The role of steric effects in the direct mutagenicity of N-acyloxy-N-alkoxyamides

Article abstract of DOI:10.1016/j.mrgentox.2006.02.003

Electrophilic N-acyloxy-N-alkoxyamides are mutagenic in Salmonella typhimurium TA100 without the need for S9 metabolic activation and they react with DNA at guanine-N7 at physiological pH. Since these are direct-acting mutagens, structural factors influen

Full text of DOI:10.1016/j.mrgentox.2006.02.003

Mutation Research 605 (2006) 51–62
The role of steric effects in the direct mutagenicity of
N-acyloxy-N-alkoxyamides
Luke E. Andrewsa, Antonio M. Boninb, Linda E. Franssona,
Ashley-Mae E. Gillsona, Stephen A. Glovera,∗
Division of Chemistry, School of Biological, Biomedical and Molecular Sciences,
a
University of New England, Armidale, NSW 2351, Australia
School of Chemistry, University of Sydney, NSW 2006, Australia
b
Received 5 September 2005; received in revised form 5 January 2006; accepted 1 February 2006
Abstract
Electrophilic N-acyloxy-N-alkoxyamides are mutagenic in Salmonella typhimurium TA100 without the need for S9 metabolic
activation and they react with DNA at guanine—N7 at physiological pH. Since these are direct-acting mutagens, structural factors
influence binding and reactivity with DNA. Mutagenicity in TA100 can be predicted by a QSAR incorporating hydrophobicity
(log P), stability to substitution reactions at nitrogen (pKa of the leaving acid) and steric effects of para-aryl substituents (Es). A
number of mutagens exhibit activities that deviate markedly from the predicted values and they fall into two classes: di-tert-butylated
N-benzoyloxy-N-benzyloxybenzamides, which – because of their size – are most probably excluded from the major groove or are
unable to achieve a transition state for reaction with DNA, and N-benzoyloxy-N-butoxyalkylamides with branching ␣-to the amide
carbonyl, which are resistant to SN2 reactions at the amide nitrogen.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Mutagenicity; Direct-acting; Ames Test; Steric effects; QSAR; Electrophiles; Amides
1. Introduction
N-Acyloxy-N-alkoxyamides are a class of direct-
acting mutagens that we have shown to react with
DNA in the major groove at guanine—N7 (G—N7) at
physiological pH [1–7]. Results have pointed to an SN2
reaction with DNA, which involves displacement of the
N-acyloxy group by the N7 lone pair of guanine [7].
The mutagenicity of a wide range of congeners (1–6)
has been measured in Salmonella typhimurium strain
TA100 and mutagenic activity has been found to be
dependent upon a number of variables.
Activity correlates with stability or pKa of the
∗
carboxylic acid leaving group in that mutagens that are
the most reactive towards SN2 reactions are the least
Corresponding author. Tel.: +61 2 67732361; fax: +61 2 67733268.
E-mail address: sglover@une.edu.au (S.A. Glover).
1383-5718/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.mrgentox.2006.02.003

52
L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
mutagenic [5,6]. Furthermore, there is an obvious depen-
dence upon hydrophobicity of the intact mutagens (as
estimated from log P, the log of the octanol/water parti-
tion coefficient). Since these mutagens react at G—N7
and require no metabolic activation, it is reasonable to
propose that at physiological pH, the hydrophobic effect
reflects binding of the molecules in the major groove
of DNA. More recently, we have determined that steric
effects also play a role and, using Taft steric parameters,
Es, for substituents attached to the para positions of ben-
zene rings on each of the three side chains, the pKa of the
departing acid and log P, a quantitative structure–activity
relationship has been derived, which predicts well the
activity of a wide range of N-acyloxy-N-alkoxyamides
(Eq. (1))[8].
Analysis of Eq. (1) indicates a strong dependence
upon log P, significant correlations with pKa and, where
present, steric effects of groups at para positions on both
the amide and benzyloxy side chains. However, statis-
tically, steric effects due to substituents on the leaving
group would appear to be less important. The depen-
dence upon hydrophobicity contrasts with the hitherto
prevailing view that a lack of correlation with log P is in
fact a characteristic of direct-acting mutagens [9–11].
The relationship in Eq. (1) provides an excellent
means of probing the effect of substructure upon muta-
genic activity and therefore upon binding to DNA. In this
paper, we describe two sets of mutagens for which the
QSAR fails in that both are found to be significantly less
mutagenic than predicted. The deviation from the pre-
dicted values can be ascribed on the one hand to distal
and, on the other, to proximal steric effects.
log TA 1001 = 0.29( 0.03) log P + 0.22( 0.09)pKa
+ 0.14( 0.04)Es1 + 0.16( 0.04)Es2
+ 0.11( 0.06)Es3 + 0.81( 0.5)
2. Materials and methods
2.1. Instrumentation
n = 41, standard error = 0.17,
Melting points were determined on a Reichert Micro-
scopic Hot-Stage and are uncorrected. Infrared spectra were
recorded on a Perkin Elmer 1725× FT spectrophotome-
ter. Nuclear Magnetic Resonance spectra (300 MHz) were
recorded in CDCl3 on a Bruker AC 300P FT NMR spectrom-
eter. ESI mass spectra were obtained on a Varian 1200L Mass
Spectrometer. Micro-analytical data were obtained from the
Research School of Chemistry, Australian National University,
Canberra.
r = 0.891, F = 29.17
(1)
ES1, ES2 and ES3 represent steric effects on the benza-
mide, benzyloxy and benzoyloxy side chains, respec-
tively.
1
Log of mutagenic activity in Salmonella typhimurium TA100 at
1 ␮mol/plate taken from the gradients of revertant colonies versus dose
in the linear range of dose response curves.

L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
53
2.2. Chemicals
phy using 10% ethyl acetate/petroleum spirit, afforded pure
N-butoxypropanamide (0.92 g, 6.34 mmol, 15%) as a pale
yellow oil; νmax (CHCl3)/cm−1 3188 (w, br), 1664 (s). 1H
NMR (CDCl3, 300 MHz) δ = 0.76 (3H, t, CH3), 0.99 (3H,
t, COCH2CH3), 1.22 (2H, sx, CH2CH3), 1.47 (2H, quintet,
OCH2CH2), 2.01 (2H, q, COCH2CH3), 3.72 (2H, t, OCH2).
13C NMR (CDCl3, 75 MHz) δ = 9.5 (q), 13.6 (q), 18.8 (t), 26.1
(t), 29.9 (t), 75.8 (t), 171.8 (s). MS(ESI): m/z = 168 (M + 23).
N-Butoxyisobutyramide. Potassium isobutyrohydroxamate
(2.01 g, 14.2 mmol), butyl bromide (1.95 g, 14.2 mmol) and
sodium carbonate (1.66 g, 15.7 mmol) were stirred overnight
in 50% aqueous methanol (150 ml) and refluxed for 24 h.
Excess methanol was removed in vacuo and the mixture
extracted with DCM, dried over anhydrous Na2SO4, filtered
and concentrated in vacuo. Purification by centrifugal chro-
matography using 10% ethyl acetate/petroleum spirit, afforded
pure N-butoxyisobutyramide (0.53 g, 3.33 mmol, 24%) as a
pale yellow oil; νmax (CHCl3)/cm−1 3188br (w), 1657 (s). 1H
NMR (CDCl3, 300 MHz) ␦ = 0.80 (3H, t, CH3), 1.02 (6H,
d, CH(CH3)2), 1.27 (2H, sx, CH2CH3), 1.51 (2H, quintet,
OCH2CH2), 2.34 (1H, m, CH(CH3)2), 3.77 (2H, t, OCH2).13C
NMR (CDCl3, 75 MHz) ␦ = 13.8 (q), 19.0 (q), 30.0 (t), 32.2 (t),
76.6 (t), 174.9 (s). MS(ESI): m/z = 182 (M + 23).
Syntheses of N-acetoxy-N-butoxybenzamide (1, X = H
[124617-84-1]), N-acetoxy-N-(4-tert-butylbenzyloxy)benza-
mide(2, Y = But [139259-93-1]) [4], N-benzoyloxy-N-
benzyloxybenzamide (3, Z = H [220168-39-8]) [5], N-benzoy-
loxy-N-(4-tert-butylbenzyloxy) benzamide (8b [37757-31-
4]) [12] and N-benzyloxy-N-(4-tert-butylbenzoyloxy) benza-
mide (8c [220168-56-9]) [5] have been described previously
as have those of the precursors to 7b, 8d and 8g, N-(4-tert-
butylbenzyloxy)-4-tert-butylbenzamide [637757-29-0] and
N-(4-tert-butylbenzyloxy)-N-chloro-4-tert-butylbenzamide
[637757-30-3] [12], 9a, N-benzyloxybenzamide [3532-25-0]
and N-benzyloxy-N-chlorobenzamide [3532-25-0] [4] and
9b, N-(4-tert-butylbenzyloxy)benzamide [93949-93-0] and
N-(4-tert-butylbenzyloxy)-N-chlorobenzamide [112403-68-6]
[4].
2.2.1. General syntheses of N-alkoxybenzamides
2.2.1.1. Method 1 [13]. The appropriate N-alkoxybenzamide
was synthesised in good yield by the reaction at room tempera-
ture (r.t.), overnight, of 1 equivalent of the potassium salt of an
appropriate hydroxamic acid [14] with 1.5 equivalents of the
relevant alkyl bromide and a 10% excess of sodium carbon-
ate in a 50% aqueous solution of methanol. After refluxing for
3 h, excess methanol was removed in vacuo and the mixture
extracted with dichloromethane (DCM), which was dried over
anhydrous Na2SO4, filtered and concentrated in vacuo. Crude
mixtures were purified by column or centrifugal chromatogra-
phy.
N-Benzyloxy-4-tert-butylbenzamide. Potassium 4-tert-
butylbenzohydroxamate (6.11 g, 26.4 mmol), benzyl bromide
(6.77 g, 39.6 mmol) and sodium carbonate (3.08 g, 29.0 mmol)
were stirred overnight in 50% aqueous methanol (150 ml) and
refluxed for 24 h. Excess methanol was removed in vacuo
and the mixture extracted with DCM, dried over anhydrous
Na2SO4, filtered and concentrated in vacuo. Purification using
centrifugal chromatography (15% EtOAc:85% petroleum
spirit), produced N-benzyloxy-4-tert-butylbenzamide in
45.6% yield. m.p. 84–86 ◦C; νmax (CHCl3)/cm−1 3300 (w)
and 1680 (s). 1H NMR (CDCl3, 300 MHz) δ = 1.35 (9H, s,
C(CH3)3), 5.07 (2H, s, OCH2Ar), 7.42 (5H, m, ArH), 7.46
(2H, t, o’-ArH), and 7.63 (2H, d, o-ArH). 13C NMR (CDCl3,
75 MHz) ␦ = 31.5 (q, Ar-C(CH3)3), 34.8 (s, Ar-C(CH3)3), 77.1
(t, OCH2Ar), 125.3 (d, m-ArC-tBu), 126.9 (d, o-ArC-tBu),
127.3 (d, o-ArC-CO), 127.4 (s, p-ArC-CO), 128.7 (d,
m-ArC-CO), 130.4 (s, ArC-CO), 140.9 (s, OCH2-ArC), 150.5
(s, ArC-C(CH3)3), and 163.0 (s, Ar-CO-NH). Calc. C 76.30,
H 7.47, N 4.94%; found: C 76.30, H 7.20, N 4.78%.
2.2.1.2. Method 2. N-Butoxybenzamide (8.30 g, 46.3 mmol)
in ethanol (45 ml) and conc. HCl (15 ml), was refluxed for
45 min. Excess ethanol was removed under reduced pressure,
water added and the mixture extracted with ether. Concen-
tration of the aqueous layer to dryness under reduced pres-
sure afforded butoxyamine hydrochloride (3.99 g, 31.8 mmol,
69%) as a white solid, which was used without further puri-
fication.
An acid chloride (neat or in an ethereal solution) was
added dropwise to an equimolar quantity of N-butoxyamine
(hydrochloride) in dry ether containing a two molar excess
of triethylamine such that the temperature remained below
5 ◦C. After stirring for 2 h, the ether layer was washed with
water, 10% NaHCO3, dried and concentrated to afford the N-
alkoxyamide in good yield. Crude mixtures were purified by
column or centrifugal chromatography.
N-Butoxyadamantane-1-carboxamide. Purification by cen-
trifugal chromatography using 10% ethyl acetate/petroleum
spirit, afforded pure N-butoxyadamantane-1-carboxamide
(0.08 g, 0.32 mmol, 25%) as
a pale yellow oil. νmax
(CHCl3)/cm−1 1681 (s). 1H NMR (CDCl3, 300 MHz) δ = 0.75
(3H, t, CH3), 1.22 (2H, sx, CH2CH3), 1.45 (2H, quintet,
OCH2CH2), 1.55 (6H, t, CHCH2CH), 1.75 (6H, d, CHCH2C),
1.84 (3H, m, CH), 3.72 (2H, t, OCH2), 9.45 (1H, s, NH). 13
C
NMR (CDCl3, 75 MHz) ␦ = 13.7 (q), 19.0 (t), 27.9 (t), 29.7
(t), 36.4 (t), 38.9 (d), 39.8 (s), 75.3 (t), 175.4 (s). MS(ESI):
m/z = 274 (M + 23).
N-Butoxypropanamide. Potassium propanohydroxamate
(5.33 g, 41.9 mmol), butyl bromide (5.74 g, 41.9 mmol) and
sodium carbonate (4.89 g, 46.1 mmol) were stirred overnight in
50% aqueous methanol (150 ml) and refluxed for 24 h. Excess
methanol was removed in vacuo and the mixture extracted
with DCM, dried over anhydrous Na2SO4, filtered and con-
centrated in vacuo. Purification by centrifugal chromatogra-
N-Butoxy-2,2-dimethylpropanamide
[64214-68-2].
Purification by centrifugal chromatography using 10%
ethyl acetate/petroleum spirit, afforded pure N-butoxy-2,2-
dimethylpropanamide (0.11 g, 0.63 mmol, 42%) as a pale
yellow oil. νmax (CHCl3)/cm−1 1683 (s). 1H NMR (CDCl3,

54
L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
300 MHz) ␦ = 0.95 (3H, t, CH3), 1.22 (9H, s, C(CH3)3), 1.42
(2H, sx, CH2CH3), 1.63 (2H, quintet, OCH2CH2), 3.91 (2H, t,
OCH2). 13C NMR (CDCl3, 75 MHz) ␦ = 13.8 (q), 19.1 (t), 27.2
(q), 30.1 (t), 38.0 (s), 76.4 (t), 176.2 (s). MS(ESI): m/z = 196
(M + 23).
N-Butoxy-3,3-dimethylbutyramide. Purification by cen-
trifugal chromatography using 10% ethyl acetate/petroleum
spirit, afforded pure N-butoxy-3,3-dimethylbutyramide
75 MHz) δ = 13.8 (q), 19.3 (t), 28.1 (t), 29.6 (t), 31.2 (t), 36.4
(d), 38.2 (s), 44.4 (s), 74.4 (t), 181.5 (s).
2.2.2.5. N-Butoxy-N-chloro-2,2-dimethylpropanamide. N-
Butoxy-N-chloro-2,2-dimethylpropanamide was produced as
a yellow oil. νmax (CHCl3)/cm−1 1727 (s). H NMR (CDCl3,
1
300 MHz) δ = 0.95 (3H, t, CH3), 1.28 (9H, s, C(CH3)3), 1.43
(2H, sx, CH2CH3), 1.66 (2H, quintet, OCH2CH2), 4.08 (2H,
t, OCH2). 13C NMR (CDCl3, 75 MHz) δ = 13.6 (q), 19.2 (t),
27.1 (q), 29.4 (t), 41.0 (s), 774.5 (t), 182.1 (s).
(0.83 g, 4.43 mmol, 70%) as
a pale yellow oil. νmax
(CHCl3)/cm−1 1653 (s). 1H NMR (CDCl3, 300 MHz) δ = 0.77
(3H, t, CH3), 0.94 (9H, s, C(CH3)3), 1.26 (2H, sx, CH2CH3),
1.47 (2H, quintet, OCH2CH2), 1.92 (2H, s, CH2C( O)N),
3.76 (2H, t, OCH2). 13C NMR (CDCl3, 75 MHz) δ = 13.7 (q),
18.9 (t), 29.7 (m), 30.0 (s), 30.7 (t), 46.1 (t), 75.9 (t), 169.5
(s). MS(ESI): m/z = 210 (M + 23).
2.2.2.6. N-Butoxy-N-chloro-3,3-dimethylbutyramide. N-
Butoxy-N-chloro-3,3-dimethylbutyramide was produced as a
yellow oil, which was used without further purification. νmax
(CHCl3)/cm−1 1742 (s). 1H NMR (CDCl3, 300 MHz) δ = 0.98
(3H, t, CH3), 1.07 (9H, s, C(CH3)3), 1.46 (2H, sx, CH2CH3),
1.68 (2H, quintet, OCH2CH2), 2.67 (2H, s, CH2C( O)N),
4.05 (2H, t, OCH2). 13C NMR (CDCl3, 75 MHz) δ = 13.8 (q),
19.2 (t), 29.6 (m), 30.0 (s), 31.4 (t), 48.0 (t), 75.1 (t), 171.8 (s).
2.2.2. Synthesis of N-alkoxy-N-chloroamides
A solution of the appropriate N-alkoxyamide and a 2–3
molar excess of tert-butyl hypochlorite [15] in CH2Cl2 was
stirred in the dark under anhydrous conditions for 3–6 h.
The progress of the reaction was monitored by thin-layer
chromatography until complete, at which time the solvent
and excess tert-butyl hypochlorite were removed in the dark
under reduced pressure at 30 ◦C to afford the N-alkoxy-N-
chloroamides in excellent yields (90–100%). The N-alkoxy-
N-chloroamides were used without further purification.
2.2.3. General synthesis of N-acyloxy-N-alkoxyamides
N-Alkoxy-N-chloroamides were acetoxylated or benzoy-
loxylated by treatment with 1.4 molar equivalents of anhydrous
sodium acetate or the appropriate sodium benzoate in dry
acetone, at room temperature, for 12–72 h in the dark. The
reaction was monitored by thin-layer chromatography. Filtra-
tion and concentration in vacuo provided the N-acetoxy or
N-benzoyloxy derivatives, frequently with quantitative con-
version. Products were obtained in a pure state or were fur-
ther purified by centrifugal chromatography (15% EtOAc:85%
petroleum spirit or 10% EtOAc:90% petroleum spirit). In all
cases, mutagens were characterised spectroscopically. For all
N-benzoyloxy-N-benzyloxy benzamides, the 1H and 13C NMR
chemical shift assignment for the benzoyloxy leaving group are
indicated by the symbol (*).
2.2.2.1. N-Benzyloxy-N-chloro-4-tert-butylbenzamide. N-
Benzyloxy-N-chloro-(4-tert-butyl)benzamide was produced
as a yellow oil, in 90.2% yield. νmax (CHCl3)/cm−1 1710
(C O). 1H NMR (CDCl3, 300 MHz) δ = 1.36 (9H, s, C(CH3)3),
5.11 (2H, s, OCH2Ar), 7.32 (5H, m, ArH), 7.44 (2H, d, ArH),
and 7.68 (2H, d, o-ArH).
2.2.2.2. N-Butoxy-N-chloropropanamide. N-Butoxy-N-
chloropropanamide was produced as a yellow oil. νmax
(CHCl3)/cm−1 1742br (s). 1H NMR (CDCl3, 300 MHz)
␦ = 0.97 (3H, t, CH3), 1.19 (3H, t, COCH2CH3), 1.43 (2H,
sx, CH2CH3), 1.67 (2H, quintet, OCH2CH2), 2.59 (2H,
q, COCH2CH3), 4.05 (2H, t, OCH2). 13C NMR (CDCl3,
75 MHz) δ = 8.7 (q), 13.7 (q), 19.1 (t), 27.6 (t), 29.5 (t), 75.3
(t), 179.2 (s).
2.2.3.1. N-Acetoxy-N-(p-tert-butylbenzyloxy)-p-tert-
butylbenzamide (7b [693809-12-0]). Purification by
centrifugal chromatography (15% EtOAc:85% petroleum
spirit) provided the title compound as a yellow oil in 84%
yield. νmax (CHCl3)/cm−1 1783 (s, ester CO) and 1725 (s,
amide CO). 1H NMR (CDCl3, 300 MHz) δ = 1.34 (9H, s,
But–Ar), 1.35 (9H, s, But–Ar), 2.09 (3H, s, COCH3) 5.19 (2H,
s, OCH2Ar), 7.37 (2H, d, ArH), 7.42 (2H, d, ArH), 7.45 (2H, d,
ArH) 7.74 (2H, d, ArH). 13C NMR (CDCl3, 75 MHz) δ = 18.8
(COCH3), 31.1 (C(CꢀH3)3), 31.3 (C(CH3)3), 34.6 (Cꢀ (CH3)3)
35.1 (C(CH3)3), 77.3 (OCH2Ar), 125.3 (oꢀ-ArCH), 125.4
(o-ArCH), 128.7 (i-ArC), 129.2 (m- and mꢀ-ArCH), 131.8
(iꢀ-ArC), 151.8 (pꢀ-ArC), 156.6 (p-ArC), 168.1 (COCH3),
174.0 (CON).1.33 (9H, s, C[CH3]3), 1.35 (9H, s, C[CH3]3),
2.08 (3H, OC(O)CH3, s), 5.17 (2H, s, OCH2Ar), 7.32-7.45
(6H, m, ArH), and 7.71 (2H, d, o-ArH).
2.2.2.3. N-Butoxy-N-chloroisobutyramide. N-Butoxy-N-
chloroisobutyramide was produced as a yellow oil. νmax
(CHCl3)/cm−1 1735 (s). 1H NMR (CDCl3, 300 MHz) δ = 0.97
(3H, t, CH3), 1.20 (6H, d, CH(CH3)2), 1.44 (2H, sx, CH2CH3),
1.68 (2H, quintet, OCH2CH2), 3.04 (1H, m, CH(CH3)2), 4.08
(2H, t, OCH2).
2.2.2.4. N-Butoxy-N-chloroadamantane-1-carboxamide.
N-Butoxy-N-chloroadamantane-1-carboxamide was produced
as a yellow oil. νmax (CHCl3)/cm−1 1721 (s). 1H NMR (CDCl3,
300 MHz) δ = 1.01 (3H, t, CH3), 1.48 (2H, sx, CH2CH3), 1.70
(2H, quintet, OCH2CH2), 1.73 (6H, t, CHCH2CH), 2.03 (9H,
m, CHCH2C and CH), 4.11 (2H, t, OCH2). 13C NMR (CDCl3,
2.2.3.2. N-Benzoyloxy-N-benzyloxy-p-tert-butylbenzamide
(8a[693809-10-8]). Purification by centrifugal chromatog-
raphy (15% EtOAc:85% petroleum spirit) provided the title

L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
55
compound as a yellow oil in 42% yield. νmax (CHCl3)/cm−1
1757 (s, ester CO) and 1724 (s, amide CO). 1H NMR (CDCl3,
300 MHz) δ = 1.33 (9H, s, C[CH3]3), 5.30 (2H, s, OCH2Ar),
7.30–7.52 (9H, m, ArH), 7.61 (1H, t, p*-ArH), 7.98 (2H,
d, o-ArH), and 8.10 (2H, d, o*-ArH). 13C NMR (CDCl3,
75 MHz) δ = 31.3 (q, Ar-C[CH3]3) 34.8 (s, Ar-C[CH3]3), 77.4
(t, OCH2Ar), 125.4 (d, m-ArC-tBu), 126.9 (d, o-ArC-tBu),
127.4 (d, ArC), 128.3 (d, m*-ArC), 128.6 (d, m-ArC), 130.1 (d,
o*-ArC), 130.4 (s, ArC-C(O)N), 130.7 (s, C*-C(O)ON), 133.4
(d, p*-ArC), 140.6 (s, OCH2-ArC), 150.8 (s, ArC-C[CH3]3),
157.3 (s, Ar-CO-NH), and 173.4 (s, OC*OAr).
Ar-C[CH3]3), 31.5 (q, ArC*[CH3]3), 34.4 (s, Ar-C[CH3]3),
35.0 (s, Ar-C*[CH3]3), 77.4 (t, OCH2Ar), 125.2 (d, m*-ArC),
125.4 (d, m-ArC), 126.8 (d, o-ArC-C(O)N), 127.1 (d, ArC),
127.6 (s, ArC-C(O)ON), 128.5 (d, m-ArC), 129.9 (d, o*-ArC),
130.1 (s, ArC-C(O)N), 139.8 (s, ArC-CH2ON), 150.7 (s,
p-ArC), 154.6 (s, p*-ArC), 160.3 (s, Ar-C(O)N), and 174.2 (s,
OC*(O)Ar).
2.2.3.6. N-(3,5-Di-tert-butylbenzoyloxy)-N-
benzyloxybenzamide
(9a
[693809-20-0]). Purification
by centrifugal chromatography (15% EtOAc:85% petroleum
spirit) provided the title compound as a yellow oil in 51%
yield. νmax (CHCl3)/cm−1 1759 (s, ester CO) and 1729 (s,
amide CO). 1H NMR (CDCl3, 300 MHz) δ = 1.33 (18H, s,
C[CH3]3), 5.31 (2H, s, OCH2Ar), 7.32–7.45 (7H, m, ArH),
7.52 (1H, t, p-ArH), 7.70 (1H, fine triplet, p*-ArH), 7.78
(2H, d, o-ArH), and 7.82 (2H, d, o*-ArH). 13C NMR (CDCl3,
75 MHz) δ = 31.4 (2q, Ar-C[CH3]3), 35.2 (s, Ar-C[CH3]3),
76.9 (t, OCH2Ar), 124.2 (d, o*-ArC), 127.1 (d, p*-ArC),
127.3 (d, o’-ArC), 127.4 (d, p-ArC), (128.6 (d, m-ArC), 129.8
(s, ArC*-C(O)ON), 131.3 (d, p-ArC), 132.9 (s, ArC-C(O)N),
139.8 (s, ArC-CH2ON), 149.7 (s, ArC-tBu), 164.3 (s,
Ar-C(O)N), and 174.2 (s, OC*(O)Ar).
2.2.3.3. N-Benzoyloxy-N-(p-tert-butylbenzyloxy)-p-tert-
butylbenzamide (8d [693809-14-2]). Purification by
centrifugal chromatography (15% EtOAc:85% petroleum
spirit) provided the title compound as a yellow oil in 74%
yield. νmax (CHCl3)/cm−1 1756 (s, ester CO) and 1730 (s,
amide CO). 1H NMR (CDCl3, 300 MHz) δ = 1.29 (9H, s,
C[CH3]3), 1.34 (9H, s, C[CH3]3), 5.26 (2H, s, OCH2Ar), 7.34
(4H, s, oꢀ,mꢀ-ArH-tBu), 7.38-7.47 (4H, m, ArH), 7.61 (1H, t,
p*-ArH), 7.74 (2H, d, o-ArH), and 7.98 (2H, d, o*-ArH). 13
C
NMR (CDCl3, 75 MHz) ␦ = 31.1 (q, Ar-C[CH3]3), 31.3 (q,
ArC[CH3]3), 34.6 (s, Ar-C[CH3]3), 35.1 (s, Ar-C[CH3]3), 77.5
(t, OCH2Ar), 125.3 (d, m-ArC), 125.4 (d, m-ArC), 125.6 (d,
o-ArC), 127.6 (d, o-ArC-C(O)N), 128.5 (d, m*-ArC), 129.1
(d, o*-ArC), 129.3 (s, ArC-C(O)N), 130.0 (s, ArC-C(O)ON),
131.9 (d, p*-ArC), 133.8 (s, ArC-CH2ON), 151.7 (s, ArC-
tBu), 156.6 (s, ArC-tBu), 164.3 (s, Ar-C(O)N), and 174.2 (s,
OC*(O)Ar).
2.2.3.7. N-(3,5-Di-tert-butylbenzoyloxy)-N-(p-tert-
butylbenzyloxy)benzamide (9b). Purification by centrifugal
chromatography (15% EtOAc:85% petroleum spirit) provided
the title compound as a yellow oil in 58% yield. νmax
(CHCl3)/cm−1 1760 (s, ester CO) and 1728 (s, amide CO).
1H NMR (CDCl3, 300 MHz) δ = 1.29 (9H, s, C[CH3]3), 1.34
(18H, s, C[CH3]3), 5.26 (2H, s, OCH2Ar), 7.29 (4H, s, oꢀ,
mꢀ-ArH-tBu), 7.38-7.44 (2H, m, ArH), 7.51 (1H, t, p-ArH),
7.66 (1H, fine triplet, p*-ArH), 7.81 (2H, d, o-ArH), and
7.82 (2H, d, o*-ArH). 13C NMR (CDCl3, 75 MHz) δ = 31.3
(C(CꢀH3)3 and C(C*H3)3), 34.6 (Cꢀ (CH3)3), 34.9 (C*(CH3)3),
77.4 (OCH2Ar), 124.3 (m’-ArCH), 125.4, 126.8, 128.3, 129.2
and 132.6 (ArCH) 131.8 and 132.1 (I-ArC or iꢀ-ArC), 151.0
and 151.7 (i*-ArC and pꢀ-ArC), 151.4 (C*(CH3)3), 164.9
(CON), 174.5 (OCOAr).
2.2.3.4. N-(p-Tert-butylbenzoyloxy)-N-(p-tert-
butylbenzyloxy)benzamide (8e [693809-16-4]). Purification
by centrifugal chromatography (15% EtOAc:85% petroleum
spirit) provided the title compound as a yellow solid in 63%
yield. νmax (CHCl3)/cm−1 1754 (s, ester CO) and 1726 (s,
amide CO). 1H NMR (CDCl3, 300 MHz) δ = 1.29 (9H, s,
C[CH3]3), 1.35 (9H, s, C*[CH3]3), 5.25 (2H, s, OCH2Ar), 7.35
(4H, s, oꢀ,mꢀ-ArH-tBu), 7.38-7.47 (4H, m, ArH), 7.53 (1H, t,
p-ArH), 7.83 (2H, d, o-ArH), and 7.90 (2H, d, o*-ArH). 13C
NMR (CDCl3, 75 MHz) ␦ = 31.0 (q, Ar-C[CH3]3), 31.3 (q,
ArC*[CH3]3), 34.6 (s, Ar-C[CH3]3), 35.2 (s, Ar-C*[CH3]3),
77.4 (t, OCH2Ar), 124.5 (d, m*-ArC), 125.4 (d, m-ArC-tBu),
125.6 (d, o-ArC), 127.0 (s, ArC*-C(O)ON), 128.2 (d, o-ArC),
129.1 (d, m-ArC), 129.9 (d, o*-ArC), 131.9 (s, ArC-C(O)N),
132.6 (s, OCH2-ArC), 151.7 (s, ArC-C[CH3]3), 157.8 (s,
p*-ArC), 164.2 (s, Ar-C(O)N), and 174.4 (s, OC*(O)Ar).
2.2.3.8. N-Benzoyloxy-N-butoxypropanamide (13). Purifica-
tion by centrifugal chromatography using 10% ethyl
acetate/petroleum spirit afforded pure N-benzoyloxy-N-
butoxypropanamide (0.41 g, 1.55 mmol, 47%) as a pale yellow
oil. νmax (CHCl3)/cm−1 1745 s (s, amide C O), 1766 (s, ester
C O). 1H NMR (CDCl3, 300 MHz) δ = 0.90 (3H, t, CH3), 1.18
(3H, t, COCH2CH3), 1.36 (2H, sx, CH2CH3), 1.67 (2H, quin-
tet, OCH2CH2), 2.61 (2H, q, COCH2CH3), 4.14 (2H, t, OCH2),
7.45 (2H, t, m-Ar-H), 7.59 (1H, t, p-Ar-H), 8.05 (2H, d, o-Ar-
H). 13C NMR (CDCl3, 75 MHz) ␦ = 8.1 (q), 13.7 (q), 19.0 (t),
27.1 (t), 30.1 (t), 75.9 (t), 127.4 (s), 128.6 (d), 130.0 (d), 134.0
(d), 164.1 (s), 180.1 (s). MS(ESI): m/z = 288 (M + 23).
2.2.3.5. N-(p-Tert-butylbenzoyloxy)-N-benzyloxy-p-tert-
butylbenzamide
(8f
[693809-16-4]). Purification
by
centrifugal chromatography (15% EtOAc:85% petroleum
spirit) provided the title compound as a yellow oil in 47%
yield. νmax (CHCl3)/cm−1 1757 (s, ester CO) and 1726 (s,
amide CO). 1H NMR (CDCl3, 300 MHz) δ = 1.32 (9H, s,
C[CH3]3), 1.35 (9H, s, C*[CH3]3), 5.29 (2H, s, OCH2Ar),
7.31-7.48 (9H, m, ArH), 7.77 (2H, d, o-ArH), and 7.93
(2H, d, o*-ArH). 13C NMR (CDCl3, 75 MHz) ␦ = 31.1 (q,
2.2.3.9. N-Benzoyloxy-N-butoxyisobutyramide (12). Purifi-
cation by centrifugal chromatography using 10% ethyl
acetate/petroleum spirit afforded pure N-benzoyloxy-N-

56
L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
butoxyisobutyramide (0.19 g, 0.68 mmol, 25%) as a pale yel-
lowoil. νmax (CHCl3)/cm−1 1736(s, amideC O), 1767(s, ester
C O). 1H NMR (CDCl3, 300 MHz) δ = 0.93 (3H, t, CH3), 1.24
(6H, d, CH(CH3)2), 1.41 (2H, sx, CH2CH3), 1.69 (2H, quintet,
OCH2CH2), 3.01 (1H, m, CH(CH3)2), 4.19 (2H, t, OCH2), 7.47
(2H, t, m-Ar-H), 7.61 (1H, t, p-Ar-H), 8.08 (2H, d, o-Ar-H).
13C NMR (CDCl3, 75 MHz) δ = 13.7 (q), 18.9 (q), 30.1 (t), 32.6
(t), 76.0 (t), 127.4 (s), 128.6 (d), 130.0 (d), 134.0 (d), 164.0 (s),
182.6 (s). MS(ESI): m/z = 302 (M + 23).
and cultured as described [16] with the exception that fresh
broth cultures were incubated at 37 ◦C in a shaking water bath
for 10 h prior to use in each assay. Top agar, supplemented
with a trace of histidine and biotin, was dispensed in 2-ml
volumes into 5-ml plastic vials and maintained at 45 ◦C in a
water bath. Before pouring on the surface of minimal agar
plates, 0.1 ml of the broth culture and 0.1 ml of the test chem-
ical dissolved in dry, analytically pure DMSO were added to
each vial. Triplicate plates at each dose level were incubated
at 37 ◦C for 72 h before counting revertant colonies with an
Artek model 880 counter. The efficacy of the strain in each set
of experiments was verified using 10 ␮g of 4-nitroquinoline-
N-oxide in 10 ␮l DMSO, which gave a 4+ reaction in a spot test
[16].
Assays at different dose levels were carried out in triplicate
together with negative controls. Responses for each compound
at 1 ␮mol plate−1 were obtained from least-squares analysis
of linear regions of the plots of mean revertants/plate versus
dose. Comparative data were derived by scaling mutagenicities
at 1 ␮mol plate−1 to that of a single standard, N-acetoxy-N-
butoxybenzamide (1, X = H) (log TA100 = 2.5) [6], which was
always analysed in parallel with new mutagens.
2.2.3.10. N-Benzoyloxy-N-butoxyadamantane-1-
carboxamide (10). Purification by centrifugal chromatogra-
phy using 10% ethyl acetate/petroleum spirit afforded pure
N-benzoyloxy-N-butoxyadamantane-1-carboxamide (0.16g,
0.43 mmol, 22%) as a pale yellow oil. νmax (CHCl3)/cm−1
1721 (s, amide C O), 1755 (s, ester C O). 1H NMR (CDCl3,
300 MHz) δ = 0.95 (3H, t, CH3), 1.45 (2H, sx, CH2CH3), 1.71
(2H, quintet, OCH2CH2), 1.76 (6H, t, CHCH2CH), 2.06 (3H,
m, CH), 2.10 (6H, d, CHCH2C), 4.20 (2H, t, OCH2). 13C
NMR (CDCl3, 75 MHz) δ = 13.8 (q), 19.2 (t), 28.0 (t), 30.1
(t), 36.5 (d), 38.0 (s), 43.2 (s), 75.0 (t), 127.9 (s), 128.6 (d),
130.0 (d), 133.7 (d), 164.1 (s), 182.2 (s). MS(ESI): m/z = 394
(M + 23).
2.4. Bimolecular rate studies
2.2.3.11. N-Benzoyloxy-N-butoxy-2,2-dimethylpropanamide
(11[693809-37-9]). Purification by centrifugal chromatog-
raphy using 10% ethyl acetate/petroleum spirit afforded
pure N-benzoyloxy-N-2,2-dimethylpropanamide (0.17 g,
0.58 mmol, 60%) as a pale yellow oil. νmax (CHCl3)/cm−1
1728 (s, amide C O), 1763 (s, ester C O). 1H NMR (CDCl3,
300 MHz) δ = 0.94 (3H, t, CH3), 1.37 (9H, s, C(CH3)3), 1.39
(2H, sx, CH2CH3), 1.69 (2H, quintet, OCH2CH2), 4.19 (2H, t,
OCH2), 7.49 (2H, t, m-Ar-H), 7.63 (1H, t, p-Ar-H), 8.10 (2H,
d, o-Ar-H). 13C NMR (CDCl3, 75 MHz) δ = 13.8 (q), 19.2
(t), 27.1 (q), 30.0 (t), 40.6 (s), 75.1 (t), 127.9 (s), 128.6 (d),
129.9 (d), 133.8 (d), 164.0 (s), 182.8 (s). MS(ESI): m/z = 316
(M + 23).
Rate constants for the bimolecular reaction between N-
methylaniline and various mutagens were determined using
1H NMR spectroscopy. Mutagen (2–10 mg) in d4-methanol
(400 ␮l) in an ultra-high precision NMR tube, was equili-
brated at the required temperature in the probe of an NMR
spectrometer. The sample was shimmed, removed from the
probe and a micropipette was used to add a minimum of
twice the molar equivalent of N-methyl aniline (2–20 ␮l).
The exact time of mixing and the initial concentrations of
both compounds were noted. After brief shimming, a series
of acquisitions were accumulated at a preset time interval,
and the extent of reaction was monitored by analysing the
disappearance of both starting materials according to peak
areas/heights of characteristic signals in the mutagen and the
N-methyl resonance of the N-methylaniline. Initial substrate
concentrations were obtained by back-extrapolation of con-
centration plots for both reagents to the initial time of mixing,
to.
2.2.3.12. N-Benzoyloxy-N-butoxy-3,3-dimethylbutyramide
(14). Purification by centrifugal chromatography using 10%
ethyl acetate/petroleum spirit afforded pure N-benzoyloxy-N-
butoxy-3,3-dimethylbutyramide (0.29 g, 0.94 mmol, 52%) as
a pale yellow oil. νmax (CHCl3)/cm−1 1735 (s, amide C O),
1763 (s, ester C O). 1H NMR (CDCl3, 300 MHz) δ = 0.83
(3H, t, CH3), 1.03 (9H, s, C(CH3)3), 1.34 (2H, sx, CH2CH3),
1.60 (2H, quintet, OCH2CH2), 2.44 (2H, s, CH2C( O)N),
4.07 (2H, t, OCH2) 7.36 (2H, t, m-Ar-H), 7.50 (1H, t, p-Ar-H),
7.99 (2H, d, o-Ar-H). 13C NMR (CDCl3, 75 MHz) δ = 13.6
(q), 19.0 (t), 29.5 (m), 30.0 (s), 30.9 (t), 45.5 (t), 75.6 (t),
127.4 (s), 128.6 (d), 129.5 (d), 133.9 (d), 163.8 (s), 177.2 (s).
MS(ESI): m/z = 330 (M + 23).
2.5. Calculations
Log P(log[octanol/waterpartitioncoefficient])valueswere
calculated by Ghose–Crippen methods as part of computation
of optimised geometries at the AM1 semi-empirical level using
SPARTAN [17].
3. Results
2.3. Mutagenicity assays
Mutagenicitiesoftwoseriesofmutagensareprovided
in Table 1 together with relevant data for related species
from earlier studies. The first series comprises congeners
S. typhimurium strain TA100 was obtained from Profes-
sor B.N. Ames, University of California, Berkeley, U.S.A.

Table 1
Mutagenicities of the N-acyloxy-N-alkoxyamides in TA100 using 1(X = H) as the normalisation standard
Set 1
␮mol/
plate
0.00
0.03
0.06
0.12
0.25
0.50
1.00
Sloped Normalised
slopee
Log
Diff.g
0.1
TA100f
1 X = H
(Set 1)a
(Set 2)b
(Set 3)c
(Set 2)
(Set 2)
(Set 1)
(Set 1)
(Set 2)
(Set 1)
(Set 1)
(Set 2)
(Set 1)
(Set 2)
186.7 (15)
211.7(13)
193.7(13)
211.7(13)
211.7(13)
186.7(15)
186.7(15)
211.7(13)
186.7(15)
186.7(15)
211.7(13)
186.7(15)
211.7(13)
194(13)
241.3(12)
353.7(8)
807.3(160
845.3(23)
626
940
676
2306
9391
2204
2827
130
319
319
319
783
3186
1123
1441
44
2.5
190.7(18)
240.3(9)
299.0(6)
235.3(10)
516.0(22)
1364.3(40)
438.3(10)
525.7(22)
233.0(10)
230.3(13)
284.0(18)
182.3(10)
206.7(6)
193.0(12)
254(11)i
203(5)i
426.3(13)
347.7(37)
516.0(22)
1500.3(65)
472.7(23)
608.7(41)
274.0(19)
237.7(10)
381.3(48)
187.0(5)
233.3(19)
188.3(12)
395(12)
795.3(31)
554.0(38)
620.3(23)
1621.0(39)
521.7(26)
708(70.5)
335.7(22)
254.3(9)
495.7(8)
199.3(6)
300.3(10)
208.0(10)
687(12)
7b
8a
8b
8c
8d
8e
8f
8g
9a
9b
327.3(57)
1217.7(50)
215.7(19)
251.014)
201.3(6)
189.0(13)
201.7(10)
204.3(9)
488.3(28)
1338.7(45)
323.3(10)
443.014)
230.0(5)
189.3(12)
220.7(6)
185.7(10)
195.0(7)
2.89
3.50
3.05
3.20
1.64
1.86
2.51
–
2.07
–
2.51
–
−0.04
−0.24
0.16
0.19
1.65
1.56
0.97
–
2.15
–
0.74
–
142
72
805
410
–
116
–
327
–
255
246
241
447
n.d.r.h
227
192.3(9)
193.3(13)
184.3(10)
n.d.r.h
693
n.d.r.h
541
522
510
10 (log P = 5.12; pKA = 4.2) (Set 3)
11 (log P = 4.37; pKA = 4.2) (Set 3)
12 (log P = 3.66; pKa = 4.2)
13 (log P = 3.1;pKa = 4.2)
14 (log P = 4.31; pKa = 4.2)
5a (Rꢀ = Ph)j
882(10)
202(14)
616(20)
716(29)
494(9)
194(13)
194(13)
194(13)
194(13)
195(8)
356(8)
358(33)
299(10)
198(17)
461(11)
478(8)
449(11)
(Set 3)
(Set 3)
(Set 3)
261(66)i
253(13)i
250(3)i
2.41
2.39
2.38
2.65
0.42
0.28
0.64
−0.17
a
1, X = H tested as reference with mutagens of Set 1.
1, X = H tested as reference with mutagens of Set 2.
1, X = H tested as reference with mutagens of Set 3.
b
c
d
e
f
Slope calculated from the gradient of the induced revertants per plate vs. the dosage in the linear range.
Members of each set scaled through its respective reference to 319 revertants per plate for 1, X = H for comparison with previous studies.
Log of the normalised slope which represents relative activity at 1 ␮mol/plate.
Computed activity from (1) – measured activity.
No dose response; non-mutagenic.
Revertants at 0.125 ␮mol/plate in Set 3.
g
h
i
j
Data from Refs. [8] and [19].

58
L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
Fig. 1. Predicted (using Eq. (1)) vs. actual mutagenicities (log10 TA100) at 1 ␮mol/plate.
7–9 in which at least one aryl ring is substituted in the
para positions with a tert-butyl group.
4. Discussion
In the second series of mutagens (10–14), the struc-
tures are characterised by their aliphatic amide side
chains and by their varying degrees of branching adja-
cent to the amide carbonyl.
Fig. 1 depicts predicted versus observed mutagenic
activities for the full set of mutagens from which the
QSAR in Eq. (1) was derived (Fig. 1, all squares). The
differences between computed and experimental activ-
ities for a range of tert-butylated mutagens are given in
Table 1. The mutagenicities of all mono 4-tert-butylated
mutagens from previous studies (1, X = But, 7a, 8c)
and those in this study (8a and 8b) are predicted well
according to the QSAR in Eq. (1), as is that for N-
acetoxy 4ꢀ-tert-butylbenzyloxy-4-tert-butylbenzamide
(7b) (Fig. 1, closed squares). In particular, the activities
of the triarylated mutagens 8a–c bearing a single
tert-butyl group are adequately predicted. Binding and
reaction of these mutagens with DNA in the major
groove would appear to be unimpeded by the presence,
on one aryl ring, of the bulky tert-butyl group. The
observed mutagenicities for the four di-tert-butylated
N-benzoyloxy-N-benzyloxybenzamides (8d–f and 9a)
deviate significantly from the predicted values (Table 1,
Fig. 1, open circles,) and only 8f can be regarded as
mutagenic, although its experimental activity is approx-
imately an order of magnitude lower than expected. The
triarylated mutagens 8g and 9b were inactive.3
In all cases, the mutagens were tested without micro-
somal activation since they are direct-acting and the
mutagens in this class that we have tested to date have
yielded similar activities both with and without S9 [6].
All the mono-tert-butylated mutagens (8a–c) and the di-
tert-butylated N-acetoxybenzyloxybenzamide (7b) were
toxic above doses of 0.25 ␮mol/plate, but gave linear
dose–response curves below this dose level. Tri-tert-
butylated mutagens 8g and 9b were inactive at any dose
and the di-tert-butylated mutagens 8d–f and 9a gave
shallow but linear dose–response curves over the full
dose range. Mutagen 8c was tested again in this series
andgaveanormalisedactivity(log TA100 = 3.2)thatwas
similar to that previously recorded (log TA100 = 3.4).
For comparison with computed activity, an average value
of log TA100 = 3.3 was used.
Mutagens 10, 11 and 13 gave excellent dose–response
curves, but 12 and 14 were toxic at higher doses
(1 ␮mol/plate).
In all cases, new mutagens were tested in conjunction
with N-acetoxy-N-butoxybenzamide (1, X = H) whose
activity at 1 ␮mol/plate was used to scale activities rela-
tive to those from previous studies.2
Lack of reactivity towards SN2 reactions at nitro-
gen would in itself account for low mutagenic activity.4
3
Mutagen 8g partially precipitated in the Ames assay however muta-
tion rates were at background level.
2
4
At 1 ␮mol/plate 1 (X = H) was scaled to 319 induced revertants and
Hydroxamic esters lacking an acyloxy group at nitrogen are inac-
the scaling factor was used to standardise mutagenicities in the same
set to those from previous studies [6,8].
tive. Steric inhibition to nucleophilic attack at nitrogen would similarly
be expected to reduce reactivity.

L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
59
Table 2
8f is less affected. While this may be attributed to the par-
ticular configuration, it is also the least reactive towards
SN2 reactions (Table 2), which, on account of the inverse
relationship between mutagenicity and reactivity, would
generate higher activity.
Bimolecular rate constants for the reaction of N-acyloxy-N-
alkoxyamides with N-methylaniline in d4-methanol
Substrate
104k303 (l mol−1 s−1
)
3, Z = H
8a
8b
8c
8d
926.2a
798.5
578.5
517.1
369.8
353.0
156.6
The mutagens 8 and 9 and most of the mutagens
we have studied to date are arylamide derivatives.
Only a handful of alkylamides have been investigated
and these were universally the acetamides (5) [8,19].
The activity of N-benzoyloxy-N-butoxyacetamide (5a,
Rꢀ = Ph) is included in Table 1 for comparison. Remark-
ably, the deviations from the predicted mutagenici-
ties increase through the series CH3 < Et < Pri < But, the
latter being totally inactive. Clearly branching at the
␣-carbon is an important factor controlling activity.
The bulky adamantamide derivative (10) shows weak
activity that deviates substantially from the predicted
activity. Comparison of the activity of the pivalamide
(11) with that of the 3,3-dimethylbutyramide (14) also
indicates a diminishing steric influence as the tert-
butyl group is moved one methylene away from the
carbonyl.
8e
8f
104k313 (l mol−1 s−1
97.2b
8.1
)
5a, Rꢀ = Ph
13
14
10
11
12
2.2
c
–
c
–
c
–
a
From Ref. [25].
From Ref. [19].
No reaction at this temperature.
b
c
However, the tert-butyl groups on 8d–g and 9a,b are
well removed from the reactive nitrogen and, as well,
SN2 reactions of mutagens 8d–f with N-methylaniline
in methanol at 303 K occur with relatively similar rate
constants to those of 8a–c (Table 2).
Unlikethemutagens8d–gand9a,btheoveralldimen-
sions of substrates 10–14 would not be expected to
impede their access to the major groove. We attribute
the systematic reduction in activities to an increase in
the activation barrier for reaction at G—N7 as a conse-
quence of branching at the ␣-carbon. The rate constants
forSN2reactionof thesesubstrateswith N-methylaniline
at 313 K (Table 2) broadly support this. The rate constant
for reaction of propanamide 13 is an order of magni-
tude smaller than that of the acetamide (5a, Rꢀ = Ph)
[19] while 12 with the isopropyl side-chain and 11
with the tert-butyl side-chain are unreactive with N-
methylaniline. In addition, while the pivalamide 11 is
unreactive, the neohexamide 14, in which the branching
is one methylene removed from the reactive nitrogen is
susceptible to SN2 reaction with N-methylaniline, albeit
with a smaller rate constant than 13. The adamantane
carboxamide mutagen (11) is also resistant to SN2 reac-
tions at nitrogen.
Clearly there are differences between reactivity and
mutagenic activity, in that the isobutyramide substrate
(12), which exhibits impaired mutagenicity, is unreac-
tive towards SN2 reactions with N-methylaniline and
the adamantamide (10), though resistant to SN2 reac-
tion with N-methylaniline at matching temperatures,
is still marginally mutagenic. However, such reactions
are also dependent upon the nature of the nucleophile
and the solvent environment. For instance, while the
tert-butyl-containing amide side-chain in 11 completely
inhibits attack by N-methylaniline at nitrogen, simi-
The steric impediment to reaction with DNA, which
is accounted for in the QSAR through Taft steric param-
eters ES1, ES2 and ES3, presupposes that the mutagen can
enter the major groove, where bulkiness of substituents
presumably impacts upon alignment of the mutagen for
reaction with guanine. The radical decrease in activ-
ity of mutagens 8d–g and 9a,b suggests that, in these
cases, the drugs cannot react with DNA since they are
either too large to enter the major groove or, once in
the groove, are unable to achieve the transition state
for reaction with G—N7. A comparison of the activ-
ity of 7b with that of 8d, or the mono tert-butylated
N-benzoyloxy-N-benzyloxybenzamides (8a–c) with the
di-tert-butylated substrates (8d–f) suggests that there is
a critical bulk beyond which, even with conformational
flexibility, inclusioninthemajorgrooveorreactionofthe
nucleotide at the electrophilic amide nitrogen becomes
difficult. The computed transition state for SN2 reaction
at nitrogen requires the “backside” attack of the nucle-
ophile on the nitrogen and approximately on line with
the bond to the acyloxy leaving group [18]. With such
constraints the transition state may be reached with one
tert-butyl group on one of the three benzene rings but
may be unattainable with two such substituents. While
all di-tert-butylated substrates exhibit no or low activity,

60
L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
with DNA [8].5 It is clear from series 8 that bulky distal
groups can have a significant influence upon the actual
groove binding with bacterial DNA.
larly substituted N-chloroamides react readily with azide
which is known to be a stronger nucleophile [20]. Dif-
ferences between experimental SN2 reaction rates and
the mutagenicity data may also be related to the influ-
ence of the hydrophobic environment within the major
groove; solvent organisation about a polar SN2 transi-
tion state accounts for substantially negative entropies
of activation in the reactions of these mutagens with
N-methylaniline [7,18,21]. Such a contribution would
presumably be different or absent in reactions at G—N7.
Overall, though, relative mutagenicities in this series
would appear to be controlled by steric inhibition of the
substitution reaction at the amide nitrogen rather than by
any steric inhibition to binding in the major groove.
These mutagens are pyramidal amides in that
their nitrogens are completely sp3 hybridised [12,22].
As a consequence, their SN2 reactivity with neu-
tral nucleophiles like N-methylaniline parallels that of
␣-haloketones with amines, which are also strongly
affected by steric effects on the ␣ꢀ-carbon [23]. In gen-
eral, the rates of SN2 reactions are strongly adversely
influenced by steric effects and branching ␤ to the
reactive centre and the same appears to be true for N-
acyloxy-N-alkoxyamides 10–14. Rates of reaction of ␣-
haloketones are also radically different with ionic nucle-
ophiles relative to neutral amino nucleophiles, which
react much more slowly. This has been attributed to
a tighter transition state with ionic nucleophiles like
acetate and azide leading to a classical SN2 transition in
which there is additional stability through orbital overlap
with the carbonyl carbon 2pz orbital [23,24].
Steric inhibition of mutagenicity by remote [26,27]
and proximal alkyl substituents [28,29] has recently been
established by Boche and coworkers who demonstrated
that the presence of increasingly bulky groups on the
remote 4ꢀ position or the 3 position of 4-nitro- (15) and 4-
aminobiphenyls (16) caused significant or complete loss
of activity in Salmonella strains TA100 and TA98. These
effects are more difficult to interpret than ours since in
both cases enzymatic activation is involved (nitroreduc-
tase in the case of 15 and cytochrome P-450 oxidases
in the case of 16) and in addition, intercalative binding
of the ultimate carcinogenic nitrenium ions, 17, formed
from both precursors is also a possibility. In the case
of the nitro compounds, the diminution in activity was
ascribed to steric inhibition of intercalation into DNA
rather than binding to nitroreductase, but steric inhibi-
tion to metabolic oxidation of the amino group of 16 was
observed. Inhibition to reaction with G-C8 was also pro-
posed to explain reduced mutagenicities. In our system,
metabolic activation is not required. Furthermore, there
is no suggestion that these simple mutagens intercalate
Boche showed that bulky groups adjacent to the nitro
or amino group also have a marked effect upon muta-
genic activity. In both cases this has been ascribed to
steric inhibition of their respective metabolic activation
steps. The respective activities of mutagens 10–14 indi-
cate that steric effects at positions close to the reactive
centre of mutagens have a marked influence upon actual
reactivity with DNA.
5. Conclusions
N-Acyloxy-N-alkoxyamides, being direct-acting
mutagens, can probe the impact of molecular structure
on interactions with DNA. Through QSAR (1) it is
possible to highlight important structural factors that
both target small molecules to DNA, or impede binding
or reaction with DNA components. In this study we
illustrate that their mutagenicity can be modified by
steric effects of two kinds. Where the steric bulk reaches
a critical level (di-tert-butylated triaryl mutagens),
regardless of their susceptibility to SN2 reactions at
nitrogen, mutagenicity is severely retarded owing to
an inability to enter the major groove and achieve the
requisite proximity to G—N7 for modification to take
place. Presumably, in these cases achievement of a
transition state for SN2 displacement at the mutagen
nitrogen would require energetically unfavourable
conformational changes that are too severe.
Mutagenicity is also moderated by lack of reactiv-
ity of the electrophilic amides. In the case of (10–14),
an increasing steric environment close to the amide
5
In Ref. [8] we reported that naphthyl containing mutagens, and one
bearing a pyrene group, produced mutagenicities that are one order of
magnitude higher than predicted by QSAR (1), which is consistent
with intercalative binding. Interestingly, biphenyl substituents appear
to give rise to only a small increase in mutagenicity, which does not
signify a strong intercalative binding to bacterial DNA in Salmonella
typhimurium TA100.

L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
61
nitrogen reduces mutagenic activity systematically and
broadly in parallel to the loss of SN2 reactivity with
N-methylaniline. This remarkable correspondence fur-
ther emphasises that the activities of these mutagens
strongly reflect their chemical and physical interactions
with DNA.
We envisage that the QSAR (1) may be used in a
similar fashion to elaborate on the influence of other
structural elements upon DNA binding. For instance, the
intercalative ability of small polycyclic aromatics, incor-
poratedintotheN-acyloxy-N-alkoxyamidestructure, has
been found to enhance mutagenicity and we have been
able to estimate that incorporation of a naphthalene or
pyrene unit appears to be worth ca. 3log P units to DNA
binding.
genicity. A review with a case study: MX compounds, Chemo-
sphere 38 (1999) 3015–3030.
[10] A.K. Debnath, A.J. Shusterman, R.L. Lopez de Compadre, C.
Hansch, The importance of the hydrophobic interaction in the
mutagenicity of organic compounds, Mutat. Res. 305 (1994)
63–72.
[11] C. Hansch, A.J. Leo, Exploring QSAR, Fundamentals and Appli-
cations in Chemistry and Biology. Part I, American Chemical
Society, Washington, DC, 1995.
[12] A.-M.E. Gillson, S.A. Glover, D.J. Tucker, P. Turner, Crys-
tal structures and properties of mutagenic N-acyloxy-N-
alkoxyamides—“most pyramidal” acyclic amides, Org. Biomol.
Chem. (2003) 3430–3437.
[13] J.H. Cooley, M.W. Mosher, M.A. Khan, Preparation and reactions
of N,Nꢀ-diacyl-N,Nꢀ-dialkoxyhydrazines, J. Am. Chem. Soc. 90
(1968) 1867–1871.
[14] A.H. Blatt (Ed.), John Wiley & Sons, New York, 1957, p. 67.
[15] M.J. Mintz, C. Walling, t-Butyl hypochlorite, Org. Synth. 49
(1963) 9.
[16] D.M. Maron, B.N. Ames, Revised methods for Salmonella muta-
genicity tests, Mutat. Res. 113 (1983) 173–215.
Acknowledgement
[17] A.K. Ghose, A. Pritchett, G.M. Crippen, Atomic physicochem-
ical parameters for three dimensional structure directed quntita-
tive structure–activity relationships. III. Modelling hydrophobic
interactions, J. Comp. Chem. 9 (1988) 80–90; MacSpartan Pro,
Wavefunction Inc., 18401 Van Karman Avenue, Suite 370, Irvine,
CA 92612, USA.
[18] S.A. Glover, SN2 reactions at amide nitrogen – theoretical mod-
els for reactions of mutagenic N-acyloxy-N-alkoxyamides with
bionucleophiles, Arkivok xii (2001) 143–160.
[19] T.M. Banks, Reactivity, Mutagenicity and DNA Damage
of N-Acyloxy-N-Alkoxyamides, University of New England,
2003.
[20] S.A. Glover, G. Mo, Hindered ester formation by SN2 azidation
of N-acetoxy-N-alkoxyamides and N-alkoxy-N-chloroamides—
novel application of HERON rearrangements, J. Chem. Soc.:
Perkin Trans. 2 (2002) 1728–1739.
The authors are grateful to the Australian Research
Council for financial support.
References
[1] R.G. Gerdes, S.A. Glover, J.F. Ten Have, C.A. Rowbottom,
N-acetoxy-N-alkoxyamides – a new class of nitrenium ion
precursors which are mutagenic, Tetrahedron Lett. 30 (1989)
2649–2652.
[2] J.J. Campbell, S.A. Glover, C.A. Rowbottom, Solvolysis and
mutagenesis of N-acetoxy-N-alkoxybenzamides – evidence for
nitrenium ion formation, Tetrahedron Lett. 31 (1990) 5377–
5380.
[3] J.J. Campbell, S.A. Glover, G.P. Hammond, C.A. Rowbottom,
Evidence for the formation of nitrenium ions in the acid-catalysed
solvolysis of mutagenic N-acetoxy-N-alkoxybenzamides, J.
Chem. Soc.: Perkin Trans. 2 (1991) 2067–2079.
[4] A.M. Bonin, S.A. Glover, G.P. Hammond, Reactive interme-
diates from the solvolysis of mutagenic O-alkyl N-acetoxy-
benzohydroxamates, J. Chem. Soc.: Perkin Trans. 2 (1994)
1173–1180.
[21] S.A. Glover, A. Rauk, A computational investigation of the
structure of the novel anomeric amide N-azido-N-methoxy-
formamide and its concerted decomposition to methyl formate
and nitrogen, J. Chem. Soc.: Perkin Trans. 2 (2002) 1740–
1746.
[22] S.A. Glover, Anomeric amides—structure, properties and reac-
tivity, Tetrahedron 54 (1998) 7229–7272.
[23] N. De Kimpe, R. Verhe´ (Eds.), The Chemistry of ␣-Haloketones,
␣-Haloaldehydes and ␣-Haloimines, John Wiley and Sons,
Chichester, 1988.
[24] J.W. Thorpe, J. Warkentin, Stereochemical and steric effects in
nucleophilic substitution of ␣-halo ketones, Can. J. Chem. 51
(1973) 927–935.
[25] J.J. Campbell, S.A. Glover, Bimolecular reactions of muta-
genic N-(Acyloxy)-N-alkoxybenzamides with aromatic amines,
J. Chem. Res. (S) 8 (1999) 474–475.
[26] M. Klein, L. Erdinger, G. Boche, From mutagenic to non-
mutagenic nitroarenes: effect of bulky alkyl substituents on the
mutagenic activity of nitroaromatics in Salmonella typhimurium.
Part II. Substituents far away from the nitro group, Mutat. Res.
467 (2000) 69–82.
[5] A.M. Bonin, S.A. Glover, G.P. Hammond,
A compari-
son of the reactivity and mutagenicity of N-benzoyloxy-
N-benzyloxybenzamides, J. Org. Chem. 63 (1998) 9684–
9689.
[6] T.M. Banks, A.M. Bonin, J.J. Campbell, S.A. Glover, G.P.
Hammond, A.S. Prakash, C.A. Rowbottom, Mutagenicity of
electrophilic N-acyloxy-N-alkoxyamides, Mutat. Res. 494 (1–2)
(2001) 115–134.
[7] T.M. Banks, A.M. Bonin, S.A. Glover, A.S. Prakash, Mutagenic-
ity and DNA damage studies of N-acyloxy-N-alkoxyamides—the
role of electrophilic nitrogen, Org. Biomol. Chem. 1 (2003)
2238–2246.
[8] L.E. Andrews, T.M. Banks, A.M. Bonin, S.F. Clay, A.-M.E.
Gillson, S.A. Glover, Mutagenic N-acyloxy-N-alkoxyamides—
probes for drug–DNA interactions, Aust. J. Chem. 57 (2004)
377–381.
[9] K. Tuppurainen, Frontier orbital energies, hydrophobicity and
steric factors as physical QSAR descriptors of molecular muta-
[27] C. Glende, M. Klein, H. Schmitt, L. Erdinger, G. Boche, Transfor-
mation of mutagenic aromatic amines into non-mutagenic species
by alkyl substituents. Part II. Alkylation far away from the amino
function, Mutat. Res. 515 (2002) 15–38.

62
L.E. Andrews et al. / Mutation Research 605 (2006) 51–62
[28] M. Klein, U. Voigtmann, T. Haack, L. Erdinger, G. Boche,
From mutagenic to non-mutagenic nitroarenes: effect of bulky
alkyl substituents on the mutagenic activity of 4-nitrobiphenyl in
Salmonella typhimurium. Part I. Substituents ortho to the nitro
group and in 2ꢀ-position, Mutat. Res. 467 (2000) 55–68.
[29] C. Glende, H. Schmitt, L. Erdinger, G. Engelhardt, G.
Boche, Transformation of mutagenic aromatic amines into non-
mutagenic species by alkyl substituents. Part I. Alkylation ortho
to the amino function, Mutat. Res. 498 (2001) 19–37.