A comparison of the reactivity and mutagenicity of n-(benzoyloxy)-n-(benzyloxy)benzamides

Article abstract of DOI:10.1021/jo980863z

A new series of N-(acyloxy)-N-alkoxybenzamides, AT-(benzoyloxy)-JV-(benzyloxy)benzamides 7 have been synthesized and have been found to be direct acting mutagens in Salmonella TA100. They undergo A_Al1 solvolysis to give N-benzoyl-N-(benzyloxy)nitrenium ions 3 under conditions of acid catalysis as well as unusual B_Al2 reactions at nitrogen with hydroxide. The latter process affords as intermediates the anomeric hydroxamic esters 4 which rearrange intramolecularly to esters in a HERON reaction. Rates of acid-catalyzed solvolysis and reaction with hydroxide ions correlate with Hammett σ values with low sensitivity (ρ = +0.32 and +0.55, respectively) in accordance with the A_Al1 and B_Al2 mechanisms. Mutagenicity for the series also appears to correlate with Hammett σ values but with low, negative sensitivity (p = -0.57), and their biological activity may be attributable to their stability under conditions of the Ames assay and hydrophobic binding to DNA, rather than their chemical reactivity.

Full text of DOI:10.1021/jo980863z

9684
J . Org. Chem. 1998, 63, 9684-9689
A Com p a r ison of th e Rea ctivity a n d Mu ta gen icity of
N-(Ben zoyloxy)-N-(ben zyloxy)ben za m id es
Stephen A. Glover* and Gerard P. Hammond
Division of Chemistry, School of Physical Sciences and Engineering, University of New England,
Armidale, New South Wales 2351, Australia
Antonio M. Bonin
Toxicology unit, Worksafe Australia, GPO Box 58, Sydney, New South Wales, 2351, Australia
Received May 7, 1998
A new series of N-(acyloxy)-N-alkoxybenzamides, N-(benzoyloxy)-N-(benzyloxy)benzamides 7 have
been synthesized and have been found to be direct acting mutagens in Salmonella TA100. They
undergo AAl1 solvolysis to give N-benzoyl-N-(benzyloxy)nitrenium ions 3 under conditions of acid
catalysis as well as unusual BAl2 reactions at nitrogen with hydroxide. The latter process affords
as intermediates the anomeric hydroxamic esters 4 which rearrange intramolecularly to esters in
a HERON reaction. Rates of acid-catalyzed solvolysis and reaction with hydroxide ions correlate
with Hammett σ values with low sensitivity (F ) +0.32 and +0.55, respectively) in accordance
with the AAl1 and BAl2 mechanisms. Mutagenicity for the series also appears to correlate with
Hammett σ values but with low, negative sensitivity (F ) -0.57), and their biological activity may
be attributable to their stability under conditions of the Ames assay and hydrophobic binding to
DNA, rather than their chemical reactivity.
In tr od u ction
ductive substituents facilitate nitrenium ion formation
and substrates with para inductive groups correlate with
Hammett σ constants with a F value of -1.61. Substrates
with positively mesomeric groups, however, undergo con-
certed decomposition to benzyl cations and the rates for
this process correlate with σ+ with a F value of 2.0.4
Under neutral conditions the mutagens are relatively
stable, although in aqueous medium they ultimately
decompose by an autocatalytic process.2,3
The fate of the N-alkoxynitrenium ions formed in these
processes has been fully explored, and the products can
all be accounted for by formation of the N-alkoxyhydrox-
amic acids 4 (from interception by water) which decom-
pose by acid-catalyzed processes to give alcohols and
aldehydes and a non-acid-catalyzed rearrangement to
give esters.4,5 Recent theoretical studies in these labora-
tories as well as experimental results suggest that the
latter process proceeds via the conjugate anion of the
hydroxamic acid 5 and is a special case of the so-called
HERON6 reaction of anomeric amides, amides that are
geminally substituted at nitrogen with two heteroatoms
(RCONXY, Figure 1).7 A heteroatom bearing a high-
energy pair of electrons such as nitrogen (Y ) N), or in
this case an oxyanion (Y ) O-), initiates rearrangement
of X from N to C through a strong anomeric overlap with
the σ* orbital of the adjacent N-X bond. Anomeric
interactions at nitrogen in bisheteroatom-substituted
amides are enhanced when X is strongly electronegative
resulting in a low-lying σ* orbital.8,9
N-Acetoxy-N-alkoxybenzamides 1 are a class of com-
pounds that we have found to be universally mutagenic
toward Salmonella TA100 without metabolic activation.1-4
To date we have synthesized some 40 of these compounds
with a variety of substituents on both the benzoyl and
alkoxy side chains. They undergo acid-catalyzed solvoly-
sis by the unusual AAl1 mechanism which involves
reversible protonation at the acetoxy carbonyl followed
by unimolecular O-N cleavage to form alkoxynitrenium
ions 3.3,4 Para substituents on the benzoyl ring in 1
support this mechanism; rates of solvolyses correlate with
Hammett σ+ constants with a modest F value of -1.4 in
accordance with incipient stabilization of positive charge
in the transition states by electron donor groups.3 Para
substituents on the benzyloxy ring of 2 exert a similar
effect since the charge deficiency in alkoxynitrenium ions
is strongly delocalized onto oxygen. Thus positively in-
(5) Hammond, G. P. Ph.D. University of New England, 1997.
(6) Heteroatom rearrangements on nitrogen (HERON). First de-
scribed in a paper presented at the Second Heron Island Conference
on Reactive Intermediates and Unusual Molecules, Heron Island,
Australia, 1994.
(7) Buccigross, J . M.; Glover, S. A. J . Chem. Soc., Perkin Trans. 2
1995, 595.
(1) Gerdes, R. G.; Glover, S. A.; Ten Have, J . F.; Rowbottom, C. A.
Tetrahedron Lett. 1989, 30, 2649.
(2) Campbell, J . J .; Glover, S. A.; Rowbottom, C. A. Tetrahedron Lett.
1990, 31, 5377.
(3) Campbell, J . J .; Glover, S. A.; Hammond, G. P.; Rowbottom, C.
A. J . Chem. Soc., Perkin Trans. 2 1991, 2067.
(4) Bonin, A. M.; Glover, S. A.; Hammond, G. P. J . Chem. Soc.,
Perkin Trans. 2 1994, 1173.
(8) Glover, S. A.; Rauk, A. J . Org. Chem. 1996, 61, 2337.
(9) Glover, S. A. Tetrahedron 1998, 54, 7229.
10.1021/jo980863z CCC: $15.00 © 1998 American Chemical Society
Published on Web 12/01/1998

N-(Benzoyloxy)-N-(benzyloxy)benzamides
J . Org. Chem., Vol. 63, No. 26, 1998 9685
N-chloro-N-(benzyloxy)benzamide and the appropriate
sodium benzoate in anhydrous acetone. Most were oils
which could be purified by centrifugal chromatography
but were only stable for long periods under dry conditions
at low temperature. Accordingly they were characterized
1
by H and 13C NMR and IR spectroscopies.
Acid -Ca ta lyzed Solvolysis Rea ction s of 7. Like
N-acetoxy-N-alkoxybenzamides 1 and 2, N-(benzoyloxy)-
N-(benzyloxy)benzamides 7 solvolyze in 25% D2O-
CD3CN. The rates were obtained using 300 MHz NMR
spectroscopy by monitoring the disappearance of the
benzyloxy methylene singlet of the starting material. The
reactions were pseudo-first-order over several half-lives
and rate constants, evaluated for each mutagen at 10 K
intervals over the temperature range of 298-338 K, and
afforded excellent Arrhenius plots from which Ea and ∆Sq
were obtained (Table 1).
The rate data at 308 K reveals that the reaction is
promoted by electron withdrawing para substituents and
analysis of Arrhenius parameters shows an excellent
isokinetic relationship (Figure 2) confirming that a
uniform mechanism is operative throughout. Substrates
with electron-withdrawing substituents have the lowest
enthalpies of activation and the tightest transition states.
The converse is true for substrates with electron-donating
groups. The positive ∆Sq values are in the range observed
for AAl1 hydrolysis of tertiary alkyl esters (+54 J K-1
mol-1 for tert-butyl acetate)17 and in contrast to those
reported for normal AAc2 hydrolysis of benzoate esters
which are strongly negative.18,19
F igu r e 1. HERON reaction in an anomeric amide.
We have also shown that para-substituted N-acetoxy-
N-butoxybenzamides react bimolecularly in methanol
with N-methylaniline, a model we have used for the
nucleosides bearing an exocyclic amino group such as
guanosine.10 These reactions involve rate-determining
nucleophilic attack at the amide nitrogen, and the
resultant N-butoxy-N-(N′-methylanilino)benzamides 6
rearrange to 1-methyl-1-phenyldiazene and butyl ben-
zoates, again through the HERON process (Figure 1, Y
) N(Me)Ph, X ) OBu).9
Recent DNA damage studies have indicated that the
mutagens react with guanine residues at N711 (its most
nucleophilic center12,13) and that two mechanisms are
possible; damage could initially be ascribed to their
ability to generate electrophilic alkoxynitrenium ions,
although, on the basis of the results we have to date, such
formation, intracellularly, would of necessity require an
acid or Lewis acid catalyst.3 Alternatively, as indicated
by their reaction with nucleophilic amines, the mutagens
may themselves behave as electrophiles in reacting with
nucleophilic centers on the nucleotide bases. These two
processes have also been proposed for the interaction of
the analogous metabolites of aromatic amines, N-aryl-
N,O-diacetylhydroxylamines.14-16
By analogy with those of their N-acetoxy analogues 1
and 2,3,4 the activation enthalpies and entropies are
determined by two processes: protonation at the benzoyl
carbonyl and N-OCOAr bond stretching as heterolysis
proceeds. Both are influenced by the para substituent but
in opposite senses.
To shed further light on the process by which N-
(acyloxy)-N-alkoxybenzamides interact with DNA, we
have synthesized a new series of mutagens, N-(benzoyl-
oxy)-N-(benzyloxy)benzamides 7a -h , and have investi-
gated their acid-catalyzed solvolysis, their reactions with
organic and inorganic bases, and their mutagenicity.
Electron-donating substituents increase the basicity of
the benzoyl carbonyl, which promotes protonation and
hence shifts the equilibrium constant Ka, to the right
(Scheme 1). However, electron-donating substituents,
while increasing the concentration of the protonated
intermediate, decrease the partial positive charge at the
carbonyl carbon thereby reducing the ease of N-O bond
heterolysis (and magnitude of k′x). These opposing effects
manifest themselves in low overall sensitivity to the
electronic effects of the para substituents. Rate constants
at 308 K (Table 1) correlate with Hammett σ substituent
constants with positive slope of low sensitivity (F ) 0.32,
r2 ) 0.87, Figure 3). A similar sensitivity has been
reported for normal acid-catalyzed solvolysis of alkyl
benzoates where substituents also influence protonation
and solvolysis steps in the opposite sense.19 Reversible
protonation of benzoic acids has a reported F+ value of
-1,20 thus a similar sensitivity in the protonation of 7
would yield a F value for the heterolysis step in the region
of +1.5 and entirely consistent with build up of negative
charge at a center indirectly conjugated to the substitu-
ent.21 Protonation to give the intermediate in Scheme 1
Resu lts a n d Discu ssion
N-(Benzoyloxy)-N-(benzyloxy)benzamides were conve-
niently prepared by a Finkelstein-type process from
(10) Campbell, J . J .; Glover, S. A. J . Chem. Soc., Perkin Trans. 2
1992, 1661.
(11) Banks, T. M.; Glover, S. A.; Prakash, A. S. Unpublished results.
(12) Sugiyama, H.; Saito, I. J . Am. Chem. Soc. 1996, 118, 7063.
(13) Pullman, A.; Pullman, B. Int. J . Quantum Chem., Symp. 1980,
7, 245.
(17) Adam, K. R.; Lauder, I.; Stimson, V. R. Aust. J . Chem. 1962,
15, 467.
(18) Normal AAc2 acid-catalyzed hydrolysis of esters have negative
∆Sq values of between 100 and 150 J K-1 mol-1. For a compilation
see, ref 19, p 514.
(14) Novak, M.; Kennedy, S. A. J . Am. Chem. Soc. 1995, 117, 574.
(15) McClelland, R. A. Tetrahedron 1996, 52, 6823.
(16) Ulbrich, R.; Famulok, M.; Bosold, F.; Boche, G. Tetrahedron
Lett. 1990, 31, 1689.
(19) Isaacs, N. S. Physical Organic Chemistry, 2nd ed.; Longman
Scientific and Technical: New York, 1995.
(20) Ritchie, C. D.; Virtanen, P. O. I. J . Am. Chem. Soc. 1972, 94,
1589.

9686 J . Org. Chem., Vol. 63, No. 26, 1998
Glover et al.
Ta ble 1. Bim olecu la r Ra te Con sta n ts a n d Ar r h en iu s Da ta for Acid -Ca ta lyzed Solvolysis of 7 a t 308 K
X
ln A
∆Sq (J K-1 mol-1
)
Ea (kJ mol-1
)
102kX (308 K)a
r2
MeO
Me
H
41.2 (1.3)
40.5 (1.2)
39.44 (0.58)
38.83 (0.84)
39.1 (1.3)
40.1 (1.2)
36.2 (1.3)
37.0 (1.7)
89.2 (11.1)
83.9 (10.0)
74.7 (4.8)
69.8 (7.0)
71.4 (10.8)
80.2 (10.0)
47.4 (10.8)
54.4 (14.1)
119.6 (3.5)
118.2 (3.0)
115.1 (1.6)
113.4 (2.2)
113.2 (3.5)
115.4 (3.3)
105.6 (3.3)
106.9 (4.4)
0.395 (0.018)
0.369 (0.015)
0.411 (0.008)
0.424 (0.010)
0.578 (0.031)
0.689 (0.030)
0.630 (0.035)
0.863 (0.076)
0.9974
0.9980
0.9992
0.9988
0.9972
0.9976
0.9970
0.9950
Cl
CHO
CF3
CN
NO2
a
Interpolated rate constants at 308 K.
Sch em e 2
F igu r e 2. Isokinetic relationship for the acid-catalyzed sol-
volysis of 7a -h .
temperature resulted in the rapid formation of benzyl
benzoates. Similarly, N-acetoxy-N-butoxy-p-chloroben-
zamide (8a ) and N-acetoxy-N-(benzyloxy)benzamide (8b)
afforded butyl p-chlorobenzoate and benzyl benzoate,
respectively. A crossover experiment using both mu-
tagens resulted in the exclusive formation of butyl
p-chlorobenzoate (46%) and benzyl benzoate (43.3%)
esters along with the hydrolysis products, p-chlorobenzoic
and benzoic acids. This result indicates that ester forma-
tion involves an intramolecular process. The formation
of esters in the acid-catalyzed solvolyses described above
involves the intermediate hydroxamic acids 4.5,7 These
hydroxamic acids can be formed from the reaction of the
mutagen with base by two processes (Scheme 2, path-
ways i and ii) and the presence of excess base would
ensure conversion to the conjugate anion 5 resulting in
the HERON reaction. Pathway ii involves a BAl2 rear-
side attack at the nitrogen, and while reactions of this
type are rare, they would in this case release para-
substituted benzoate anions and should proceed under
moderate influence of the para substituents on the
leaving group. Anomeric weakening of the N-OCOAr
bond would be expected with electron-withdrawing para
substituents since these would lower the energy of the
F igu r e 3. Hammett plot for acid-catalyzed solvolysis of 7a -h
at 308 K.
Sch em e 1
would result in a significant lowering in energy of the
N-OCO+HAr σ* orbital and the strong nO-σ*NO ano-
meric interaction at nitrogen leads, in this case, to
elimination of benzoic acid.9 The effect can be viewed as
being similar to that observed when thermally stable
N-alkoxy-N-chlorobenzamides 9 are treated with Lewis
acids. Complexation of chlorine with silver ions also
results in the formation of N-alkoxy-N-acylnitrenium
ions, and we and others have utilized this reaction in the
formation of novel benzoxazines and benzoxazepines as
well as N-alkoxybenzolactams by intramolecular cycliza-
tion of the electrophilic nitrenium ions onto aromatic
rings.22-27
(22) Glover, S. A.; Rowbottom, C. A.; Scott, A. P.; Schoonraad, J . L.
Tetrahedron 1990, 46, 7247.
(23) Glover, S. A.; Goosen, A.; McCleland, C. W.; Schoonraad, J . L.
J . Chem. Soc., Perkin Trans. 2 1984, 2255.
(24) Glover, S. A.; Goosen, A.; McCleland, C. W.; Schoonraad, J . L.
Tetrahedron 1987, 43, 2577.
Rea ction s of 7 w ith Hyd r oxid e Ion s. Treatment of
7a -h , with dilute aqueous sodium hydroxide at room
(25) Kikugawa, Y.; Kawase, M. J . Am. Chem. Soc. 1984, 106, 5728.
(26) Kawase, M.; Kitamura, T.; Kikugawa, Y. J . Org. Chem. 1989,
54, 3394.
(21) log(kX/kH) ) log(KX/KH) + log(k′X/k′H) ) -1.0σ+ + Fσ and a plot
of (log(kX/kH) + σ+) vs σ yields a better correlation (r2 ) 0.94) and a F
value for the heterolysis step in the region of +1.5.
(27) Kikugawa, Y.; Shimada, M. Chem. Lett. 1987, 1771.

N-(Benzoyloxy)-N-(benzyloxy)benzamides
J . Org. Chem., Vol. 63, No. 26, 1998 9687
F igu r e 5. Hammett correlation for the base hydrolysis of
7a ,c-e,g,h .
F igu r e 4. Pseudo-first-order kinetic plot for the reaction of
7a with excess base at 275.4 K.
Ta ble 3. Am es Mu ta gen icity Levels of 7 a t 1 µm ol/p la te
Ta ble 2. Bim olecu la r Ra te Con sta n ts for th e Rea ction of
Hyd r oxid e w ith 7 a t 275.4 K
X
revertantsa
a
X
kX2 (L mol-1 s-1
)
r2
log(kX2 /kH2 )
MeO
Me
1661.7
2758.5
2511.1
2949.6
1528.8
841.1
Me
H
Cl
CHO
CN
NO2
2.27 (0.39)
3.16 (0.65)
3.82 (0.69)
4.91 (0.59)
6.97 (0.09)
8.14 (1.26)
0.9716
0.9998
0.9990
0.9860
0.9998
0.9541
-0.144 (0.135)
0.000 (0.149)
0.082 (0.105)
0.191 (0.118)
0.343 (0.100)
0.410 (0.129)
But
H
Cl
CHO
CN
NO2
1104.3
606.9
a
Bimolecular rate constants were obtained from reactions at
different concentrations of hydroxide and were determined by
linear regression analysis of plots of k′ vs [HO-].
a
Relative mutagenicities at 1 µmol/plate. The mutagenicity data
was obtained over the concentration range 0-0.25 µmol, and the
slopes (mutagenicities at 1 µmol/plate) were calculated from the
linear dose-response region for each component. Results in
different sets of assays were scaled relative to the mutagenicity
of 7a . Mutagen 7i (X ) But), not synthesized for the original rate
studies, was available for Ames testing and is included as a
member of the series.
σ* orbital and BAl2 substitution by base rather than rate-
determining attack at the benzoyloxy carbonyl (the BAc2
process) may be more favorable in these cases.9
Pathway i involves the normal BAc2 mechanism and
attack at the carbonyl of the benzoyloxy moiety, a process
which should be under significantly greater control of a
para substituent on the ring. Rates of base-catalyzed
solvolyses of benzoate esters correlate with Hammett σ
values with F ) 2.0-2.4 and reflect the greater impor-
tance of attack at the carbonyl carbon.19 In these systems,
anomeric effects at nitrogen should disfavor this process
since negative charge would be delocalized onto the
acyloxy group.
N-methylaniline and N-acetoxy-N-butoxybenzamides 1,10
there is unequivocal evidence for attack of N-methyl-
aniline through nitrogen at the benzamide nitrogen.
Products are esters formed from the HERON reaction.7,29
The susceptibility of N-(acyloxy)-N-alkoxybenzamides
to nucleophilic attack at nitrogen suggests that this
process could play a role in the mutagenesis of this class
of compounds; DNA nucleotides are inherently nucleo-
philic at nitrogen and oxygen centers.13,30-33
Mu ta gen icity of 7. The mutagenicity levels for 7 were
obtained in Salmonella TA100 without metabolic activa-
tion. All substrates gave satisfactory dose response curves
from which induced revertants at 1 µmol/plate were
derived (Table 3). Relative mutagenicities show a possible
correlation with Hammett σ constants with low sensitiv-
ity (Figure 6). A line of best fit (r2 ) 0.79) has a negative
gradient with F ) -0.55. While sensitivities to substit-
uents in reactivity studies are similar, nucleophilic
substitution at nitrogen (with HO-, methylaniline28) and
acid-catalyzed solvolysis are both favored by electron-
withdrawing substituents resulting in Hammett plots of
positive slope. This structure-activity relationship there-
fore suggests that the ease of neither process plays a
significant role in determining mutagenicity.
To distinguish between pathways i and ii, the rates
of base solvolyses were determined for members of
series 7.
The bimolecular reaction of 7a at low temperature
(275.4 K) and at low hydroxide concentration was moni-
tored by analytical HPLC, and pseudo-first-order kinetics
was obtained by ensuring that the relative concentration
of base was significantly greater than the concentration
of mutagen (Figure 4). Under these conditions d[7a ]/dt
) -k′[7a ] where k′ ) k2 [HO-] and is the pseudo-first-
order reaction rate constant.
The pseudo-first-order rate constant k′ was linearly
dependent upon the concentration of base and the bimo-
lecular rate constants for the reaction of hydroxide with
7a ,c-e,g,h , kX2 , at 275.4 K are given in Table 2.
The least-squares fit of log(kX2 /k2H) versus Hammett σ
values gave F ) 0.55 (Figure 5) which is inconsistent with
normal BAc2 hydrolysis process but in accordance with a
BAl2 mechanism (Scheme 2, pathway ii), an SN2 process
at nitrogen where substituents on the leaving group exert
only a weak influence on the developing negative charge
in the transition state. Bimolecular rate constants for the
reaction of N-methylaniline with series 7 in methanol
have recently been determined and also yield a σ cor-
relation with modest, positive sensitivity (F ) 1.8).28 In
those reactions, as was the case for the reaction between
The activity of chemical mutagens is a function of
several properties: lipophilicity, which plays a role in
transport across cell walls; stability toward side reactions
(28) Campbell, J . J .; Glover, S. A. Unpublished Results.
(29) The HERON reaction pathway has recently been supported by
density functional calculations at the B3LYP/6-31G(D) level. Glover,
S.; Mo, G.; Rauk, A. Tetrahedron. In press.
(30) Harvey, R. G.; Geacintov, N. E. Acc. Chem. Res. 1988, 21, 66.
(31) Gniazdowski, M.; Cera, C. Chem. Rev. 1996, 96, 619.
(32) Ford, G. P.; Smith, C. T. J . Comput. Chem. 1989, 10, 568.
(33) Blackburn, G. M.; Kellard, B. Chem. Ind. 1986, 607, 687, 770.

9688 J . Org. Chem., Vol. 63, No. 26, 1998
Glover et al.
National Institute of Occupational Health and Safety in
Sydney. Acetonitrile used was HiPerSolv, far UV grade (BDH).
Ether refers to anhydrous diethyl ether stored over sodium
wire. Dichloromethane (DCM) and acetone were distilled and
dried over 4 Å molecular sieves. Ethyl acetate (EtOAc) and
methanol (MeOH) were distilled before use. Anhydrous sodium
sulfate was used for drying all organic mixtures. Flash
chromatography was executed on columns loaded with Kie-
selgel 60 (Merck). Thin-layer chromatography was performed
on aluminum sheets precoated with 0.2 mm of silica gel 60
F254 (Merck). Para-substituted benzoic acids and deuterioac-
etonitrile (99.5%) were purchased from Aldrich. The syntheses
of butyl N-acetoxy-p-chlorobenzohydroxamate (8a ), benzyl
N-acetoxybenzohydroxamate (8b), and benzyl benzohydrox-
amate have been described previously.3,4,36 1H NMR coupling
constants (J ) are reported in hertz.
F igu r e 6. Correlation of mutagenicities of 7 at 1 µmol/plate
with Hammett σ constants.
Sod iu m P a r a -Su bstitu ted Ben zoa te Sa lts. The ap-
propriate benzoic acid was treated with slightly less than a
molar equivalent of aqueous sodium carbonate at room tem-
perature. The filtrate was collected from the suspension and
evaporated to dryness in an oven. Dried salts were used
without further purification.
Ben zyl N-Ch lor oben zoh yd r oxa m a te. Benzyl benzohy-
droxamate (3.41 g, 15 mmol) and tert-butyl hypochlorite (4.87
g, 45 mmol) in DCM were stirred for 5 h in the dark. Removal
of solvent in vacuo provided the title compound which was used
immediately without further purification: ΙR (CDCl3) 1718
(CO) cm-1; 1H NMR (CDCl3) δ 5.09 (2H, s), 7.26 (2H, m), 7.30
(3H, m), 7.40 (2H, t), 7.54 (1H, t), 7.68 (2H, d).
with adventitious, intracellular reagents the substrate
encounters en route to the target DNA; residence time
at the DNA and their reactivity with nucleotides when
bound to DNA. If the putative correlation in Figure 6 is
real, it suggests that mutagenicity may be controlled
either by the basicity of the benzoate carbonyl oxygen,
which would be enhanced with para electron-releasing
groups or, alternatively, mutagenic activity may be
inversely related to overall chemical reactivity. In other
words it may be dependent upon their survival in the
aqueous environment of the Ames assay.
Gen er a l Syn th esis of Ben zyl N-(4-Su bstitu ted ben zoyl-
oxy)ben zoh yd r oxa m a tes. Benzyl N-chlorobenzohydroxam-
ates were stirred in the dark at room temperature, with 1.4
molar equiv of the appropriate anhydrous sodium benzoate
salts in dry acetone, for 12-72 h. The reaction was monitored
by 1H NMR, and filtration and concentration provided the
benzyl N-(benzoyloxy)benzohydroxamate derivatives in high
yields. Yields were determined by analytical HPLC analysis.
Ben zyl N-(Ben zoyloxy)ben zoh yd r oxa m a te (7a ). So-
dium benzoate (0.91 g, 6.3 mmol) was stirred at room tem-
perature with benzyl N-chlorobenzohydroxamate (1.18 g, 4.5
mmol) in acetone for 48 h. Purification by flash chromatogra-
phy (88% hexane:12% EtOAc) provided the title compound
Con clu sion s
N-(Benzyloxy)-N-(benzoyloxy)benzamides 7 are alkoxy-
nitrenium ion sources under acidic aqueous/organic
conditions. Like their N-acetoxy counterparts, they un-
dergo solvolysis by the AAl1 rather than the AAc2 mech-
anism. They are also susceptible to SN2 reactions at
nitrogen in base yielding the reactive hydroxamic esters
4. Both processes are facilitated by electron-withdrawing
para substituents on the benzoyloxy ring, and in the case
of acid-catalyzed solvolysis, the heterolysis step (charac-
terized by k′ in Scheme 1) is more important than the
protonation step (characterized by Ka in Scheme 1). The
anomeric effect at nitrogen can be viewed as favoring
both reaction processes.
All members of series 7 have been found to be mu-
tagenic in Salmonella TA100 without metabolic activa-
tion. Para substituents on the benzoyloxy ring appear to
affect mutagenicity in the opposite sense which suggests
that their biological activity may be related either to their
basicity or stability, rather than their ease of conversion
to electrophilic alkoxynitrenium ions or susceptibility to
nucleophilic attack at nitrogen. Studies to be published
elsewhere have also pointed to a significant hydrophobic
effect and residence time on DNA may thus be a
significant factor.5,34 In both respects, the mutagenic
activity of N-(benzoyloxy)-N-(benzyloxy)benzamides 7
bears some resemblance to the antitumor activity of the
duocarmycins where cytotoxicity is inversely correlated
with rates of acid-catalyzed solvolysis and reactivity with
DNA appears to be driven by hydrophobic binding.35
1
(60%): ΙR (CDCl3) 1758, 1731 (CO) cm-1; H NMR (CDCl3) δ
5.26 (2H, s), 7.26-7.40 (9H, m), 7.46 (1H, t, J ) 8, p-Ar), 7.54
(1H, t, p′′-Ar), 7.77 (2H, d, J ) 8), 7.91 (2H, d, J ) 8); 13C
NMR (CDCl3) δ 77.44 (t), 127.07 (s), 128.18 (d), 128.35 (d),
128.49 (d), 128.54 (d), 128.97 (d), 129.11 (d), 129.85 (dt), 131.51
(s), 132.67 (dt), 133.92 (dt), 134.64 (s), 164.13 (s), 174.28 (s).
Ben zyl N-(4-Meth oxyben zoyloxy)ben zoh yd r oxa m a te
(7b). Purification by flash chromatography (84% hexane:16%
EtOAc) provided the title compound (39%): ΙR (CDCl3) 1750
(ester CO), 1718 (amide CO) cm-1 1H NMR (CDCl3) δ 3.73
;
(3H, s), 5.26 (2H, s), 6.81 (2H, d), 7.25-7.47 (8H, m), 7.75 (2H,
d, J ) 8), 7.86 (2H, d, J ) 8); 13C NMR (CDCl3) δ 55.18 (q),
77.31 (t), 113.73 (d), 118.92 (s), 128.01 (d), 128.20 (d), 128.36
(d), 128.80 (d), 128.98 (d), 131.55 (s), 131.94 (d), 132.43 (d),
134.69 (s), 163.70 (s), 164.04 (s), 174.25 (s).
Ben zyl N-(4-Meth ylben zoyloxy)ben zoh ydr oxam ate (7c).
Purification by flash chromatography (88% hexane:12% EtOAc)
provided the title compound (47%): ΙR (CDCl3) 1756 (ester
1
CO), 1733 (amide CO) cm-1; H NMR (CDCl3) δ 2.38 (3H, s),
5.26 (2H, s), 7.29 (2H, d), 7.3-7.4 (8H, m), 7.47 (1H, t, J ) 8),
7.77 (2H, d, J ) 8), 7.81 (2H, d, J ) 8); 13C NMR (CDCl3) δ
21.61 (q), 77.58 (t), 124.27 (st), 128.15 (d), 128.34 (d), 128.50
(d), 128.99 (d), 129.11 (d), 129.23 (d), 129.94 (d), 131.64 (dt),
132.59 (s), 134.75 (s), 144.94 (s), 164.21 (s), 174.36 (s).
Ben zyl N-(4-Ch lor oben zoyloxy)ben zoh ydr oxam ate (7d).
Purification by flash chromatography (88% hexane:12% EtOAc)
provided the title compound (83%): ΙR (CDCl3) 1759 (ester
Exp er im en ta l Section
Gen er a l Meth od s. Ames tests were carried out by Dr. A.
Bonin in the Department of Environmental Toxicology at the
1
CO), 1734 (amide CO) cm-1; H NMR (CDCl3) δ 5.26 (2H, s),
(34) Mutagenicity appears to increase with increasing aromatic
character, i.e., N-acetoxy-N-alkoxybenzamides 1 < N-acetoxy-N-ben-
zyloxybenzamides 2 < N-benzoyloxy-N-benzyloxybenzamides 7. In
addition, N-acetoxy-N-butoxy-2-naphthamide (5500 revertants) is more
than 10 times more mutagenic than N-acetoxy-N-butoxy-2-benzamide
(477 revertants) in ΤΑ100 at 1 µmol/plate.
7.26-7.40 (9H, m), 7.47 (1H, t, J ) 8), 7.77 (2H, d, J ) 8),
(35) Boger, D. L. Acc. Chem. Res. 1995, 28, 20.
(36) Cooley, J . H.; Bills, W. D.; Throckmorton, J . R. J . Org. Chem.
1960, 25, 1734.

N-(Benzoyloxy)-N-(benzyloxy)benzamides
J . Org. Chem., Vol. 63, No. 26, 1998 9689
Ta ble 4. Acid -In d ep en d en t Ra te Con sta n ts for Acid -Ca ta lyzed Solvolysis of 7a -h a t Va r iou s Tem p er a tu r es
ln k
T (K)
NO2
CN
CF3
CHO
Cl
H
Me
MeO
298
308
318
328
338
338
-6.210
-4.571
-3.535
-2.363
-0.932
-6.368
-5.231
-3.671
-2.607
-1.377
-6.423
-4.971
-3.747
-2.173
-0.924
-6.5989
-5.3054
-3.6629
-2.3597
-1.3213
-6.9093
-5.5661
-3.9811
-2.7931
-1.5270
-6.9466
-5.5069
-4.1918
-2.7380
-1.5083
-1.4561
-7.0311
-5.7158
-4.2095
-2.8198
-1.4166
-7.0759
-5.6648
-3.8927
-2.7562
-1.3925
Ta ble 5. P seu d o-Un im olecu la r Ra te Con sta n ts for Rea ction of 7 w ith Hyd r oxid e a t 275.4 K
NO2 Cl Me CN
H
CHO
104[HO-]
103k′
(s-1
104[HO-]
103k′
(s-1
104[HO-]
103k′
(s-1
104[HO-]
103k′
(s-1
104[HO-]
103k′
(s-1
104[HO-]
103k′
(s-1
(mol L-1
)
)
(mol L-1
)
)
(mol L-1
)
)
(mol L-1
)
)
(mol L-1
)
)
(mol L-1
)
)
9.12
7.54
6.23
3.51
2.422
1.418
1.380
0.508
4.80
3.55
2.67
2.03
3.484
2.505
2.141
1.080
5.40
4.48
3.58
2.72
1.524
1.225
0.657
0.579
18.00
11.72
8.03
4.646
2.449
1.406
1.024
5.54
3.69
2.21
2.585
1.275
0.271
7.38
5.54
3.69
2.943
2.225
1.132
5.35
7.82 (2H, d, J ) 8); 13C NMR (CDCl3) δ 77.62 (t), 125.51 (st),
128.16 (d), 128.32 (d), 128.54 (d), 128.80 (d), 128.94 (d), 129.08
(d), 131.15 (dd), 131.32 (s), 132.73 (dt), 134.52 (s), 140.32 (s),
163.25 (s), 174.2 (s).
spectrometer (298-338 K). Substrate (10-40 mg) in CD3CN
(400 µL) was diluted with D2O such that, after addition of an
appropriate volume of a solution of sulfuric acid in D2O (typi-
cally 0.5-1.5 molar), the ratio of CD3CN:D2O was constant at
3.81:1. The acid solution was mixed into the mixture to initiate
reaction immediately prior to insertion in the probe. 1H NMR
spectra were acquired automatically at preprogrammed in-
tervals, and the peak areas for the benzylic methylene singlet
(and acetic acid) were obtained by integration. Arrhenius
studies were carried out at an appropriate acid concentration
to enable data collection at each temperature. A minimum of
five temperatures between 298 and 338 K were used for each
substrate, and rate data is presented in Table 4.
Rea ction w ith Hyd r oxid e. The progress of the reaction
between hydroxide ion and mutagenic substrate was monitored
by analytical HPLC. Typically, a 0.4 × 10-6 M solution of
mutagen in 25% aqueous acetonitrile was treated with a 10-
30 M excess of sodium hydroxide such that pseudo-unimolecu-
lar kinetics were valid for the disappearance of mutagen. The
temperature for all solvolysis reactions was held at 275.4 (
0.1 K by circulating water around the reaction vessel supplied
from a large reservoir. Approximately 10 µL of sample was
analyzed by HPLC injection as the reaction progressed.
Naphthalene was used as the internal reference. Pseudo-
unimolecular rate constants at various concentrations of base
are presented in Table 5.
Ben zyl N-(4-F or m ylb en zoyloxy)b en zoh yd r oxa m a t e
(7e). Purification by flash chromatography (85% hexane:15%
EtOAc) provided the title compound (50%): ΙR (CDCl3) 1761
(ester CO), 1735 (amide CO), 1707 (CHO) cm-1 1H NMR
;
(CDCl3) δ 5.27 (2H, s), 7.26 (3H, m), 7.39 (4H, m), 7.49 (1H, t),
7.78 (2H, d, J ) 8), 7.89 (2H, d, J ) 8), 8.06 (2H, d, J ) 8),
10.06 (H, s); 13C NMR (CDCl3) δ 77.87 (t), 128.31 (d), 128.45
(d), 128.70 (d), 129.08 (d), 129.19 (d), 129.45 (d), 130.45 (dd),
131.26 (st), 132.10 (st), 132.97 (dt), 134.48 (s), 139.63 (s), 163.21
(s), 174.07 (s), 191.28 (d).
Ben zyl N-(4-(Tr iflu or om eth yl)ben zoyloxy)ben zoh ydr ox-
a m a te (7f). Purification by flash chromatography (85% hex-
ane:15% EtOAc) provided the title compound (78%): ΙR
(CDCl3) 1765 (ester CO), 1734 (amide CO) cm-1 1H NMR
;
(CDCl3) δ 5.27 (2H, s), 7.26 (3H, m), 7.39 (4H, m), 7.52 (1H, t),
7.66 (2H, d, J ) 8), 7.78 (2H, d, J ) 8), 8.02 (2H, d, J ) 8); 13
C
NMR (CDCl3) δ 77.91 (t), 125.55 (dq), 128.35 (d), 128.49 (d),
128.74 (d), 129.12 (d), 129.23 (d), 129.64 (s), 130.33 (dd), 130.59
(s), 131.32 (s), 133.0 (dt), 134.54 (s), 163.07 (s), 174.12 (s).
Ben zyl N-(4-Cyan oben zoyloxy)ben zoh ydr oxam ate (7g).
Purification by flash chromatography (86% hexane:14% EtOAc)
provided the title compound (34%): ΙR (CDCl3) 2234, 2236
(CN),1763 (ester CO), 1732 (amide CO) cm-1; 1H NMR (CDCl3)
δ 5.26 (2H, s), 7.25-7.44 (8H, m), 7.55 (1H, t, J ) 8), 7.73
(2H, d, J ) 8), 7.78 (2H, d, J ) 8), 8.00 (2H, d, J ) 8); 13C
NMR (CDCl3) δ 78.05 (t), 117.29 (st), 117.62 (s), 128.42 (d),
128.54 (d), 128.92 (d), 129.18 (d), 129.56 (d), 130.39 (d), 131.21
(s), 132.30 (d), 133.15 (d), 134.46 (s), 162.70 (s), 174.04 (s).
Ben zyl N-(4-Nitr oben zoyloxy)ben zoh yd r oxa m a te (7h ).
Purification by flash chromatography (88% hexane:12% EtOAc)
provided the title compound (55%): ΙR (CDCl3) 1763 (ester
Cr ossover Exp er im en t. A solution of butyl N-acetoxy-p-
chlorobenzohydroxamate (0.013 mmol) and benzyl N-acetoxy-
benzohydroxamate (0.012 mmol) in 25% aqueous acetonitrile
(50 mL) was equilibrated for 1 h at 298 K. Dilute aqueous
NaOH (170 µL, 0.9 M) was added, and the reaction was
complete within 60 s. Analytical HPLC analysis of the reaction
mixture revealed the presence of benzyl benzoate (43.3%) and
butyl p-chlorobenzoate (46.3%) and the complete absence of
the crossover esters, benzyl p-chlorobenzoate and butyl ben-
zoate.
1
CO), 1732 (amide CO) cm-1; H NMR (CDCl3) δ 5.27 (2H, s),
7.27-7.43 (8H, m), 7.54 (1H, t, J ) 8), 7.79 (2H, d, J ) 8),
8.05 (2H, d, J ) 8), 8.21 (2H, d); 13C NMR (CDCl3) δ 78.01 (t),
123.55 (d), 128.32 (d), 128.43 (d), 128.71 (d), 129.08 (d), 129.18
(d), 130.98 (d), 131.07 (dd), 132.67 (s), 133.12 (dt), 134.39 (s),
150.84 (s), 162.37 (s), 173.93 (s).
Ack n ow led gm en t. This work was funded by the
Australian Research Council and the Commonwealth
Government through an Australian Postgraduate Award
to G.P.H. We are also grateful to Worksafe Australia
for mutagenicity testing (the views in this article are
those of the authors and do not necessarily reflect those
of the National Occupation Health and Safety Commis-
sion).
Ben zyl N-(4-ter t-Bu tylben zoyloxy)ben zoh yd r om a m a te
(7i). Purification by flash column chromatography (85% hex-
ane:15% EtOAc) afforded pure benzyl N-(p-tert-butylbenzoyl-
oxy)benzohydroxamate (44%): IR (CDCl3) 1756 (ester CO),
1
1738 (amide CO) cm-1; H NMR (CDCl3) δ 1.32 (9H, s), 5.26
(2H, s), 7.25-7.44 (8H, m), 7.55 (1H, t), 7.73 (2H, d), 7.78 (2H,
d), 8.00 (2H, d); 13C NMR (CDCl3) δ 30.99 (q), 34.89 (s), 77.58
(t), 124.27 (st), 128.15 (d), 128.34 (d), 128.50 (d), 128.99 (d),
129.11 (d), 129.23 (d), 129.94 (d), 131.64 (dt), 132.59 (s), 134.75
(s), 144.94 (s), 164.21 (s), 174.36 (s).
Kin etic Stu d ies. Acid -Ca ta lyzed Solvolysis. The acid-
catalyzed solvolysis of the substrates was monitored in the
variable-temperature probe of the Bruker AC300P NMR
1
Su p p or tin g In for m a tion Ava ila ble: 300 MHz H NMR
spectra for new compounds 7a -i prepared in this work (8
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
J O980863Z