Reaction of N-acyloxy-N-alkoxyamides with biological thiol groups

Article abstract of DOI:10.1071/CH10470

Mutagenic N-acyloxy-N-alkoxyamides 1 react with thiols by an S_N2 process at nitrogen with displacement of carboxylate. They react with glutathione 4 in [D6]DMSO/D₂O and methyl and ethyl esters of cysteine hydrochloride, 11 and 12, in [D4]methanol but the intermediate N-alkoxy-N-(alkylthio)amides undergo a rapid substitution reaction at sulfur by a second thiol molecule to give hydroxamic esters and disulfides. Arrhenius activation energies and entropies of activation obtained for a series of different N-benzyloxy-N-(4-substitutedbenzoyloxy)benzamides 13-17 were similar to those found for the S_N2 reaction of the same series with N-methylaniline. Entropies of activation were strongly negative in keeping with polar separation and attendant solvation in the transition state, and in keeping with this, bimolecular reaction rate constants at 298K correlated with Hammett σ constants with a positive ρ-value of 1.1. The structure of model N-methoxy-N-(methylthio)acetamide has been computed at the B3LYP/6-31G(d) level and exhibits properties atypical of other anomeric amides with more electronegative atoms at nitrogen. Relative to N,N-bisoxyl substitution, the combination of a sulfur and an oxygen atom at the amide nitrogen results in a relatively small reduction in amide resonance.

Full text of DOI:10.1071/CH10470

RESEARCH FRONT
Full Paper
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2011, 64, 443–453
Reaction of N-Acyloxy-N-alkoxyamides with Biological
Thiol Groups-
A B
,
A
Stephen A. Glover
and Meredith Adams
A
Department of Chemistry, School of Science and Technology, University of New England,
Armidale, NSW 2351, Australia.
B
Corresponding author. Email: sglover@une.edu.au
Mutagenic N-acyloxy-N-alkoxyamides 1 react with thiols by an SN2 process at nitrogen with displacement of carboxylate.
They react with glutathione 4 in [D6]DMSO/D2O and methyl and ethyl esters of cysteine hydrochloride, 11 and 12, in [D4]
methanol but the intermediate N-alkoxy-N-(alkylthio)amides undergo a rapid substitution reaction at sulfur by a second
thiol molecule to give hydroxamic esters and disulfides. Arrhenius activation energies and entropies of activation obtained
for a series of different N-benzyloxy-N-(4-substitutedbenzoyloxy)benzamides 13–17 were similar to those found for the
SN2 reaction of the same series with N-methylaniline. Entropies of activation were strongly negative in keeping with polar
separation and attendant solvation in the transition state, and in keeping with this, bimolecular reaction rate constants
at 298 K correlated with Hammett s constants with a positive r-value of 1.1. The structure of model N-methoxy-N-
(methylthio)acetamide has been computed at the B3LYP/6–31G(d) level and exhibits properties atypical of other
anomeric amides with more electronegative atoms at nitrogen. Relative to N,N-bisoxyl substitution, the combination of a
sulfur and an oxygen atom at the amide nitrogen results in a relatively small reduction in amide resonance.
Manuscript received: 21 December 2010.
Manuscript accepted: 17 February 2011.
O
N
O
N
O
N
O
C
Introduction
R1
R
Ph
R2
Bu
N-Acyloxy-N-alkoxyamides 1 are molecular electrophiles that
react directly with DNA, inducing DNA damage.[1–5] They
belong to the wider class of so-called ‘anomeric amides’, 2,
bearing two electronegative atoms at the nitrogen, which toge-
ther impart very different amide properties (Chart 1).[6,7] The
combined electronegativity of the heteroatoms in 2 results
in a highly pyramidalized nitrogen and extensive loss of amide
resonance. In addition, strong anomeric effects operate between
the heteroatoms through the nitrogen and, where one atom or
group is significantly more electron-demanding than the other,
as is the case in N-acyloxy-N-alkoxyamides 1, both SN1 and SN2
reactions at nitrogen are possible.[8–16]
N-Acyloxy-N-alkoxyamides have been shown to be direct-
acting mutagens towards Salmonella typhimurium TA100 and
their activity in the Ames reverse mutation assay can be
predicted with some accuracy from quantitative structure–
activity relationships.[1–5] We have established an understand-
ing of the manner in which activity is influenced by the primary
and secondary structure of the mutagens. Their overall hydro-
phobicity is important, as are steric effects, which we have
shown to influence binding to DNA.[1–3] In addition, from
studies of a limited series of mutagens bearing different
benzoyloxyl groups 3, it was apparent that activity correlated
with the pKA of the corresponding benzoic acid, and
therefore inversely with the stability of the corresponding
O
X
O
O
Y
C R3
Z
O
O
1
2
3
X, Y ꢀ Cl, OR, NR2
Z ꢀ H, MeO, Ph, But, Me,
Cl, CHO, CN, NO2
Chart 1.
carboxylate.[1,3,5] Although the reaction with DNA is most
likely an SN2 process at guanine-N7,[4,16] stability to other
intracellular nucleophiles, which by reacting at the electrophilic
amide nitrogen would consume or deactivate the drugs and
thereby deplete mutagen concentrations, would appear to be an
important factor determining their level of mutagenic activity.
N-Acyloxy-N-alkoxyamides are susceptible to SN2 reactiv-
ity with a variety of inorganic nucleophiles such as azide[13] and
hydroxide,[14] as well as aromatic amines,[11,15,17] which we
have used as models for DNA reactivity. N-methylaniline reacts
bimolecularly with a wide range of mutagens bearing different
substituents on the amide, alkoxyl and acyloxyl side chains.[1]
As the N-acyloxy-N-alkoxyamide configuration is analogous to
that of a-haloketones, where a leaving group is adjacent to a
carbonyl and bulky a0-acyl side chains strongly impede SN2
yThis paper is dedicated to the memory of Athel Beckwith whose influence, both through his chemistry and through his generous spirit, has been felt by so many
who crossed his path.
Ó CSIRO 2011
10.1071/CH10470
0004-9425/11/040443

RESEARCH FRONT
444
S. A. Glover and M. Adams
displacement of halogen,[18,19] SN2 reactivity with aniline is
similarly inhibited by branching a to the amide carbonyl (at R1
in 1).[2] Branching adjacent to the alkoxyl oxygen (R2 in 1) is
also inhibitory.[15] However, steric effects are not important
on the acyloxyl group (R3 in 1) where reaction rate constants
correlate inversely with the pKAs of the departing carboxylic
acid.[15]
We have computed the reaction pathways for the
model bimolecular reactions of ammonia with N-formyloxy-
N-methoxyformamide (1, R1 ¼ R3 ¼ H, R2 ¼ Me)[12] and
N-acetoxy-N-methoxyacetamide (1, R1 ¼ R2 ¼ R3 ¼ CH3),[15]
which are found to involve transition-state geometries typical
of SN2 processes at nitrogen, with a good degree of polar
separation and long bonds to both the incoming ammonia and
carboxylate groups.
Glutathione (GSH) 4 is a biological reducing and conjugating
agent with a manifold of physiological functions.[20] It is present
in cells in concentrations up to 5 mmol and reacts with a range of
xenobiotics, in particular molecular electrophiles or metabolites
(Chart 2). Its role in this regard is to render such molecules
soluble and excretable.[21,22] Conjugation to electrophilic spe-
cies is achieved through the sulfur of the cysteinyl residue of
glutathione and the adduct can then be fed into the mercapturic
acid pathway, a major route of detoxification. The product of
this pathway is the N-acetylated conjugated cysteine, or mer-
capturic acid 5, which is excreted. Cysteine by itself and other
biological thiols do not serve as acceptors in this role.[22]
Thiols are known to protect against mutagenesis and carci-
nogenesis but DNA damage is intrinsic to the mode of action of
many electrophilic anticancer agents. As a consequence, thiols
can also lead to drug resistance and less efficient chemotherapy.
It has been noted that in some forms of cancer, the cells are
glutathione-deficient and these tumours are more sensitive to the
action of anticancer agents than cells with normal intracellular
levels of this molecule.[21,23] This indicates that glutathione is
able to intercept xenobiotic compounds that enter the cell and
render them inactive. Accumulation of glutathione or glu-
tathione conjugates, both critical to maintenance of the GSH
to oxidized GSH (GSSG) composition in cells, have been found
to confer drug resistance.
Preliminary work with MM96 L melanoma cells has indi-
cated that several N-acyloxy-N-alkoxyamides 1 cause mem-
brane damage, possibly through an interaction with glutathione
in cells resulting in a membrane pump effect (P. G. Parsons,
unpubl. data). This prompted a computational study at the
BP/DN*//6–31G* level on the possible interaction of thiols with
1. The model methanethiol was found to react smoothly
with N-formyloxy-N-methoxyformamide (Scheme 1), replacing
formate in a classical SN2 process resembling that of ammonia
with the same substrate.[12] Importantly, the process involves a
good deal of charge separation as predicted by the changes in
group charges (Dq) between the reactants and transition state
(Dq(OCHO) ¼ ꢀ0.72; Dq(CH3SH) ¼ þ0.4)), which forms in an
asynchronous process and requires a low active energy (EA) of
just 17.6 kJ molꢀ1 in the gas phase.
In this paper, we show chemically that N-acyloxy-N-
alkoxyamides 1 are indeed reactive with glutathione and with
methyl and ethyl esters of cysteine and that their reactivity
resembles that of N-methylaniline and other aromatic amines.
The data presented herein point to a classical SN2 process with
charge characteristics in accordance with the model reaction.
We show also that under reaction conditions, N-alkoxy-N-
(alkylthio)amides (2, X ¼ OR, Y ¼ SR) are unstable, reacting
further with thiols at the thioalkyl sulfur. In the process, thiols
are oxidized to the disulfide forms.
Results and Discussion
To establish whether glutathione (GSH) 4 reacts with N-
acyloxy-N-alkoxybenzamides, initial studies were carried out
by NMR at room temperature using, as substrate, N-acetoxy-N-
butoxybenzamide 6 in [D6]DMSO/D2O. The composition (ratio
of DMSO:D2O) was critical because GSH is water-soluble and
the hydrophobic mutagen is soluble in [D6]DMSO. A ratio of
2.5:1 was found to give a homogeneous reaction mixture.
Fig. 1a illustrates the 1H NMR spectrum of glutathione 4 in
[D6]DMSO/D2O. The methylene hydrogens adjacent to the side
chain thiol group, Ha, resonate as a diastereotopic pair between
d 2.6 and 2.9 and the cysteine methine, Hb, is a broadened triplet
at d 4.4, to high frequency of the large water peak at d 4.1. The
spectrum showed trace impurity peaks in the vicinity of d 4.5
and 3.1 and the new peaks were more developed in the 1H NMR
spectrum of glutathione when left for 1 h in [D6]DMSO/D2O,
(2.5:1) in a beaker of hot water (Fig. 1b). These were assigned to
the oxidized form of cysteine as DMSO is known to facilitate
oxidation of thiols to disulfides.[24] By comparison with fully
oxidized glutathione under the same conditions (Fig. 1c), they
were assigned to one of the diasteriotopic protons, Ha, adjacent
to sulfur at d 3.1 and the cysteinyl methine, Hb, at d 4.5 in the
H2N
O
O
O
H3C
O
HOOC
HN
OH
HN
N
H
COOH
E
S
HS
E ꢀ electrophile
4
5
Chart 2.
H
δꢁS
CH3
HSCH3
ꢁ
2.06 Å
O
O
SHCH3
O
O
OCH3
ꢂO
OCH3
N
ꢁ
N
N
H
H
H
2.41 Å
OCH3
H
H
O
O
O
O
δꢂ
H
Scheme 1.

RESEARCH FRONT
SN2 Reactions of Thiols at Amide Nitrogen of N-Acyloxy-N-alkoxyamides
445
disulfide form 8. The second diastereotopic proton of the
oxidized form overlaps with one of the equivalent pair in the
monomeric glutathione at d 2.8. On oxidation, other changes
were noted, particularly for the glutamyl methine proton, Hf at
d 3.43, which shifted slightly to higher frequency.
An NMR study of the behaviour of 6 and glutathione 4 in
[D6]DMSO/D2O indicated rapid consumption of both reactants
(Fig. 2). The products were the oxidized form of glutathione 8,
the hydroxamic ester N-butoxybenzamide 7 and acetic acid
(Scheme 2).
(c)
Hc
Hf
Hd
He
Hb
Ha
Ha
S
X
a
(b)
CH2
CHb
O
H
N
ꢂ
e
O2C
CH2
CO2H
d
c
CHf
CH2
N
H
CH2
O
NHꢁ3
X ꢀ H, S-glutathiyl
Hc
(a)
Hd
He
Hf
Hb
Ha Ha
4
3
2
ppm
Fig. 1. 1H NMR spectrum of (a) reduced glutathione 4; (b) partially oxidized glutathione; and (c) oxidized glutathione 8 in 2.5:1 [D6]DMSO/D2O at 303 K.
(b)
c
a
O
CH2
CH3
PhCO
d
b
N
H
CH2
CH2
Ha
Hd
Hc Hb
(a)
c
a
O
CH2
CH3
PhCO
d
b
N
CH2
CH2
He
e
OCOCH3
Ha
Hd
Hc
Hb
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
ppm
Fig. 2. 1H NMR spectrum of the reaction mixture of glutathione 4 and N-acetoxy-N-butoxybenzamide 6 in [D6]DMSO/D2O at 303 K after (a) 173 s, and
(b) 1835 s.

RESEARCH FRONT
446
S. A. Glover and M. Adams
O
O
N
H
O
4
O
O
N
O
7
AcOH
CH3
[D6]DMSO/D2O
HOOC
H2N
O
H
N
6
N
COOH
H
O
S
2
8
Scheme 2.
H
S
R
R
S
S
R
O
O
R
8
–AcOH
OBu
OBu
H
N
N
S
OAc
O
S
R
OBu
Ph
N
H
6
9
7
RS ꢀ glutathiyl
Scheme 3.
1
Comparison of H NMR spectra taken after 3 and 30 min
clearly shows that the triplet due to the oxymethylenic protons
Hd at d 4.18, to low frequency of the water peak at d 4.3, the
quintet due to Hc at d 1.55, the sextet of Hb at d 1.27 and methyl
triplet at d 0.85 due to Ha corresponding to the butyl resonances
of 6 in Fig. 2a, were replaced by a corresponding set of mutiplets
in Fig. 2b at d 3.92, 1.63, 1.4 and 0.93 respectively. The new
butyl resonances corresponded to those of the parent hydro-
xamic ester 7. The singlet due to the acetoxyl methyl group of 6
at d 2.15 (He in Fig. 2a) was replaced by a singlet at d 2.0 in
Fig. 2b, shown by spiking to coincide with the methyl group of
acetic acid. The preponderance of oxidized glutathione 8 after
30 min is evident from the side chain methylene proton region
(d 2.7–3.5) and the glutamyl methine resonances at d 4.45 and
4.63 (Fig. 2b).
80
70
60
50
40
30
20
10
0
k
2 ꢀ 0.0327
r2 ꢀ 0.9973
0
500
1000
1500
2000 2500
Time [s]
Fig. 3. Integral data fitted to Eqn 1 for reaction of N-acetoxy-N-
butoxybenzamide 6 and glutathione 4 in [D6]DMSO/D2O at 308 K.
and b the time-dependent concentrations (in mol Lꢀ1) of
N-acetoxy-N-butoxybenzamide 6 and glutathione 4 respec-
tively. k2 is the second-order rate constant.
On monitoring the NMR integrals with time, glutathione 4
was clearly consumed at twice the rate of N-acetoxy-N-
butoxybenzamide 6. Considering the nature of the products,
the reaction that best accounts for the products invokes initial,
rate-determining attack by glutathione at the nitrogen of
N-acetoxy-N-butoxybenzamide 6, giving the N-butoxy-N-(S-
glutathiyl)benzamide intermediate 9 (Scheme 3). Rapid, non-
rate-determining reaction of 9 with a second molecule of
glutathione yields the disulfide 8 and hydroxamic ester 7.
Careful integration of the acetoxyl methyl signal of 6 at
d 2.16 and, for glutathione, the difference between the total
integral d 2.7–3.0 and twice the integral of the low-field
diasteriotopic proton resonances between d 3.15 and 3.35
confirmed consumption of GSH at twice the rate of mutagen,
and the reaction was shown to be bimolecular with rate constants
at 303 K in the region of 2.5 ꢁ 10ꢀ2 L molꢀ1 sꢀ1. Fig. 3 gives a
representative fit to the second-order rate law in Eqn 1 for a
reaction at 308 K. A and B are the initial concentrations, and a
k2t ¼ lnðAb=BaÞ=ðB ꢀ 2AÞ
ð1Þ
The slope of this line of best fit gave the second-order rate
.
constant under these conditions of k3208 ¼ 3.3 ꢁ 10ꢀ2 L molꢀ1 sꢀ1
The reaction is clearly bimolecular and most probably
proceeds according to Scheme 3 with rate-determining forma-
tion of adduct 9. Since glutathione is a first, rather than the
second order reactant, the second step involving attack of
glutathione on 9 must be very fast. While these micro-scale
reaction mixtures were not buffered to constant pH, the genera-
tion of acetic acid and hydroxamic ester in the reaction clearly
has no observable catalytic or inhibitory effect on the rate.
Presumably any reduction in pH would serve to maintain the
thiol in its neutral form (pKA 4 9).

RESEARCH FRONT
SN2 Reactions of Thiols at Amide Nitrogen of N-Acyloxy-N-alkoxyamides
447
O
Clꢂ
Clꢂ
O
O
O
Ph
CH2Ph
O
O
N
O
C
H3N
OR
AcN
HS
H3N
OH
OR
Ph
CH2Ph
O
HN
Z
HS
S
O
2
10
11 R ꢀ Me
12 R ꢀ Et
13
14
15
16
17
18
19
20
Z ꢀ H
R ꢀ Me
R ꢀ Et
Z ꢀ OMe
Z ꢀ Me
Z ꢀ Cl
Z ꢀ NO2
Chart 3.
(b)
ꢂ
Hd
Cl
O
Hc
O
Hb
H3N
CHe
CH2
d
Ph
OCH3
O
CO2H
a
N
H
CH2
f
S
2
Ha
Hc
Hb
Hf
He
(a)
Hc
O
ꢂ
Cl
O
Ph
O
Hd
a
H3N
N
CH2
O
CHe
d
OCH3
Hf
O
f
CH2
SH
Hb
He
Ha
b
c
H
H
8
7
6
5
4
3
ppm
Fig. 4. 1H NMR spectra from the reaction of excess L-cysteine methyl ester hydrochloride 11 and N-benzoyloxy-N-benzyloxybenzamide 13 in [D4]methanol
(a) shortly after mixing, and (b) after complete conversion of 13 to benzoic acid and 18 and partial conversion of 11 to 19.
Quantitatively, the method proved to be unreliable. Variation
ofstarting concentrationsofbothN-acetoxy-N-butoxybenzamide
6 and glutathione 4 produced some scatter in the bimolecular rate
constants at 303K, which we attribute to the lability of both the
mutagen and particularly glutathione under the reaction condi-
tions, resulting in some uncertainty in the initial concentrations
of both reagents at the time of data acquisition. This impacts on
both the log term and the denominator in Eqn 1. In addition,
glutathione oxidizes slowly to the disulfide form throughout
the reaction process. Nonetheless, over five runs with an r2
greater than 0.985, an average rate constant of 1.9 ꢂ 0.8 ꢁ
10ꢀ2 L molꢀ1 sꢀ1 was obtained.
To explore the generality of this reaction, N-acyloxy-N-
alkoxyamides were subjected to reaction with other biological
thiols. Furthermore, in order to make comparisons with
kinetic data for the reaction of N-acyloxy-N-alkoxyamides
with aromatic amines,[1,2,10,11,15] conditions analogous to those
reactions were sought, namely reaction of a suitable thiol in
[D4]methanol.
Cysteine, the active residue in glutathione, was itself insoluble
in DMSO/H2O and methanol, and its N-acetyl derivative 10, a
natural thiol, thoughsolubleinboth DMSO/H2O and inmethanol,
was curiously unreactive with N-acyloxy-N-alkoxyamides
(Chart 3). A reaction mixture containing both 10 and

RESEARCH FRONT
448
S. A. Glover and M. Adams
Clꢂ
(b)
O
H3N
S
d
OCH2cCH3
CHb
Hd
a
CH2
2
Hc
Ha
Hb
Clꢂ
O
(a)
Hd
H3N
HS
d
OCH2cCH3
CHb
CH2
a
Hc, Hb
Ha
8
7
6
5
4
3
2
ppm
Fig. 5. 1H NMR from the reaction of L-cysteine ethyl ester hydrochloride 12 and N-benzoyloxy-N-benzyloxybenzamide 13 in [D4]methanol (a) shortly after
mixing, and (b) after conversion of 13 to benzoic acid and 18 and partial conversion of 12 to 20.
N-benzoyloxy-N-benzyloxybenzamide 13 in [D4]methanol at
303 K was monitored by NMR for 45min and remained unaltered
(see Accessory Publication). No reaction could be detected on
addition of either D2O or acetic acid to raise the polarity. In
[D6]DMSO/D2O, 13 reacted over several hours with a 10-fold
excess of N-acetylcysteine.
However, in [D4]methanol, a non-oxidizing solvent, both
L-cysteine methyl and ethyl esters 11 and 12 as the hydrochlor-
ides reacted with N-acyloxy-N-alkoxyamides in a similar
fashion to glutathione in [D6]DMSO/D2O.B
Figs 4 and 5 illustrate 1H NMR spectra of reaction mixtures
from the treatment of the N-benzoyloxy-N-benzyloxybenzamide
13 with a four-fold excess of the hydrochlorides of L-cysteine
methyl ester 11 and L-cysteine ethyl ester 12 in [D4]methanol.
Fig. 4 shows that the singlet at d 5.4 and doublets at d 7.9
and 8.0 (Fig. 4a) assigned to the benzylic methylene, Ha, and
ortho hydrogens on the benzamide and benzoyloxyl rings of
13, Hb and Hc, respectively, are replaced by a doublet at d 8.2
due to ortho hydrogens of benzoic acid, together with a singlet
close to d 5.1 and a doublet at d 7.85 of N-benzyloxybenza-
mide 18 (Hb, Ha and Hc respectively in Fig. 4b). Chemical
shifts of both products were confirmed by comparison with
those of authentic materials. The methylene doublet and
methine triplet of L-cysteine methyl ester 11 at d 3.2 and
4.5 (Hf and He in Fig. 4a) are replaced by a pair of diaster-
eotopic protons in the region of d 3.4 to 3.5 and a triplet at
d 4.6 (Hf and He in Fig. 4b). In addition, a second methyl
singlet is evident at d 4.0. These are consistent with the
oxidized form of the methyl ester 19.
Fig. 5 illustrates the same conversion of N-benzoyloxy-N-
benzyloxybenzamide 13 to benzoic acid and N-benzyloxyben-
zamide 18 and analogous changes in chemical shifts for the
methine and side-chain methylene of the L-cysteine ethyl ester
12. In this case, the methine, Hb, which overlaps with the ethyl
methylene group, Hc, at d 4.5 and the methylene protons, Ha at
d 3.25 in Fig. 5a, shift slightly to higher frequency in Fig. 5b,
where two ethyl triplet resonances are also visible at d 1.5. These
new resonances are assigned to the oxidized form of L-cysteine
ethyl ester 20.
The ABX splitting patterns at d 3.5 in Figs 4b and 5b that are
ascribed to the oxidized forms of both esters were reproduced
with similar shifts to higher frequencies on prolonged standing
of either thiol in [D6]DMSO/D2O, a medium that catalyzes the
oxidation of thiols to their disulfide forms.[24]
The rate of disappearance of starting mutagen 13 and both the
L-cysteine methyl and ethyl esters 11 and 12 could be monitored
in [D4]methanol by integration of the methylene singlet of
N-benzoyloxy-N-benzyloxybenzamide at
d 5.4 and the
BThe diminished reactivity of N-acetylcysteine relative to the methyl and ethyl esters of cysteine could be ascribed to intramolecular hydrogen bonding
between the carboxylic acid group and the thiol sulfur, which would reduce the nucleophilicity of the sulfhydryl group and would not be possible in the esters.
However, we have been unable to support this with computations at the B3LYP/6–31G* level.

RESEARCH FRONT
SN2 Reactions of Thiols at Amide Nitrogen of N-Acyloxy-N-alkoxyamides
449
Table 1. Arrhenius activation energies (EA), entropies of activation (DSz) and rate constants (k2) at 298 K for reaction of N-benzoyloxy-N-
benzyloxybenzamides 13–17 with L-cysteine ethyl ester hydrochloride 12 in [D4]methanol
Substrate (Z)
lnA
EA [kJ molꢀ1
]
DSz [J Kꢀ1 molꢀ1
]
104k2298 [L molꢀ1 sꢀ1
]
r
13 (H)
9.46 (1.17)
21.89 (3.84)
10.43 (1.89)
14.36 (2.49)
15.30 (3.02)
36.5 (2.7)
68.5 (9.3)
38.6 (4.5)
45.3 (5.5)
45.2 (6.7)
ꢀ183 (10)
ꢀ80 (32)
51.3
31.3
0.9862
0.9819
0.9738
0.9716
0.9786
14 (OMe)
15 (Me)
16 (Cl)
ꢀ175 (16)
ꢀ142 (21)
ꢀ134 (25)
57.0
200.0
523.2
17 (NO2)
respective cysteine side-chain methylenes of the methyl and
ethyl esters at d 3.2 and 3.25 respectively. In both cases, the
esters were consumed at approximately twice the rate of muta-
gen, and considering that the products, hydroxamic ester 18 and
benzoic acid, were formed exclusively from N-benzoyloxy-N-
benzyloxybenzamide, a similar reaction process to that depicted
in Scheme 3 for glutathione and N-acetoxy-N-butoxybenzamide
is proposed.
Of the two thiols, the product mixture from the ethyl ester
hydrochloride was more amenable to NMR analysis. The rates
of reaction of L-cysteine ethyl ester 12 with N-benzoyloxy-
N-benzyloxybenzamides 13–17 in [D4]methanol, studied under
pseudo-first-order conditions using a 10-fold excess of the
thiol, afforded good fits to the pseudo-first-order rate law
(Eqn 2), where [M] is the concentration of N-benzoyloxy-N-
benzyloxybenzamide and k0 ¼ k2[12]0.
1.2
1
NO2
0.8
0.6
0.4
0.2
0
Cl
y ꢀ 1.13x ꢁ 0.1211
r ꢀ 0.985
Me
H
OMe
ꢂ0.2
ꢂ0.4
ꢂ0.5
0
0.5
1
σ
Fig. 6. Hammett correlation for rate constants at 298 K for reaction of
L-cysteine ethyl ester 12 and N-benzoyloxy-N-benzyloxybenzamides 13–17
in [D4]methanol.
ln½Mꢃt ¼ ln½Mꢃ0 ꢀ k0t
ð2Þ
Underpseudo-first-orderconditions, thereactions proceededmore
rapidly and adventitious oxidation of the cysteine ethyl ester was
less evident in [D4]methanol. Bimolecular reaction rate constants,
k2, at different temperatures therefore afforded good Arrhenius
plots from which activation energies, entropies of activation
and rate constants at 298 K were determined. Data for 13
togetherwiththatfor the series ofN-(4-substitutedbenzoyloxy)-N-
benzyloxybenzamides 14–17 are presented in Table 1.
O
R1
R2O
O
OR2
R1
NR32
NR32
N
N
Activation energies and entropies of activation are of the
same order as those found previously for SN2 reactions of the
same series of mutagens with neutral N-methylaniline.[1,11] Both
nucleophilic reactions are characterized by large, negative DSz
values in accord with a high degree of charge separation and a
strong increase in solvation at the transition state. The computed
charge separation at the transition state for the reaction of
methanethiol with N-formyloxy-N-methoxyformamide sup-
ports this.[12] In particular, the high degree of carboxylate
character should be stabilized by benzoyloxyl groups bearing
electron-withdrawing substituents and the relative rate constants
at 298 K show this to be the case.
Fig. 6 indicates that the rate constants correlate with
Hammett s values, with r ¼ þ1.13 (r ¼ 0.985), in keeping with
a development of benzoate character in the transition state. A
previous study of the SN2 reactions of N-methylaniline
with N-benzoyloxy-N-benzyloxybenzamides correlated with
Hammett s constants with reaction constant, r ¼ þ1.7. The
initial, rate-determining steps in the two reaction processes are
not dissimilar. However, the fate of the intermediates in the two
reactions is quite different. Whereas the N-alkoxy-N-(alkylthio)
amides undergo a fast, secondary SN2 reaction at sulfur giving
21
22
23
Scheme 4.
the disulfide and hydroxamic ester, N-benzyloxy-N-
aminoamides 21 undergo, instead, the novel HERONC intramo-
lecular rearrangement to esters 22 and aminonitrenes 23
(Scheme 4).[1,6,10,25–29]
To date, no stable examples of N-alkoxy-N-(alkylthio)
amides have been reported in the literature. The reaction of
thiols at the anomeric nitrogen of N-acyloxy-N-alkoxyamides
has been modelled theoretically at the pBPDN(d)//HF/6–31G(d)
level and is predicted to lead to a stationary point for
the substitution product.[1,6,12] The intermediate N-alkoxy-N-
(alkylthio)amides from the reactions of N-acyloxy-N-alkoxy-
amides with thiols are, however, clearly unstable under the
reaction conditions, reacting with a second thiol molecule at
sulfur.
B3LYP/6–31G(d) calculations on model N-methoxy-N-
(methylthio)acetamide 24 (Chart 4) predicted four possible
conformations, in each of which the nitrogen is slightly
CHeteroatom Rearrangements on Nitrogen.

RESEARCH FRONT
450
S. A. Glover and M. Adams
O
Me
N
MeO
O
N
O
O
O
Me
Me
Me
Me
MeO
SMe
Me
OMe
SMe
OMe
HN
HN
N
Me
24
25
26
27
28
Chart 4.
(a)
(b)
C2
1.215 Å
C1
O2
C4
114.3°
1.408 Å
122.0°
1.420 Å
O1
N
1.717 Å
116.1°
S
C2–C1–N–O1 ꢀ ꢂ21.2ꢃ
O2–C1–N–S ꢀ 12.2ꢃ
O2–C1–N–O1 ꢀ 160.2ꢃ
C2–C1–N–S ꢀ ꢂ169.3ꢃ
C4–O1–N–S ꢀ ꢂ86.6ꢃ
C3–S–N–O1 ꢀ ꢂ79.0ꢃ
C3
Fig. 7. (a) B3LYP/6–31G(d) optimized geometry, and (b) LUMO of N-methoxy-N-(methylthio)acetamide 24.
pyramidal. The oxygen atoms may be cis or trans with respect
N-methoxy-N-(methylthio)acetamide 24 and N-methoxyaceta-
mide 26 have been estimated at 57 and 60 kJ molꢀ1 respectively,
,75 and 78% that of N,N-dimethylacetamide 25 (76 kJ molꢀ1),
whereas N-(methylthio)acetamide 27 has a resonance energy
some 90% that of N,N-dimethylacetamide 25. For comparison,
the resonance energy of N,N-dimethoxyacetamide 28 is com-
puted to be only 36% that of N,N0-dimethylacetamide. It is clear
that the combined electronegativity of substituents at nitrogen
strongly influences the resonance energy of amides by reducing
p-character of the nitrogen lone pair and localizing the electron
pair on nitrogen. The comparatively low electronegativity of
sulfur (2.58) relative to oxygen (3.44) results in the least impact
on amide resonance.
At the B3LYP/6–31G(d) level, the reaction of methanethiol
with 24 giving 26 and dimethyl disulfide (Scheme 5) is com-
puted to be exothermic by 54.4 and 45.4 kJ molꢀ1 with zero-
point energy and enthalpy contributions. In addition, the LUMO
of 24, shown in Fig. 7b, has both S–N s* and C¼O p* character,
suggesting that nucleophiles would react either at sulfur or at the
carbonyl carbon. On this basis, the reaction of a second thiol
molecule at the sulfur atom would not be unexpected.
to the N–C bond and the methyl substituents can be exo or endo
to the pyramid prescribed by the nitrogen and the three atoms
attached to it. The lowest-energy conformation has oxygens
trans and the thiomethyl and methoxyl methyls respectively
endo and exo (Fig. 7a). Winkler–Dunitz[30,31] pyramidality and
twist indices xN and t for this geometry were 328 and 4.58
respectively and the average angle at nitrogen is 117.58.D
Relative to other anomeric amides, substitution at nitrogen
with both sulfur and oxygen results in a much smaller deviation
from planarity; theoretical and experimental xN values of N,N-
bisoxo-substituted amides are closer to 608 or greater, with
average angles at nitrogen around 1088.[1,6,11,32,33] Nonetheless,
the pyramidal character suggests some loss of amide resonance
and, although the N–C bond lengths in anomeric amides are also
affected by the degree of hybridization at nitrogen,[6] the N–C
˚
˚
bond (1.408 A) is longer and C¼O bond (1.215 A) shorter than
found for N,N-dimethylacetamide 25 (respectively 1.378 and
˚
1.227 A) but very similar to those obtained for the lowest-energy
conformer of the parent hydroxamic ester, N-methoxyacetamide
˚
26 (respectively 1.394 and 1.219 A) at the same level of theory.
The computed gas-phase C¼O vibrational frequencies of
N,N-dimethylacetamide 25, model 24 and N-methoxyacetamide
26 were 1696, 1733 and 1738 cmꢀ1 respectively after appro-
priate scaling.[34] Clearly, the thiomethyl substituent has a
relatively small influence on the amide properties of the model
hydroxamic ester. In a recent study, we have estimated the
resonance energies of a range of mono- and bisheteroatom-
substituted amides using B3LYP/6–31G(d) (S. A. Glover
and A. Rosser, unpubl. data). The resonance energies of
The Me–O–N–SMe dihedral angle in 24 is ꢀ86.68 as
opposed to ꢀ79.08 for the Me–S–N–OMe dihedral angle,
suggesting a stronger anomeric effect in this direction. Anome-
ric interactions are not only influenced by the energies of the
lone pairs and the s* orbitals involved, but also by the sizes of
the respective orbitals, which could account for the observed
preference; the disparity between the sizes of valence p-orbitals
on sulfur and nitrogen would disfavour an nS–s* overlap
NO
when compared with an nO–s* anomeric interaction.[7]
NS
DA xN value of 608 would represent an sp3 hybridized nitrogen and a t value of 08 represents zero twist about the N–CO bond.

RESEARCH FRONT
SN2 Reactions of Thiols at Amide Nitrogen of N-Acyloxy-N-alkoxyamides
451
O
N
O
Me
MeO
Me
CH3SH
CH3SSCH3
ΔHreact ꢀꢂ45.4 kJmolꢂ1
SMe
OMe
HN
24
26
Scheme 5.
A strong anomeric effect and attendant charge redistribution
are largely responsible for the HERON rearrangement process
in both N-alkoxy-N-amino amides, where the high-energy lone
pair on the amino nitrogen destabilizes the N–O bond,[27,28,35,36]
and in N-acyloxy-N-alkoxyamides where the N–OAcyl bond
is weakened and carboxylate can stabilize developing negative
charge in the migration transition state.[37] In N-alkoxy-N-
(alkylthio)amides 24, HERON migrations would be disfavoured
by the high degree of amidicity that must be lost.
Experimental
Materials and Methods
NMR spectra were recorded in CDCl3, [D6]DMSO/D2O or
[D4]methanol on a Bruker Avance 300P FT NMR spectrometer
with a 5-mm 1H inverse/BB z gradient probe, operating at
300.13 MHz (1H) or 75.46 MHz (13C). Centrifugal chromato-
graphic separations were performed on a 7294T model Harrison
Research Chromatotron with plates coated with 2.0 mm of silica
gel 60 F254 (Merck). Flash chromatography columns were
charged with silica gel 60, 230–400 mesh. Glutathione, cysteine,
N-acetylcysteine, cysteine methyl ester hydrochloride (MEC)
and cysteine ethyl ester hydrochloride (EEC) were all available
commercially in greater than 98% purity and were used without
further purification.
Conclusion
N-Acyloxy-N-alkoxyamides are nitrogen electrophiles that
react with DNA by an SN2 reaction but, in line with previous
model studies with aromatic amines, they are also susceptible
to nucleophilic displacement of the N-acyloxyl group by
extraneous nucleophiles. Such reaction processes diminish
mutagenic activity by depletion of the concentration of elec-
trophilic reagent. This study has demonstrated that thiols,
notably glutathione, would also consume the mutagens in cells
replete with this substance. Together with methyl and ethyl
esters of the cysteine, the thiols replace carboxylate in a classical
SN2 reaction at the amide nitrogen but the intermediate N-
alkoxy-N-(alkylthio)amides are susceptible to a fast, secondary
reaction at sulfur by thiol, leading to disulfides and hydroxamic
esters. This represents disparate behaviour when compared with
other anomeric amides, such as N-amino-N-alkoxyamides or N-
acyloxy-N-alkoxyamides themselves, which tend to undergo the
HERON reactions.
Arrhenius activation energies and entropies of activation are
similar to those obtained for SN2 reactions of N-acyloxy-N-
alkoxyamides with aromatic amines, and the bimolecular reac-
tion processes are likely to be similar, in accord with previous
findings from computational studies. This is further supported
by a similar, positive Hammett correlation with a series of
N-benzyloxy-N-(4-substitutedbenzoyloxy)benzamides.
Computational Methods
Fully optimized ground states of amides 24–28 were computed
at the B3LYP/6–31G(d) level using MacSpartan08.[38]
Carbonyl stretch frequencies were obtained by frequency
analysis at B3LYP/6–31G(d) level and are scaled by 0.9614
according to Scott and Radom.[34] Geometries, energies and
unscaled vibrational frequencies are provided as an Accessory
Publication.
Synthesis of N-acyloxy-N-alkoxyamides
Synthesis of N-acetoxy-N-butoxybenzamide 6,[8] N-benzoyloxy-
N-benzyloxybenzamide 13,[8] N-(4-methoxybenzoyloxy)-N-
benzyloxybenzamide 14, N-(4-methylbenzoyloxy)-N-benzyloxy-
benzamide 15, N-(4-chlorobenzoyloxy)-N-benzyloxybenzamide
16 and N-(4-nitrobenzoyloxy)-N-benzyloxybenzamide 17[14]
have been described previously, as have the hydroxamic esters
N-butoxybenzamide 7 and N-benzyloxybenzamide 18.[8] All
N-acyloxy-N-alkoxyamides were purified by centrifugal or flash
column chromatography using hexane/ethyl acetate as eluent
before use.
Calculations on model structures indicate that the intermedi-
ates from these reaction processes retain a good deal of amide
character, and they are not dissimilar to hydroxamic esters.
This contrasts with N,N-bisoxyl-substituted amides such as
N-acyloxy-N-alkoxyamides and N,N-dialkoxyamides, which
possess highly pyramidalized nitrogens and radically reduced
amideresonance.ThetheoreticalpropertiesofmodelN-methoxy-
N-(methylthio)acetamideaswellasthepreferredsecondarymode
of reaction with thiols can be rationalized in terms of the lower
electronegativityofsulfur comparedwithnitrogenandoxygen. In
this case, the combined electronegativity of oxygen and sulfur
appears to be insufficient to overcome the stabilization gained
from amide resonance. A HERON rearrangement of N-alkoxy-
N-(alkylthio)amides would require loss of amide character,
whereas SN2 displacement of hydroxamic ester leading to dis-
ulfide formation would result in retention of a similar degree of
resonance-stabilization in the hydroxamic ester.
Kinetic Studies
General Method[2,10,11,15]
Kinetic studies for the determination of both bimolecular
and pseudo-unimolecular rate constants involved monitoring
the reaction between N-acyloxy-N-alkoxybenzamides and the
appropriate thiol. For each of the studies, the reagents were
added separately to an NMR tube, either mutagen or thiol first,
followed by acquisition of a 1H NMR spectrum. The tube was
ejected for addition of the second reagent and the reaction
mixture was capped, sealed and mixed by inverting the tube
several times before return to the thermostatted probe of the
NMR. After optimum shim was achieved, an automated pro-
gram was initiated that acquired and stored successive spectra
of the reaction at predetermined time intervals. The exact time of
mixing and the concentrations of the starting reagents were
recorded. The extent of reaction was ultimately measured
by recording the areas of peaks characteristic of the reagents
in processed spectra. For bimolecular rate studies, the
This study suggests that in mammalian cells, which have
high concentration of GSH, N-acyloxy-N-alkoxyamides would
be consumed and are likely to be less mutagenic.

RESEARCH FRONT
452
S. A. Glover and M. Adams
Table 2. Bimolecular rate constants (k2) for reactions of N-acetoxy-N-butoxybenzamide 6 with glutathione 4 in
DMSO/D2O at 303 K
6 [mmol Lꢀ1
]
4 [mmol Lꢀ1
]
[4]/[6]
Graphical ratioA
104k2303 [L molꢀ1 sꢀ1
]
r2
30.3B
29.5B
43.8B
26.3
89.8
91.1
2.96
3.09
2.20
3.89
3.74
3.86
3.00
4.08
3.56
2.94
3.60
4.17
336
174
612
127
138
176
0.989
0.996
0.966
0.995
0.989
0.985
96.3
102.2
98.9
26.5
26.5
102.2
ARatio from extrapolation of concentration versus time plots back to t0 using a fourth-order polynomial fit.
BMutagen added first.
Table 3. Bimolecular rate constants (k ) for the reaction of cysteine ethyl ester 12 with N-benzoyloxy-N-benzyloxybenzamides 13 17 in
]
2
[D4]methanol at different temperatures
13 (Z ¼ H)
14 (Z ¼ OMe)
15 (Z ¼ Me)
16 (Z ¼ Cl)
17 (Z ¼ NO2)
T [K]
104k2
T [K]
104k2
T [K]
104k2
T [K]
104k2
T [K]
104k2
[L molꢀ1 sꢀ1
]
[L molꢀ1 sꢀ1
]
[L molꢀ1 sꢀ1
]
[L molꢀ1 sꢀ1
]
[L molꢀ1 sꢀ1
]
253
263
273
283
293
303
313
3.83
6.47
273
283
303
313
1.78
11.53
263
273
283
293
303
313
7.29
15.06
28.90
28.30
66.97
158.22
243
253
263
273
283
293
4.79
5.90
253
263
273
283
20.95
50.93
16.37
21.81
30.75
90.09
93.68
49.13
13.00
41.02
55.03
220.09
73.11
104.23
235.60
disappearance of both products was monitored using suitable,
discrete resonances for each reactant. The actual initial
starting concentration of each reagent was determine by back-
extrapolation of the concentration versus time plot to t0, using
a fourth-order polynomial. For pseudo-unimolecular studies,
the concentration of substrate was monitored from a suitable
resonance and the ratio of thiol to mutagen was ,10:1.
were made up. The N-acetoxy-N-butoxybenzamide solution
(350 mL) was loaded into a high-precision NMR tube. A proton
spectrum was acquired at optimum shim, after which GSH
solution (150 mL) was micropipetted into the NMR tube. The
sample was returned to the probe for data acquisition as
described above.
The reaction was repeated a number times by both methods,
yielding reaction rate constants showing some variability
(Table 2) but with an average value in the region of
Reactions with Glutathione in Aqueous [D6]DMSO
260.5 ꢂ 188 ꢁ 10ꢀ4 L molꢀ1 sꢀ1 E
.
The bimolecular rate constant was determined by monitoring
the changes in the concentration of N-acetoxy-N-butoxybenzamide
6 and glutathione 4 with time in micro-reaction mixtures in
[D6]DMSO/D2O.
Reactions of L-Cysteine Ethyl Ester 12 with
N-Benzoyloxy-N-benzyloxybenzamides 13–17
in [D4]Methanol
Method One
The pseudo-first-order rate constant was determined by
monitoring changes in concentration of N-benzoyloxy-N-
benzyloxybenzamides 13–17 with time in micro-reaction mix-
tures with L-cysteine ethyl ester 12. The ratio between thiol and
mutagen was set to approximately 10:1.
A stock solution of L-cysteine ethyl ester hydrochloride
(0.145 g, 78.1 mmol) was made up in [D4]methanol (1 mL).
N-Benzoyloxy-N-benzyloxybenzamides 13–17 (,2 mg) were
loaded into an ultra-high-precision NMR tube and dissolved
in [D4]methanol (450 mL). The sample was shimmed, and a 1H
NMR spectrum acquired. The sample was retrieved from the
magnet and 50 mL of the stock solution of L-cysteine ethyl ester
hydrochloride was added by micropipette. After return to the
probe, acquisition of data utilized the method described above.
DMSO (350 mL) and D2O (150 mL) were mixed in a high-
precision NMR tube followed by the addition of glutathione 4
(10.0–25.0 mg). The sample was shimmed and a 1H NMR
spectrum was acquired. Mutagen 6 was then weighed by dif-
ference into the NMR tube ([4]:[6] typically 3:1), followed by
data acquisition by the method described above. The extent of
reaction and relative reactant ratios were obtained from areas of
the methyl singlet at d 2.13 of the mutagen and the diastereotopic
AB spin systems of the thiol.
Method Two
Stock solutions of GSH (15.9 mg in 200 mL) in D2O and
N-acetoxy-N-butoxybenzamide (16.0 mg in 1.8 mL) in DMSO
EThis variability can be attributed to a number of difficulties with the method. Theoretically, the ratio of the y-intercepts of the glutathione and the mutagen
obtained by back extrapolation to time zero, the time of mixing, should correspond to the ratio of the concentrations of the starting materials. The graphical
ratios were found to differ from the measured concentration ratio of the starting reagents, affecting both the log term and the denominator in Eqn 1. We attribute
this either to oxidation of glutathione or decomposition of N-acetoxy-N-butoxybenzamide under the reaction conditions.

RESEARCH FRONT
SN2 Reactions of Thiols at Amide Nitrogen of N-Acyloxy-N-alkoxyamides
453
Arrhenius Studies
[13] S. A. Glover, G. Mo, J. Chem. Soc., Perkin Trans. 2 2002, 1728.
doi:10.1039/B111250N
[14] S. A. Glover, G. P. Hammond, A. M. Bonin, J. Org. Chem. 1998, 63,
Kinetic studies of the reaction of L-cysteine ethyl ester
hydrochloride 12 and amides 13–17 were undertaken in
[D4]methanol over a range of temperatures, which varied
depending on the substrate. The sample and probe were allowed
to equilibrate at each temperature for at least 10 to 15 min before
acquisition. Rate constants at associated temperatures are given
in Table 3.
9684. doi:10.1021/JO980863Z
[15] K. L. Cavanagh, S. A. Glover, H. L. Price, R. R. Schumacher, Aust.
J. Chem. 2009, 62, 700. doi:10.1071/CH09166
[16] T. M. Banks, PhD Thesis, Reactivity, Mutagenicity and DNA damage
of N-Acyloxy-N-alkoxyamides. University of New England, Armidale,
2003.
[17] J. J. Campbell, S. A. Glover, J. Chem. Soc., Perkin Trans. 2 1992,
Accessory Publication
1661. doi:10.1039/P29920001661
[18] N. De Kimpe, R. Verhe´, The Chemistry of a-Haloketones, a-Haloal-
dehydes and a-Haloimines (Eds S. Patai, Z. Rappoport) 1988, p. 38
(John Wiley & Sons: Chichester, UK).
A publication is available from the Journal’s website that
illustrates:
[19] J. W. Thorpe, J. Warkentin, Can. J. Chem. 1973, 51, 927. doi:10.1139/
V73-137
[20] J. Vina, Glutathione: Metabolism and Physiological Functions 1990
1. NMR spectrum in [D4]methanol showing lack of reaction of
N-acetylcysteine 10 with N-benzoyloxy-N-benzyloxybenz-
amide 13.
(CRC Press, Inc.: Boca Raton, FL).
2. B3LYP/6–31G(d)-computed energies (with zero point
energy (ZPE) and enthalpy corrections), geometries and
carbonyl stretch frequencies for N-methoxy-N-(methylthio)
acetamide 24, N,N-dimethylacetamide 25, N-methoxyaceta-
mide 26, N-(methylthio)acetamide 27 and N,N-dimethoxya-
cetamide 28 as well as computed energies (with ZPE and
enthalpy corrections) and geometries for methanethiol and
dimethyl disulfide.
[21] S. C. Mitchell, Biological Interactions of Sulfur Compounds 1996
(Taylor & Francis: London).
[22] A. L. Fluharty, in The Chemistry of the Thiol Group, Part 2 (Ed.
S. Patai) 1974, p. 615 (John Wiley and Sons: London).
[23] K. Zhang, P. Mack, K. P. Wong, Int. J. Oncol. 1998, 12, 871.
[24] A. Ohno, S. Oae, in Organic Chemistry of Sulfur (Ed. S. Oae) 1977,
p. 159 (Plenum Press: New York).
[25] S. A. Glover, Truly Australian – the HERON reaction. Chemistry in
Australia 2007, 74(7), 6.
[26] S. A. Glover, in Merck Index, Organic Name Reactions ONR-43 (Ed.
M. J. O’Neil) 2006 (Merck & Co., Inc.: Whitehouse Station, NJ).
[27] S. A. Glover, A. Rauk, J. M. Buccigross, J. J. Campbell, G. P.
Hammond, G. Mo, L. E. Andrews, A.-M. E. Gillson, Can. J. Chem.
2005, 83, 1492. doi:10.1139/V05-150
[28] S. A. Glover, G. Mo, A. Rauk, Tetrahedron 1999, 55, 3413.
doi:10.1016/S0040-4020(98)01151-X
[29] J. M. Buccigross, S. A. Glover, J. Chem. Soc., Perkin Trans. 2 1995,
References
[1] S. A. Glover, in Advances in Physical Organic Chemistry (Ed.
J. Richard) 2008, Vol. 42, pp. 35–123 (Elsevier: London).
[2] L. E. Andrews, A. M. Bonin, L. E. Fransson, A.-M. E. Gillson, S. A.
Glover, Mutat. Res. 2006, 605, 51.
[3] L. E. Andrews, T. M. Banks, A. M. Bonin, S. F. Clay, A.-M. E. Gillson,
S. A. Glover, Aust. J. Chem. 2004, 57, 377. doi:10.1071/CH03276
[4] T. M. Banks, A. M. Bonin, S. A. Glover, A. S. Prakash, Org. Biomol.
Chem. 2003, 1, 2238. doi:10.1039/B301618H
[5] A. M. Bonin, T. M. Banks, J. J. Campbell, S. A. Glover, G. P.
Hammond, A. S. Prakash, C. A. Rowbottom, Mutat. Res. 2001,
494, 115.
[6] S. A. Glover, in The Chemistry of Hydroxylamines, Oximes and
Hydroxamic Acids, Part 2 (Eds Z. Rappoport, J. F. Liebman) 2009,
pp. 839–923 (Wiley: Chichester, UK).
595. doi:10.1039/P29950000595
[30] J. D. Dunitz, X-Ray Analysis and Structure of Organic Molecules 1979
(Cornell University Press: London).
[31] F. K. Winkler, J. D. Dunitz, J. Mol. Biol. 1971, 59, 169. doi:10.1016/
0022-2836(71)90419-0
[32] A. Rauk, S. A. Glover, J. Org. Chem. 1996, 61, 2337. doi:10.1021/
JO9521817
[33] A.-M. E. Gillson, S. A. Glover, D. J. Tucker, P. Turner, Org. Biomol.
Chem. 2003, 1, 3430. doi:10.1039/B306098P
[7] S. A. Glover, Tetrahedron 1998, 54, 7229. doi:10.1016/S0040-4020
(98)00197-5
[8] J. J. Campbell, S. A. Glover, G. P. Hammond, C. A. Rowbottom,
J. Chem. Soc., Perkin Trans. 2 1991, 2067. doi:10.1039/P29910002067
[9] A. M. Bonin, S. A. Glover, G. P. Hammond, J. Chem. Soc., Perkin
Trans. 2 1994, 1173. doi:10.1039/P29940001173
[10] J. J. Campbell, S. A. Glover, J. Chem. Res. (S) 1999, 8, 474.
doi:10.1039/A903555I
[11] J. J. Campbell, S. A. Glover, J. Chem. Res. (M) 1999, 8, 2075.
[12] S. A. Glover, Arkivoc, 2001, Part xii, Issue in Honour of O. S. Tee.
http://www.arkat-usa.org/get-file/19825/, published on the web
29/1/2001.
[34] A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502. doi:10.1021/
JP960976R
[35] J. M. Buccigross, S. A. Glover, G. P. Hammond, Aust. J. Chem. 1995,
48, 353.
[36] S. A. Glover, A. Rauk, J. Org. Chem. 1999, 64, 2340. doi:10.1021/
JO982048P
[37] J. P. Johns, A. van Losenoord, C. Mary, P. Garcia, D. Pankhurst,
A. Rosser, S. A. Glover, Aust. J. Chem. 2010, 63, 1717. doi:10.1071/
CH10350
[38] Spartan ’08 for Macintosh 2008, Ver. 1.2.1 (Wavefunction Inc.:
Irvine, CA).