Soft antimicrobial agents: Synthesis and activity of labile environmentally friendly long chain quaternary ammonium compounds

Article abstract of DOI:10.1021/jm030829z

A series of soft quaternary ammonium antimicrobial agents, which are analogues to currently used quaternary ammonium preservatives such as cetyl pyridinium chloride and benzalkonium chloride, were synthesized. These soft analogues consist of long alkyl chain connected to a polar headgroup via chemically labile spacer group. They are characterized by facile nonenzymatic and enzymatic degradation to form their original nontoxic building blocks. However, their chemical stability has to be adequate in order for them to have antimicrobial effects. Stability studies and antibacterial and antiviral activity measurements revealed relationship between activity, lipophilicity, and stability. Their minimum inhibitory concentration (MIC) was as low as 1 μg/mL, and their viral reduction was in some cases greater than 6.7 log. The structure-activity studies demonstrate that the bioactive compounds (i.e., MIC for Gram-positive bacteria of < 10μg/mL) have an alkyl chain length between 12 and 18 carbon atoms, with a polar headgroup preferably of a small quaternary ammonium group, and their acquired inactivation half-life must be greater than 3 h at 60 °C.

Full text of DOI:10.1021/jm030829z

J . Med. Chem. 2003, 46, 4173-4181
4173
Soft An tim icr obia l Agen ts: Syn th esis a n d Activity of La bile En vir on m en ta lly
F r ien d ly Lon g Ch a in Qu a ter n a r y Am m on iu m Com p ou n d s
Thorsteinn Thorsteinsson,†,‡ M a´ r M a´ sson, Karl G. Kristinsson, Martha A. Hj a´ lmarsd o´ ttir,
†
§
§
|
,†
Hilmar Hilmarsson, and Thorsteinn Loftsson*
Faculty of Pharmacy, University of Iceland, Hagi, Hofsvallagata 53, IS-107 Reykjavik, Iceland, Decode Genetics Inc.,
Reykjavik, Iceland, Department of Clinical Microbiology, Landspitali University Hospital, IS-101 Reykjavik, Iceland, and
Institute of Biology, University of Iceland, Hagi, Hofsvallagata 53, IS-107 Reykjavik, Iceland
Received March 26, 2003
A series of soft quaternary ammonium antimicrobial agents, which are analogues to currently
used quaternary ammonium preservatives such as cetyl pyridinium chloride and benzalkonium
chloride, were synthesized. These soft analogues consist of long alkyl chain connected to a
polar headgroup via chemically labile spacer group. They are characterized by facile nonen-
zymatic and enzymatic degradation to form their original nontoxic building blocks. However,
their chemical stability has to be adequate in order for them to have antimicrobial effects.
Stability studies and antibacterial and antiviral activity measurements revealed relationship
between activity, lipophilicity, and stability. Their minimum inhibitory concentration (MIC)
was as low as 1 µg/mL, and their viral reduction was in some cases greater than 6.7 log . The
structure-activity studies demonstrate that the bioactive compounds (i.e., MIC for Gram-
positive bacteria of <10 µg/mL) have an alkyl chain length between 12 and 18 carbon atoms,
with a polar headgroup preferably of a small quaternary ammonium group, and their acquired
inactivation half-life must be greater than 3 h at 60 °C.
In tr od u ction
This report discloses the synthesis and evaluation of
new soft analogues of cetylpyridinium chloride and
benzalkonium chloride possessing greater efficiency,
The shortage of new antibacterial drugs and increas-
ing resistance of bacteria to antimicrobial agents has
1
8
enhanced bacterial activity, and chemical stability. The
been of some concern. Long chain quaternary am-
monium compounds exert antibacterial activity against
both Gram-positive and Gram-negative bacteria, as well
as against some pathogenic species of fungi and proto-
selection of building blocks was partly based on com-
pounds found in marine lipids, with polar quaternary
ammonium as a headgroup and a chemically labile
spacer. It was expected that this kind of chemically
labile compounds would be more environmental friendly
2
zoa. These quaternary ammonium compounds, in gen-
eral, have toxic effects toward mammalian cells. In
humans and animals they are considered too toxic for
systemic applications, but acceptable for topical applica-
tions. The sustained toxicity and environmental impact
of these antibacterial compounds are related to their
chemical stability. To overcome these limitations, a
series of chemically labile derivatives of long chain
quaternary ammonium compounds, so-called soft ana-
logues, have been synthesized and tested both in vitro
9
than currently used hard compounds. In short, the
main object of this study was to synthesize antimicrobial
compounds that are relatively labile (“soft”) and envi-
ronmentally friendly and hopefully less likely to induce
bacterial resistance toward these agents.
Ch em istr y
and in vivo.3,4 Soft compounds can be defined as biologi-
Quaternary ammonium compounds with various forms
of quaternary ammonium headgroups, spacers, and
lipophilic alkyl chains were synthesized. The majority
of these compounds are made from natural building
blocks such as fatty acids, alcohols, or amides. Seven of
these compounds were commercially available (used as
a intermediates or for reference) and few have previ-
cally active compounds that are readily degraded into
nontoxic and biologically inactive products in vivo, as
well as in the environment. The concept of soft anti-
bacterial agents was proposed by Bodor over 20 years
5
ago, whereby the environmental impact and toxicity
of cetylpyridinium chloride was reduced by a soft drug
approach. Although these analogues did not possess
sufficient chemical stability to cause acceptable anti-
bacterial effect, the concept was shown to be effective.
Similar compounds such as labile L-carnitine esters
have also been shown to possess antibacterial activity.
5
,10
ously been described.
The headgroup was either
pyridinium or small trialkyl moiety (trimethyl, triethyl,
or tributyl), the spacer was either ester or amide group,
consisting of one to three carbons, and the lipophilic
alkyl chain consisted in most cases of a C12 to C18
chain, which has been shown to afford the best bioac-
6
7
2
,6
tivity. The spacer group was designed to be labile in
order for the compound to be converted to original
nontoxic building blocks, both chemically and enzymati-
cally. The relationship between the nature of the labile
spacer group and the antibacterial activity was inves-
*
To whom correspondence should be addressed. Tel: +354-525-
4
464. Fax: +354-525-4071, E-mail: thorstlo@hi.is.
†
Faculty of Pharmacy, University of Iceland.
Decode Genetics Inc.
Landspitali University Hospital.
Institute of Biology, University of Iceland.
‡
§
|
1
0.1021/jm030829z CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/05/2003

4
174 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19
Thorsteinsson et al.
Ta ble 1. Compounds Synthesized and Evaluated in This Study
Sch em e 1. Example of the Synthetic Route and Route
of Deactivation for the Compounds
Resu lts a n d Discu ssion
The antibacterial properties of the compounds were
determined with strains of Enterococcus faecalis, Sta-
phylococcus aureus, Escherichia coli, and Pseudomonas
aeruginosa. The bacterial strains represent important
Gram-positive and Gram-negative species. Their anti-
bacterial activity was assessed by measuring minimum
inhibitory concentration (MIC) and minimum lethal
concentration (MLC) (Table 2). Some of the soft anti-
bacterial agents possessed antibacterial activity com-
parable to that of their parent hard compounds, i.e.,
cetylpyridinium chloride (3a ) and benzalkonium chlo-
ride (5). A mixture of several different antibacterial
agents was also tested, denoted as 6a -f in Table 2, and
their composition is given in Table 3. However, the
antibacterial activity of the mixtures appeared to be
comparable to that of the individual compounds.
tigated. The general synthetic and degradation route
of the compounds is shown in Scheme 1. The NMR and
the elemental analysis of several compounds (2b, 2c,
3
o, 3p , 3v, 3w , and 4b) showed the presence of free fatty
acids impurities (<5%), which lacked antibacterial
activity. Some analogues were synthesized from mix-
tures of fatty acids obtained from cod liver oil. Since cod
liver fatty acids are high in unsaturated fatty acids, one
aspect of this study was to investigate the effect of
degree of unsaturation on the antibacterial effect.
Quaternary ammonium compounds 1a , 2a , 3d , 3n ,
3
q, 3x, and 4a , intermediates with no alkyl chain,
lacked antibacterial activity. Compounds 3b, 3r , and 3z
were also inactive. We tested all of the free fatty acids,
the possible degradation products, and solvents used in

Soft Antimicrobial Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19 4175
Ta ble 2. Antibacterial Activity for the Compounds in µg/mL
Enterococcus faecalis
Staphylococcus aureus
ATCC 25923
Escherichia col
ATCC 25922
Pseudomonas aeruginosa
ATCC 29212
ATCC 27853
no.
MIC
MLC
MIC
MLC
MIC
MLC
MIC
MLC
1
1
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
5
6
6
6
6
6
6
a
b
a
b
c
d
e
a
b
c
d
e
f
g
h
i
k
l
m
n
o
p
q
r
>2000
16
>2000
64
>2000
16
>2000
16
>2000
64
>2000
64
>2000
250
>2000
250
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
32
8
32
16
16
2
16
4
16
64
125
250
250
64
250
250
not determined
15
<0.5
320
-
7.5
<0.5
3200
50
>2000
-
4
3200
50
250
16
3200
100
-
-
500
500
3200
-
1000
3200
a
8
3200
200
3200
400
10
3200
3200
>2000
>2000
>2000
>2000
>2000
>2000
>2000
2
100
<6.25
500
1
1
8
50
25
100
200
200
>2000
>2000
8
500
1
2
16
-
-
16
1000
32
500
62
60
1000
32
500
>2000
250
>2000
1000
8
8
4
7.5
7.5
16
8
62
-
-
250
250
250
250
250
125
2.0
15
8
>2000
4
-
-
-
-
250
500
>2000
g2000
>2000
250
64
500
g2000
>2000
250
>2000
>2000
>2000
>2000
>2000
4
1
16
4
4
1
125
125
<0.25
250
250
500
500
12800
>12800
>12800
>12800
>12800
4
<0.25
>1000
250
12800
>12800
>12800
>12800
>12800
8
6400
>12800
>12800
12800
>12800
4
12800
>12800
>12800
12800
>12800
4
12800
>12800
>12800
6400
>12800
125
1000
>1000
>1000
>12800
12800
>12800
>12800
12800
>12800
125
>2000
>1000
>1000
>12800
12800
>12800
>12800
6400
12800
>12800
>12800
6400
s
t
u
v
w
x
y
z
>12800
>12800
125
125
8
<0.25
>1000
250
4
1000
>2000
>1000
>1000
>12800
>1000
>1000
>1000
>1000
>12800
500
250
>12800
>12800
>12800
>12800
a
b
not determined
4
4
1
16
8
4
16
125
16
32
32
64
2
2
1
8
8
4
8
16
32
8
2
8
32
16
32
16
250
125
125
125
16
125
500
250
125
125
125
125
250
64
125
500
500
250
250
250
500
250
64
1000
1000
64
c
a
a
b
c
d
e
f
8
500
500
32
16
32
32
64
g2000
g
g2000
500
g2000
g2000
g2000
1000
1000
1000
a
3
a ) Cetylpyridinium chloride and 5 ) benzalkonium chloride.
Ta ble 3. Composition in Percentage for the Mixture
Compounds 6a -f
lack of chemical stability) have been tested previous-
10-12
ly.
The novel compounds killed the bacteria at
compound 3e
3g
3h
4b
4c
2b
2c
3w
3v
relatively fast rate (Figure 1), and the relatively small
difference in values for MIC and MLC confirms that
these compounds are bactericidal. Cod liver derivatives
6
6
6
6
6
6
a
b
c
d
e
f
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
43.5 56.5
58.1 41.9
-
-
23.4
39.6
-
-
-
-
-
-
-
59.9 40.1
-
-
-
-
(2e, 3l, and 3m ) were also highly active, but not as
24.6 50.8 24.6
26.1
60.4
-
23.5
-
-
-
-
-
27.0
-
active as the compounds with single defined alkyl
chains, probably due to the fact that these are mixtures
of very active long alkyl chain compounds and relatively
inactive short alkyl chain compounds. There was no
difference in antibacterial activity between compounds
containing saturated alkyl chains and corresponding
unsaturated chains.
-
-
-
-
the synthesis, and they had negligible antibacterial
activity (MIC > 250 µg/mL); therefore, the activity was
due to the antibacterial properties of the compounds and
not to that of impurities or degradation products. The
long chain compounds 1b, 2,b, 2e, 3i, 3l, and 3m were
able to inhibit bacterial growth and even kill some
bacteria at as low concentration as 10 µg/mL. However,
compounds 2c, 3e, 3g, 3h , 3p , 3k , 3o, 3v, 3w , 4b, and
Selected compounds were tested for anti-HSV-1 activ-
ity at concentrations up to 100 µg/mL (Table 4). Com-
pound 4b showed the greatest antiviral activity with
g6.7 log (about 5 million-fold) reduction in virus titer
compared to the control. At a 50 µg/mL concentration,
three compounds 2c, 3e, and 4b had lost most of their
activity, while the others still caused significant reduc-
tion in a virus titer. Compound 3a (the parent hard
compound) was the most active at low concentrations
and could be diluted to 12.5 µg/mL without losing its
4
c were inhibitory at concentrations <2 µg/mL, which
is comparable to the parent hard compounds cetylpyri-
dinium chloride (3a ) and benzalkonium chloride (5). The
betaine ester analogues 3s, 3t, and 3u , with the ester
linkage reversed vs the novel compounds herein (i.e.,
the alkyl chain is a fatty alcohol residue, essential in
comparison with the fatty acids derivatives due to their

4
176 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19
Thorsteinsson et al.
F igu r e 1. Effect on same type of bacteria against time (b) compared with control (O) for compounds 3g (A, C, E) and 3h (B, D,
F). Amount used was 4 times the MIC (4 µg/mL for Staph. aureus).
a
Ta ble 4. Antiviral Activity of Compounds against HSV-1 after 2 h Treatment at 37 °C. Virus Titer (Log10 TCID50/0.1 mL) after
Treatment with the Following Concentrations of Compounds (µg/mL)b
compound
100
50
25
12.5
6.25
NDc
6.43 ( 0.12
7.03 ( 0.29
6.90 ( 0.35
6.47 ( 0.25
ND
control
2
3
3
3
3
3
4
c
a
e
h
k
v
b
0.80 ( 0.17
5.73 ( 0.25
2.5T
6.73 ( 0.25
1.63 ( 0.58
0.90 ( 0.17
1.5T
6.33 ( 0.29
1.5T
7.00 ( 0.30
1.5T
7.00 ( 0.30
7.10 ( 0.30
7.30 ( 0.00
7.10 ( 0.35
6.57 ( 0.12
7.00 ( 0.30
7.20 ( 0.17
d
2.5T
1.63 ( 0.12
1.30 ( 0.36
0.80 ( 0.17
1.5T
6.73 ( 0.25
2.90 ( 1.22
2.90 ( 0.17
5.40 ( 0.17
6.63 ( 0.12
6.80 ( 0.36
6.10 ( 0.69
6.07 ( 0.40
6.57 ( 0.40
7.00 ( 0.30
e
0.5 ( NA
6.00 ( 0.30
7.00 ( 0.30
a
Mean ((standard deviation) of three measurements for each compound. b Final concentrations of compounds. c ND, not determined.
d
T, not possible to measure due to toxic effects on Vero cells. e NA, not available.
activity; however, it was also the most toxic toward Vero
cells. Compounds 2d and 3h showed high antiviral
activity at 25 µg/mL (reduced the virus titer about 3 to
Accelerated degradation studies were performed in
aqueous buffer solutions at pH 6.0 and 60 °C. The
observed pseudo-first-order constants (kobs) shown in
Table 5 indicate that both the structure of the spacer
group as well as the lipophilicity affect great variations
between the compounds depending on how labile spacer
4
log ), without showing any toxicity toward the Vero
cells, even at the highest concentration tested, i.e., 100
µg/mL.

Soft Antimicrobial Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19 4177
Ta ble 5. Observed Degradation Rate (kobs) in Aqueous Buffer
Solutions, at pH 6.0 at 60 °C, ( Standard Deviation, n ) 3.
Other Compounds Were Intermediates or Not Structurally
Similar and Not Active
antibacterial activity will only be adequate if its chemi-
cal stability is sufficient to allow the compound to
express its activity for sufficient duration of time.
Increasing the lipophilicity further, beyond what is
observed in the present data set, is not expected to
increase activity of the compounds. Other investigators
have shown that activity will decrease when the alkyl
compound
kobs (h-1)
compound
kobs (h-1
)
a
1
2
2
2
3
3
3
3
3
3
3
3
b
b
c
d
a
c
d
e
f
0.2052 ( 0.0026
0.2070 ( 0.0043
0.1805 ( 0.0121
3k
3p
3o
3s
3t
3u
3v
3w
3y
4b
4c
5
ND
ND
a
0.0107 ( 0.0018
0.4911 ( 0.0272
5.2817 ( 0.3801
8.4202 ( 1.7302
0.0933 ( 0.0142
0.1071 ( 0.0099
0.6111 ( 0.0202
0.3690 ( 0.0136
0.5299 ( 0.0447
1
5-18
a
chains are longer than C18.
In addition, a further
ND
a
increase in stability is not expected to increase antibac-
terial activity, as the most stable compounds in the
present data set are just as active as the hard com-
pounds, benzalkonium chloride (5) and cetylpyridinum
chloride (3a ).
ND
0.3333 ( 0.0157
0.0402 ( 0.0019
0.2034 ( 0.0111
0.6213 ( 0.0044
0.0546 ( 0.0057
0.0864 ( 0.0065
0.0473 ( 0.0037
g
h
i
a
ND
Con clu sion
a
No degradation was observed during 72 h.
Compounds were synthesized, possessing various
alkyl chain lengths, various polar quarter ammonium
headgroups, and different spacers. The bactericidal
activities of some of the compounds were very good, and
the most active compounds (MIC for Gram-positive
bacteria <10 µg/mL) have alkyl chain length between
12 and 18 carbon atoms, a polar headgroup (preferably
small quaternary ammonium group), and a half-life in
aqueous solutions of at least 3 h at 60 °C. These
compounds could replace currently used compounds
resulting in less residue impact on the environment,
with multipurpose usage against bacterial and viral
infections.
Ta ble 6. Observed and Predicted Antibacterial Activity, and
Data Used to Derive SAR
Log(1/MIC)
observed
predicted
∆
CLogP
Log t1/2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
b
c
e
f
g
h
i
o
p
s
-1.20
-1.51
-0.90
-2.69
0.00
-0.80
-2.90
-2.08
-2.38
-0.37
-0.29
-0.02
0.04
-0.41
1.40
1.17
-0.31
0.37
4.90
2.03
1.97
0.01
3.94
4.92
4.70
2.46
4.43
-1.13
3.78
4.76
4.27
3.27
4.86
6.37
8.34
0.52
0.31
0.33
0.62
1.10
0.90
1.17
1.81
2.36
0.15
-0.90
-1.10
0.62
0.92
0.81
0.27
0.12
0.00
0.29
-0.90
-0.60
0.60
-4.10
-4.10
-4.10
0.42
-0.60
0.60
-0.30
-0.30
-0.88
-0.64
-0.86
-0.70
-1.04
-1.12
1.31
0.24
1.03
0.31
-0.17
1.46
-3.40
-3.06
-2.98
-0.89
-0.85
-0.43
-0.61
-0.13
Exp er im en ta l Section
t
u
y
v
w
b
c
Melting points were determined on a Gallenkamp melting
point apparatus. 1H NMR and C NMR spectra were mea-
sured on a 250 MHz Bruker AC 250P NMR spectrometer with
13
tetramethylsilane (Me
or D
are reported in parts per million (δ) relative to Me
4
Si) as an internal reference and CDCl
O as a solvent. Both H NMR and C NMR spectral data
3
1
13
2
4
Si.
Elemental analysis was performed at the Microanalytical
Laboratorium Institut for Physical Chemistry at University
of Vienna, Austria. Compounds 1a , 1b, 3a , and 5 are com-
mercially available and not synthesized in this study, com-
pounds 2a , 3n , 3q, 3x, and 4a are commercial available
intermediates that were synthesized for this study. All chemi-
cals were of synthetic, analytical or HPLC grade. All chemicals
used are commercial available and were from Merck, Sigma-
Aldrich, ICN or Rathburn. The refined cod liver fatty acids
were donated from Lysi ehf, Reykjavik.
Tr ieth yl-(2-h yd r oxyeth yl)a m m on iu m Br om id e (2a ).
Equimolar amounts of triethylamine (2.8 g, 28 mmol) and
bromoethanol (3.5 g, 28 mmol) were mixed at 0 °C in a cooling
water bath and then heated slowly until the solution turned
to solid. The mixture was cooled to room temperature, anhy-
drous ether was added, and the solid was washed thoroughly
with ether. Recrystallization from ethyl acetate:methanol (50:
is connecting the alkyl chain with the trialkylammo-
nium group and also a small variation between the alkyl
chain lengths. Stability studies in both types of agar
revealed no differences between the two agars, with
regard to the observed degradation rate and relative
rate compared to the half-lives obtained at 60 °C and
pH 6.0. Half-life (t1/2), of the most unstable active
compound (3c), was 7 h in both types of agar and up to
3
6 h for compound 3v at 37 °C. The stability relation-
ship was done using the MIC values of Staphylococcus
aureus, and there appear to be a relationship between
activity and half-life of the compounds (R2 ) 0.50)
according to our calculations, but it is not highly
significant (p ) 0.111), and relationship between activity
and the ClogP values (R2 ) 0.25, p ) 0.133) is less
significant. However, when all those factors were com-
bined in structure-activity relationship (SAR), there is
a strong relationship between all these parameters. The
relationship follows the following equation:
5
0) afforded white crystals (5.4 g, yield 85%). Mp 210-212 °C.
1
H NMR (D
Hz), 3.22 (t, 2H, J ) 6 Hz), 1.28 (t, 9H, J ) 7.5 Hz). C NMR
(CDCl ) δ 59.9, 55.9, 49.1, 9.2. Anal. (C 20BrNO) C, H, N.
2
O) δ 3.98 (t, 2H, J ) 6 Hz), 3.36 (q, 6H, J ) 7.5
13
3
8
H
Gen er a l P r oced u r e for Com p ou n d s 2b ,c. Equimolar
amounts of corresponding fatty acid chloride (6 mmol) and
compound 2a were mixed, sonicated, and heated at reflux in
2 2
CH Cl overnight. The mixture was cooled to room tempera-
Log (1/MIC) ) 1.316 Log t1/2 + 0.350 CLog P -
ture and evaporated, anhydrous ether was added, and the solid
was isolated by filtration and washed thoroughly with ether.
The crude compound was recrystallized in a mixture of ethyl
acetate:EtOH (10:1), affording white crystals. Yields 89-93%.
3
.202 (stdev ) 0.91)
2
R ) 0.73, p ) 0.004
(
2-Dod eca n oyloxyet h yl)t r iet h yla m m on iu m Br om id e
1
(
3
2b). H NMR (CDCl ) δ 4.53 (t, 2H, J ) 7.0 Hz), 3.85 (t, 2H,
Comparison of measured values and calculated values
is shown in Table 6. The correlation shows that even
though a given compound has adequate lipophilicity, its
J ) 7.0 Hz), 3.58 (q, 6H, J ) 7.2 Hz), 2.30 (t, 2H, J ) 7.8 Hz),
1.56 (m, 2H), 1.40 (t, 9H, J ) 7.2 Hz), 1.22 (m, 16H), 0.84 (t,
3H, J ) 7.0 Hz). 13C NMR (CDCl
) δ 172.7, 57.1, 55.8, 54.3,
3

4
178 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19
Thorsteinsson et al.
4
1
5
6.0, 33.7, 31.8, 29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 24.5, 22.6,
4.0, 8.1. Mp 143-147 °C. Anal. (C20H42BrNO ) H, N; C: calcd,
2
8.81; found, 57.18.
Hz), 8.12 (q, 2H, J ) 7.5 Hz), 5.30 (t, 2H, J ) 5 Hz), 4.58 (t,
2H, J ) 5 Hz), 2.22 (m, 2H), 1.43 (m, 2H), 1.16 (bs, 8H), 0,80
(t, 3H, J ) 6.7 Hz). 13C NMR (CDCl
) δ 172.9, 145.5, 128.2,
62.6, 60.4, 33.6, 31.4, 28.8, 24.7, 22.4, 20.9, 13.9. Oil at rt.
-(2-Hexa d eca n oyloxyeth yl)p yr id in iu m Br om id e (3g).
H NMR (CDCl ) δ 9.53 (d, 2H, J ) 7.5 Hz), 8.55 (q, 1H, J )
.5 Hz), 8.12 (q, 2H, J ) 7.5 Hz), 5.38 (t, 2H, J ) 5 Hz), 4.60
t, 2H, J ) 5 Hz), 2.22 (m, 2H), 1.44 (m, 2H), 1.19 (bs, 24H),
3
(
2-Hexadecan oyloxyeth yl)tr ieth ylam m on iu m Br om ide
1
(
2c). H NMR (CDCl
3
) δ 4.52 (t, 2H, J ) 7.0 Hz), 3.84 (t, 2H,
1
J ) 7.0 Hz), 3.59 (q, 6H, J ) 7.2 Hz), 2.29 (t, 2H, J ) 7.8 Hz),
1
3
1
3
4
2
.56 (m, 2H), 1.39 (t, 9H, J ) 7.2 Hz), 1.21 (m, 24H), 0.83 (t,
H, J ) 7.0 Hz). 13C NMR (CDCl
) δ 172.7, 57.1, 55.8, 54.3,
6.0, 33.7, 31.8, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 24.5,
2.6, 14.0, 8.1. Mp 157-162 °C. Anal. (C24 50BrNO ) H, N; C:
7
3
(
13
0
6
2
.82 (t, 3H, J ) 6.8 Hz). C NMR (CDCl
2.6, 60.4, 33.2, 31.7, 29.7, 29.6, 29.5 29.4, 29.3, 29.2, 29.1,
2
9.0, 28.9, 24.2, 22.5, 14.1. Mp 78-82 °C. Anal. (C23 40BrNO )
3
) δ 172.8, 145.8, 128.1,
H
2
calcd, 62.05; found, 60.42.
H
Gen er a l P r oced u r e for Com p ou n d s 3b, 3r , a n d 3z.
Equimolar amounts (30 mmol) of the corresponding acyloxy-
alkyl halide and pyridine were mixed and heated to 70-90 °C
for 3-6 h in water bath. The mixture was cooled to room
temperature, anhydrous ether was added, and the mixture was
triturated in anhydrous ether overnight. The solid was col-
lected by filtration, washed thoroughly with ether, and purified
by silica gel chromatography. Yields 72-86%.
H, N; C: calcd 62.43; found 63.18. 1-(2-Eicosa n oyloxyeth yl)-
1
p yr id in iu m Br om id e (3h ). H NMR (CDCl ) δ 9.53 (d, 2H,
3
J ) 7.5 Hz), 8.51 (q, 1H, J ) 7.5 Hz), 8.12 (q, 2H, J ) 7.5 Hz),
5
1
.41 (t, 2H, J ) 5 Hz), 4.61 (t, 2H, J ) 5 Hz), 2.24 (m, 2H),
.46 (m, 2H), 1.21 (bs, 28H), 0.84 (t, 3H, J ) 6.3 Hz). C NMR
13
3
(CDCl ) δ 172.9, 145.7, 128.1, 62.6, 60.4, 33.7, 31.8, 29.7, 29.6,
2
8
9.5, 29.4, 29.3, 29.2, 29.1, 29.0, 28.9, 24.5, 22.6, 14.0. Mp 79-
3 °C. Anal. (C25 44BrNO )C, H, N.
-(2-(cis-9)-Oct a d ecen oyloxyet h yl)p yr id in iu m Br o-
H
2
Acetyloxym eth ylp yr id in iu m Ch lor id e (3b). 1H NMR
1
(CDCl
3
) δ 9.43 (d, 2H, J ) 7.5 Hz), 8.55 (q, 1H, J ) 7.5 Hz),
m id e (3i). Equimolar amounts of oleic acid (1.0 g, 3.5 mmol)
and compound 3d (0.7 g, 3.5 mmol) were mixed with an excess
.20 (q, 2H, J ) 7.5 Hz), 6.80 (s, 2H), 1.21 (s, 3H). 1 C NMR
3
8
(CDCl
3
) δ 173.0, 148.2, 144.9, 129.1, 79.7, 33.3, 17.0.
(35%) of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDAC)
1
-Eth yloxyca r bon ylm eth ylp yr id in iu m Ch lor id e (3r ).
H NMR (CDCl ) δ 9.48 (d, 2H, J ) 6.5 Hz), 8.49 (q, 1H, J )
.5 Hz), 8.10 (q, 2H, J ) 7.0 Hz), 6.55 (s, 2H), 1.60 (m, 2H),
(
0.9 g) in CH Cl and a catalytic amount of 4-(dimethylamino)-
2
2
1
3
pyridine (DMAP) (0.01 g, 0.1 mmol) and heated at reflux
overnight. The mixture was cooled to room temperature and
concentrated under reduced pressure. Cold anhydrous ether
was added, and the mixture was stirred for 2 h. An impure
solid was isolated by filtration and washed thoroughly with
7
0
.85 (t, 3H, J ) 7.0 Hz).
1
1
-Eth yloxyca r bon yleth ylp yr id in iu m Ch lor id e (3z). H
NMR (CDCl
3
) δ 9.45 (d, 2H, J ) 6.0 Hz), 8.45 (q, 1H, J ) 7.5
Hz), 8.05 (q, 2H, J ) 7.1 Hz), 6.38 (m, 1H), 4.17 (d, 3H, J )
7
cold ether. This was purified by silica gel chromatography to
.0 Hz), 3.62 (m, 2H), 0.80 (t, 3H, J ) 7.5 Hz).
-Dod eca n oyloxym et h ylp yr id in iu m Ch lor id e (3c).
1
afford a yellowish oil (1.3 g, 81% yield). H NMR (CDCl
9
3
) δ
1
.40 (d, 2H, J ) 7.5 Hz), 8.47 (q, 1H, J ) 7.5 Hz), 8.03 (q, 2H,
5
Adopted from Bodor et al. Equimolar amounts (32 mmol) of
the corresponding acyloxyalkyl chloride and paraformaldehyde
were mixed and heated between 90 and 100 °C for 3-6 h. The
resulting crude chloromethyl ester was purified by silica gel
chromatography. A mixture of 1-dodecanoyloxymethyl chloride
and pyridine was heated at 90 °C for 3h. The mixture was
cooled to room temperature, anhydrous ether was added, and
the mixture was stirred in anhydrous ether overnight. The
J ) 7.5 Hz), 6.09 (m, 1H), 5.23 (m, 2H, J ) 5 Hz), 5.16 (m,
1
H), 4.50 (t, 2H, J ) 5 Hz), 2.11 (m, 2H), 1.35 (m, 2H), 1.09
13
(
bs, 20H), 0.70 (t, 3H, J ) 6.3 Hz). C NMR (CDCl ) δ 172.6,
3
1
2
45.6, 145.3, 129.2, 62.3, 60.1, 55.1, 33.7, 31.8, 29.6, 29.5, 29.4,
9.3, 29.2, 29.1, 29.0, 24.5, 22.6, 14.0.
Gen er a l P r oced u r e for Com p ou n d s 2d a n d 3k . Equi-
molar amounts of palmitoyl chloride (2.0 g, 7.3 mmol) and
-chloroethylammonium chloride (0.9 g, 7.3 mmol) were mixed
and heated at 70 °C for 24 h. CH Cl was added and heated
2
solid was isolated by filtration and washed thoroughly with
1
2
2
ether. Mp 71-75 °C and yield 68%. H NMR (CDCl
3
) δ 9.44
at reflux for 24 h. The mixture was cooled to a room temper-
(d, 2H, J ) 7.5 Hz), 8.57 (q, 1H, J ) 7.5 Hz), 8.18 (q, 2H, J )
ature and purified by silica gel chromatography, affording a
7
0
1
2
.5 Hz), 6.82 (s, 2H), 2.33 (m, 2H), 1.54 (m, 2H), 1.18 (bs, 16H),
1
1
3
clear oil (1.8 g, 79% yield). H NMR (CDCl
3
) δ 5.90 (bs, 1H),
.81 (t, 3H, J ) 6.7 Hz). C NMR (CDCl ) δ 172.6, 147.7, 145.6,
3
3
.61 (m, 4H), 2.20 (m, 2H), 1.20 (m, 12H), 0.87 (t, 3H, J ) 6.5
28.2, 79.7, 33.4, 31.8, 29.5, 29.4, 29.2, 29.1, 29.0 28.9, 24.2,
2.5, 14.0.
13
Hz). C NMR (CDCl
3
) δ 140.1, 44.2, 41.1, 36.7, 31.0, 29.7, 29.3,
2
9.1, 29.0, 25.7, 22.7, 14.1. Equimolar amounts of triethy-
1
-(2-H yd r oxyet h yl)p yr id in iu m Br om id e (3d ). Equi-
molar amounts of pyridine (2.5 g, 32 mmol) and bromoethanol
3.9 g, 32 mmol) were mixed and heated slowly until the
lamine or pyridine were mixed with the product and refluxed
overnight. The mixture was cooled to a room temperature,
anhydrous ether was added, and the solid was isolated by
filtration and washed thoroughly with ether. Purification by
chromatography afforded white crystals. Yields 73-81%.
(
solution turned to a solid. The mixture was cooled to room
temperature, anhydrous ether was added, and the solid was
washed thoroughly with ether. Recrystallization in ethanol
afforded white crystals (6.1 g, 95% yield). Mp 99-103 °C. H
NMR (D O) δ 8.83 (dd, 2H, J ) 7 Hz), 8.58 (q, 1H, J ) 7.5
1
Tr ieth yl-(2-h exadecan oylam in oeth yl)am m on iu m Ch lo-
-
1
1
r id e (2d ). IR (KBr) 1651 cm (carbamyl), H NMR (CDCl ) δ
3
2
Hz), 8.08 (q, 2H, J ) 7.5 Hz), 4.72 (t, 2H, J ) 5.5 Hz), 4.07 (t,
H, J ) 5.5 Hz). Anal. (C 10BrNO) C, H, N.
Gen er a l P r oced u r e for Com p ou n d s 3e-h . Equimolar
4.43 (t, 2H, J ) 7.5 Hz), 3.93 (t, 2H, J ) 6.5 Hz), 3.14 (q, 6H,
J ) 7.0 Hz), 2.56 (m, 2H), 1.40 (t, 9H, J ) 7.5 Hz), 1.18 (m,
2
7
H
2
4H), 0.83 (t, 3H, J ) 6.5 Hz). Mp 109-113 °C.
amounts of corresponding fatty acid chloride (10 mmol) and
compound 3d were mixed and heated at 70 °C for 3-6 h. The
mixture was cooled to room temperature, anhydrous ether was
added, and the solid was isolated by filtration and washed
thoroughly with ether. The crude product was crystallized from
1-(2-H exa d eca n oyla m in oet h yl)p yr id in iu m Ch lor id e
-
1
1
(3k ). IR (KBr) 1651 cm (carbamyl), H NMR (CDCl ) δ 9.47
3
(d, 2H, J ) 7.5 Hz), 8.59 (t, 1H, J ) 7.0 Hz), 8.41 (t, 1H, J )
7.5 Hz), 8.00 (t, 2H, J ) 7.5 Hz), 6.25 (t, 2H, J ) 6.3 Hz), 3.90
(q, 2H, J ) 5.0 Hz), 2.12 (t, 2H, J ) 7.5 Hz), 1.39 (m, 2H),
1.25 (m, 24H), 0.84 (t, 3H, J ) 5.0 Hz). 13C NMR (CDCl ) δ
dichloromethane:diethyl ether (1:1). Yields 86-91%.
3
1
1
-(2-Dod eca n oyloxyeth yl)p yr id in iu m Br om id e (3e). H
174.6, 145.7, 144.9, 127.8, 60.6, 39.7, 36.0, 31.8, 29.8, 29.6, 29.3,
NMR (CDCl
Hz), 8.12 (q, 2H, J ) 7.5 Hz), 5.39 (t, 2H, J ) 5 Hz), 4.61 (t,
H, J ) 5 Hz), 2.22 (m, 2H), 1.44 (m, 2H), 1.18 (bs, 16H), 0.81
3
) δ 9.56 (d, 2H, J ) 7.5 Hz), 8.53 (q, 1H, J ) 7.5
29.1, 29.0, 25.6, 22.6, 14.1. Mp 123-127 °C. Anal. (C23
H41(Br0.7-
OH0.3)NO ) C, H, N. Compounds 2e and 3l: equimolar
2
2
(
amounts of corresponding quaternary ammonium compound
(3q or 2a ) (4 mmol) and refined cod liver fatty acids were mixed
with excess (35%) of EDAC in CH Cl and a catalytic amount
2 2
of DMAP and heated at reflux overnight. The mixture was
cooled to room temperature and evaporated, anhydrous ether
was added, and the solid was isolated by filtration and washed
thoroughly with ether. The crude compound was purified by
1
3
t, 3H, J ) 6.3 Hz). C NMR (CDCl
3
) δ 172.8, 145.7, 145.6,
28.1, 62.6, 60.4, 33.2, 31.7, 29.4, 29.3, 29.2, 29.1, 29.0 28.9,
4.2, 22.5, 14.0. Mp 78-83 °C. Anal. (C19 32BrNO )N; C: calcd
9.06; found 59.64, H: calcd, 8.35; found 8.87.
-(2-Octa n oyloxyeth yl)p yr id in iu m Br om id e (3f). 1
) δ 9.42 (d, 2H, J ) 7.5 Hz), 8.55 (q, 1H, J ) 7.5
1
2
5
H
2
1
H
NMR (CDCl
3

Soft Antimicrobial Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19 4179
silica gel chromatography, affording a clear or yellowish oil.
a white solid (except hexanoyl derivative was yellowish oil).
Yields 56-68%.
Yields 76-82%.
(
2-Cod liver oil eth yl)tr ieth yla m m on iu m Br om id e (2e).
1-Hexyloxyca r bon ylm eth ylp yr id in iu m Ch lor id e (3s).
1
1
H NMR (CDCl ) δ 5.42 (m, 3H), 4.75 (t, 2H, J ) 7.0 Hz), 3.85
3
H NMR (CDCl ) δ 9.50 (d, 2H, J ) 6.0 Hz), 8.48 (q, 1H, J )
3
(
2
2
t, 2H, J ) 7.0 Hz), 3.53 (q, 6H, J ) 7.5 Hz), 2.70 (m, 1H),
7
)
.65 Hz), 8.05 (q, 2H, J ) 7.15 Hz), 6.38 (s, 2H), 4.17 (t, 2H, J
.00 (m, 2H), 1.58 (m, 2H), 1.40 (t, 9H, J ) 7.5 Hz), 1.22 (m,
0H), 0.85 (t, 3H, J ) 7.0 Hz).
7.0 Hz), 1.63 (m, 2H), 1.24 (m, 4H), 0.85 (t, 3H, J ) 7.0 Hz).
Mp oil at rt.
-Hexa d ecyloxyca r bon ylm eth ylp yr id in iu m Ch lor id e
1
-(2-Cod liver oil eth yl)p yr id in iu m Br om id e (3l). 1
H
1
NMR (CDCl
Hz), 8.21 (t, 1H, J ) 7.5 Hz), 5.42 (m, 4H), 4.65 (m, 2H), 3.45
m, 4H), 2.00 (m, 2H), 1.58 (m, 2H), 1.22 (m, 22H), 0.85 (t,
H, J ) 7.0 Hz).
-(2-Cod liver oil sa t. eth yl)p yr id in iu m Br om id e (3m ).
Same as for 3l except the refined cod liver fatty acids had been
perhydrogenated with H and palladium on carbon catalysis.
Yield (54%). H NMR (CDCl ) δ 9.55 (d, 2H, J ) 7.5 Hz), 8.55
t, 1H, J ) 7.0 Hz), 8.12 (t, 1H, J ) 7.5 Hz), 5.41 (m, 1H), 4.62
m, 2H), 3.46 (m, 1H), 1.97 (m, 2H), 1.49 (m, 2H), 1.21 (m,
4H), 0.84 (t, 3H, J ) 7.0 Hz).
-(3-Hyd r oxyp r op yl)p yr id in iu m Ch lor id e (3n ). Equi-
3
) δ 9.50 (d, 2H, J ) 7.5 Hz), 8.52 (t, 1H, J ) 7.0
1
(
3t). H NMR (CDCl
3
) δ 9.50 (d, 2H, J ) 6.7 Hz), 8.47 (q, 1H,
J ) 7.75 Hz), 8.04 (q, 2H, J ) 7.75 Hz), 6.43 (s, 2H), 4.17 (t,
(
3
2
7
H, J ) 7.0 Hz), 1.64 (m, 2H), 1.24 (m, 24H), 0.87 (t, 3H, J )
.0 Hz). Mp 113-114 °C.
1
1
-Oct a d ecyloxyca r bon ylm et h ylp yr id in iu m Ch lor id e
1
(
3u ). H NMR (CDCl
J ) 7.75 Hz), 8.04 (q, 2H, J ) 7.75 Hz), 6.43 (s, 2H), 4.17 (t,
3
) δ 9.52 (d, 2H, J ) 6.7 Hz), 8.46 (q, 1H,
2
1
3
2
7
H, J ) 7.0 Hz), 1.64 (m, 2H), 1.24 (m, 28H), 0.86 (t, 3H, J )
.0 Hz). Mp 38-39 °C.
Com p ou n d s 3v-w . Chloroacetyl chloride was mixed in
excess amount with corresponding fatty amine (6 mmol) and
heated between 60 and 70 °C for 4 h. The excess of chloroacetyl
chloride was removed in a vacuum. The obtained chloro amide
and pyridine was mixed and heated at 60 °C for 4 h. The
mixture was cooled to room temperature, anhydrous ether was
added, and the mixture was triturated in anhydrous ether.
The white solid was isolated by filtration and washed thor-
oughly with ether. Purified by silica gel chromatography, yields
(
(
2
1
molar amounts of pyridine (2.5 g, 32 mmol) and 3-chloro-1-
propanol (3.0 g, 32 mmol) were mixed and slowly heated to 70
C and stirred for 3 h. The mixture was cooled to room
°
temperature, anhydrous ether was added, and the solid was
washed thoroughly with ether. Recrystallization in ethyl
acetate:MeOH (14:1) afforded white crystals (4.6 g, 83% yield).
Gen er a l P r oced u r e for Com p ou n d s 3o-p . Equimolar
amounts of corresponding fatty acid chloride (5 mmol) and
compound 3n were mixed and heated at 70 °C for 3-6 h. The
mixture was cooled to room temperature, anhydrous ether was
added, and the solid was isolated by filtration and washed
thoroughly with ether. Purified by silica gel chromatography
and recrystallized in mixture of ethyl acetate:MeOH (10:1)
affording white crystals. Yields 82-89%.
(
70-81%).
1-Dod ecylca r ba m oylm eth ylp yr id in iu m Ch lor id e (3v).
-
1
1
IR (KBr) 1671 cm (carbamyl). H NMR (CDCl ) δ 9.34 (bs,
3
3H), 8.43 (t, 1H, J ) 7.5 Hz), 8.03 (t, 2H, J ) 7.5 Hz), 5.95 (s,
2H), 3.19 (q, 2H, J ) 7.5 Hz), 1.54 (m, 2H), 1.22 (m, 18H),
0.85 (t, 3H, J ) 6.4 Hz). 13C NMR (CDCl
) δ 163.5, 146.0, 145.0,
127.6, 62.5, 40.3, 31.9, 29.5-29.0, 27.0, 22.6, 14.1. Mp 83-87
°C. Anal. (C19 O) H, N; C: calcd 69.94; found 65.05
1-Octadecylcar bam oylm eth ylpyr idin iu m Ch lor ide (3w).
3
1
-(3-Dod eca n oyloxyp r op yl)p yr id in iu m Ch lor id e (3o).
H NMR (CDCl ) δ 9.61 (d, 2H, J ) 7.5 Hz), 8.47 (t, 1H, J )
.5 Hz), 8.10 (t, 2H, J ) 7.5 Hz), 5.08 (t, 2H, J ) 7.5 Hz), 4.13
t, 2H, J ) 6.3 Hz), 2.39 (t, 2H, J ) 6.3 Hz), 2.12 (t, 2H, J )
.5 Hz), 1.43 (m, 2H), 1.18 (m, 16H), 0.78 (t, 3H, J ) 6.3 Hz).
H33ClN
2
1
3
7
(
-1
1
IR (KBr) 1651 cm (carbamyl). H NMR (CDCl
3
) δ 9.35 (bs,
H), 8.43 (t, 1H, J ) 7.5 Hz), 8.03 (t, 2H, J ) 7.5 Hz), 5.94 (s,
H), 3.20 (q, 2H, J ) 7.5 Hz), 1.56 (m, 2H), 1.23 (m, 30H),
3
2
0
1
7
1
3
C NMR (CDCl
1.7, 30.7, 29.4-28.9, 24.6, 22.5, 14.0. Mp 149-152 °C. Anal.
34ClNO ) H, N; C: calcd, 67,49; found 66. 71. 1-(3-
Hexa d eca n oyloxyp r op yl)p yr id in iu m Ch lor id e (3p ).
NMR (CDCl ) δ 9.54 (d, 2H, J ) 5.0 Hz), 8.51 (t, 1H, J ) 7.5
3
) δ 173.4, 145.5, 145.1, 128.4, 60.5, 59.0, 33.9,
13
.86 (t, 3H, J ) 6.4 Hz). C NMR (CDCl
27.6, 62.5, 40.3, 31.9, 29.6-29.0, 27.0, 22.6, 14.1. Mp 93-97
C. Anal. (C25 O) H, N; C: calcd. 70.64; found 68.86.
-(2-Ca r boxyeth yl)p yr id in iu m Ch lor id e (3x). A mixture
3
) δ 163.4, 146.0, 145.0,
3
(
C
20
H
2
°
H45ClN
2
1
H
1
3
of chloropropionic acid (3.5 g, 32 mmol) in equimolar amounts
of pyridine (2.5 g, 32 mmol) was heated to 70-80 °C until the
mixture had turned solid. Washed thoroughly with ether and
recrystallization in 50/50 mixture of ethyl acetate and ethanol
Hz), 8.13 (t, 2H, J ) 7.5 Hz), 5.09 (t, 2H, J ) 6.3 Hz), 4.16 (t,
2
H, J ) 6.3 Hz), 2.42 (t, 2H, J ) 6.3 Hz), 2.16 (t, 2H, J ) 7.5
13
Hz), 1.48 (m, 2H), 1.22 (m, 24H), 0.83 (t, 3H, J ) 6.3 Hz).
C
NMR (CDCl ) δ 173.4, 145.5, 145.2, 128.4, 60.5, 59.0, 33.9, 31.7,
3
afforded white crystals (5.4 g, 90% yields). Mp 157-160 °C.
3
0.8, 29.4, 29.3, 29.1, 29.0, 24.6, 22.5, 14.0. Mp 129-133 °C.
Anal. (C24 42ClNO ) N; C: calcd 69.96; found 66.71, H: calcd.
0.27; found 10.77.
-Ca r boxym eth ylp yr id in iu m Ch lor id e (3q). A mixture
1
H NMR (D O) δ 8.90 (q, 2H, J ) 5.5 Hz), 8.54 (q, 1H, J ) 7.5
2
H
2
Hz), 8.05 (q, 2H, J ) 7.5 Hz), 4.87 (t, 2H, J ) 6 Hz), 3.17 (t,
1
2
5
H, J ) 6 Hz). 13C NMR (D
9.3, 36.9. Anal. (C 10ClNO
-(2-Hexa d ecyloxyca r bon yleth yl)p yr id in iu m Ch lor id e
3y). Equimolar amounts of hexadecanoic acid (1.0 g, 3.9
mmol) and compound 3x (0.7 g, 3.9 mmol) were mixed with
an excess (35%) of EDAC (1.0 g) in CH Cl and a catalytic
2
O) δ 176.0, 148.5, 147.3, 130.6,
2
) C, H, N.
1
8
H
of chloroacetic acid (2.5 g, 26 mmol) in equimolar amounts of
pyridine (2.1 g, 26 mmol) was heated to 70-80 °C until all
the mixture had turned solid. Washed thoroughly with ether
1
(
and recrystallization from ethanol which afford white crystals
1
2
2
(
2
5
4.6 g, 92% yield). Mp 118-122 °C. H NMR (D
2
O): δ 8.84 (q,
amount of DMAP (0.01 g, 0.1 mmol), heated to reflux over-
night. The mixture was cooled to room temperature and
concentrated under reduced pressure. Cold anhydrous ether
was added, and the mixture was stirred for 2 h. The solid was
isolated by filtration and washed thoroughly with cold ether.
H, J ) 7 Hz), 8.62 (q, 1H, J ) 8 Hz), 8.12 (q, 2H, J ) 8 Hz),
13
.43 (s, 2H). C NMR (D
2
O): δ 176.4, 149.0, 147.6, 130.8, 59.5.
Gen er a l P r oced u r e for Com p ou n d s 3s-u . Chloroacetyl
chloride in excess amount was mixed with corresponding fatty
alcohol (10 mmol) and heated between 60 and 70 °C for 4 h.
The resultant chloro ester was purified by silica gel chroma-
tography and crystallized in MeOH (except the hexyl chloro
ester was liquid). Structure verified by NMR. Cetyl chloro
Purified by silica gel chromatography afforded white solid (1.1
1
g, 67% yield). H NMR (CDCl
3
) δ 9.48 (d, 2H, J ) 7.0 Hz), 8.45
(
q, 1H, J ) 7.7 Hz), 8.03 (q, 2H, J ) 7.7 Hz), 5.43 (t, 2H, 5.0
Hz), 4.27 (t, 2H, J ) 5.0 Hz), 1.62 (m, 2H), 1.22 (m, 24H), 0.88
t, 3H, J ) 7.0 Hz). Mp 97-101 °C. Anal. (C24 42ClNO ‚H O):
C, N; H: calcd. 10.31; found 10.77.
1
ester: H NMR (CDCl
H), 1.65 (m, 2H), 1.25 (m, 24H), 0.83 (t, 3H, J ) 7.0 Hz). Steryl
chloro ester: H NMR (CDCl
s, 2H), 1.66 (m, 2H), 1.24 (m, 28H), 0.83 (t, 3H, J ) 7.0 Hz).
A mixture of chloroester and pyridine was mixed and heated
at 60 °C for 4 h. The mixture was cooled to a room tempera-
ture, anhydrous ether was added, and the mixture was
triturated in anhydrous ether. The solid was isolated by
filtration and washed thoroughly with ether, which afforded
3
) δ 4.18 (t, 2H, J ) 7.0 Hz), 4.05 (s,
(
H
2
2
2
1
3
) δ 4.20 (t, 2H, J ) 7.0 Hz), 4.07
(
Tr ibu tyl-(2-h yd r oxyeth yl)a m m on iu m Br om id e (4a ). A
mixture of bromoethanol (3.5 g, 28 mmol) in equimolar
amounts of tributylamine (5.2 g, 28 mmol) was heated to 70-
80 °C and stirred overnight. Washed thoroughly with ether,
1
afford white solid (6.7 g, yields 78%). H NMR (CDCl
3
) δ 4.62
(t, 2H, J ) 5.5 Hz), 4.07 (t, 2H, J ) 5.5 Hz), 3.42 (t, 6H, J )

4
180 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19
.5 Hz), 1.72 (m, 6H), 1.43 (m, 6H), 0.97 (t, 9H, J ) 7.5 Hz).
Thorsteinsson et al.
7
value 1 µg/mL for the doubling dilutions. The media used for
MIC measurements was Mueller Hinton broth (Oxoid) and
blood agar (heart infusion agar (Oxoid) with 5% defibrinated
horse blood) for measurements of MLC. Measurements are
expressed in µg/mL. The broth dilutions and the agar plates
were incubated for 24h, at 37 °C, in ambient air. The bacterial
strains, Enterococcus faecalis (ATCC 29212), Staphylococcus
aureus (ATCC 25923), E. coli (ATCC 25922), and Pseudomonas
aeruginosa (ATCC 27853), selected were type strains from the
ATCC (American Type Culture Collection), representing clini-
cally important species recommended by the NCCLS as quality
control strains for susceptibility testing.
An tivir a l Test. Vero cells (African green monkey kidney
cell line) were grown in RPMI-1640 media with 2.0 mM
L-glutamine, 0.05 mg/mL gentamicin, 0.375% (w/v) sodium
bicarbonate, and 10% heat-inactivated fetal bovine serum
FBS). The maintenance medium (MM) for Vero cell mono-
layers was RPMI-1640 with 2% FBS. All media was obtained
from GIBCO, Paisley, Scotland. Herpes simplex virus type 1
1
3
C NMR (CDCl
7 °C.
3
) δ 59.9, 58.3, 49.4, 29.3, 21.1, 13.5. Mp 63-
6
Gen er al P r ocedu r e Com pou n ds 4b,c. Equimolar amounts
of corresponding fatty acid chloride (6 mmol) and compound
a were mixed, sonicated, and refluxed in CH Cl overnight.
4
2
2
The mixture was cooled to room temperature and evaporated,
anhydrous ether was added, and the solid was isolated by
filtration and washed thoroughly with ether. The obtained
compound was recrystallized in mixture of ethyl acetate:MeOH
(10:2), affording white crystals. Yields 64-76%.
Tr ibu tyl-(2-d od eca n oyloxyeth yl)a m m on iu m Br om id e
1
(
4b). H NMR (CDCl
3
) δ 4.52 (bs, 2H), 3.91 (bs, 2H), 3.41 (t,
6
H, J ) 7.7 Hz), 2.26 (t, 2H, J ) 7.4 Hz), 1.69 (m, 6H), 1.55-
(
m, 2H), 1.40 (m, 6H), 1.20 (m, 24H), 0.97 (t, 9H, J ) 7.1 Hz).
1
3
C NMR (CDCl
9.7, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 24.7, 24.5, 22.9,
0.0, 13.6, 13.5. Mp 60-64 °C. Anal. (C26 54BrNO ) H, N; C:
3
) δ 172.7, 59.5, 57.5, 34.0, 31.6, 29.9, 29.8,
(
2
2
H
2
calcd. 63.39; found 62.16.
(
HSV-1) strain MacIntyre (American Type Culture Collection,
Rockville, MD) was grown in Vero cells. Virus stocks with an
Tr ibu tyl-(2-d od eca n oyloxyeth yl)a m m on iu m Br om id e
1
(
4c). H NMR (CDCl
3
) δ 4.53 (bs, 2H), 3.92 (bs, 2H), 3.42 (t,
H, J ) 7.9 Hz), 2.28 (t, 2H, J ) 7.6 Hz), 1.70 (m, 6H),
.57(m, 2H), 1.41 (m, 6H), 1.22 (m, 16H), 0.98 (t, 9H, J ) 7.2
infectivity titer of 106 -10 TCID50 (50% Tissue Culture
Infective Dose) per 100 µL were used in the experiments. The
virus was titrated by inoculation of 10-fold dilutions in MM
into monolayers of Vero cells in 96-well microtiter tissue
culture plates (Nunc, Roskilde, Denmark). One hundred
microliters of each virus dilution was inoculated into quadru-
plicate wells. The plates were incubated at 37 °C in a
.5
7.5
6
1
1
3
Hz). C NMR (CDCl
9.2, 29.1, 24.5, 22.5, 20.0, 13.6, 13.5. Mp 79-83 °C. Anal.
62BrNO ) C, H, N.
HP LC Mea su r em en ts. The HPLC system consisted of
3
) δ 172.7, 59.5, 57.6, 34.0, 31.7, 29.4, 29.3,
2
(C
30
H
2
Merck Hitachi L-7100 gradient solvent delivery system with
a L-7400 UV detector, using (a) 150 mm, 4.6 mm i.d., 5 µm
bead, C18 reverse-phase column; (b) cyanocolumn 150 mm,
2
humidified incubator with 5% CO in the air and examined
for cytophatic effect daily for 5 days. Virus titers were
calculated by the method of Reed and Muench (1938). Stock
solutions of the compounds were made in Hanks balanced salt
solution (HBSS) at a concentration of 1 mg/mL. Two-fold
dilutions of each stock solution were made in MM, and each
dilution was mixed with an equal volume (100µL) of HSV-1
in 12 × 75 mm polystyrene round-bottom tubes (Falcon). The
mixtures were incubated at 37 °C in a water bath for 2 h. The
action of the compounds was then stopped by diluting the
mixtures 10-fold in MM. Virus mixed with MM served as a
control. The virus titers of the mixtures were determined by
an inoculation of 10-fold dilutions into cell culture as previously
described. The difference in the titer (log ) of the compound-
virus mixture and the titer (log ) of the control mixture i.e.,
the reduction in viral infectivity, was used as a measure of
the antiviral activity of the compounds.
4
.6 mm i.d., 5 µm; (c) 75 mm, 4.6 mm i.d., 3 µm bead, C18
reverse-phase column; (d) LiChrosorb NH , 5 µm. The detection
2
was performed at 230 or 254 nm wavelength, and the flow rate
was 1.5 mL/min. The mobile phases and retention times for
each compound were as follows: (3c) (a) gradient system of
acetonitrile, 0.015% octasulfonic acid in
acetonitrile in 15 min), 4.0 min, (3e) acetonitrile, acetic acid,
O (90:0.5:9.5), 4.1 min; (b) (3s) MeOH, 0.05 M phosphoric
acid (92:8), 2.1 min. (3t) 2.6 min. (3u ) 3.6 min. (3q) 3.0 min.
c) MeOH, 1% acetic acid in H O (55:45), (3g) 2.8 min. (3h )
.6 min. (3i) 3.3 min. (3w ) (81:19), 2.9 min. (3v) (68:32), 3.4
min. (3o) (70:30), 2.1 min. (3p ) (75:25), 3.8 min. (d) 0.015%
octasulfonic acid in H O, MeOH (80:20) (3d ) 2.5 min.
2
H O (40-100%
H
2
(
3
2
2
Degr a d a tion Ra te Stu d ies. Stock solution (0.1 mg/mL)
was prepared in ethanol, and 30 µL of this solution was added
to 1.5 mL an aqueous buffer solution. The degradation was
measured by periodic HPLC analysis of buffered solutions
heated on the sample rack. The degradation rate constants
Ack n ow led gm en t. This study was supported in
part by a research grant from the Icelandic Research
Council. Special thanks to Margret Thorsteinsdottir and
Gisli Bragason at Encode for LC-MS/MS measurements.
(kobs) were obtained by linear regression of the peak integra-
tion. The degradation rates of the compounds were determined
at pH 6 in 10 mM phosphate buffer at 60 °C ((0.2 °C). Stability
of selected compounds (3c, 3e, and 3v) was also determined
in diluted (1:4) agar at 37 °C.
Compounds that did not have a good chromophore were
measured at Encode Inc. in a LC-MS/MS Quattro Ultima from
Micromass with Hewlett-Packard 1100 gradient pump and
HCT PAL autosampler from CTC Analytics.
Lip op h ilicity Ca lcu la tion . Octanol-water partition coef-
ficient (ClogP) was calculated online (www.syrres.com) accord-
ing to structure. Two compounds had sufficient aqueous
solubility for the octanol-water partition coefficients deter-
mination and were used to compare to the calculation. These
compounds were measured according to the shake flask
method. Literature values for the known compounds were
also used to compare to the calculated values. The calculations
were less than 10% from the experimental values.
An tiba cter ia l Test. The antibacterial tests were performed
at the Department of Clinical Microbiology, Landspitali Uni-
versity Hospital. The method used was a microtiter dilution,
performed according to the NCCLS standard. It was adapted
to the specific needs related to the solubility of the compounds.
In the case of poor water solubility, as small a concentration
of dimethyl sulfoxide (Sigma) as possible was used as a solvent,
and in one of the test series it was not possible to obtain the
Refer en ces
(
1) Pool, K. Overcoming antimicrobial resistance by targeting
resistance mechanism. J . Pharm. Pharmacol. 2001, 53, 283-
294.
(2) Petrocci, A. N. Surface-Active Agents: Quaternary Ammonium
compounds. In Disinfection, Sterilization and Preservation.;
Block, S. S., Ed.; Lea & Febiger Pub.: Philadelphia, 1983; pp
3
09-329.
(3) Bodor, N.; Kaminski, J . J . Soft drugs. 2. Soft alkylating
compounds as potential antitumor agents. J . Med. Chem. 1980,
2
3, 566-569.
(
(
(
4) Bodor, N.; Woods, R.; Raper, C.; Kaminski, J . J . Soft drugs. 3. A
new class of anticholinergic agents. J . Med. Chem. 1980, 23,
4
74-480.
5) Bodor, N.; Kaminski, J . J .; Selk, S. Soft drugs 1. Labile
quaternary ammonium salts as soft antimicrobials J . Med.
Chem. 1980, 23, 469-474.
1
3
6) Calvani, M.; Critelli, L.; Gallo, G.; Giorgi, F.; Gramiccioli, G.;
Santaniello, M.; Scafetta, N.; Tinti, M. O.; De Angelis, F.
L-Carnitine esters as “soft”, broad-spectrum antimicrobial am-
philes. J . Med. Chem. 1998, 41, 2227-2233.
(7) Bodor, N.; Buchwald, P. Soft Drug Design: General Principles
1
4
and Recent Applications. Med. Res. Rev. 2000, 20, 58-101.
(
8) Bodor, N. Soft drugs: Principles and methods for the design of
safe drugs Med. Res. Rev. 1984, 3, 449-469.
9) Cooper, J . C. Review of the Environmental Toxicity of Quater-
nary Ammonium Halides Ecotox. Environ. Safe. 1988, 16, 65-
71.
(

Soft Antimicrobial Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19 4181
(
10) Ahlstr o¨ m, B.; Thompson, R. A.; Edebo, L. The effect of hydro-
carbon chain length, pH, and temperature on the binding and
bactericidal effect of amphiphilic betaine esters on Salmonella
typhimurium. APMIS 1999, 107, 318-324.
(15) Ari e¨ ns, E. J . Design of safer chemicals. In Drug Design; Ari e¨ ns,
E. J ., Ed.; Academic Press: New York, 1980; Vol. IX, pp 1-46.
(16) Cutler, R. A.; Cimijotti, E. B.; Okolowich, T. J .; Wetterau, W.
F.; Alkylbenzyl-dimethylammonium chlorides- a comparative
(
11) Ahlstr o¨ m, B.; Thompson, R. A.; Edebo, L. Loss of bactericidal
capacity of long-chain quaternary ammonium compounds with
protein at lowered temperature. APMIS 1999, 107, 606-614.
12) Edebo, L.; Ahlstr o¨ m, B.; Allenmark, S.; Bertilsson, M.; J ennische,
E.; Lange, S.; Lindstedt, M.; Thompson, R. A.; Betain esters:
Quaternary ammonium compounds with time-limited activity.
In Industrial Applications of Surfactants III; Karsa, D. R., Ed.;
Springer-Verlag: New York, 1992; pp 184-207.
study of the odd and even chain homologues, In C.S.M. A
Proceedings of the 53rd Annual Meeting; 1966; pp 102-113.
(17) Shelton, R. S.; Van Campen, M. G.; Tilford, C. H.; Lang, H. C.;
Nisonger, L.; Bandelin, F. J .; Rubenkoenig, H. L. Quaternary
ammonium salts as germicides. I. Nonacylated quaternary
ammonium salts derived from aliphatic amines. J . Am. Chem.
Soc. 1946, 68, 753-755.
(
(
18) J ono, K.; Takayama, T.; Kuno, M.; Higashide, E. Effect of alkyl
chain length of benzalkonium chloride on the bactericidal
activity and binding to organic materials, Chem. Pharm. Bull.
(13) Sangster, J . Octanol water partition coefficients; J ohn Wiley and
Sons Ltd: Chichester, 1997; pp 57-59.
(
14) NCCLS, Methods for Dilution Antimicrobial Suseptibility Tests
for Bacteria That Grow Aerobically; approved Standard, 5th ed.;
NCCLS document M7-A5 (ISBN 1-56238-394-9); NCCLS: Wayne,
PA, 2001.
1
986, 34, 42154224.
J M030829Z