Enhanced activity of fluorinated quaternary ammonium surfactants against Pseudomonas aeruginosa

Article abstract of DOI:10.1016/j.ejmech.2008.07.032

A novel series of fluorinated quaternary bisammonium surfactants has been synthesized in view to optimize their antimicrobial activities against Pseudomonas aeruginosa. As compared with commercial references, most of the new surfactants synthesized exhibit an enhanced activity which is discussed as a function of the nature of the spacer group between the quaternized nitrogen atoms and of the nature of the connector function between the nitrogen atoms and the perfluorinated carbon chains. It appears that the fluorinated "Gemini" surfactants bearing an amide connector can be an interesting alternative to hydrocarbon ammonium salts as preservatives and disinfectants against P. aeruginosa.

Full text of DOI:10.1016/j.ejmech.2008.07.032

European Journal of Medicinal Chemistry 44 (2009) 1615–1622
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Enhanced activity of fluorinated quaternary ammonium surfactants against
Pseudomonas aeruginosa
a
a
b
a,
*
Lionel Massi , Frederic Guittard , Richard Levy , S. G e´ ribaldi
a
Institut de Chimie de Nice FR 3037 CNRS, Laboratoire de Chimie des Mat e´ riaux Organiques et M e´ talliques, CMOM, UFR Sciences, Universit e´ de Nice-Sophia Antipolis,
Parc Valrose, 06108 Nice Cedex 2, France
Rohm and Haas France, Laboratoires Europ e´ ens, D e´ partement Process Chemicals and Biocides, 371, rue Beethoven, Sophia Antipolis, 06565 Valbonne, France
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 May 2008
Received in revised form 16 July 2008
Accepted 17 July 2008
Available online 5 August 2008
A novel series of fluorinated quaternary bisammonium surfactants has been synthesized in view to
optimize their antimicrobial activities against Pseudomonas aeruginosa. As compared with commercial
references, most of the new surfactants synthesized exhibit an enhanced activity which is discussed as
a function of the nature of the spacer group between the quaternized nitrogen atoms and of the nature of
the connector function between the nitrogen atoms and the perfluorinated carbon chains. It appears that
the fluorinated ‘‘Gemini’’ surfactants bearing an amide connector can be an interesting alternative to
hydrocarbon ammonium salts as preservatives and disinfectants against P. aeruginosa.
Ó 2008 Elsevier Masson SAS. All rights reserved.
Keywords:
Bisammoniums
Fluorinated surfactants
Minimal inhibitory concentration
Preservatives
Pseudomonas aeruginosa
Quaternary ammonium salts
1. Introduction
However, preservatives and disinfectants based on QACs are
widely used in the hospital environment in which P. aeruginosa is
Quaternary ammonium compounds (QACs) are among the most
common products used as preservatives and disinfectants in
human health-related domains such as disinfecting in hospital,
disinfectants for hard surface cleaning, disinfectants for contact
lenses, etc. [1]. However, this type of products suffers from a lack of
activity against one of the Gram (ꢀ) bacteria that have developed
the most important antimicrobial resistance phenomenon today,
Pseudomonas aeruginosa [2,3]. Indeed this organism possesses the
ability to metabolize quaternary ammonium compounds as
a source of carbon and nitrogen and it is well known that selective
culture media for Pseudomonas species contain hexadecyl trimethyl
ammonium bromide as only this species of bacteria can live in such
medium [4]. This extreme difficulty to prevent contamination from
this bacterium with QACs is found even with recent generations of
salts exhibiting strong activity and wide antibacterial spectra [5].
mostly a nosocomial pathogen [6–13]. So, it would be of the first
importance to find new classes of these cheap antibacterial agents
that will be also effective against P. aeruginosa.
Moreover, among the different classes of QACs used as bacteri-
cides, those presenting two quaternised nitrogen atoms, each
bearing an hydrophobic tail and called ‘‘Gemini’’ surfactants, seem
to exhibit more effective antimicrobial activities than their mono-
ammonium analogues as showed in several recent works [14–16].
To build our strategy for the design of new antimicrobial agents
with QAC structure able to possess an enhanced activity against
P. aeruginosa, we have to consider the specific nature of Gram (ꢀ)
bacterial membrane. Various mechanisms of antimicrobial resis-
tance can take place in this complex structure and especially for
P. aeruginosa [17]. On the other hand, the general behavior of bio-
films acting as a diffusion barrier and the specific mechanisms
employed by these microbial communities to resist the action of
classical antibacterial agents such as the hydrocarbon ‘‘Gemini’’
quarternary ammonium salts have to be considered. Taking account
of these parameters, we hypothesized that an important change in
the structural parameters of this type of surfactants could mislead
this Gram (ꢀ) bacteria in the expression of its various mechanisms
of resistance used as well in its planktonical nature as in its biofilm
forms. In this aim, our strategy is to radically change the hydrophilic
and lipophilic properties of classical hydrocarbon ‘‘Gemini’’
Abbreviations: cfu, colony forming unit; CMCs, critical micelle concentrations;
CPC, cetyl pyridinium chloride; DCC, dicyclohexylcarbodiimide; MIC, minimal
inhibitory concentration; QACs, quaternary ammonium compounds; r.t., room
temperature.
*
Corresponding author. Tel.: þ33 (0) 4 92 07 61 12; fax: þ33 (0) 4 92 07 61 56.
E-mail addresses: Lionel.Massi@unice.fr (L. Massi), Frederic.Guittard@unice.fr
(
F. Guittard), RLevy@rohmhaas.com (R. Levy), Serge.Geribaldi@unice.fr (S. G e´ ribaldi).
0223-5234/$ – see front matter Ó 2008 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2008.07.032

1616
L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
surfactants, first by replacement of the lipophilic hydrocarbon tails
of these antibacterial agents with semi-fluorinated tails connected
to the quaternized nitrogen atoms using various chemical active
functions, and secondly by introduction of various spacers,
different in nature and in length, between the two quaternized
nitrogen atoms.
- CH
+
3
C H
Cl
CH2
n
2n+1
CH
Cl
CH3
3
-
N
C H
n 2n+1
N
+
C H
n' 2n'+1
CH3
1
7, Empigen bac 50 with n = 14 (50
n = 12 (40 ) and n = 10 (10
)
18, Catigene 818 with n = 8 and n' = 10 (50
n= n' = 8 (25 ) and n = n' = 10 (25
)
)
)
In the present work, we describe the synthesis of ‘‘Gemini’’
fluorinated surfactants 1–16 (Fig. 1) and the evaluation of their
antimicrobial activity using measurement of minimal inhibitory
concentrations (MICs) in comparison with commercial references,
CH3
+
N
CH
-
3
-
O
Cl
Cl +
CH CH
C
H
CH (CH ) CH2
N
2
2
12 25
2
2 14
17–20, and with a hydrocarbon analogue of 3, noted 21 (Fig. 2).
19, CPC
20, Domiphen bromide
2
. Results and discussion
O
H
-
Br
+
N
N
N
+
Br
C H
8 17
2.1. Chemical synthesis
C H
-
N
H
8
17
O
From our previous works concerning the study of antimicrobial
2
1
properties of heterogeneous series of fluorinated ‘‘Gemini’’ QACs,
we observed that the most ‘critical’ molecular parameter for their
biological activity seems to be the nature of connector Q, the
Fig. 2. Structures of the hydrocarbon surfactants 17–21 used as standards.
fluorinated QACs with an amide connector Q ¼ NHC(O)CH
2
exhib-
iting the best bacteriological activities against various microor-
ganisms (Gram (þ) bacteria: Staphylococcus aureus ATCC 9144;
yeast: Candida albicans ATCC 2091 and moulds: Aspergillus niger
ATCC 6275) [18,19]. In order to confirm this hypothesis on P. aeru-
ginosa and to attempt to optimize the other structural parameters
of fluorinated ‘‘Gemini’’ surfactants, we decided on the one hand to
extend the series with the amide connector by varying the length of
azide to give 2-perfluorohexylethylazide under phase transfer
catalysis conditions. The reduction reaction using hydrazine hydrate
catalysed by Raney nickel gives the corresponding amine which
reacts with bromoacetic acid using dicyclohexylcarbodiimide (DCC)
as coupling reagent to give the expected building block 22. Then, the
0
0
reaction of building block 22 with an half-equivalent of N,N,N ,N -
tetramethyl- -alkanediamine leads to QACs 1–7. The same
reaction with an half-equivalent of N,N,N ,N -tetrametylcystamine,
spacer (CH
2
)
n
(Compounds 1–7 : Block 1 in Fig. 1) and its nature and
a,u
0
0
rigidity (Compounds 8–11: Block 2 in Fig. 1), and on the other hand
to modify largely the nature of the connector Q keeping the spacer
R as a constant parameter (Compounds 3 and 12–16: Block 3 in
Fig. 1).
0
0
trans-N,N,N ,N -tetramethyl-2-butene-1,4-diamine, 1,3-bis(dime-
thylamino)-2-propanol, and 1,4-diazabicyclo[2.2.2]octane, gives
QACs 8–11, respectively.
Schemes 1 and 2 show that QACs 1–12 are synthesized from
building blocks 22 and 23, while 13–16 are obtained from the
intermediates 24–27, respectively.
The building block 22 is obtained from the commercially avail-
able 2-perfluorohexyethyl iodide. This latter reacts with sodium
The commercially available 2-perfluorohexylethanethiol serves
as a starting material for synthesis of building block 23 leading to
QAC 12. First, the thiol is reacted under inert atmosphere to poly-
oxymethylene and triethylamine in catalytic amounts. Then,
phosphorus tribromide is used to convert the intermediate alcohol
+
Br
+ Br
C6F
13
N
N
C6F13
Q
R
Q
Q
R
Q
R
1
2
3
4
5
NHC(O)CH2
(CH )
8
9
NHC(O)CH CH CH2
2
2
S
S CH2 CH2
CH2
2 2
NHC(O)CH2
CH2
NHC(O)CH2
NHC(O)CH2
NHC(O)CH2
NHC(O)CH2
(CH )
2 3
CH2 CH(OH) CH2
1
0
1
NHC(O)CH2
NHC(O)CH2
(CH )
2 4
+
+
1
N
N
(CH )
2 6
BLOCK 2
1
2
S
CH2
No connector
C(O)NHCH CH
2
(CH2)4
(CH2)4
(CH2)8
(CH2)10
(CH2)12
1
1
3
4
(CH2)4
6
7
NHC(O)CH2
2
1
5
NHC(O)NHCH CH
(CH2)4
2
2
NHC(O)CH2
(CH2)4
16
NHC(O)OCH CH
2
2
BLOCK 1
BLOCK 3
Fig. 1. Structures of the ‘‘Gemini’’ fluorinated surfactants 1–16 synthesized.

L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
1617
C6F
13
SH
I
C6F
13
a
d
e
Br
+
Br
N+
O
O
C6F
13
13
N3
C6F
13
N
C6F13
S
OH
N
H
N
C6F
13
H
b
c
1
1
C6F
NH3
N
N
f
H
S
Br
C6F
13
N
C6F
13
Br
23
2
2
O
H C
CH3
CH3
3
H C
CH3
3
N
(CH2)n
N
N
R N
H C
CH3
H C
3
3
f
f
Br
Br
O
O
Br
Q
Br
+
+
C6F
+
+
13
N (CH ) N
2 n
C6F
13
C6F13
N
R
N
C6F13
Q
N
N
H
H
1
2
3
4
Q = NHC(O) n =2
5 Q = NHC(O) n =8
"
"
"
n =3
n =4
n =6
n =4
6
7
"
"
n =10
n =12
S
8
R =
R =
S
9 R =
1
0
12 Q = S
OH
Reagents and conditions: (a) NaN3, H O, ALIQUAT 336 (cat.), (b) 90°C; CH (CH ) OCH CH OCH CH OH,
2
3
2
3
2
2
2
2
NH NH .H O, Raney Ni (cat.), 35°C; (c) DCC, BrCH CO H, CH Cl , 20°C; (d) N , Et N, (CH O) , 0 to 20°C,
2
2
2
2
2
2
2
2
3
2
n
then H O, 0°C; (e) PBr , Et O, 12h, r.t;; (f) corresponding diamine, Et O, 35°C.
2
3
2
2
Scheme 1. Synthesis of Gemini fluorinated surfactants 1–12.
in bromide, affording 23. The reaction of 23 with an half-equivalent
of N,N,N ,N -tetramethyl butanediamine leads to 12.
the ‘‘Gemini’’ surfactant homologue of compound 3 according to
the synthetic pathway shown in Scheme 3.
0
0
As shown in Scheme 2, 2-perfluorohexylethyliodide is the
starting material to obtain intermediates 24–27. Then, 24–27 are
reacted with a half-equivalent of 1,4-dibromobutane to give 13–16,
respectively, with good yields. The two first steps for the synthesis
of 24 are the same that for 22. The third step is an Eschweiler and
Clarke methylation reaction of primary amines using acetic acid
and formaldehyde to afford compound 24 [20]. Concerning 25 and
2.2. Antimicrobial activity and surface activity
Antibacterial activity of QACs 1–21 was evaluated using
measurement of minimal inhibitory concentrations (MICs)
expressed in
mmol/L (see Section 4). These measurement tests were
realised on P. aeruginosa (CIP A22) Class II type Gram. Benzalko-
nium chloride (17), Catigene 818 (18), Cetyl pyridinium chloride
(CPC) (19), and Domiphen bromide (20), were taken as commercial
references. Results from the antibacterial evaluation of compounds
1–21 are summarized in Table 1 along with the critical micelle
concentrations (CMCs) measured using a conductivity technique
2
6, the two first steps are the same. 3-Perfluorohexylpropanoyl
chloride is synthesized by reacting 3-perfluorohexylpropanoic acid,
itself obtained from the reaction of carbon dioxide on the organo-
magnesium derivative of 2-perfluorohexylethyliodide, with phos-
phorus pentachloride under inert atmosphere. The obtained acyl
chloride can then react either with N,N-dimethylaminoethylamine
affording the hydrochloride derivative of compound 25 or with
trimethylsilyl azide to give first the perfluorohexylethylisocyanate
through a Curtius rearrangement which in turn reacts with dime-
thylethanolamine or N,N-dimethylaminoethylamine to afford the
carbamate derivative 26 and the urea derivative 27, respectively.
In order to observe the impact of the fluorinated chain on the
antimicrobial activity of such surfactants, we have synthesized 21,
(see Section 4) and expressed in mmol/L. The MIC values presented
are the average of 8–12 measurements for each compound, and the
uncertainties are in the order of 10%. Multiple range tests for bio-
logical activity by products (Duncan analysis: comparison of aver-
ages regarding their standard errors) allowed the ranking of the
molecules according to their efficacy for the microorganism tested.
Results from the multiple range tests of compounds from Blocks 1, 2
and 3 are summarized in Table 2.

1618
L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
a
N3
I
C6F
13
C6F
13
d
b
CO H
2
C6F
13
e
NH2
O
g
C6F
13
N=C=O
C6F
13
C6F13
Cl
f
c
h
H
O
CH3
CH3
CH3
CH3
N
X
CH3
C6F13
N
N
N
C6F
13
N
C6F13
CH3
O
H
2
4
25
26 X = O
27 X = NH
i
i
_
_
4
_
_
4
Br (CH ) Br
Br (CH ) Br
2
2
_
_
Br
Br
Br
Br
+
_
_
+
_
_
+
N
(CH2)4
N
C6F13
N+ (CH2)4 N
C6F13
C6F
13
C6F13
Q
Q
1
3
1
1
1
4 Q = C(O)NH
5 Q = NHC(O)NH
6 Q = NHC(O)O
Reagents and conditions: (a) NaN3, H O, ALIQUAT 336 (cat.), 90°C; (b) CH (CH ) OCH CH OCH CH OH,
2
3
2
3
2
2
2
2
NH NH .H O, Raney Ni (cat.), 35°C; (c) HCO H, HCOH, 0 to 90°C; (d) N , Mg, Et O, 35°C; dry CO , then
2
2
2
2
2
2
2
+
H O ; (e) N , PCl , 80°C; (f) N , (CH ) NCH CH NH , Et O, 20°C; (g) (CH ) SiN , N , r.t.; (h) N , (CH ) NCH
2
3
2
5
2
3
2
2
2
2
2
3
3
3
2
2
3
2
CH OH or (CH ) NCH CH NH , Et O, 20°C; (i) Br(CH ) Br, CHCl , 61°C.
2
3
2
2
2
2
2
2
4
3
Scheme 2. Synthesis of Gemini fluorinated surfactants 13–16.
In comparison with the reference compounds 17–20 for which
the best activity is exhibited by CPC (19) with an MIC of 17.60 mol/
L, all the fluorinated ‘‘Gemini’’ surfactants synthesized, 1–16, are
particularly active antibacterial agents against P. aeruginosa with
MIC ranging from 1.88 to 9.25
times larger than those of references. The comparison of activities
mol/L)
mmol/L that is an activity from 2 to 10
m
of compounds 3 (MIC ¼ 5.94
mmol/L) with 21 (MIC ¼ 51.16
m
demonstrates that this enhanced activity is due to the presence of
O
O
DCC, r.t.
CH Cl
Br
C H
NH2
+
8
17
Br
C H
N
8
17
HO
H
2
2
Et O
2
(Me) N(CH ) NMe
2
2 4
2
reflux
_
O
Br
H
+
N
+
N
N
C H
8
17
C H
_
N
8
17
H
Br
O
2
1
Scheme 3. Synthesis of the Gemini hydrocarbon surfactant 21.

L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
1619
Table 1
Antibacterial activity of QACs 1–21 against Pseudomonas aeruginosa according to
minimal inhibitory concentrations (MICs) expressed in mol/L and the corre-
sponding critical micelle concentrations (CMCs) of 1–16 expressed in mol/L
6
5
4
3
,5
5
m
m
,5
4
Biocide no. MIC (
mmol/L) CMC (
m
mol/L) Biocide no. MIC (
m
mol/L) CMC ( mol/L)
m
1
2
3
4
5
6
7
8
9
5.80
5.19
5.94
5.52
2.89
2.83
2.76
4.95
3.67
9.25
5.82
0.591
0.546
0.758
0.588
0.361
0.296
0.269
0.451
0.559
0.541
0.466
12
13
14
15
16
17
18
19
20
21
4.59
7.31
1.88
5.38
3.69
47.72
50.67
17.60
40.82
51.16
Insoluble
2.39
0.547
0.532
0.540
,5
3
2,5
2
0
2
4
6
8
10
12
n
1
0
Fig. 3. Variations of MIC for the series 1–7 as a function of the number n of methylene
groups in the spacer R.
11
the perfluorinated chain. This result could be considered as
surprising because although the introduction of a perfluorinated
chain in the structure of surfactants increases their surface activity,
the lipophobicity of surfactants is also increased that could
decrease the interaction with the external membrane of the Gram
The third structural parameter studied is the length of the
hydrocarbon chain as spacer R between the ammonium groups.
Fig. 3 shows the variation of MIC for the series 1–7 as a function of
the number n of methylene groups in the spacer.
Two plateaus are observed, the first for n ꢁ 6 and the second for
(
–) bacteria.
n ꢂ 8. This observation can be related to that of Zana and Colleagues
þ
þ
In our previous study, we found the connector Q to have a crit-
for hydrocarbon ‘‘Gemini’’ surfactants C
m
H
2mþ1N (CH
3
)
2
(CH
2
)
n
N
ꢀ
ical impact on the antimicrobial activity. In view of these previous
results, it appears that the most active molecules against various
germs have a connector amide, urea or carbamate while connector
ester and thioester give QACs with a poor antibacterial activity [18].
The present results are in total agreement with our previous
observation since, on one hand compound 3 with an amido
connector presented an enhanced activity against P. aeruginosa
(CH
3
)
2
C
m
H
2mþ1, 2Br . These authors found that at constant m, the
CMC of these surfactants changes with the value of n, a small
maximum of CMC being observed for n ¼ 5–6 that reflects the
contribution of a change of conformation of the surfactant at the
interface with increasing spacer chain length [22,23]. This implies
that a linear relationship must exist between the CMC and the MIC
at least for the compounds of the homogeneous series 1–7. Indeed,
a linear relationship is found to exist between the CMC of surfac-
tants 1–7 and their MIC (Fig. 4) that is in agreement with the results
obtained for the first time in 1955 from a homogeneous series of
quaternary ammonium chlorides [24] and then observed for
various series of antibacterial agents [25–28]. It is noteworthy that
no relationship can be established between the MIC and the CMC
values neither for the series with variable spacers 8–11 nor for the
series with variable connectors, 12–16. This can be explained by the
complexity and the great number of components of the antibac-
terial effect of quaternary ammonium salts in which the surfactant
effect at the surface of the bacterial cell represents only a small part
of the whole effect of the antibacterial agent. In the present case,
the variation of the length of alkyl spacers in series 1–7 modifies
only the interfacial effect of the surfactant reflected by the variation
of CMC while the other components of the antibacterial effect are
not modified as a whole. On the other hand, the variation of the
nature of spacers R in the series of compounds 8–11 or of the nature
of the connector Q in the series 12–16 modifies not only the
primary surfactant effect but also the other components of the
antibacterial effect that are not reflected by the CMC values.
(
MIC ¼ 5.94
m
mol/L) compared to compound 13 without amido
connector (MIC ¼ 7.31
m
mol/L) and on the other hand, molecules
with amide (14), urea (15) or carbamate (16) connector present
very high activities across the series. These results accounts for an
important role played by hydrogen bonds in the biological activity
against P. aeruginosa and are in good agreement with the models
established for the prediction of antibacterial activity of various
classes of antibacterial agents which state that hydrogen bonding is
one of the most important descriptors [21].
However, the ‘‘Gemini’’ surfactant 12 with a sulphide connector
shows also a high activity (MIC ¼ 4.59
mmol/L) although it is not
a hydrogen bond donor but only a weak hydrogen bond acceptor.
What is striking here is that this sulphide derivative shows an
important activity against P. aeruginosa while the thioester deriv-
ative (Q ¼ SC(O)) in which a sulphur lone pair is implied in
a conjugation with the carbonyl group and not available to be an
hydrogen bond acceptor presents no activity (MIC > 800
19].
The second structural parameter we studied here is the nature
mmol/L)
[
of the spacer R between the two ammonium groups. The compar-
ison of compound 3 with compounds 8–11 demonstrates the
modulating effect of this spacer on the antibacterial property. At
this stage, it is impossible to propose an explanation from the few
different spacers we studied in this work.
8
7
6
5
4
3
2
1
0
Table 2
Multiple range tests (DUNCAN analysis) for QACs of Blocks 1, 2 and 3 against Pseu-
domonas aeruginosa according to minimal inhibitory concentrations (MICs)
expressed in mmol/L
Block
Ranking (by compounds)
Block 1
6 ꢂ 5 ꢂ 7 > 1 ꢂ 2 ꢂ 4 > 3
2
.83 ꢂ 2.89 ꢂ 3.27 > 5.40 ꢂ 5.79 ꢂ 5.81 >7.07
Block 2
Block 3
8 > 9 ꢂ 11 >3 > 10
.67 > 5.04 ꢂ 5.82 > 7.07 > 9.81
14 ꢂ16 ꢂ 15 ꢂ12 ꢂ 3 ꢂ13
.54 ꢂ 3.68 ꢂ 5.38 ꢂ 5.52 ꢂ 7.07 ꢂ 7.31
0,2
0,3
0,4
0,5
0,6
0,7
0,8
3
CMC µM/L
2
Fig. 4. Linear relationships between the CMC and the MIC of surfactants 1–7.

1620
L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
3
. Conclusion
4.1.5. Synthesis of building block 25 (Scheme 2)
-Perfluorohexylpropanoic acid is obtained under nitrogen
3
In this work, we first show the large increase of antibacterial
activity of quaternary bisammonium salts against P. aeruginosa
when a perfluorinated carbon chain is introduced as tails of these
‘Gemini’’ surfactants. These fluorosurfactants exhibit an important
atmosphere from the reaction of dry carbon dioxide on the orga-
nomagnesium derivative prepared from 2-perfluorohexylethyliodide
and magnesium (step d). (e) 3-Perfluorohexylpropanoyl chloride is
synthesized in 90% yields by reacting 3-perfluorohexylpropanoic
acid with phosphorus pentachloride under inert atmosphere (step
‘
antibacterial activity as compared to commercial ammonium salt
preservatives and to a hydrocarbon homologue. Secondly, we
demonstrate that if the length of the alkyl connector between the
quaternized nitrogen allows to modulate the antibacterial effect of
an homogeneous series of Gemini surfactants by varying their
interfacial action with the bacterial cell surface, the variation of the
nature of the connector between the charged nitrogen atoms and
the fluorinated tails, particularly the presence of hydrogen bond
donor group, and of the nature of the spacer between the two
quaternized nitrogen atoms modifies more largely the antibacterial
effect. These results show the multiplicity of factors occurring in
the antibacterial effect of the ‘‘Gemini’’ ammonium surfactants.
ꢃ
e). Then (step f), N,N-dimethylaminoethylamine is added at 20 C
and under nitrogen atmosphere to the acyl chloride synthesized
previously affording the hydrochloride derivative of compound 25
in 73% yields.
4.1.6. Synthesis of building blocks 26 and 27 (Scheme 2)
In the step g, 2-perfluorohexylethyl isocyanate is prepared
ꢃ
under nitrogen atmosphere at 70 C in 80% yields through a Curtius
rearrangement of the derivative obtained at room temperature
from reaction of trimethylsilyl azide on the acyl chloride synthe-
sized previously. Then (step h), dimethylethanolamine or dime-
thylaminoethylamine is added under inert atmosphere to the
isocyanate to afford the carbamate derivative 26 in 80% yields and
the urea derivative 27 in 90% yields, respectively.
To obtain the ‘‘Gemini’’ surfactants 1–12, the building blocks 22 or
23 synthesized previously were reacted with an half-equivalent of the
corresponding diamine (step f in Scheme 1) while ‘‘Gemini’’ surfactants
4
4
4
. Experimental section
.1. Chemical synthesis
.1.1. General
13–16 are obtained from reaction between the corresponding
dibromoalkanes and the building blocks 24, 25, 26 or 27.
a-u
NMR data were collected on Br u¨ ker Advance 200 spectrometer.
Chemical shifts are reported in parts per million (ppm) relative to
tetramethylsilane (TMS), and spin multiplicities are given as s
4.1.7. General procedure for preparation of Gemini surfactants 1–12
(Scheme 1)
1
(
singlet), d (doublet), t (triplet), q (quartet), and m (multiplet). H
NMR results are given only for final compounds. The mass spectra
of ‘‘Gemini’’ surfactants were recorded on a Finnigan Mat LCQ
Classic spectrometer fitted with an Electrospray source API1 ESI-MS
mass spectrometer in the positive mode. In this case, only the
cationic part of the molecule was detected. As these cations are
In the step f, the building block 22 or 23 is dissolved in diethyl ether
and a half-equivalent of the appropriate diamine is added to the
solution. Themixtureisstirred atrefluxfor 12 h. The mixture isfiltrated
and the residue is washed with diethyl ether. This procedure affords the
surfactants 1–12 with a yields ranging from 80 to 90%.
bischarged (excepted for 21), the observation of the sole
þþ
[
M ꢀ 2Br] /2 peak (where M is the mass of the cationic part of the
4.1.8. General procedure for preparation of Gemini surfactants
13–16 (Scheme 2)
In the step i, the building block 25, 26 or 27 is dissolved in
chloroform (isopropanol for building block 24), and a half-equiva-
molecule) is in agreement with the structure. Elemental analysis
was performed by analytical laboratory at the Department of
Chemistry of University of Nice-Sophia Antipolis on a Thermo-
Electron Eager 300 CHNSO apparatus.
lent of a-u dibromobutane is added to the solution. The mixture is
As shown in Schemes 1 and 2, the Gemini surfactants 1–12 and
3–16 were synthesized from building blocks 22 and 23, and 24–27,
respectively. The synthesis of the hydrocarbon surfactant 21 used as
stirred at reflux for 24 h (72 h for 24). The solvent is evaporated, and
the residue is washed with diethyl ether. This procedure affords the
surfactants 13–16 with a yields ranging from 60 to 80%.
1
reference is resumed in Scheme 3.
4.1.9. Synthesis of the Gemini hydrocarbon surfactant
4.1.2. Synthesis of building blocks 22 (Scheme 1)
21 (Scheme 3)
In the first step (a), 2-perfluorohexylethylazide is obtained by
Octylamine dissolved in methylene chloride is added dropwise
to a solution of DCC and bromoacetic acid in methylene chloride
and stirred 3 h at room temperature. After filtration and purifica-
tion by column chromatography the octylbromoacetamide is
obtained with 91% yields. Then, this precursor is added to N,N,N ,N -
tetramethyl-1,4-butanediamine dissolved in refluxing diethyl ether
to provide the Gemini hydrocarbon surfactant 21 in 93% yields.
reacting sodium azide on 2-perfluorohexylethyliodide in phase
transfer catalysis conditions. (b) Then, the 2-perfluorohexylethyl-
amine is obtained from the reduction of 2-perfluorohexylethylazide
by hydrazine hydrate catalysed by Raney nickel (step b). 2-Per-
fluorohexylethylamine is reacted with DCC and bromoacetic acid to
afford compound 22 in 88% yields by coupling reaction (step c).
0
0
4.1.3. Synthesis of building block 23 (Scheme 1)
4.2. Characterizations of final compounds
In the step d, 2-perfluorohexylethanethiol is reacted under inert
atmosphere with polyoxymethylene and triethylamine in catalytic
amount. Then (step e), phosphorus tribromide is used to convert the
alcohol function in bromide, affording the compound 23 in 75% yields.
Characterization of compounds 1–4 and 21 is given elsewhere
[18].
4
.2.1. Gemini surfactant 5
þþ
1
4.1.4. Synthesis of building block 24 (Scheme 2)
m/z 504.3 (M ꢀ 2Br) /2; H NMR (methanol-d
4
) d 1.44 (8H, m,
The steps (a) and (b) are the same that those presented previ-
[N–CH
2
CH
2
CH
2
CH
2
]
2
), 1.85 (4H, m, [N–CH
13–CH
2
CH
2
]
2
), 2.52 (4H, tt,
3
3
þ
ously. Then (step c), 2-perfluorohexylethylamine is added to
a solution of acetic acid and formaldehyde according to the
Eschweiler and Clarke methylation reaction of amines to afford
compound 24 in 85% yields.
J
HH ¼ 6.9 Hz, JHF ¼ 19.2 Hz, [C
6
F
2
]
2
), 3.32 (12H, s, [CH
3
–N –
þ
CH
[C(O)–CH
3
]
2
), 3.61 (8H, m, [CH
2
–NH–C(O)]
2
þ [ N–CH
2
CH
2
]
2
), 4.17 (4H, s,
: C 32.89, H
þ
2
–N ]
2
). Anal. Calcd for C32
H42Br
2
F
26
N
4
O
2
3.62, N 4.79; found: C 32.83, H 3.64, N 4.75.

L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
1621
þ
), 3.47 (4H, t, 3
4
.2.2. Gemini surfactant 6
(12H, s, [CH
CH
[NH–CH
3
–N –CH
3
]
2
J
HH ¼ 6.6 Hz, [NH–CH
2
–
þþ
1
þ
3
m/z 518.2 (M ꢀ 2Br) /2; H NMR (methanol-d
4
)
d
1.48 (12H, m,
2
]
2
), 3.58 (4H, m, [N –CH ), 3.70 (4H, t,
–CH ). Anal. Calcd for C30H38Br F N O : C 31.60, H 3.36,
N 4.91; found: C 31.50, H 3.38, N 4.89.
2
–CH
2
]
2
J
HH ¼ 6.6 Hz,
[
N–CH
J
2
CH
2
CH
2
CH
2
CH
2
]
2
), 1.90 (4H, m, [N–CH
13–CH
–NH–C(O)]
þ [ N–CH
2
CH
2
]
2
), 2.57 (4H, tt,
2
2
]
2
2 26 4 2
3
3
þ
HH ¼ 7 Hz,
J
HF ¼ 19.3 Hz, [C
6
F
2
]
2
), 3.41 (12H, s, [CH
3
–N –
þ
CH
3
]
2
), 3.66 (8H, m, [CH
2
2
2 2 2
CH ] ), 4.23 (4H, s,
þ
[C(O)–CH
2
–N ]
2
). Anal. Calcd for C34
H
46Br
2
F
26
N
4
O
2
: C 34.68, H
4.2.11. Gemini surfactant 15
þþ
1
3
.94, N 4.76; found: C 34.65, H 3.93, N 4.74.
m/z 505.2 (M ꢀ 2Br) /2; H NMR (methanol-d
4
) d 2.01 (4H, m,
þ
3
3
[
N –CH
CH ), 3.29 (12H, s, [CH
NH(CO)NH] –CH
þ [NH–CH
Anal. Calcd for C30
30.96, H 3.33, N 7.22.
2
–CH
2
]
2
), 2.49 (4H, tt,
J
HH ¼ 7.3 Hz,
J
HF ¼ 19.0 Hz, [C
6
F
13
–
–
þ
4.2.3. Gemini surfactant 7
2
]
2
3
–N –CH
3
]
2
), 3.57 (16H, m, [CH
2
þþ
1
þ
m/z 532.3 (M ꢀ 2Br) /2; H NMR (methanol-d
4
)
d
1.38 (16H, m,
2
2
]
2 2
2
þ [NH–CH –CH
2
]
2
2 2 2
þ [N –CH –CH ] ).
þ
[
(
[
N –CH
4H, tt,
CH
2
CH
2
CH
2
CH
2
CH
2
CH
2
]
2
), 1.82 (4H, m, [N–CH
2
CH
2
]
2
), 2.50
2 26 6 2
H40Br F N O : C 30.79, H 3.44, N 7.18; found: C
3
3
JHH ¼ 6.7 Hz,
J
HF ¼ 19.3 Hz, [C
6
F
13–CH
–NH–C(O)]
2
]
2
), 3.32 (12H, s,
þ
þ
3
–N –CH
3
]
2
), 3.59 (8H, m, [CH
2
2
þ [ N–CH
2
CH
2
]
2
),
þ
4
3
.18 (4H, s, [C(O)–CH
2
–N ]
2
). Anal. Calcd for C36
H50Br
2
26
F N
4
O
2
: C
4.2.12. Gemini surfactant 16
þþ
1
5.31, H 4.11, N 4.57; found: C 34.51, H 4.11, N 4.56.
m/z 506.1 (M ꢀ 2Br) /2; H NMR (methanol-d
4
) d 1.95 (4H, m,
þ
3
3
[
N –CH
CH
NH(CO)O–]
m, [O–CH
2
–CH
2
]
2
), 2.47 (4H, tt,
J
HH ¼ 7.3 Hz,
J
HF ¼ 19.0 Hz, [C
6
F
13
–
–
þ
3
4
.2.4. Gemini surfactant 8
2
]
2
), 3.22 (12H, s, [CH
), 3.55 (4H, t,
–CH ), 4.55 [O–CH
: C 30.73, H 3.27, N 4.78; found: C 31.01, H 3.21, N
3
–N –CH
HH ¼ 8.0 Hz [N –CH
3
]
2
), 3.46 (4H, t, J ¼ 7.3 Hz [CH
2
þþ
1
3
þ
m/z 508.2 (M ꢀ 2Br) /2; H NMR (methanol-d
4
)
d
2.52 (4H, tt,
2
J
2 2 2
–CH ] ), 3.78 (4H,
3JHH ¼ 7.0 Hz, JHF ¼ 19.6 Hz, [C
3
F
13–CH
]
), 3.39 (16H, m, [CH
–N –
þ
]
–CH
]
.
Anal. Calcd for
6
2
2
3
2
2
2
2
2
2
3
CH
4
N ]
3
]
2
þ [CH
2
–CH
2
–S]
2
), 3.61 (4H, t, JHH ¼ 7.0 Hz, [CH
2
–NH–C(O)]
2
),
–
C
30
H38Br F
2 26
N
4
O
4
3
.02 (4H, t,
J
HH ¼ 7.5 Hz, [CH
2
–CH
2
–S]
2
), 4.30 (4H, s, [C(O)–CH
: C 28.59, H 2.91, N 4.76, S
2
4.79.
þ
2
). Anal. Calcd for C28
H
34Br
2
F
26
N
4
S
2
O
2
5
.45; found: C 28.64, H 2.92, N 4.67, S 5.41.
4.2.13. Gemini surfactant 21
þ
1
m/z 563.4, 565.3 (M-Br) ; H NMR (methanol-d
4
)
d
1.00 (6H, t,
4
.2.5. Gemini surfactant 9
[CH
[CH
3
–CH
2
]
2
), 1.40 (20H, m, [CH
3
(CH
CH
2 5 2
) ] ), 1.62 (4H, m,
þþ
1
þ
m/z 475.2 (M ꢀ 2Br) /2 H NMR (methanol-d
4
)
d
2.51 (4H, tt,
2
CH NH]
2
2
), 201 (4H, m, [N –CH
2
2
]
2
), 3.31 (4H, t, [CH
2
–
–
3JHH ¼ 7.7 Hz, JHF ¼ 19.5 Hz, [C
3
F
13–CH
]
), 3.37 (12H, s, [CH
–NH–C(O)]
–CH ¼ CH] ), 6.51 (2H, m, [CH
: C 30.29, H 2.90, N
–N –
þ
NHC(O)]
CH
), 3.42 (12H, s, [CH
–N –CH
þ
]
), 3.89 (4H, m, [ N–CH
þ
6
2
2
3
2
3
3
2
2
3
þ
CH
3
]
2
), 3.58 (4H, t,
J
HH ¼ 7.7 Hz, [CH
2
2
), 4.20 (4H, s,
2
]
2
), 4;09 (4H, s, [C(O)–CH
2
–N ]
2
). Anal. Calcd for
þ
[C(O)–CH
2
–N ]
2
), 4.45 (4H, m, [CH
2
2
2
–
28 2 4 2
C H60Br N O : C 52.17, H 9.38, N 8.69; found: C 52.28, H 9.42, N
8.65.
CH ¼ CH]
2
). Anal. Calcd for C28
H
2
32Br F
26
N
4
O
2
5
.05; found: C 30.34, H 2.84, N 5.11.
4
.3. Antimicrobial evaluation
4
.2.6. Gemini surfactant 10
þþ
1
m/z 477.2 (M ꢀ 2Br) /2; H NMR (methanol-d
4
)
d
2.47 (4H, tt,
), 3.46 and 3.41 (6H þ 6H,
–NH–C(O)] ),
HH ¼ 15.2 Hz,
–CH(OH)]). Anal. Calcd for
: C 29.10, H 2.89, N 5.03; found: C 29.29, H 2.80, N
Antibacterial activity of QACs was evaluated using measurement
3JHH ¼ 7.3 Hz, JHF ¼ 19.5 Hz, [C
3
F
13–CH
]
6
2
2
of MIC [29] expressed in mmol/L. These measurement tests were
realised on P. aeruginosa (CIP A22) Class II type Gram. Bacteria were
incubated on TSA (Trypcase Soy Agar) slants for 18 h at 30 C. MIC
tests are run with the third generation of bacteria; samples are
taken during the exponential phase of bacterial growth. This third
generation of bacterial strains obtained is incubated in an M9
minimal phosphate medium suspension. Benzalkonium chloride
þ
3
2
3
s, [CH
3
–N –CH
3
]
2
), 3.55 (4H, t,
–CH(OH)]
J
HH ¼ 7.3 Hz, [CH
2
2
2
.82 (4H, m, [CH
2
2
), 4.36 (2H þ 2H, 2d,
J
ꢃ
þ
[C(O)–CH
2
–N ]
2
), 5.05 (1H, m, [CH
2
C
5
27
H32Br
2
F
26
N
4
O
3
.09.
4.2.7. Gemini surfactant 11
Ò
Ò
(Empigen Bac 50 purchased from Albright & Wilson ), Catigene
þþ
1
m/z 460.1 (M ꢀ 2Br) /2. H NMR (methanol-d
4
) d 2.61 (4H, tt,
Ò
8
18 (purchased from Stepan ), Domiphen bromide and Cetyl pyr-
3JHH ¼ 7.1 Hz, JHF ¼ 19.1 Hz, [C
3
F13–CH
]
), 3.71 (4H, t, JHH ¼ 7.1 Hz,
3
6
2
2
Ò
idinium Chloride (CPC) (both purchased from ACIMA ) were taken
as commercial references. Initial solutions of antimicrobial were
prepared at a 500 ppm concentration of active ingredient in
distilled water. Bacterial inocula were prepared at 5.10 –5.10 cfu/
mL (cfu: colony forming unit). Adjustment was made by optical
density measurement (absorbance 0.05 at
6
for our experiments realised automatically dilutions of the tested
anti-microbial agent solution in an inoculated 96 wells micro-
titration plate. After 24 h of incubation at 30 C, growth (or its lack)
þ
þ
[
[
CH
C(O)–CH
2
–NH–C(O)]
2
), 4.47 (12H, s, [N (CH
). Anal. Calcd for C26
2
–CH
2
)
3
N ]), 4.55 (4H, s,
þ
2
–N ]
2
H26Br
2
F
26
N
4
O
3
: C 28.91, H 2.43,
N 5.19; found: C 28.95, H 2.45, N 5.16.
6
7
4.2.8. Gemini surfactant 12
a
wavelength of
þþ
1
m/z 465.1 (M ꢀ 2Br) /2. H NMR (methanol-d
4
)
d
1.93 (4H, m,
Ò
Ò
60 nm). The Biomek 1000 (Beckman-Coulter ) apparatus used
þ
3
3
[N –CH
2
CH
2
]
2
), 2.68 (4H, tt,
JHH ¼ 7.3 Hz,
J
HF ¼ 19.0 Hz, [C
6 13
F –
þ
þ
CH
CH
C
2
]
2
), 3.19 (16H, m, [CH
3
–N –CH
3
]
2
þ [CH
2
–S]
2
), 3.61 (4H, m, [ N–
þ
2
CH
2
]
2
), 4.88 (4H, s, [S–CH
2
–N ]
2
). Anal. Calcd for
ꢃ
26
H
32Br
2
26
F N
2
2
S : C 28.64, H 2.96, N 2.57, S 5.88; found: C 28.64, H
was determined visually. The lowest concentration at which no
growth was observed (turbidity) was taken as the MIC. Each
measurement is repeated at least 8 times to provide MIC values that
are a mean value of a sufficient number of experiments.
2
.96, N 2.44, S 5.77.
4.2.9. Gemini surfactant 13
þþ
1
m/z 419.2: [M ꢀ 2Br] /2; H NMR (methanol-d
4
) d 2.02 (4H, m,
þ
3
3
[
N –CH
CH ), 3.31 (12H, s, [CH
CH
C 28.88, H 2.83, N 2.81; found: C 28.95, H 2.85, N 2.69.
2
-CH
2
]
2
), 2.99 (4H, tt,
J
HH ¼ 8.0 Hz,
J
HF ¼ 19.0 Hz, [C
6
F
13
–
–
:
þ
þ
2
]
2
3
þ
–N –CH
3
]
2
), 3.62 (8H, m, [N –CH
2
4.4. Statistics
2
]
2
þ [C
6
F
13–CH
2
–CH
2
–N ]
2
). Anal. Calcd for C24
H
28Br
2
F
26
N
2
A full statistic treatment of the antimicrobial activity was carried
on. A full analysis of variance (ANOVA) was used to determine the
main effects on biological activity. Compounds from Blocks 1, 2 and
3 (Fig. 1) were tested and multiple range tests were done allowing
the ranking of the molecules with respect to their activity. Such
4.2.10. Gemini surfactant 14
þþ
1
m/z 490.1 (M ꢀ 2Br) /2; H NMR (methanol-d
4
)
d
1.96 (4H, m,
þ
[N –CH
2
–CH
2
]
2
), 2.55 (8H, m, [CH
2
–(CO)]
2
þ [C
6
F
13–CH ), 3.21
2 2
]

1622
L. Massi et al. / European Journal of Medicinal Chemistry 44 (2009) 1615–1622
tests allow highlighting statistically significant differences in MIC
values.
[11] J.A.A. Hoogkamp-Korstanje, J. Roelofs-Willemse, Int. J. Antimicrob. Agents 21
2003) 547–556.
(
[
12] F. Gunseren, L. Mamikoglu, S. Ozturk, M. Yucesoy, K. Biberoglu, N. Yulug,
M. Doganay, B. Sumerkan, S. Kocagoz, S. Unal, S. Cetin, S. Calangu, I. Koksal,
H. Leblebicioglu, M. Gunaydin, J. Antimicrob. Chemother. 43 (1999) 373–378.
13] M.C. Miranda, F. Perez, T. Zuluaga, M.R. Olivera, A. Correa, S.L. Reyes,
M.V. Villegas, Biomedica 26 (2006) 424–433.
4.5. CMC measurements
[
[
The CMCs were measured with a Tacussel CDM 210 con-
14] F.M. Menger, J.S. Keiper, Angew. Chem. Int. Ed. 39 (2000) 1907–1920.
ꢃ
ductimeter. Samples were thermostated at 25 C. Conductivity cells
[15] G. Ronsin, A.J. Kirby, S. Rittenhouse, G. Woodnutt, P. Camilleri, J. Chem. Soc.,
Perkin Trans. 2 (2002) 1302–1306.
ꢃ
ꢃ
were calibrated using 0.1 M potassium chloride solution at 25 C.
[
[
16] F. Zhou, T. Maeda, H. Nagamune, H. Kourai, Biocontrol Sci. 9 (2004) 61–67.
17] R. Beumer, S. Bloombfield, M. Exner, G.M. Fara, J.K. Nath, E. Scott, International
Scientific Forum on Home Hygiene, London, 2000.
All the samples were measured at a constant temperature of 25 C.
The CMCs were determined from the breakpoint of each con-
ductimetry (function of the concentration) curve obtained.
[
[
[
18] L. Massi, F. Guittard, S. Geribaldi, R. Levy, Y. Duccini, Int. J. Antimicrob. Agents
21 (2003) 20–26.
19] L. Massi, F. Guittard, R. Levy, Y. Duccini, S. Geribaldi, Eur. J. Med. Chem. 38
(2003) 519–523.
20] E. Farkas, C.J. Sunman, J. Org. Chem. 50 (1985) 1110–1112.
References
[
1] J.M. Ascenzi, Handbook of Disinfectants and Antiseptics, Marcel Dekker Inc.,
[21] M.T.D. Cronin, A.O. Aptula, J.C. Dearden, J.C. Duffy, T.I. Ntezeva, H. Patel,
P.H. Rowe, T. Wayne Schultz, A.P. Worth, K. Voutzoulidis, G. Sch u¨ rmann,
J. Chem. Inf. Comput. Sci. 42 (2002) 869–878.
[22] R. Zana, M. Benrraou, R. Rueff, Langmuir 7 (1991) 1072–1075.
[23] E. Alami, G. Beinert, P. Marie, R. Zana, Langmuir 9 (1993) 1465–1467.
[24] J.A. Cella, L.A. Harriman, D.N. Eggenberger, H.J. Harwood, J. Am. Chem. Soc. 77
(1955) 4264–4266.
[25] T.R.R. Kurup, L.S.C. Wan, L.W. Chan, Pharm. Acta Helv. 66 (1991) 274–280.
[26] F. Devinsky, I. Lacko, D. Mlynarcik, V. Racansky, L. Krasnec, L. Tenside, Deter-
gents 22 (1985) 10–15.
[27] J. Pernak, A. Skrzypczak, M.B. Bogacki, Chem. Pharm. Bull. 43 (1995)
2019–2020.
New York, 1996.
[
[
2] L.B. Rice, Clin. Infect. Dis. 43 (Suppl. 2) (2006) S100–S105.
3] T.F. Mah, B. Pitts, B. Pellock, G.C. Walker, P.S. Stewart, G.A. O’Toole, Nature 246
(
2003) 306–310.
[
[
[
4] V.I. Brown, E.J.L. Lowbury, J. Clin. Pathol. 18 (1965) 752–756.
5] J. Pernak, J. Rogoza, I. Mirska, Eur. J. Med. Chem. 36 (2001) 313–320.
6] M. Thuong, K. Arvaniti, R. Ruimy, P. de la Salmoni e` re, A. Scanvic-Hameg,
J.C. Lucet, B. Regnier, J. Hosp. Infect. 53 (2003) 274–282.
[
[
7] H. Giamarellou, Int. J. Antimicrob. Agents 16 (2000) 103–106.
8] P.-R. Hsueh, W.-H. Chen, K.-T. Luh, Int. J. Antimicrob. Agents 26 (2005)
463–472.
[
9] E. Kucukates, Jpn. J. Infect. Dis. 58 (2005) 228–231.
[28] J. Pernak, A. Skrzypczak, Eur. J. Med. Chem. 31 (1996) 901–903.
[29] M.K. Bruch, in: S.S. Block (Ed.), Disinfection, Sterilization, Preservation, fourth
ed. Lea & Febiger, Philadelphia, 1991, pp. 1028–1046.
[
10] M. Arvanitidou, A. Vayona, N. Spanakis, A. Tsakris, J. Appl. Microbiol. 95 (2003)
180–185.