Mutagenicity of N-acyloxy-N-alkoxyamides as an indicator of DNA intercalation part 1: Evidence for naphthalene as a DNA intercalator

Article abstract of DOI:10.1039/c6ob00162a

N-Acyloxy-N-alkoxyamides are direct-acting mutagens in S. typhimurium TA100 with a linear dependence upon log P that maximises at log P₀ = 6.4. Eight N-acyloxy-N-alkoxyamides (2-9) bearing a naphthalene group on any of the three side-chains and with log P₀ < 6.4 have been demonstrated to be significantly and uniformly more mutagenic towards S. typhimurium TA100 than 50 mutagens without naphthalene. The activity enhancement of 2-9 is likely due to intercalative binding of naphthalene to bacterial DNA as a number are also active in TA98, a frame-shift strain of S. typhimurium, which is modified by intercalators. DNA damage profiles for naphthalene-bearing mutagens confirm enhanced reactivity with DNA when naphthalene is incorporated and a different binding mode when compared to mutagens without naphthalene. The effect is independent of whether the naphthalene is attached to an electron-donating alkyl or electron-withdrawing acyl group, alkyl tether length or, in the case of 6 and 7, the point of attachment to naphthalene. A new quantitative structure activity relationship has been constructed for all 58 congeners incorporating log P and an indicator variable, I, for the presence (I = 1) or absence (I = 0) of naphthalene and from which the activity enhancing effect of a naphthalene has been quantified at between three and four log P units. Contrary to conventional views, simple naphthalene groups could target molecules to DNA through intercalation.

Full text of DOI:10.1039/c6ob00162a

View Article Online
View Journal
Organic &
Biomol ec ular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: T. M. Banks, S. A.
Glover, S. F. Clay and R. R. Schumacher, Org. Biomol. Chem., 2016, DOI: 10.1039/C6OB00162A.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/obc

Page 1 of 39
Organic & Biomolecular Chemistry
Mutagenicity of N-Acyloxy-N-alkoxyamides as an indicator of DNA
View Article Online
DOI: 10.1039/C6OB00162A
intercalation
†
Part 1: Evidence for naphthalene as a DNA intercalator
*
*
Tony. M. Banks, Samuel F. Clay, Stephen A. Glover and Rhiannon S. Schumacher.
Chemistry Department, School of Science and Technology, University of New England, Armidale
New South Wales 2351, Australia.
Abstract
N-Acyloxy-N-alkoxyamides are direct-acting mutagens in S. typhimurium TA100 with a linear
dependence upon LogP that maximises at LogP = 6.4. Eight N-acyloxy-N-alkoxyamides (2-9)
o
bearing a naphthalene group on any of the three side-chains and with LogP < 6.4 have been
o
demonstrated to be significantly and uniformly more mutagenic towards S. typhimurium TA100
than 50 mutagens without naphthalene. The activity enhancement of 2-9 is likely due to
intercalative binding of naphthalene to bacterial DNA as a number are also active in TA98, a frame-
shift strain of S. typhimurium, which is modified by intercalators. DNA damage profiles for
naphthalene-bearing mutagens confirm enhanced reactivity with DNA when naphthalene is
incorporated and a different binding mode when compared to mutagens without naphthalene. The
effect is independent of whether the naphthalene is attached to an electron-donating alkyl or
electron-withdrawing acyl group, alkyl tether length or, in the case of 6 and 7, the point of
attachment to naphthalene. A new quantitative structure activity relationship has been constructed
for all 58 congeners incorporating LogP and an indicator variable, I, for the presence (I=1) or
absence (I=0) of naphthalene and from which the activity enhancing effect of a naphthalene has
been quantified at between three and four LogP units. Contrary to conventional views, simple
naphthalene groups could target molecules to DNA through intercalation.
†
Electronic supplementary information (ESI) is available.
1

Organic & Biomolecular Chemistry
Page 2 of 39
Introduction
View Article Online
DOI: 10.1039/C6OB00162A
N-Acyloxy-N-alkoxyamides 1 are members of the class of anomeric amides, amides substituted at
1, 2
nitrogen with two heteroatoms.
These amides are unusual in that the nitrogen is strongly
pyramidal and this, together with the inductive effect of the two heteroatoms, results in a much
1
-5
weaker resonance between the nitrogen lone pair and the carbonyl. In this case, the amides are
electrophilic agents, reacting with nucleophiles at the nitrogen with displacement of the acyloxyl
4
, 6-12
group.
O
R1
R2
O
N
O
O
R3
1
Since they were first synthesised, they have been shown to be direct-acting mutagens and a wide
number that have been tested in the Ames reverse mutation assay are mutagenic towards Salmonella
4
, 7, 13-22
typhimurium TA100 without the requirement for metabolic activation.
Well over a hundred
congeners have now been synthesised and tested and several quantitative structure activity
relationships (QSAR's) have evolved that can explain, with reasonable accuracy, their mutagenic
4
, 18-22
activity in TA100.
For congeners with LogP < 6, a QSAR based on the activity of fifty
congeners, equation (1), relates the activity to hydrophobicity (LogP) of the intact mutagen, the
stability to extraneous S 2 reactions through pK of the acyloxyl group and, in pertinent cases,
N
A
1
2
through Taft steric parameters to the size of para substituents on benzyloxy (E ), benzoyloxy (E )
s
s
3
s
21
2
and benzamide (E ) side chains. The leave-one-out cross validation coefficient, Q of 0.7 indicates
good predictive ability.
1
S
2
3
Log TA100 = 0.28 (±0.03) LogP + 0.18 (±0.08) pK + 0.13 (±0.03) E + 0.15 (±0.04) E + 0.11 (±0.05) E
S
A
s
+
1.02 (±0.4)
2
2
2
n = 50, r = 0.7968, adj. r = 0.7736, s = 0.16, F = 34.5; LOOCV Q =0.6987
(1)
The mutagenic activity of 1 undoubtedly derives from their electrophilic behaviour towards DNA,
as our DNA damage studies have shown that they damage DNA at guanosine-N7 (N7-G) and
1
8, 23-27
adenosine-N3 (N3-A), the most nucleophilic centres in duplex DNA.
Such damage renders
1
8
DNA susceptible to cleavage reactions from which the reactive bases can be deduced.
2

Page 3 of 39
Organic & Biomolecular Chemistry
The QSAR in equation (1) permits identification of other structural factors that impact upon
View Article Online
DOI: 10.1039/C6OB00162A
1
mutagenicity. For instance, branching in R on the carbon α to the amide carbonyl in 1
results in loss of mutagenicactivity, which is accounted for by a corresponding resistance to SN2
reactivity at nitrogen. In those substrates, the limiting factor in Fig. 1 is a low rate of reactivity with
4
, 20
DNA bases, ie a low kM.
In other cases, large mutagens with dimensions similar to those of the
major groove of DNA show radically reduced activity when compared to the predictions based on
1
3
equation (1). Such mutagens are the triaryl N-benzoyloxy-N-benzyloxybenzamides (1, R =R =Ph,
2
R =Bn) bearing at least two para tert-butyl groups. In these cases, complexation with DNA is
4
, 20
limiting due to a low KM,
although their higher hydrophobicity and a reduced K also plays a
L
2
2
role.
DNA
DNA
Cell membrane
KM
kM
Damaged
DNA
Drug
Drug
Drug
KL
Cytosol
Complex
Fig. 1 Elemental processes in reaction of N-acyloxy-N-alkoxyamides with DNA.
3
Branching at R on the α-carbon on the acyloxyl side chain also results in radically reduced activity
but these mutagens readily undergo S 2 reactions at nitrogen, a process unimpeded by steric bulk
N
on the leaving group. Mutagens in this class are relatively small so K cannot be limiting. Here we
m
have concluded that steric bulk prevents binding in the DNA grooves with an orientation that
4
, 21
permits S 2 reaction at the amide nitrogen.
Finally, a series of N-acyloxy-N-alkoxyamides with
N
2
11
branching in R close to the alkoxyl group oxygen are slow to undergo S 2 reaction at nitrogen,
N
but are nonetheless as mutagenic as predicted by equation (1). In this case, we have suggested that
SN2 reactivity is easier in the hydrophobic environment of the major groove where the disadvantage
4
, 21
of desolvation (a powerful rate limiting factor in solution reactions) is not operative.
.
We have recently established that where N-acyloxy-N-alkoxyamides are much more hydrophobic
2
2
(
LogP > 6), a bilinear relationship holds (equation (2)) with maximal LogP, LogP of 6.4.
0
3

Organic & Biomolecular Chemistry
Page 4 of 39
LogTA100 = 0.243 (±0.02) LogP - 0.667 (±0.12) Log(βP + 1) + 0.106 (±0.08) pKA + 0.088
View Article Online
DOI: 10.1039/C6OB00162A
1
2
3
(
±0.03) E + 0.092 (±0.04) E + 0.022 (±0.05) E + 1.466 (±0.41)
S S S
2
2
2
Logβ = -6.639, n = 55, r = 0.7458, adj. r = 0.714, s = 0.18, F = 19.7017; LOOCV Q = 0.5991,
(2)
The positive LogP dependency in this QSAR relates to hydrophobic binding to cytosolic DNA
affecting K in Fig. 1, while above LogP the negative dependence most probably relates to normal
M
0
lipid entrapment, in this case in the bacterial membrane, thereby diminishing K in Fig. 1.
L
In earlier studies, we found that N-acetoxy-N-butoxy-2-naphthamide 2 exhibited mutagenicity,
which is about an order of magnitude higher than is predicted by equation (1) and equation (2), an
4
, 17-19, 22
observation supported by DNA damage studies.
Relative to several N-acyloxy-N-
alkoxybenzamides, N-acetoxy-N-butoxy-2-naphthamide, with a significantly lower LogP, clearly
1
8
induced more damage to plasmid DNA at N7-G. In this paper we demonstrate that incorporation
of naphthyl group on any of the three side chains of 1 results in similarly enhanced levels of
mutagenicity and DNA damage, which we demonstrate is most probably owing to intercalation of
naphthyl at specific sites of DNA.
4

Page 5 of 39
Organic & Biomolecular Chemistry
Results and Discussion
View Article Online
DOI: 10.1039/C6OB00162A
The enhanced DNA damage observed with 2, as well as its increased mutagenic activity was
attributed to the potential for the naphthyl moiety to intercalate between the bases of DNA, resulting
in stronger drug-DNA interactions and increased K in Fig. 1. Through intercalative binding, an N-
M
acyloxy-N-alkoxyamide would have a longer residence time on DNA than if DNA binding was
simply under classical diffusion and hydrophobic control. It was postulated that this extended
28
residence time could increase the time dependant probability for reaction with the bases of DNA.
To establish whether this tenet holds for other naphthalene-substituted mutagens, a series of
congeners, 3-9, each bearing a single naphthalene unit on one of the three side chains of 1 were
synthesised and subjected to our standard mutagenicity assay using point mutation strain S.
typhimurium TA100.
O
O
O
O
O
O
O
O
N
O
N
O
N
N
O
O
O
OAc
O
2
3
4
5
O
O
O
O
O
O
O
N
N
N
O
N
OAc
OAc
OAc
OAc
7
8
9
6
Synthetic procedures
The N-acyloxyl derivatives 13 of the appropriate hydroxamic ester 11 were obtained according to
our established protocol (Scheme 1) whereby the N-chloro derivative 12 was treated with the
7
, 11, 15-18, 21, 22
required sodium carboxylate in anhydrous acetone.
Hydroxamic esters 11 were
synthesised by reaction of the potassium salt of the hydroxamic acid 10 with the appropriate alkyl
2
9
bromide, or for 2, by reaction of n-butoxyamine with 2-naphthoyl chloride. The synthesis of 8 and
required 2-(2'-naphthyl)ethyl and 3-(2'-naphthyl)propyl bromides, which were obtained by side
chain elongation of 2-naphthylmethyl bromide (Scheme 2) and 2-naphthaldehyde respectively
Scheme 3).
9
(
5

Organic & Biomolecular Chemistry
Page 6 of 39
2
O
t
O
3
O
O
Bu OCl
R Br
R CO Na
2
OR2
OR2
2
View Article Online
O-K+
OR
R1
N
H
R1
N
R1
N
R1
DOI: 10.1039/C6OB00162A
N
H
Dry acetone
3
Cl
O CR
2
10
11
12
13
Scheme 1
OH
Br
Br
CN HCl
1. LiAlH4
. HBr
KCN
O
2
Scheme 2
O
O
O
1
. LiAlH4
CH (CO H)
2
Pd/C
2
2
H
OH
OH
Br
α-picoline
H2
2. HBr
Scheme 3
1
3
Table 1 IR carbonyl absorption frequencies (CHCl ) and C chemical shifts (CDCl ) for 2 and 9
3
3
Compound
Amide C=O
Ester C=O
-1
13
-1
13
IR (cm )
1724
1737
1740
1739
1739
1740
1735
1740
C (CDCl3)
IR (cm )
C (CDCl3)
1
7
2
3
4
5
6
7
8
9
174.0
176.5
176.3
176.2
176.4
176.3
176.3
176.2
1792
1757
1789
1798
1793
1793
1793
1794
168.0
164.3
168.3
164.6
167.9
168.0
168.0
167.9
1
3
Table 1 summarises the main infrared carbonyl stretch frequencies and carbonyl C NMR shifts for
mutagens 3-9 together with corresponding data for 2. Their amide carbonyl stretch frequencies are
-1
typical of all other acetamide analogues, which are found between 1735 -1745 cm and somewhat
4
higher than aromatic amides such as 2. The data are characteristic of N-acyloxy-N-alkoxyamides
where the very high amide carbonyl stretch frequency is due to the much reduced resonance owing
to the pyramidal nature of the amide nitrogen and to destabilisation of the charge-separated, single
bond form of the carbonyl on account of the gross electronegativity of the amide nitrogen group.
-1
The alkanoyloxyl carbonyls are uniformly high at around 1790 cm and, typically, aryl conjugation
6

Page 7 of 39
Organic & Biomolecular Chemistry
1
3
in 3 results in a lower frequency. The amide and ester carbonyl carbon C NMR shifts are also
View Article Online
DOI: 10.1039/C6OB00162A
characteristic. In this class of amides, most acetamide carbonyls resonate around δ176, while ester
4
carbonyls generally resonate in the region of δ160-170.
7

Organic & Biomolecular Chemistry
Page 8 of 39
Mutation studies
View Article Online
DOI: 10.1039/C6OB00162A
Activity towards S. typhimurium TA100
Mutagens 2 – 9 were subjected to the Ames assay in S. typhimurium TA100 without microsomal
activator S9, since all N-acyloxy-N-alkoxyamides are direct-acting. The standard, N-acetoxy-N-
butoxybenzamide 14, which we have consistently used to scale different test batches and which has
provided a means of normalising activity of all mutagens in all our studies on N-acyloxy-N-
alkoxyamides to date, was tested in conjunction with each batch of mutagens. The dose-response
data for naphthalene bearing mutagens 2 to 9 together with that for 14 is presented in Tables 2-4.
a
Table 2 Dose response data for 2, 3, 6, 7 and 14 in S. typhimurium TA100 without S9.
µmol/plate 14
µmol/plate 2
3
6
µmol/plate 7
0
0
0
1
1
.000
.250
.500
.000
.500
211.3 (12) 0.000
441.0 (6) 0.015
634.3 (30) 0.030
938.7 (69) 0.060
[T]
209.7 (12) 209.7 (12) 209.7 (12) 0.000
224.7 (12) 238.3 (15) 241.7 (15) 0.060
418.0 (34) 517.0 (24) 420.3 (32) 0.125
744.7 (57) 759.3 (21) 680.0 (23) 0.250
0.500
211.3 (12)
727.7 (57)
1157.3 (59)
1442.0 (69)
1962.7 (40)
7554.7
Gradient
717.71
9482.5
9844.4
8281.3
a
Counts displayed are average of three plates per dose with standard deviation in parentheses;T indicates toxicity
at this dose
a
Table 3 Dose response data for 8, 9 and 14 in S. typhimurium TA100 without S9.
µmol/plate 14
µmol/plate 8
9
0
0
0
0
1
2
.000
.120
.250
.500
.000
.000
165 (24) 0.000
165 (24)
794 (13)
165 (24)
987 (37)
315 (28) 0.060
379 (21) 0.120
678 (43) 0.250
1311 (62) 0.500
1081 (90) 1.000
1143.1
1276 (37) 1453 (61)
2044 (59) 2024 (69)
2768 (129) 1773 (102)
1651 (101) [T]
Gradient
9252.8
10731.0
a
Counts displayed are average of three plates per dose with standard deviation in parentheses;T indicates
toxicity at this dose
8

Page 9 of 39
Organic & Biomolecular Chemistry
a
Table 4 Dose response data for 4, 5 and 14 in S. typhimurium TA100 without S9.
View Article Online
DOI: 10.1039/C6OB00162A
µmol/plate
14
µmol/plate
0.000
4
5
0
0
0
0
1
.000
.125
.250
.500
.000
91 (9)
91 (9)
91 (9)
127 (16)
146 (28)
200 (21)
312 (7)
217.0
0.015
110 (14)
143 (10)
190 (21)
219 (14)[T]
1694.7
148 (16)
166 (20)
217 (11)
271 (27)[T]
1980.4
0.030
0.060
0.125
Gradient
a
Counts displayed are average of three plates per dose with standard deviation in parentheses;T indicates
toxicity at this dose
All compounds showed a good dose response with some deviation from linearity at concentration
above 0.5 µmol/plate owing to toxicity (see ESI). The gradients were obtained from the linear
region of the dose-response plots and scaled as usual to N-acetoxy-N-butoxybenzamide 14 (319
revertants/plate at 1 µmole). The scaled activity of N-acetoxy-N-butoxy-2-naphthamide in this study
1
7
(LogTA100 = 3.62) was almost identical to that previously determined (LogTA100 = 3.56)
All members of this naphthyl series are significantly more mutagenic than non-naphthyl substituted
benzamides and acetamides upon which the linear QSAR in equation (1) is based. Computed LogP
values, pK 's (of the carboxylic acid corresponding to the leaving group) required for application of
A
equation (1) or equation (2), gradients and scaled gradients (activities at 1 µmole/plate) are
‡
22
presented in Table 5. All LogP values for 2-9 are well below LogP (LogP = 6.4), therefore it
O
O
is possible to apply the linear QSAR (equation (1)) to give the predicted mutagenicities, which are
also presented in Table 5 together with the differences between theoretical and experimental
activities.
‡
1
2
3
Taft steric parameters ES ,Es and ES are not applicable to 2-9 and are given a value of zero.
9

Organic & Biomolecular Chemistry
Page 10 of 39
Table 5 Experimental and predicted mutagenicity for 2-9 in S. typhimurium TA100 when
View Article Online
LogTA100 is calculated using linear QSAR in equation (1).
DOI: 10.1039/C6OB00162A
Scaled
Experimental Predicted
a
2
R
c
Compound LogP pKA Gradient
Diff.
b
Gradient LogTA100
LogTA100
2.86
2
3
4
5
6
7
8
9
3.44 4.76 9483
3.44 4.16 9844
3.38 4.24 1695
3.80 4.57 1980
2.03 4.76 8281
2.03 4.76 7555
2.31 4.76 9253
2.73 4.76 10733
0.95 4214
0.95 4375
0.99 2491
0.95 2911
0.97 3680
0.90 3358
0.98 2583
0.99 2995
3.62
3.64
3.40
3.46
3.53
3.57
3.41
3.48
-0.76
-0.89
-0.65
-0.53
-1.06
-1.10
-0.87
-0.82
2.75
2.75
2.93
2.47
2.47
2.54
2.66
a
Induced revertants at 1µmol/plate
b
c
Log of scaled gradient
Predicted LogTA100 – Experimental LogTA100
The experimental and predicted activities for the naphthalene series, together with the training set of
2
1, 22
5
0 compounds without naphthalene as a substituent,
are illustrated in Fig. 2 and show that the
mutagenicities of all naphthalene-containing compounds are significantly under-predicted by
equation (1). From differences in Table 5, the average discrepancy between experimental and
predicted activities at 1µmol/plate is in the order of 1 LogTA100 unit. There is clearly an additional
factor enhancing the mutagenic activity of N-acyloxy-N-alkoxyamides bearing naphthalene groups,
which within this series, is operative irrespective of which side chain is functionalised, length of
tethering groups and electronic effects; naphthalene is substituted by both +I alkyl groups (4-9) as
well as –M carbonyl groups (2 and 3). 1- and 2-naphthylmethoxy substituents in 6 and 7 also
produce similar enhancement.
The positive hydrophobic correlation with mutagenicity for the compounds previously tested (grey
squares in Fig. 2) and embodied in the modest LogP term in equation (1), has been ascribed to a
hydrophobic interaction between DNA and the groove-bound mutagens. However, in the case of the
naphthalene series it is likely that the increased binding due to the naphthyl rings is due to
intercalation in the DNA matrix. This type of binding involves π-stacking and charge transfer, both
of which result in much stronger drug-DNA interactions and longer residence time on DNA. With
substrates 2-9, K (Fig. 1) would be larger leading to a higher probability of reaction with DNA.
M
1
0

Page 11 of 39
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C6OB00162A
4
.00
.50
3
0
0
1
A
3.00
5
2
T
g
o
L
3
4
9
d
e
t
c
i
2.50
2.00
1.50
8
d
7
e
r
6
P
1.5
2
2.5
3
3.5
4
Experimental LogTA100
Fig. 2 Predicted vs experimental mutagenicities for N-acyloxy-N-alkoxyamides 2-9 (solid squares)
and training set (open squares) in S. typhimurium TA100 when predicted by the linear QSAR in
equation (1)
Activity towards S. typhimurium TA98
To support the TA100 results, four of the naphthalene series (2, 3, 6 and 7) with mono-substitituted
naphthalene on the three possible side chains were also tested in S. typhimurium TA98, a frameshift
mutation detection strain. Since the space requirement for an intercalated PAH (3.4 Å) is similar to
30-36
37
that required by a single base pair,
intercalation of a PAH between the bases of DNA reinstates
3
6, 38
the wild type reading frame in S. typhimurium through a +1 frame shift.
Furthermore, to see
3
9
significant reversion of the TA98 strain, a mutagen must be bifunctional. Mutagenic potency of an
intercalator can be increased 10 to 100-fold through its attachment to a side-chain capable of a
3
6, 38-40
covalent reaction with DNA, resulting in localisation of the intercalator.
Accordingly, N-
acyloxy-N-alkoxyamides bearing an intercalating PAH would be well set up to produce a positive
response with the TA98 strain.
Initial studies with TA98 were carried out with N-acetoxy-N-butoxy-2-naphthamide, 2, together
with four other substrates for comparison. 14 and 15 are N-acyloxy-N-alkoxyamides that are
17-22
mutagenic in TA100,
but are lacking a naphthalene group, while 16 and 17 contain naphthalene
1
1

Organic & Biomolecular Chemistry
Page 12 of 39
but lack a reactive nitrogen. Both of these lack a leaving group at nitrogen though N,N-
View Article Online
, 41, 42
DOI: 10.1039/C6OB00162A
5
dialkoxyamide 16 has a similar, pyramidal nitrogen and reduced resonance to that in 2.
Hydroxamic ester 17 has a largely planar, strongly conjugated nitrogen. In earlier studies we have
13
shown that an N,N-dialkoxybenzamide and an N-alkylhydroxamic ester are non-mutagenic in
1
8
TA100. Dose-response data for TA98 studies are presented in Table 6.
O
O
O
O
OBu
OBu
OBu
N
OBu
N
N
N
OAc
O
O
OEt
Me
14
15
16
17
a
Table 6 Dose-response data for mutagenicity of 2, 3, 6 and 7 and 14-17 in S. typhimurium TA98
b
c
2
R
Test Set
µmol/plate
0.06 0.125
Gradient
0
.00
0.03
0.25
2
2
2
2
2
2
1
1
3
1
6
7
1
A
B
C
D
E
F
47 (7) 59 (6) 79 (10) 112 (5) 160 (17)
31 (1) 59 (8) 71 (14) 93 (19) 166 (16)
459
0.992
0.986
0.996
0.988
0.996
0.991
0.43
511
40 (7) 59 (8) 66 (4)
33 (4) 46 (8) 65 (3)
26 (3) 36 (3) 51 (5)
24 (2) 32 (2) 38 (1)
33 (4)
98 (16) 148 (11)
95 (7) 138 (11)
67 (14) 122 (11)
424
420
762
45 (1)
40 (8)
34 (7)
29 (5)
24 (4)
35 (7)
34 (1)
22 (3)
72 (4)
38 (2)
29 (5)
29 (5)
25 (2)
40 (3)
61 (6)
23 (4)
371
4
5
D
D
E
E
F
18 (20)
33 (4)
-16 (-17)
25 (15)
0.57
26 (3) 28 (3)
27 (2)
0.566
0.018
0.924
0.942
0.370
6
7
26 (3) 24 (1) 27 (1)
-3 (-2)
24 (2) 29 (1)
24 (2) 29 (5)
31 (4)
30 (6)
114 (141)
287 (355)
-15 (-18)
F
F
24 (2) 25 (2) 29 (2)
a
Counts displayed are average of three plates per dose with standard deviation in parentheses
b
Measurements for 2 were carried out independently in Sets A-C and in conjunction with 14 and 15 in Set D, 3 and 16
in Set E and 6, 7 and 17 in Set F
c
Activities at 1 µmol/plate scaled to the activity of 2 = 459 in Set A in parentheses.
N-Acetoxy-N-butoxy-2-naphthamide 2, bearing both an electrophilic nitrogen as well as
intercalating naphthalene, was strongly and reproducibly active in TA98 over the dose range with
1
2

Page 13 of 39
Organic & Biomolecular Chemistry
LogTA98 in excess of 2 in each case. Like 2, naphthalene-bearing mutagens 6 and 7 were also
View Article Online
DOI: 10.1039/C6OB00162A
active (LogTA98 > 2). All three possess a localising electrophilic centre and induce +1 frameshift
mutations in the TA98 strain. However, 3 with the naphthalene on the leaving group is completely
inactive in TA98, which we ascribe to the fact that the leaving group bearing the naphthalene is
displaced by electrophilic attack on DNA resulting in loss of localisation and the duality of the
mutagen-DNA interaction that enhances the TA98 response. Both naphthalene derivatives 16 and
1
1
7 were completely inactive reflecting their lack of an electrophilic nitrogen. The mutagens 14 and
5, lacking a naphthyl group, while active in TA100, were as expected inactive in TA98.
Comparison of the activity of 2 with inactivity of 14 indicates that the naphthyl, as opposed to the
phenyl aromatic substituent, interacts with DNA in a different manner.
Based on the significantly enhanced activity of all the naphthalene-bearing mutagens in S.
typhimurium TA100 and the activity of those tested in S. typhimurium TA98, we conclude that
naphthyl groups most probably intercalate with DNA bases, thereby increasing the binding
equilibrium constant K in Fig. 1 and residence time on DNA, with attendant increase in probability
M
of electrophilic reaction with a proximal DNA base.
O
O
O
O
O
CONH2
CONH2
OR
O
O
MeO
MeO
18
19
20a R=H
0b R=Me
2
There is significant debate in the literature as to the intercalative ability of naphthalenes. Many
studies exploit the intercalation of naphthalene mono- and di- imides, well-known and established
43-45
threading intercalators.
The naphthalene unit is broadly discussed as only acting as an
4
6
intercalator when facilitated by the presence of positively charged side-chains, or acyl
4
6, 47
substitution.
There is recent evidence that derivatives of the “left half” of Azinomycin B
modified to remove aziridine alkylating activity 18 and which have an acylated side-chain
naphthalene motif similar to that of the N-acyloxy-N-alkoxyamides 2 and 3 bind intercalatively to
4
8
DNA. However the unsubstituted naphthalene system in the same structure 19 was shown to be
4
8
ineffective in DNA binding. The crucial role played by substituents in the intercalation of the
naphthalene moiety in their study suggested that π-electron density distribution is likely to make a
significant contribution to the intercalative potential of the naphthalene group. Computational
1
3

Organic & Biomolecular Chemistry
Page 14 of 39
studies by Alcaro et al. have been unable to clearly define the role of the naphthalene moiety in
View Article Online
4
9, 50
51, 52
DOI: 10.1039/C6OB00162A
suggests that there is no intercalation of the naphthalene
Azinomycin,
but Coleman et al.
moieties in 18, 20a and 20b. Casely-Hayford is equivocal on the intercalative binding ability of
53
naphthalene component of azinomycins. The results reported here appear to be the first to show an
activating effect of a simple naphthalene moiety over and above straightforward hydrophobic
groove-binding. While attractive electrostatic interactions are important with polarised intercalators
35
such as polycyclic heteroaromatics or neutral polycyclics bearing a positively charged side chains,
these studies clearly show that a neutral, at best weakly polarised naphthyl group can intercalate to
an extent that influences interaction and thereby reactivity with DNA bases. Notably, naphthyl
groups with both electron-withdrawing acyl substituents (in 2 and 3) and electron-donating alkyl
substituents (in 4-9) behave similarly in this study. The finding also represents a departure from the
2
accepted view that PAH's with at least three fused rings provide an optimum surface area of 28Å
3
0, 54-56
for intercalation.
Secondary interactions such as hydrogen bonding can also promote
intercalation of substructure and such bonding to the nitrogens of N-acyloxy-N-alkoxyamides could
occur as they are probably more basic relative to normal amides on account of their amine-like
character. However, the properties of the nitrogen in 16, which was inactive in TA98 studies, are
similar to those of 2. Intercalation of the naphthalene groups of N-acyloxy-N-alkoxyamides 2-9
would possibly be driven by hydrophobicity and π-π-stacking resulting in van der Waals dispersion
and weak electrostatic effects as well as HOMO-LUMO interactions.
Quantitative structure-activity relationships
QSARs in equation (1) and equation (2) can be modified to incorporate an indicator variable I
which has the value I=1 for 2-9, where naphthalene is present and I=0 for all other mutagens in our
training set. Such an indicator variable has been used in QSAR as a descriptor to define the
difference between congeners in a series of compounds with PAH of three or more fused rings (I =
5
7, 58
1
), and congeners with two or less rings (I = 0).
The revised linear and bilinear QSARs are presented in equation (3) and equation (4).
Log TA100 = 0.26 (±0.03) LogP + 0.17 (±0.08) pKA
1
2
3
+
0.12 (±0.03) E + 0.14 (±0.04) E + 0.08 (±0.05) E + 0.83 (±0.06) I + 1.12 (±0.41)
S s S
2
2
2
n = 58, R = 0.85, adj. R = 0.83, s = 0.16, F = 48.2; LOO CV Q = 0.85
(3)
1
S
LogTA100 = 0.23 (±0.02) LogP - 0.65 (±0.12) Log(βP + 1) + 0.11 (±0.08) pK + 0.09 (±0.03) E
A
2
3
+
0.09 (±0.04) E + 0.01 (±0.05) E + 0.85 (±0.07) I + 1.48 (±0.40)
S S
2
2
2
Logβ = -6.705, n = 63, R = 0.81, adj. R = 0.79, s = 0.18, F = 29.6; LOO CV Q = 0.76
(4)
1
4

Page 15 of 39
Organic & Biomolecular Chemistry
From the coefficients in equation (4) a revised LogP value of 6.44 can be calculated, slightly
O
View Article Online
DOI: 10.1039/C6OB00162A
higher than 6.37 calculated from Equation 2
Calculated and predicted LogTA100 from equation (3) and equation (4) using I = 1 are given in
Table 7. The differences between experimental and predicted mutagenicities are significantly
diminished when compared to activities from equation (1) (Table 5).
Table 7 Experimental and predicted mutagenicities for 2-9 in S. typhimurium TA100 when
LogTA100 is calculated using linear QSAR in equation (3) and bilinear QSAR in equation (4)
Mutagen
Linear QSAR
Bilinear QSAR
b
b
Experimental
Predicted
Difference
0.09
Predicted
Difference
0.06
a
2
3
4
5
6
7
8
3.59
3.68
3.58
3.57
3.74
3.31
3.31
3.39
3.50
3.65
3.59
3.58
3.72
3.33
3.33
3.39
3.49
3.64
3.4
-0.06
0.17
-0.05
0.18
3.46
3.53
3.57
3.41
3.48
0.28
0.24
-0.22
-0.26
-0.02
0.02
-0.20
-0.24
-0.02
0.01
9
a
An averaged value for 2 is used in derivation of equation (3) and equation (4)
b
Predicted LogTA100 – Experimental LogTA100
Fig. 3 illustrates the fit of the predicted data for naphthalene-containing mutagens based on the
linear QSAR in equation (3). Data produced from the bilinear QSAR in equation (4) yields a
similar plot since the Log(βP+1) terms are negligible when LogP < LogPo.
1
5

Organic & Biomolecular Chemistry
Page 16 of 39
View Article Online
DOI: 10.1039/C6OB00162A
4.00
3.50
3.00
5
2
3
4
9
6
0
8
7
0
1
A
T
g
o
L
d
e
t
c
i
d
e
r
P
2.50
2
.00
.50
1
1.5
2
2.5
3
3.5
4
Experimental LogTA100
Fig. 3 Predicted vs experimental mutagenicities for N-acyloxy-N-alkoxyamides 2-9 (solid squares)
and training set (open squares) in S. typhimurium TA100 when predicted by the linear QSAR in
equation (3).
The correlations with LogP and I in equation (3) and equation (4) are highly significant (P-values of
-
13
-17
1
0 and 10 respectively; see ESI), which makes it possible to quantify the changes in activity due
to intercalation by naphthalene. From a comparison of the respective coefficients of dependence
upon LogP and I in equation (3) and equation (4), the presence of one naphthalene group generates
increased activity equivalent to between 3 and 4 LogP. Though LogP of 2-9 fall well below the
LogP for N-acyloxy-N-alkoxyamides they exhibit mutagenic activity equivalent to mutagens with
O
an ‘effective’ LogP somewhat above LogPO.
1
6

Page 17 of 39
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C6OB00162A
4
.5
3
3
2
1
0
0
1
A
T
g
o
L
.5
2
Effective LogP
Actual LogP
.5
0
2
4
6
8
10
12
LogP
Fig. 4 Plot of the linear (equation (5), broken line) and bilinear (equation (6), solid line) dependence
of LogTA100 on LogP and the experimental activities of naphthalene-containing compounds 2-9 at
calculated (black triangles) and effective LogP range (open triangles).
This is depicted in Fig. 4. LogP is the most important element of the linear and bilinear QSARs in
equation (1) and equation (2) and satisfactory linear and bilinear dependencies upon LogP alone
have previously been derived for the N-acyloxy-N-alkoxyamides in our training set (equation (5)
22
and equation (6) respectively). From equation (6), a LogP of 6.38 has been found. Fig. 4 shows
0
the relationship between these two functions of LogP over the range LogP = 1-10, together with the
experimental activities for naphthalene-bearing mutagens 2-9 (solid triangles).
LogTA100 = 0.19 (±0.02) LogP + 2.08 (±0.08)
2
2
2
n = 50, r = 0.67, adj. r = 0.67, s = 0.19, F = 98.5; LOOCV Q = 0.6381
(5)
(6)
LogTA100 = 0.21 (±0.02) LogP – 0.47 (±0.10) Log(βP + 1) + 2.02 (±0.08)
2
2
2
Logβ = -6.466, n = 55, r = 0.68, adj. r = 0.66, s = 0.19, F = 36.0; LOOCV Q = 0.64
It is clear from Fig. 4 that the mutagenic activities of N-acyloxy-N-alkoxyamides-containing
naphthalene groups are significantly higher than predicted by the bilinear function of LogP. When
1
7

Organic & Biomolecular Chemistry
Page 18 of 39
mapped onto the extrapolated linear relationship, the naphthalene substrates have mutagenicities
View Article Online
DOI: 10.1039/C6OB00162A
that would be generated if LogP were in the range of 6.5-8.5, an increase on their calculated
hydrophobicity of about 4 LogP units. At these LogP, activities would normally be sharply reduced
22
due to lipid entrapment. However, in the case of 2-9, incorporation of naphthalene into the
structure of N-acyloxy-N-alkoxyamides results in LogP values which are low enough to avoid the
negative effects associated with high LogP, such as membrane localisation and reduced active
concentrations, while generating activity that would be associated with much greater LogP values.
Importantly, in this system, a mutagen bearing a naphthalene with a maximal LogP ≈ LogP (6.4)
O
could display activity equivalent to a molecule with a virtual LogP of around 9-10. In drug terms,
5
9
activity well above the Lipinski limit for N-acyloxy-N-alkoxyamides becomes possible. When
actual LogP > LogP , lipid entrapment would be expected to negate the activity enhancing effect of
0
22
naphthalene substitution. In support of this, we recently showed that the mutagenic activity of a
highly hydrophobic analogue of naphthamide 2, bearing N-octyloxyl and N-octanoyloxyl groups
and with a computed LogP = 7.9 is well predicted by bilinear QSAR in Equation 2 (experimental
and predicted LogTA100 were 3.0 and 3.1, respectively).
1
8

Page 19 of 39
Organic & Biomolecular Chemistry
DNA damage studies
View Article Online
DOI: 10.1039/C6OB00162A
The cellular target of N-acyloxy-N-alkoxyamides is undoubtedly DNA to which they bind and with
which they ultimately react electrophilically. In DNA damage studies on plasmid DNA at pH 7.8,
naphthamide 2 has already been shown to react more prolifically than several benzamides with
18
higher hydrophobicity, reacting at N7-G in the major groove and N3-A in the minor groove. In
32
these studies a 375 base pair fragment of plasmid pBR322 DNA, which is 3' end–labelled with P,
is treated with mutagens. The DNA is susceptible to cleavage by hot piperidine treatment at
damaged bases and labelled fragments are detectable along with sequencing lanes in
autoradiograms following electrophoresis. The damage is compared with reference profiles for
unreacted DNA. Using the same DNA damage studies, a direct comparison between 2 and 14 has
been made, which demonstrates their very different damage profiles. In addition, using a range of
naphthalene-bearing mutagens 3, 6-7, these studies confirm reactivity that is comparable to, or
higher than that of 2.
The DNA damage profiles for naphthamide 2 and its benzamide analogue, 14 at three pH's are
compared in Fig. 5. While the purine damage for benzamide 14 and naphthamide 2 was mainly
guanine damage at neutral and basic pH and typical of other groove-binding N-acyloxy-N-
18
alkoxyamides, benzamide 14 exhibited additional, more intense adenine damage at low pH,
noticeably at A33, A46, A47, A56, A61, A92 and A105 (Fig. 5, lane 7). However, the DNA
damage profile for the naphthamide 2 at low pH (Fig. 5, lane 4) differs very significantly from that
of 14. Damage at purines immediately preceded by a 3'- cytosine and in one case, a 3'-thymine, are
strongly damaged. This is evident at C39/G40, C42/A43, T52/G53, C69/G70, C77/G78, C82/A83,
C101/G102, C144/G145, C150/G151 and C161/G162 (Fig. 5, red labels). The corresponding bases
3
'
5'
were at most weakly affected by 14 at this pH. The preponderance of C- G sequence selectivity
3
'
5'
would suggest a preference for naphthalene intercalation between CG|| GC base pairs leading to
5'
reaction at the N7- G in the major groove. The strong A43 and A83 bands would indicate a second
3'
5'
5'
preference for intercalation between CG|| AT leading to N3- A damage in the minor groove.
1
9

Organic & Biomolecular Chemistry
Page 20 of 39
View Article Online
DOI: 10.1039/C6OB00162A
G162
C161
G151
C150
G145
C144
A139
A132
A117
G116
G115
A114
G112
G109
G106
A105
G104
G102
C101
G97
G93
A92
A83
C82
G81
G80
G78
C77
G74
A73
G70
C69
G68
A65
G64
A62
A61
A56
G53
T52
G51
A50
A47
A46
A43
C42
G40
C39
A38
A34
A33
G32
Lane = 1 2 3 4 5 6 7 8 9
Substrate = – – – 2 2 2 141414
6.6 7.4 8.6
6.6 7.4 8.6
pH
6.6 7.4 8.6
Fig. 5 Strand cleavage patterns obtained by the chemical treatment of the 3´-end-labelled EcoR1 to
BamH1 fragment of pBR322 DNA with 2 (lanes 4-6) and 14 (lanes 7-9). All lanes incubated for 16
hours. Lane 1-3 are the DMSO control. Lanes 1, 4, 7= pH 6.6, lanes 2, 5, 8 = pH 7.4 and lanes 3, 6,
1
8
9
= pH 8.6. Base labels from previous sequencing of the DNA fragment.
2
0

Page 21 of 39
Organic & Biomolecular Chemistry
Fig. 6 shows the strand cleavage pattern for 2, 3 and 7 with naphthalene on different side chains and
View Article Online
DOI: 10.1039/C6OB00162A
at three different pH's. Damage was more significant at pH 6.6 in each case. While at pH 7.4 and
8
.6, strand cleavage occurs due to damage at N7-G, at the lower pH, characteristic damage at both
N7-G and N3-A is evident (G40, A43, G53, G70, G78, A83 and G102). From lanes 4, 7 and 10 it
is clear that cleavage patterns for all three are similar and that 3 and 7 are more reactive than 2.
From depletion of the uncut fragment and larger fragments at the top of lane 7 in the autoradiogram,
as well as the high intensity of smaller fragments towards the bottom of this lane, the degree of
damage is noticeably greater for 3 where multiple hits are evident. In turn, damage detected for 7 is
more intense than that for 2.
2
1

Organic & Biomolecular Chemistry
Page 22 of 39
View Article Online
DOI: 10.1039/C6OB00162A
A132
G131
A117
G116
G112
G109
G106
G104
G102
A83
C82
G81
G80
G78
G74
G70
G68
A65
G64
G53
G51
A47
A46
A43
G40
C39
A34
A33
Lane = 1 2 3 4 5 6 7 8 9 10 1112
Substrate = – – – 2 2 2 3 3 3
7 7 7
6.6 7.4 8.6
6
.6 7.4 8.6
pH
6.6 7.4 8.6
6.6 7.4 8.6
Fig. 6 Strand cleavage patterns obtained by the chemical treatment of the 3´-end-labelled EcoR1 to
BamH1 fragment of pBR322 DNA with 2 (lanes 4-6), 3 (lanes 7-9) and 7 (lanes 10-12). All lanes
incubated for 16 hours. Lane 1-3 are the DMSO control. Lanes 1, 4, 7, 10 = pH 6.6, lanes 2, 5, 8, 11
=
pH 7.4 and lanes 3, 6, 9, 12 = pH 8.6. Base labels from previous sequencing of the DNA
1
8
fragment.
2
2

Page 23 of 39
Organic & Biomolecular Chemistry
Fig. 7 shows a comparison of the DNA damage of mutagens 6, 8 and 9 with that of 7. Like 7, all
View Article Online
DOI: 10.1039/C6OB00162A
three are more reactive at lower pH. At higher pH, damage is again at N7-G and from comparison
of lanes 9 and 10 with lanes 7 and 8 at pH 7.4 and lanes 14 and 15 with lanes 12 and 13 at pH 8.6
the more hydrophobic mutagens 8 and 9 bearing tethered naphthalene appear to be more reactive.
At lower pH, 7 would appear to be more reactive than 6, 8 and 9. The damage pattern is by and
large independent of position of substitution at naphthalene or of the nature of the tether. In this
experiment, in addition to the weaker but characteristic damage at low pH, all four exhibit much
strong fragmentation at A34/33, A65, and A94, which may be the result of multiple hits as
evidenced by the depletion of DNA at the origin and preponderance of smaller fragments. This
consequential damage has been noted in a previous study of N-benzoyloxy-N-benzyloxybenzamide
1
3
2
18
(
1, R =R =Ph, R =Bn). In addition to this, the nature and variability of DNA damage in these
6
0
studies can be attributed to revertible N3-A damage, which is time and pH dependent, so direct
comparison of band intensities in autoradiograms from different studies is unwise. However, while
Fig. 6 and Fig. 7 are not directly comparable, consideration of the comparisons in each study
indicates that the most reactive mutagen at low pH is 3 followed by 7, and both are more reactive
than 2, 6, 8 and 9. From experimental data in Table 5 this ordering broadly correlates with their
mutagenicities in point mutation studies with S. typhimurium TA100 where mutagenic activities of
3
and 7 are marginally greater than that of 6, 8 and 9.
2
3

Organic & Biomolecular Chemistry
Page 24 of 39
View Article Online
DOI: 10.1039/C6OB00162A
G151
A152
G138
G131
A139
A132
G130
G123
G121
G118
G116
G115
G112
G109
G106
G104
A117
G102
G97
G93
A94
A83
G82
G81
G78
G74
G70
G68
T66
A65
G64
A57
A56
G53
G51
A43
G40
G32
A34
A33
Lane = 1 2
3
4 5 6
7
8
9 10 1112 13 1415
Substrate
pH
–
6
7
8 9 –
6
7
8 9 – 6 7 8 9
6.6
7.4
8.6
Fig. 7 Strand cleavage patterns obtained by the chemical treatment of the 3´-end-labelled EcoR1 to
BamH1 fragment of pBR322 DNA with 6 (lanes 2, 7, 11), 7 (lanes 3, 8, 13), 8 (lanes 4, 9, 14) and 9
(
lanes 5, 10, 15). All lanes incubated for 16 hours. Lane 1, 6, 11 are the DMSO control. Lanes 1-5 =
pH 6.6, lanes 2-10 = pH 7.4 and lanes 11-15 = pH 8.6. Base labels from previous sequencing of the
18
DNA fragment.
2
4

Page 25 of 39
Organic & Biomolecular Chemistry
Conclusion
View Article Online
DOI: 10.1039/C6OB00162A
Compelling evidence has been presented showing that mutagenic activity of N-acyloxy-N-
alkoxyamides in point mutation strain S. typhimurium TA100 is strongly and uniformly enhanced
when a naphthalene group is present on any of the three side chains. We attribute this to a hitherto
unrecognised ability of naphthalene to increase residence time in the major and minor grooves of
DNA through intercalation. This is supported by activity in the frame-shift strain S. typhimurium
TA98. DNA damage studies on a range of structures bearing naphthalene demonstrate comparable
enhanced, sequence specific damage at guanine and adenine upon naphthyl substitution.
Comparison of damage profiles of N-acetoxy-N-butoxy-2-naphthamide and its benzamide analogue
confirms a different DNA binding mode, which is indicative of preferred intercalation between
3'
5'
3'
5'
CG—G C or CG—A T base pairs.
Based upon our extensive data for activity in TA100, we have established a working QSAR that
incorporates an indicator variable, I, for the presence (I=1) or absence (I=0) of naphthalene
intercalator. We will presently demonstrate that the fitting of mutagenic activity of N-acyloxy-N-
alkoxyamides towards TA100 to this new QSAR is a reasonably accurate test of whether other,
substituted naphthyl substituents facilitate intercalation or whether other PAH substructures can
intercalate. For instance, several mutagens with a biphenyl substituent, and the related fluorenyl as
a substituent, fit our QSAR with I=0 from which we conclude that both biphenyl and fluorenyl do
not intercalate. On the other hand, mutagens with fluorenone and pyrene fit with I=1 indicating that
these substructures, like naphthalene do intercalate. Since we have always evaluated the activity of
new mutagens in TA100 in conjunction with standard 14, with appropriate synthesis of substituted
N-acyloxy-N-alkoxyamides and 14, the methods described above could be executed in other
laboratories to discriminate between intercalating and non-intercalating substructures.
The results presented herein point very strongly to the enhanced association of molecules to
prokaryotic DNA when a naphthalene unit is incorporated as a substituent. In these studies, the
value to binding with DNA is worth between 3-4 LogP units such that a mutagen bearing
naphthalene with LogP close to optimum LogP can have DNA binding that would be generated by
0
LogP well in excess of LogP without lipid entrapment. We suggest that the enhanced binding to
0
DNA due to naphthalene intercalation may be put to similar advantage in the design of drugs, which
like N-acyloxy-N-alkoxyamides, target DNA and may as a consequence result in activity beyond
their limiting LogP or Lipinski limit.
0
2
5

Organic & Biomolecular Chemistry
Page 26 of 39
Acknowledgements
View Article Online
DOI: 10.1039/C6OB00162A
The authors are indebted to Dr Tony Bonin for carrying out Ames mutagenicity testing in TA100
and for valuable assistance in establishing the assay in our laboratories at the University of New
England. Earlier aspects of this work were carried out with financial assistance from the Australian
Research Council.
2
6

Page 27 of 39
Organic & Biomolecular Chemistry
Experimental
View Article Online
DOI: 10.1039/C6OB00162A
Chemicals
Infrared spectra were recorded on a Perkin–Elmer1600 series Fourier-transform (FT)-IR
spectrophotometer solutions in chloroform. Mass spectra were recorded on a Varian 1200L liquid
chromatograph-mass spectrometer at a 30-V capillary voltage and 350 °C, 138 kPa drying gas
temperature and pressure, in HiPerSolv acetonitrile/water. TOF ESI accurate mass determinations
were obtained from the Mass Spectral Unit of the Australian National University. Nuclear magnetic
resonance spectra were recorded in CDCl on a Bruker Avance 300P FT NMR spectrometer with a
3
1
5
-mm 1H inverse/BroadBand probe with a z-gradient, operating at 300.13 MHz ( H), 75.46 MHz
13
15, 17
(
C). Syntheses of mutagens 2 and 14 have been described elsewhere
Mutagens 3-9 and 15
were all synthesised as part of this study. For DNA damage studies 40% (19:1)
Acrylamide/bisacrylamide was purchased from Bio-Rad Laboratories Pty Ltd. pBr322, Klenow
fragment, BamH1, EcoR1, BSA (bovine serum albumin) and NE buffer 2, were purchased from
3
2
(
New England Biolabs) Genesearch Pty Ltd. α-dATP([α- P]-deoxyadenosine-5'-triphosphate
10µCi/µl), isoblue stabilized) was purchased from ICN Biomedicals Inc.
(
General Synthesis of N-alkoxyamides
Condensation of the potassium salt of acetohydroxamic acid with an appropriate alkyl bromide,
prepared from the parent alcohol (see ESI), and a 10% excess of sodium carbonate in 50% aqueous
2
9
methanol gave the appropriate N-alkoxyamide in good yield.
61
Potassium acetohydroxamate. Potassium hydroxide (56.10 g, 1.00 mol) in boiling methanol
(
140 ml) was added to hydroxylamine hydrochloride (46.70 g, 0.67 mol) in boiling methanol (240
ml) and the mixture cooled in an ice bath for 5 minutes. Ethyl acetate (29.08g, 0.33 mol) was added
with shaking the mixture filtered immediately and reduced to dryness in vacuo to give potassium
acetohydroxamate as a straw-coloured viscous oil (31.00 g, 83%) which was used without further
purification.
N-Butoxyacetamide. Potassium acetohydroxamate (10.00 g, 0.0884 mol), butyl bromide (12.11 g,
0
.0884 mol) and sodium carbonate (10.3 g, 0.01 mol) were stirred overnight in 50% aq. MeOH (160
ml) and refluxed for 3 hours. Removal of MeOH in vacuo followed by extraction with chloroform
provided the crude hydroxamate. N-Butoxyacetamide was obtained as a straw coloured oil after
purification by repeated fractional distillations under reduced pressure (4.9 g, 42%). b.p. 70°C @
2
7

Organic & Biomolecular Chemistry
Page 28 of 39
-
1
0
.2mm Hg; νmax.(neat)/cm 3190br (NH), 1678s (C=O), 1370s, 1086s; δ (300 MHz; CDCl ) 1.05
H
3
View Article Online
DOI: 10.1039/C6OB00162A
(
3H, t, CH ), 1.50 (2H, m, CH CH ), 1.72 (2H, m, OCH CH ), 2.11 (3H, br d, COCH ), 4.01 (2H, br
3 2 3 2 2 3
s, OCH ), 8.40 (1H, br s, NH); δ ( 75 MHz; CDCl ) 13.7 (q), 18.9 (t), 19.7 (br), 29.9 (t), 76.3 (t),
2
C
3
+
+
+
1
67.9 (s); m/z (ESI) 170 ([M+K ], 50%), 154 ([M+Na ], 100%); m/z (EI) 131.094093 ([M ],
C H NO requires 131.094629).
6
13
2
N-(2-Naphthylmethoxy)acetamide. 2-Bromomethylnaphthalene (10.00 g, 45.2 mmol), potassium
acetohydroxamate (7.68 g, 67.8 mmol) and sodium carbonate (7.2 g, 68 mmol) were stirred
overnight in 50% aq. MeOH (160 ml) and refluxed for 4 hours. Removal of MeOH in vacuo
followed by extraction with DCM provided impure N-(2-naphthylmethoxy)acetamide. Purification
was achieved by centrifugal chromatography with ethyl acetate/pet spirit. N -(2-
Naphthylmethoxy)acetamide was obtained as a white solid (3.7 g, 38%), mp 113-115°C (from
benzene); (Found: C, 72.01%; H, 6.17%; N, 6.34%. C H NO requires C, 72.54%; H, 6.09%; N,
1
3
13
2
–
1
6
1
.51%); ν (CHCl )/cm 3401w (NH), 3235br (NH), 1691s (C=O); δ (300 MHz; CDCl , 50°C)
max 3 H 3
.92 3H, br, CH ), 5.03 (2H, br, CH ), 7.50-7.53 (3H, m, Ar-H), 7.83-7.88 (4H, m, Ar-H), 8.32 (1H,
3
2
br, NH); δ (75 MHz; CDCl , 50°C) 19.6 (q br), 78.7 (br), 126.4 (d br), 125.5 (d), 127.7 (d), 128.0
C
3
1
3
(
d), 128.5 (br), 132.6 (s very br), 133.2 (s), 133.5 (s), 168.2 (very br). C incomplete due to extreme
+
line broadening even at 50°C; m/z (ESI) 238 ([M+Na ], 100%), 141 (15%).
N-(1-Naphthylmethoxy)acetamide. 1-Bromomethylnaphthalene (10.00 g, 45.2 mmol), potassium
acetohydroxamate (7.68 g, 67.8 mmol) and sodium carbonate (7.2 g, 68 mmol) were stirred
overnight in 50% aq. MeOH (160 ml) and refluxed for 4 hours. Removal of MeOH in vacuo
followed by extraction with DCM provided impure N-(1-naphthylmethoxy)acetamide. Purification
was achieved by centrifugal chromatography with ethyl acetate/pet spirit. N -(1-
Naphthylmethoxy)acetamide was obtained as a white solid (2.8 g, 29%), mp 119-121°C (from
–1
benzene); ν (CHCl )/cm 3401w (NH), 1695s (C=O); δ (300 MHz; CDCl ) 1.88 (3H, br s, CH ),
max
3
H
3
3
5
.39 (2H, s, CH ), 7.38-7.54 (4H, m, Ar-H), 7.92 (2H, m, Ar-H), 8.31 (1H, br s, Ar-H); δ (75 MHz;
2 C
CDCl , 50°C) 19.7 (v br), 77.2 (v br), 124.2 (v br), 125.1 (d), 126.2 (d), 126.8 (d), 128.6 (br), 128.8
3
1
3
(
br), 130.0 (br s), 133.8 (s), 167.6 (v br). C incomplete due to extreme line broadening even at
+
+
+
5
0°C; m/z FAB 216 [M+H ], 238 ([M+Na ],100%), 141 (13%), (Found: 215.09435 ([M ],
C H NO requires 215.09462).
1
3
13
2
N-(2-(2'-Naphthyl)ethoxy)acetamide. 2-(2'-Naphthyl)ethyl bromide (2.20 g, 9.3 mmol), potassium
acetohydroxamate (1.06 g, 9.4 mmol) and sodium carbonate (1.2 g, 11 mmol) were stirred
overnight in 50% aq. MeOH (160 ml) and refluxed for 4 hours. Removal of MeOH in vacuo
2
8

Page 29 of 39
Organic & Biomolecular Chemistry
followed by extraction with DCM provided impure N-(2-(2'-naphthyl)ethoxy)acetamide.
View Article Online
DOI: 10.1039/C6OB00162A
Purification was achieved by centrifugal chromatography with ethyl acetate/pet spirit. N-(2-(2'-
Naphthyl)ethoxy)acetamide was obtained as a white solid. (1.09 g, 51%), mp 146-149°C (from
–l
benzene); ν (CHCl )/cm 3401 (NH), 1695s (C=O), 1368s; δ (300 MHz; CDCl ) 1.90 (3H, br s,
max
3
H
3
CH ), 3.14 (2H br t, ArCH ), 4.21 (2H br s, OCH ), 7.36 (1H, d), 7.47 (2H, m), 7.69 (1H, s), 7.80
3
2
2
(
(
3H, m), 8.34 (1H, br s, NH); δ (75 MHz; CDCl ) 19.6 (br q), 34.7 (t), 76.8 (br t), 125.5 (d), 126.1
C 3
d), 127.3 (d), 127.3 (d), 127.5 (d), 127.6 (d), 128.2 (d), 132.3 (s), 133.6 (s), 135.4 (br s), 168.1 (br
+
+
+
s); m/z (ESI) 268 ([M+K ], 30%), 252 ([M+Na ], 100%), 155 (10%), m/z (EI) 229.110193 ([M ],
C H NO requires 229.110279).
1
4
15
2
N-(3-(2'-Naphthyl)-1-propoxy)acetamide. A solution of 3-(2'-naphthyl)-1-propyl bromide (1.7 g,
.7 mmol) potassium acetohydroxamate (0.7 g, 6.7 mmol) and sodium carbonate (0.85 g, 8 mmol)
6
were stirred overnight in 50% aq. MeOH (40 ml) and refluxed for 4 hours. Removal of MeOH in
vacuo followed by extraction with DCM provided impure N-(3-(2'-naphthyl)-1-
propyloxy)acetamide. Purification was achieved by centrifugal chromatography with ethyl
acetate/pet spirit. N-(3-(2'-Naphthyl)-1-propyloxy)acetamide as a light brown solid (1.06 g, 68 %),
mp 90-91°C (from benzene); (Found: C, 73.63%; H, 7.01%; N, 5.61%. C H N O requires C,
2
0
24
2
4
-l
7
4.05%; H, 7.04%; N, 5.76%); ν (CHCl )/cm 3401 (NH), 1694s (C=O), 1369; δ (300 MHz,
max 3 H
CDCl ) 1.95 (3H, br d, CH ), 2.08 (2H, qu, CH CH CH ) 2.90 (2H br t, ArCH ), 3.95 (2H br s,
3
3
2
2
2
2
OCH ), 7.34 (1H, d), 7.46 (2H, m), 7.64 (1H, s), 7.80 (3H, m), 8.39 (1H, br s, NH); δ (75 MHz
2
C
CDCl ) 19.6 (br q), 29.4 (t), 32.1 (t), 75.8 (br t), 125.3 (d), 126.0 (d), 126.5 (d), 127.2 (d), 127.4 (d),
3
+
1
2
27.6 (d), 128.1 (d), 132.1 (s), 133.6 (s), 138.8 (br s), 168.0 (br s); m/z (ESI) 266 ([M+Na ], 100%),
+
44 ([M+H ], 10).
General procedure for chlorination of N-alkoxyamides
N-Chlorination of N-alkoxyamides was achieved in quantitative yields by stirring with a 3-fold
0
excess of neat tert-butyl hypochlorite at room temperature or 0 . Rarely did the reaction require
longer than 10-15 minutes for completion. Removal of excess tert-butylhypochlorite and tert-
butanol under reduced pressure afforded N-alkoxy-N-chloroamides generally as clean yellow oils,
which required no further purification and were characterised spectroscopically. N-Chlorination was
-1
evident from characteristically high carbonyl stretch frequencies in the range 1720-1745 cm ,
resonance of the methylenoxyl protons slightly downfield of the corresponding protons in the parent
N-alkoxyamide and carbonyl resonances in the region of 175 ppm, 7 ppm downfield of the parent
2
9