium ion. In order to further investigate
the increased production of 7-methylguanine from CMA,
we attempted to trap the methanediazonium ion as
methanol by incubation of this triazene at 37 °C in a
sealed ampule in buffer, both with and without DNA
present. The results from the NMR spectrometry showed
some (1.5%) methanol in reaction mixtures without DNA
and no methanol when DNA was present, presumably
because the methanediazonium ion was scavenged by the
DNA. No more than a trace of methanol could be found
in a similar experiment from CMC. Thus it appears that
CMA gives a higher yield of the methanediazonium ion
than does CMC. While this yield is small, the high
alkylation efficiency of this species accounts for what
might otherwise seen as unusually low 7-HOETG +
7-ClEtG/7MeG ratio for CMA. We further showed that
the presence of different DNA polymers had no effect on
the rate of triazene decomposition, as compared with the
decomposition of the same triazenes in buffer alone.
However, the presence of the polymer did not increase
the rate.
Ack n ow led gm en t. This research was sponsored by
the National Cancer Institute, DHHS, under Contract
NO1-CO-46000 with ABL. R.H.S. was supported in part
by the National Science Foundation (CHE-9215925). The
authors wish to thank Mr. J ohn Klose for the NMR
spectral determinations and Dr. Grzegorz Czerwinski for
his assistance in the interpretation of the NMR data. The
contents of this publication do not necessarily reflect the
views or policies of the Department of Health and Human
Services, nor does mention of trade names, commercial
products, or organizations imply endorsement by the U.S.
Government.
Refer en ces
(1) Smith, R. H., J r., Scudiero, D. A., and Michejda, C. J . (1990) 1,3-
Dialkyl-3-acyltriazenes, a novel class of antineoplastic alkylating
agents. J . Med. Chem. 33, 2579-2583.
(2) Kroeger-Koepke, M. B., Michejda, C. J ., and Smith, R. H., J r.
(1992) Alkylation of DNA by 1,3-dialkyl-3-acyltriazenes: correla-
tion of biological activity with chemical behavior. Chem. Res.
Toxicol. 5, 541-547.
One of the interesting, but still unexplained, findings
of the aforementioned alkylation study using the pBR322
fragment (17) was that the nonalkylating portion of a
triazene could influence the ultimate alkylation spec-
trum. In that study, two different (chloroethyl)triazenes
produced different patterns of alkylation, suggesting that
they did not act through a common intermediate (chlo-
roethyl diazohydroxide), as had been previously postu-
lated was the case in the alkylation of DNA by the
(chloroethyl)nitrosoureas (20-24). In our study we have
observed that the ratio of product from CMC to CMA for
any given sequence paralleled the ratio (13) of the rate
of decomposition of the two compounds, suggesting the
same decomposition pathway was operative for both
compounds. This is not the case, however, for the
(hydroxyethyl)triazenes. HMA alkylated the 7-position
of guanine from 0.5 to 2.3 times as much as did HMC,
depending on the sequence. HMA, however, decomposes
at a rate which is 0.2 times that of HMC. Moreover, in
all of the sequences examined except calf thymus DNA,
the amount of methylation from HMA vs HMC appeared
to be elevated at the expense of hydroxyethylation (see
ratios in Table 3). This correlates with the small amount
of methanol detected from HMA, but not HMC, in the
NMR experiments. The fact that HMA also showed this
increased amount of methylation means that the pres-
ence of the acetyl group alone may possibly be a sufficient
condition for deacylation to occur. However, the exact
reason(s) for an acetyl vs a carbethoxy acyl group on the
triazene affecting the balance of decomposition via dif-
ferent pathways awaits further experimentation.
(3) Oakberg, E. F., and Crosthwait, C. D. (1993) The effect of ethyl-,
methyl-, and hydroxyethyl-nitrosoureas on the mouse testis.
Mutat. Res. 108, 337-344.
(4) Pelfene, A., Mirvish, S. S., and Gold, B. (1976) Brief communica-
tion: induction of malignant bone tumors in rats by 1-(2-
hydroxyethyl)-1-nitrosourea. J . Natl. Cancer Inst. 56, 445-446.
(5) Kohn, K. W. (1977) Interstrand cross-linking of DNA by 1,3-bis-
(2-chloroethyl)-1-nitrosourea and other 1-(2-haloethyl)-1-nitro-
soureas. Cancer Res. 37, 1450-1454.
(6) Lown, J . W., and Chauhan, S. M. S. (1981) Mechanism of action
of (2-haloethyl)nitrosoureas on DNA. Isolation and reactions of
postulated 2-(alkylimino)-3-nitrosooxazolidine intermediates in
the decomposition of 1,3-bis(2-chloroethyl)-3-(4′-trans-methylcy-
clohexyl)-1-nitrosourea. J . Med. Chem. 24, 270-279.
(7) Tong, W. P., Kirk, M. C., and Ludlum, D. B. (1982) Formation of
the cross-link, 1-[N3-deoxyguanosinyl]-ethane, in DNA treated
with N,N′-bis-(2-chloroethyl)-N-nitrosourea (BCNU). Cancer Res.
42, 3102-3105.
(8) Smith, R. H., J r., Mehl, A. F., Hicks, A., Denlinger, C. L., Kratz,
L., Andrews, A. W., and Michejda, C. J . (1986) 1,3-Dimethyl-
3-acyltriazenes: synthesis and chemistry of a novel class of
biological methylating agents. J . Org. Chem. 51, 3751-3757.
(9) Smith, R. H., J r., Mehl, A. F., Shantz, D. L., Chmurny, G. N.,
and Michejda, C. J . (1988) Novel cross-linking agents, 1-(2-
chloroethyl)-3-methyl-3-acyltriazenes. J . Org. Chem. 53, 1467-
1471.
(10) Koepke, S. R., Kroeger-Koepke, M. B., Bosan, W., Thomas, B. J .,
Alvord, W. G., and Michejda, C. J . (1988) Alkylation of DNA in
rats by N-nitrosomethyl-2-(hydroxyethyl) amine: dose response
and persistence of the alkylated lesions in vivo. Cancer Res. 48,
1537-1542.
(11) Smith, R. H., J r., Wladkowski, B. D., Herling, J . A., Pfaltzgraff,
T. D., Pruski, B., Klose, J ., and Michejda, C. J . (1992) 1,3-
Dialkyl-3-acyltriazenes: Products and rates of decomposition in
acidic and neutral aqueous solutions. J . Org. Chem. 57, 654-
661.
(12) Briscoe, W. T., Anderson, S. P., and May, H. E. (1990) Base
sequence specificity of three 2-chloroethylnitrosoureas. Biochem.
Pharmacol. 40, 1201-1209.
(13) Gibson, N. W., Hartley, J . A., Strong, J . M., and Kohn, K. W.
(1986) 2-Chloroethyl(methylsulfonyl)methanesulfonate (NSC-
338947), a more selective DNA alkylating agent than the chlo-
roethylnitrosoureas. Cancer Res. 46, 553-557.
DNA modifications by acyltriazenes are subject to
multiple factors. One of these is the structure of the acyl
group, which appears to offer considerable potential as
a means of modulating the activity of acyltriazenes. The
acyl group exerts its major influence through its elec-
tronic characteristics, which serve to modulate the rate
and mechanism by which acyltriazenes decompose into
reactive products. In the extreme, the acyl group can
change the fundamental chemistry of the acyltriazenes,
as had been observed in the case of 1-(2-chloroethyl)-3-
(14) Ahmed, M., and Hadi, S. M. (1988) Sequence preference in
alkylation of DNA by N-methyl-N-nitrosourea and N-ethyl-N-
nitrosourea. Med. Sci. Res. 16, 1173-1174.
(15) Hartley, J . A., Gibson. N. W., Kohn, K. W., and Mattes, W. B.
(1986) DNA sequence selectivity of guanine-N-7 alkylation by
three antitumor chloroethylating agents. Cancer Res. 46, 1943-
1947.
+
+

348 Chem. Res. Toxicol., Vol. 9, No. 1, 1996
Kroeger Smith et al.
(16) Briscoe, W. T., and Duarte, S. P. (1988) Preferential alkylation
by 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) of guanines with
guanines as neighboring bases in DNA. Biochem. Pharmacol. 37,
1061-1066.
(17) Hartley, J . A., Mattes, W. B., Vaughan, K., and Gibson, N. W.
(1988) DNA sequence specificity of guanine N7-alkylations for a
series of structurally related triazenes. Carcinogenesis 9, 669-
674.
(18) Pullman, A., and Pullman, B. (1981) Molecular electrostatic
potential of the nucleic acids. Q. Rev. Biophys. 14, 289-380.
(19) Koepke, M. B. K., Schmiedekamp, A. M., and Michejda, C. J .
(1994) Theoretical investigation of reaction pathways of 3-me-
thyloxadiazolinium ion and 1,2,3-oxadiazole: correlation with
experimental findings. J . Org, Chem. 59, 3301-3306.
(20) Lown, J . W., Koganty, R. R., and Chauhan, S. M. S. (1984)
Characterization of alkyldinitrogen species implicated in the
carcinogenic, mutagenic and anticancer activities of N-nitroso
compounds: Occurence, biological effects and relevance to human
cancer. IARC Sci. Publ. 57, 689-693.
(21) Lown, J . W., Koganty, R. R., Bhat, U. G., and Chauhan, S. M. S.
(1886) Isolation and characterization of electrophiles from 2-ha-
loethylnitrosoureas forming cytotoxic DNA cross-links and cyclic
nucleotide adducts and the analysis of base site-selectivity by ab
initio calculations. In The Role of Cyclic Nucleic Acid Adducts in
Carcinogenisis and Mutagenesis (Singer, B., and Bartsch, H.,
Eds.) IARC Publications No. 70, pp 139-136, International
Agency for Research on Cancer, Lyon.
(22) Lown, J . W., Koganty, R. R., Bhat, U. G., Sapse, A.-M., and Allen,
E. B. (1986) Mechanism of interstrand cross-linking of DNA by
anticancer 2-haloethylnitrosoureas. Drugs Exp. Clin. Res. 12,
463-473.
(23) Lown, J ., and Chauhan, S. M. S. (1981) Synthesis of specifically
15N- and 13C-labeled antitumor (2-haloethyl)nitrosoureas. The
study of their conformations in solution by nitrogen-15 and
carbon-13 nuclear magnetic resonance and evidence for stereo-
electronic control in their aqueous decomposition. J . Org. Chem.
46, 5309-5321.
(24) Lown, J . W., Chauhan, S. M. S., Koganty, R. R., and Sapse, A.-
M. (1984) Alkyldinitrogen species implicated in the carcinogenic,
mutagenic, and anticancer activities of N-nitroso compounds:
Characterization by 15NMR of 15N-enriched compounds and
analysis of DNA base site selectivity by ab initio calculations.
J . Am. Chem. Soc. 106, 6401-6408.
(25) Rouzer, C. A., Thompson, E. J ., Skinner, T. L., Heavner, P. A.,
Bartolini, W. P., Mitchell, K., Kurz, E., Smith, R. H., J r., and
Michejda, C. J . (1993) An unexpected pathway for the metabolic
degradation of 1,3-dialkyl-3-acyltriazenes. Biochem. Pharmacol.
46, 165-173.
TX9500742
+
+