New cationic vesicles prepared with double chain surfactants from arginine: Role of the hydrophobic group on the antimicrobial activity and cytotoxicity

Article abstract of DOI:10.1016/j.colsurfb.2016.01.020

Cationic double chain surfactants have attracted much interest because they can give rise to cationic vesicles that can be used in biomedical applications. Using a simple and economical synthetic approach, we have synthesized four double-chain surfactants with different alkyl chain lengths (LANHC. _x). The critical aggregation concentration of the double chain surfactants is at least one order of magnitude lower than the CMC of their corresponding single-chain LAM and the solutions prepared with the LANHC. _x contain stable cationic vesicles. Encouragingly, these new arginine derivatives show very low haemolytic activity and weaker cytotoxic effects than conventional dialkyl dimethyl ammonium surfactants. In addition, the surfactant with the shortest alkyl chain exhibits good antimicrobial activity against Gram-positive bacteria. The results show that a rational design applied to cationic double chain surfactants might serve as a promising strategy for the development of safe cationic vesicular systems.

Full text of DOI:10.1016/j.colsurfb.2016.01.020

Accepted Manuscript
Title: New cationic vesicles prepared with double chain
surfactants from arginine: Role of the hydrophobic group on
the antimicrobial activity and cytotoxicity
Author: A. Pinazo V. Petrizelli M. Bustelo R. Pons M.P.
Vinardell M. Mitjans A. Manresa L. Perez
PII:
S0927-7765(16)30020-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.colsurfb.2016.01.020
COLSUB 7597
To appear in:
Colloids and Surfaces B: Biointerfaces
Received date:
Revised date:
Accepted date:
23-9-2015
11-1-2016
12-1-2016
Please cite this article as: A.Pinazo, V.Petrizelli, M.Bustelo, R.Pons, M.P.Vinardell,
M.Mitjans, A.Manresa, L.Perez, New cationic vesicles prepared with double
chain surfactants from arginine: Role of the hydrophobic group on the
antimicrobial activity and cytotoxicity, Colloids and Surfaces B: Biointerfaces
http://dx.doi.org/10.1016/j.colsurfb.2016.01.020
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

NEW CATIONIC VESICLES PREPARED WITH DOUBLE CHAIN
SURFACTANTS FROM ARGININE: ROLE OF THE HYDROPHOBIC GROUP
ON THE ANTIMICROBIAL ACTIVITY AND CYTOTOXICITY
1
4
1
1
2
2
A. Pinazo , V. Petrizelli , M. Bustelo , R. Pons , M.P. Vinardell , M. Mitjans , A.
3
1*
Manresa , L. Perez
1
Department of Chemical and Surfactant Technology, IQAC, CSIC, C/JordiGirona 18-26,
8034, Barcelona, Spain
0
2
Departament of Physiology, Facultat de Farmàcia, Universitat de Barcelona, Av. Joan XIII
s/n, 08028, Barcelona, Spain
Departament of Microbiology, Facultat de Farmàcia, Universitat de Barcelona, Av. Joan
3
XIII s/n, 08028, Barcelona, Spain
While in Department of Chemical and Surfactant Technology, IQAC, CSIC
4
*
Corresponding author:
Lourdes Pérez Muñoz
C/JordiGirona 18-26, 08034, Barcelona, Spain
Tf : (+34) 934006164
Fax: (+33) 34 93 2045904
E-mail: lourdes.perez@cid.csic.es
Statistical Summary
Number of words: 5929
Number of tables: 4
Number of figures: 4
Table of contents graphic
VERY LOW HAEMOLYTIC ACTIVITY
Arginine
Self assembly
CATIONIC VESICLES
+
Biological activity
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
ANTIMICROBIAL PROPERTIES
+
+
+
+
+
+ +
+
+
+
+
++

MODERATE CYTOTOXICITY
ABSTRACT
Cationic double chain surfactants have attracted much interest because they can give
rise to cationic vesicles that can be used in biomedical applications. Using a simple and
economical synthetic approach, we have synthesized four double-chain surfactants with
different alkyl chain lengths (LANHC ). The critical aggregation concentration of the
x
double chain surfactants is at least one order of magnitude lower than the CMC of their
corresponding single-chain LAM and the solutions prepared with the LANHC contain
x
stable cationic vesicles. Encouragingly, these new arginine derivatives show very low
haemolytic activity and weaker cytotoxic effects than conventional dialkyl dimethyl
ammonium surfactants. In addition, the surfactant with the shortest alkyl chain exhibits
good antimicrobial activity against Gram-positive bacteria. The results show that a
rational design applied to cationic double chain surfactants might serve as a promising
strategy for the development of safe cationic vesicular systems.
Keywords : cationic vesicles, arginine-based surfactants, antimicrobial activity, cytotoxicity
INTRODUCTION
Development of dialkyl cationic surfactants has attracted considerable interest because
of their use in biomedical applications [1]. Due to their molecular geometry, these types

of amphiphiles give rise to niosomes with a cationic charge that can be used as drug
delivery systems or gene therapy [2]. These therapeutic applications require the use of
stable surfactants under sterilization conditions that present low haemolytic or cytotoxic
activity.
Several cationic lipids such as DOTMA (1,2-di-O-octadecenyl-3-
trimethylammonium propane) or DOTAP (1,2-dioleoyl-3-trimethylammonium-propane)
have been used to prepare cationic niosomes for pharmaceutical applications [3], but the
production of such lipids on a large scale is very expensive. Alternatively, dialkyl
dimethyl ammonium bromide surfactants can be prepared at reasonable cost and these
surfactants also form cationic niosomes and show antimicrobial activity [4]. However,
their use is seriously questioned from an environmental point of view as quaternary
ammonium surfactants are toxic to aquatic organisms and not biodegradable [5-6-7].
In general, quaternary ammonium surfactants are more toxic than surfactants with the
cationic charge delocalized in a heterocyclic ring or guanidine group [8]. Over the last
years, our research group has studied N-acyl arginine derivative surfactants that consist
of one alkyl chain linked to the α-amino group of the arginine methyl ester. These
surfactants present physicochemical properties similar to those of classic cationic
surfactants, but they show high biodegradability, low toxicity against aquatic
microorganisms and moderate toxicity against human cell lines. Additionally, these
compounds can be easily prepared from renewable raw materials [9].
In this work, we have prepared new double-chain cationic surfactants from arginine
α
using the single-chain N -lauroyl-arginine methyl ester (LAM) as starting material
(Figure 1). The new compounds consist of one arginine as the cationic polar head and
two alkyl chains linked to the arginine through amide bonds (Figure 1). For the newly
prepared surfactants, we have studied the micellization process, their ability to form
cationic niosomes in water solutions as well as the size distribution and zeta potential of

the cationic vesicles. Moreover, their antimicrobial activity and cytotoxicity have been
determined in order to explore their potential applications. This work may represent a
step forward in the development of non-toxic and stable colloidal systems to be used in
pharmaceutical applications.
MATERIALS AND METHODS
This section briefly describes the methods used in this work, and more details are
reported as supplementary information.
The progress of the reactions as well as the purity of the final compounds was
monitored by HPLC, Merck-Hitachi D-2500 model using a UV-VIS detector L-4250 at
1
13
2
15 nm. The structures of the pure compounds were examined by H and C nuclear
1
magnetic resonance analyses, recorded on a Varian spectrometer at 499.803( H) and
1
3
1
25.233 ( C) MHz, respectively. Chemical shifts (δ) are reported in parts per million
(ppm) downfield from tetramethylsilane (TMS). Mass spectroscopy (MS) spectra with
fast atom bombardment (FAB) or electrospray techniques were carried out with a VG-
QUATTRO from Fisons Instruments. (S.I.2 and S.I.3)
Conductivity
Conductivity was measured using an Orion Cond. Cell 011010A with platinized
platinum electrodes in conjunction with a Thermo Orion 550A with a cell constant of
-1
0
.998 cm . Samples for conductivity measurements were prepared by weight in
Millipore ultra-pure water.( S.I.3.)
Surface tension
The surface tension was measured using a home-made pendant drop surfactometer. A
surfactant solution drop is created at the end of a straight cut Teflon tube with a 0.8 mm
internal diameter and a 1.58 mm external diameter The image of the drop is taken using
a web cam. The image is corrected for spherical aberration. The droplet contour is taken

at the point of maximum intensity slope. This contour is fitted to the Laplace –Young
equation [10] using a Golden Section Search Algorithm that has the centre coordinates,
an angular correction for the vertical alignment, the radius of the droplet and the
interfacial tension as parameters.
Vesicles preparation
Surfactants, adequately weighed, were dissolved in methanol. The organic solvent was
vacuum dried and the resulting lipid film was hydrated under sonication (30 minutes at
5
0 ºC) with deionized water (Table 1).
TABLE 1
Size distribution and ζ -potential
The ζ-potential of the formulations was measured by laser Doppler electrophoresis
using the Zetasizer 2000 (Malvern Instruments Ltd., Malvern, U.K.), at 25°C. Analyses
were done in triplicate.
Dynamic Light Scattering (DLS) was used to measure the size distribution of all
formulations. The DLS unit was a Malvern Zeta Nanosizer, working at 632.8 nm and
2
5ºC. The scattering intensity was measured at a 173º angle to the incident beam [11].
At least 10 runs were performed for each sample. The measured correlation was used to
calculate the size distribution. The size distribution profile was determined with the
CONTIN algorithm [12] which allows working with monomodal as well as multimodal
distributions.
Antimicrobial activity
a) MIC determinations
Antimicrobial activity was determined “in vitro” on the basis of the minimum inhibitory
concentration (MIC) [13], defined as the lowest antimicrobial agent concentration that
inhibits the development of visible growth after incubation at 37 ºC for 24 hours (S.I.5).

Cytotoxicity evaluation
Cell line cultures. 3T3 (murine Swiss albino fibroblasts), HaCaT (spontaneously
immortalized human keratinocytes) and human epithelial cervical cancer HeLa cell lines
(Eucellbank, Barcelona).
Chemical exposure. Stock solutions of surfactants were prepared in DMSO at a
maximum concentration of 10 mg/mL. Serial dilutions were subsequently prepared in
DMEM (Dulbecco's Modified Eagle's Medium) supplemented with 5% FBS (phosphate
buffer saline), 2 mM L-glutamine, and 1% antibiotic solution (final concentration of
DMSO in culture medium░<░1%). After removal of the medium culture, 3T3, HaCaT
or HeLa cells were exposed to the surfactants in multi-well plates. Controls containing
only culture medium were included in each plate, and they were independent for each of
the different surfactants tested. Plates were incubated at 37 ºC, 5% CO for 24 hours.
2
The cytotoxicity of surfactants against these three cell lines was determined using a
tetrazolium-based assay (MTT) [14] and the NRU assay. (S.I.6)
Haemolysis assay. We followed an adaptation of the method described by Pape et al.
[15]. ( S.I.6)
Synthesis
Synthesis of LANHC . 0.0035 moles of LAM, synthesized as reported in the literature
n
[16], were weighed into a round flask. Then 0.014 moles of the corresponding fatty
amine were added. The mixture was heated to the melting point of the amine and was
stirred at this temperature for three hours. After the completion of the reaction, the
mixture was cooled at room temperature and several crystallizations in
methanol/acetonitrile were carried out to remove the excess of fatty amine. The
1
13
identification of the products was performed by HPLC, HNMR and CNMR, Mass
Spectroscopy and Elemental analysis (Supporting Information)

RESULTS AND DISCUSSION
Cationic vesicles have great potential in new biomedical applications. They can be used
as antimicrobial systems, in gene delivery and as vesicles for drugs. The aim of this
work is to design and synthesise new biocompatible surfactants that give rise to
nontoxic cationic vesicles in the aqueous medium.
Synthesis
Surfactants based on renewable raw materials are characterized by their low negative
impact on the environment. In this work, we have prepared one single-chain surfactant
and four double-chain surfactants using renewable resources and a clean and rapid
synthetic methodology.
The LAM can be classified as biocompatible and biodegradable due to its biological
properties. It has been synthesized following the procedure previously reported by us.
[
9]. The compounds with two alkyl chains were obtained by simply heating the LAM
with the corresponding amine without using any solvent (Figure 1). In these conditions,
the percentage of conversion is always higher than 90% after 2 hours and the reaction
does not require the use of any activating agent. This approach allowed us to perform
the synthesis in a short time and on a large scale and at a reasonable cost.
These cationic double chain surfactants have amide linkages between the polar head and
the hydrophobic chains. Usually, the introduction of an ester or amide linkage between
the alkyl chain and the hydrophilic moiety improves the biodegradation of surfactants
[17].
FIGURE 1
Critical aggregation concentration
The concentration at which the surfactant molecules start to form aggregates in aqueous
solutions was determined by tensiometry and conductivity. The LANHC surfactants
x

are sparingly soluble in water; therefore, the most concentrated samples had to be
prepared by sonication of the aqueous solution during 10 minutes at 45 ºC. All samples
give rise to bluish solutions, indicating the presence of big aggregates. Previous studies
showed that the single-chain LAM formed transparent solutions that contained
spheroidal micelles [18]. The critical packing parameter (CCP) considers the geometry
of the molecule to predict what type of structures can be formed when monomers start
to aggregate [19]. Usually, double-chain surfactants have a cylindrical shape with CPP
values between 0.5 and 1. This shape explains why these surfactants have low
spontaneous curvature and tend to form planar structures like lamellar bilayers and
vesicles while the conically shaped single-chain surfactants form spherical micelles.
According to the above considerations, the surfactants studied in this work are expected
to form vesicles or big aggregates, therefore the concentration at which these surfactants
start to form aggregates in solution will be denoted as the critical aggregation
concentration (CAC).
TABLE 2
The surface tension/concentration plots show only one breakpoint in the 0.01-0.1 mM
concentration range corresponding to the CAC values (Table 2). These results suggest
that vesicles are formed directly at low concentrations. Double chain surfactants that
form micelles and vesicles show two transition points in the surface tension curves [20].
It has been reported that surfactants with one arginine or lysine and two alkyl chains
also form vesicles in aqueous solutions [21]. The didodecyl dimethylammonium
bromide, a widely studied double chain cationic, forms lamellar bilayers or vesicles, and
the critical aggregation is denoted as CVC (critical vesicular concentration) [22-23-24].
The incorporation of the second alkyl chain into the LAM enhances the hydrophobic
character of the molecule and consequently causes a significant decrease of the CAC

(
Table 2). The difference in the CAC between LANHC and LAM is more than one
12
order of magnitude. The CAC of gemini surfactants from arginine with two alkyl chains
of twelve carbon atoms is also higher than that found for these double-chain derivatives.
Para et al. [25] also found that the surface activity of the gemini surfactant was
somewhere in between the surface activity of the monomeric and that of the dimeric
surfactant.
The CAC in aqueous medium varies little with the change in alkyl chain length of one
of the hydrophobic chains. First, the CAC slightly decreases from n=10 to n=14 and
then it rises for n=18. For single-chain ionic surfactants, the CAC decreases abruptly as
the number of carbon atoms in the hydrophobic group increases. However, for very
hydrophobic surfactants the CAC does not decrease so rapidly with increase in the
length chain, and when the chain exceeds 18 carbons, it may remain substantially
unchanged [26] The presence of a carbon-carbon double bond in LANHC gives rise to
1
8
a bulky hydrophobic group. The increase of the CAC is probably due to the difficulty of
incorporating the bulky hydrophobic group in the interior of the micelle.
Table 2 also contains the effectiveness of these surfactants in decreasing the surface
tension of water (π) as well as the minimum area per molecule at the interface (Am)
(S.I.4). Double chain surfactants are more effective reducing the surface tension than
the single-chain LAM. The presence of two alkyl chains results in close packing at the
interface and consequently the surface tension at the CAC decreases. Among the
LANH compounds, the effectiveness decreases by increasing the length of one alkyl
x
chain. The Amin of the double-chain derivatives is similar to that of LAM. This suggests
that, due to the presence of two hydrophobic chains, the LANHCx compounds are
tightly packed at the interface.
FIGURE 2

The curve of conductivity against surfactant concentration at 25 and 50 ºC for
LANHC14 is shown in Figure 2. Similar curves have been obtained for the other
surfactants, and the CAC values obtained at both temperatures are included in Table 2.
The CAC of LANHC18 was measured by conductivity at 50 ºC only since it was
difficult to achieve stable values of conductivity at 25 ºC. Only one breakpoint was
detected in all cases confirming that these surfactants form big aggregates directly.
Moreover, the temperature does not affect the CAC values. The CAC values obtained
by conductivity are slightly higher than those obtained by surface tension
measurements. The discrepancy between surface tension measurements and
determinations by bulk techniques to obtain the CMC/CAC is present in the literature.
[27-28].
Cationic vesicles prepared with double chain arginine surfactants
In this work we have explored the capability of cationic dialkyl arginine derivatives to
form vesicles. These aggregates were prepared following the procedure described in the
experimental section.
After preparation, vesicles were left for 3 hours at room temperature and then the
diameter and size distribution was measured at 25 ºC using dynamic light scattering
measurements (DLS). All solutions were then heated again for 60 minutes at 50ºC and
the diameter and size distribution were evaluated at 50 ºC. Afterwards, the formulations
were kept at room temperature for 6 months.
FIGURE 3
The image in Figure 3 shows the dispersions at three different concentrations (1, 2.5 and
5
mm). The samples´ visual appearance provides valuable information. It is known that

a change in the turbidity of a surfactant solution can be related to the change in the
amount and/or the size of the aggregates. The LANHC10 forms a bluish translucent
solution at 1 mM. By increasing the concentration, the solutions became milky and
turbid (Figure 3). All solutions of LANHC appeared as bluish and viscous dispersions
1
2
with the typical aspect of vesicular formation. The appearance of LANHC solutions is
1
4
similar to those based on LANHC , with a slight intensification of turbidity. After the
1
2
second heating cycle, all samples appear less turbid. Finally, some evidence of phase
separation has been observed after 1 month for the most concentrated samples of
LANHC and after 2 months for the LANHC 1 mM solution. Formulations based on
1
0
10
LANHC LANHC and LANHC remained stable for 6 months.
1
2,
14
18
TABLE 3
Table 3 shows the particle size and size distribution data obtained for LANHC based
x
formulations and Figure S1 (Supporting information) shows the size distribution
profiles. At room temperature, LANHC10 solutions contain large vesicles
(diameters░>1000 nm) and at 50ºC the diameter of the aggregates is still high. At room
temperature LANHC solutions show high polydispersity and contain giant vesicles
1
2
(diameters░>1000 nm). However, at 50ºC these samples show only one peak with an
average diameter in the range 150-300 nm. Samples based on LANHC show a very
1
4
similar behaviour. However, the sample prepared with LANHC18 at 1 mM is a
translucent solution that contains small aggregates (about 60 nm) at 25 and 50 ºC. From
the results, it is not possible to establish how the alkyl chain length affects the vesicle’s
size. For 1 mM samples, the vesicle’s size decreases as the alkyl chain increases,
especially for n=18. However, the more concentrated formulations appear not to follow
this trend.

These results correlate well with the stability of samples. Formulations that contain very
large vesicles show phase separation after 1 or 2 months while those that contain
medium size vesicles remain stable for more than 6 months. The stability can also be
related to the repulsive interactions among aggregates with positive charges. As
expected, large positive ζ-potential values have been obtained for vesicles prepared with
these surfactants (61.4 mV for LANHC , 48.3 mV for LANHC , 36.6 mV for
1
0
12
LANHC14 and 18 for LANHC ). The positive charge density of the aggregates is
18
strong enough to ensure that no coagulation occurs for LANHC , LANHC and
1
2
14
LANHC formulations. In the case of LANHC the large sizes observed can result in
1
8
10
large van der Waals attractive forces between vesicles that allow these aggregates to
overcome the repulsive electrostatic forces and induce phase separation. It has been
described that vesicles prepared with mechanical forces, like ultrasonication, are
2
9
metastable.
For example, sonicated DDAB vesicles with high z potentials were
unstable, resulting in phase separation [30]. However, when the size of these DDAB
vesicles is reduced, the aggregates are stable for extended periods of time as a result of
repulsive electrostatic forces.
The preparation procedure also can strongly affect the size of the vesicle. Different
vesicle sizes have been reported for the same compound. Feitosa el al. reported DDAB
vesicles prepared at room temperature with a diameter of 195 nm [31]. For DDAB and
dioctadecyl dimethylammonium bromide (DODAB) the diameters of spontaneously
formed vesicles were 70 nm [32] and 674 nm [33] respectively. After bath sonication of
the dispersion, the diameter of the DODAB vesicles dropped to 44 nm [34], and to 66
nm after tip sonication [35]. Kondo el al. [30] showed that the DDAB vesicles´ diameter
decreased with ultrasonication time, reaching a limiting value of approximately 17 nm
after about 1.5 h.

To summarize, it can be stated that these compounds form large and stable vesicles at
room temperature. The size of niosomes formulated with these compounds exhibits
strong temperature dependence. DLS measurements at 25 ºC show a major population
of very big aggregates in the micrometre range with a large polydispersity. When heated
to 50 ºC the polydispersity, as well as, the size of the aggregates decrease drastically.
This behaviour could be due to the state of the niosomes. In general, vesicular
dispersions scatter more light in the gel than in the liquid-crystalline state owing to
formation of large or more complex vesicular structures [31].
Antimicrobial activity
In this work, we have evaluated how the incorporation of a second alkyl chain affects
the antimicrobial activity of LAM [16]. Table 4 shows that the antimicrobial activity of
these new surfactants is lower than that of LAM, probably due to their high
hydrophobicity. The low antimicrobial activity of the new surfactants can also be
related to their aggregation behaviour. For LAM the MIC values occur at concentrations
below the CMC. Therefore, monomers are the species that interact with the cells.
However, the aggregation of LANHC takes place at very low concentrations, and they
x
form vesicles instead of micelles. Hence, in this case, the vesicles have to interact with
the cellular membranes. Recent studies have shown that for colloidal suspensions
prepared with arginine-based gemini surfactants the antimicrobial and haemolytic
activity increases as the aggregate’s size decreases [36]. Vieira et al. who studied the
antimicrobial and haemolytic activity of DODAB (dioctadecyl dimethylammonium
bromide) bilayer fragments and DODAB large vesicles and also found that small
aggregates penetrate more deeply into the cell surface, and, consequently have lower
MIC and HC50 values [37]. It has also been reported that niosomes containing

monoalkyl surfactants from methylprolinolinium show good antimicrobial activity
while niosomes formulated with the dialkyl homologue do not show biological activity.
These authors suggested that the lateral homogeneity of the niosomes´ bilayer could be
another important parameter that controls their biological activity [38-39]. Whereas
niosomes prepared with single chain surfactants had Tm values around 27 ªC, double
chain cationic surfactants gave rise to niosomes with a high Tm (40 ºC). Therefore, at
3
7 ºC the single chain surfactants are in the LC phase while the double chain ones are in
the solid ordered phase, and the bilayers have poor lateral miscibility.
TABLE 4
The antimicrobial activity of these new surfactants depends on hydrophobicity, although
the correlation between hydrophobicity and antimicrobial activity is not linear. The
LANHC , the less hydrophobic and most active surfactant, shows antimicrobial
1
0
activity against most of the microorganism tested. However, the LANHC12 and
LANHC show a moderate activity only against Gram-positive bacteria and LANHC
1
4
18
did not show activity against any of the microorganisms.
As expected, Gram-negative bacteria were more resistant than Gram-positive bacteria.
These bacteria have a hydrophilic surface due to the side chains of the LPS which
hinders hydrophobic molecules from entering the bilayer [40].
Cytotoxicity
The concentration that induces haemolysis of 50% of the erythrocytes (HC ) was
5
0
determined and quantified from plots of percentage of haemolysis as a function of
surfactant concentration.
The low haemolytic activity of LANHC , LANHC and LANHC14 is noteworthy
1
0
12
(
Table 5). For cationic surfactants, these HC concentrations are exceptionally high.
50
The surface of red blood contains negative charges, which prevents their self-

aggregation and makes them as colloids suspended in blood [41]. For this reason,
cationic surfactants can interact well with the negatively charged surface of
erythrocytes.
TABLE 5
Compared with the LAM, (HC = 144µM) [42] the haemolytic activity of these
50
compounds is exceptionally low (Table 5). This could be ascribed to the size and
stability of the aggregates that they form. Notice that, except the LANHC , all
1
8
LANHC surfactants form large and stable giant vesicles at 25 ºC. In general, single-
x
chain surfactants are less haemolytic than the corresponding double-chain surfactants
with a gemini structure [43]. However, the haemolytic character of gemini surfactants
from arginine with alkyl chains of 12 carbon atoms and long spacer chains is
considerably lower than that of LAM, probably due to the small size of the gemini
aggregates [36]. Carmona-Ribeiro et al. also found that solutions of the single chain
derivative hexadecyl trimethylammonium bromide (HC =0.7 µM) are more haemolytic
5
0
than those of large vesicles formed by the double-chain compound dioctadecyl
dimethylammonium bromide (HC = 4.0 µM) [37].
5
0
It should also be considered that the vesicles’ stability can also affect their haemolytic
character. It has been described that haemolysis was induced immediately after the
addition of dialkyl dimethylammonium surfactants with relatively short alkyl chains (10
carbon atoms) while haemolysis was gradually induced for long alkyl chains (16 and 18
carbon atoms). This behaviour was ascribed to the stability of the vesicles formed by
these compounds. Haemolysis is induced by the penetration of the monomer into the
erythrocyte membrane. Vesicles of dialkyl dimethylammonium with short chains appear
unstable and may easily become monomers. However, dialkyl dimethylammonium
surfactants with long hydrophobic chains form more stable vesicles or aggregates,

because of that the presence of monomers was reduced and haemolytic activity
decreased [44].
The haemolytic activity of these surfactants increased considerably by increasing the
length of the second alkyl chain. The influence of the hydrophobic domain (alkyl chain
length and number of hydrophobic chains) on cellular toxicity is not well established in
the literature. There are some surfactant analogues in which haemolysis increases with
the elongation of the alkyl chain [45-46] whereas, in other groups, haemolysis decreases
as the alkyl chain increases [47]. Fogt et al. [48] and Lv et al. [8] also reported that for
aliphatic chains, single-chain cationic lipids are more toxic and less efficient than their
double-chain counterparts whereas Tang and Hughes [49] demonstrated that the single-
chain 6-lauroxyhexyl ornithinate presented lower cytotoxicity than DOTAP, a double-
chain cationic surfactant widely used in gene therapy. It is, therefore, essential to
accurately characterize the physicochemical and biological properties of new surfactants
to obtain data that allow correlating results with toxic effects and help to develop novel
and safe cationic lipids.
To summarize, it can be stated that the haemolytic activity of LAM [42] is lower than
that of standard monoQuat surfactants with comparable CMC values and the haemolytic
character of gemini surfactants from arginine with long spacer chains [36] is much
lower than that of bis(Quats) gemini surfactants [50] or gemini surfactants based on
pyridinium salts [51]. Our results also indicate that the haemolytic activity of arginine-
based double chain surfactants is also much lower than that of double chain surfactants
with the cationic charge on a quaternary ammonium [37]. Then it can be concluded that
for similar chemical structures, cationic surfactants from arginine, in which the cationic
charge is located on the guanidine group, are less haemolytic than QUATS.

The cytotoxic effects of these surfactants were evaluated with 3T3, HaCaT and HeLa
cells using two different endpoint assays (Table 5). The combination of different cell
lines and cytotoxicity assays gives information concerning general and specific
toxicological mechanisms [52].
All these cationic surfactants induced a clear dose-response relationship, which allowed
us to determine the half maximal inhibitory concentration or IC . The toxicity depends
5
0
on both the cell line tested and the end point assay. HaCat, 3T3 and HeLa cells are
much more fragile than red blood cells to the vesicles formed by these surfactants.
Moreover, the IC values obtained using the MTT assay are lower than those obtained
5
0
with the NRU assay. These differences can be ascribed to the cytotoxicity mechanisms
evaluated by the different biological assays. Haemolysis depends on the interaction of
surfactants with the plasma membrane of a cell. The MTT assay is a measurement of
cellular metabolic activity within the mitochondrial compartment while the NRU assay
measures membrane integrity. Taking into account the cell damage evaluated by every
assay, our results suggest that the aggregates formed by the LANHC derivatives
x
interact first with the mitochondrial compartment affecting cellular metabolism while
the plasma membrane is less affected. These differences also suggest that the
internalization of the compound is produced before cell membrane damage. Similar
behaviour has been described for single and double-chain surfactants from lysine [53] as
well as for cationic nanovesicular systems containing lysine-based surfactants [54].
The toxicity showed by these surfactants is lower than those reported for double-chain
surfactants from arginine and lysine with a gemini structure [42-43]. However, double
chain surfactants from arginine with a glycerolipid structure exhibited lower toxicity
possibly due to their lower chemical stability [42]. The chemical bond linking the head
group to the two hydrophobic chains can affect toxicity. Glycerolipid surfactants from

arginine contain two alkyl chains linked to the head group by ester bonds. These ester
bonds can be easily hydrolysed giving rise to non-toxic species.
Our results clearly show that the incorporation of the second alkyl chain on LAM
increases the cytotoxicity against the 3T3, HaCat and HeLa cell lines. This behaviour
contrasts with that observed for the haemolytic activity and indicates that the
mechanisms that dominate the interaction of LAM or double-chain derivatives
aggregates ( micelles for LAM or vesicles for the double-chain compounds) with red
blood cells and the other cell lines are different. These results are in line with previous
studies that reported that cytotoxic effects of carrier systems differ depending on the cell
lines used due to the nature of these cells [55].
It seems that cytotoxicity slightly decreases by increasing the length of one of the
hydrophobic chains. In fact, the lowest cytotoxicity was obtained for the LANHC .
1
8
Notice that vesicles formed by this surfactant have the lowest surface cationic charge.
Cell membranes have large negatively charged domains. It has been described that
cationic vesicles interact with cell membranes neutralizing the sialic acid groups and
changing the surface charge from negative to positive [56]. Then, because of
electrostatic interactions a great affinity of vesicles with high cationic charge density is
expected.
The IC values suggest that the aggregate’s size does not affect the cytotoxicity against
5
0
the three cell lines tested. The influence of aggregate’s size on cellular toxicity is not
clearly explained in the literature. In fact, some researchers have proposed that cells
more easily take up smaller particles [57-36] while other studies suggest that the size of
nanoparticles does not influence cytotoxicity [58].

In general, dialkyl dimethylammonium bromides with similar alkyl chain lengths
showed higher cytotoxicity than these arginine derivatives. For example, the IC of
5
0
DDAB against erythroleukemic cells was 0.85 µM [59].
It should be noticed that overall, vesicles formed by pure cationic lipids can exhibit
toxicity, but this can be reduced adding another compound to the vesicular system. At
1
00 µM the cell viability of vesicles prepared with pure DODAB (dioctadecyl
dimethylammonium bromide), DHDAB (dihexadecyl dimethylammonium bromide)
and DMDAB (dimiristoyl dimethylammonium bromide) is about 30%, 20% and 5%
respectively. However, at this same concentration, the cell viability rises up to more
than 60% when the vesicles are formulated with these cationic surfactants together with
a neutral phospholipid [60]. The haemolytic activity of negatively charged vesicles
prepared
with
mixtures
of
diacylglycerol
arginine-based
surfactants/phosphatidylglycerols was much lower than those prepared with the cationic
6
1
surfactant alone. Moreover, it has been reported that cationic niosomes prepared with
mixtures of DODAB, DPPC and cholesterol showed very low toxicity against B F
1
6 10
cell lines (IC > 3.12 mM) [1b]. For this reason, studies of the effect of adding neutral
5
0
phospholipids on cationic vesicles prepared with these surfactants are underway. The
aim is to know if this procedure reduces the cellular toxicity of these cationic
formulations.
CONCLUSIONS
Four novel double-chain surfactants with different hydrophobic chain lengths have been
α
prepared from the biocompatible N -Lauroyl arginine methyl ester (LAM). The
synthetic approach used allows for the preparation of these new compounds on a large
scale, at a reasonable cost and low ecological impact.

The CAC of these new cationic surfactants is much smaller than that shown by the
single-chain homolog LAM. Because of their molecular architecture consisting of one
polar head and two alkyl chains, they form cationic vesicles at very low concentrations.
Our results indicate that by introducing a second alkyl chain on the single-chain
surfactant LAM it is possible to obtain new compounds that form stable cationic
vesicles in aqueous solutions and present very low haemolytic activity. Moreover,
cationic vesicles prepared with the LANHC10 also have good antimicrobial activity
against Gram-positive bacteria. In vitro studies with 3T3, HaCaT and HeLa cells show
that these double arginine surfactants exhibit weaker cytotoxic effects compared to
conventional dialkyl dimethylammonium surfactants. These compounds can be suitable
candidates to be used in biotechnological and pharmaceutical applications because of
both the synthetic procedure required for their preparation as well as their properties.
Finally, it should be remarked that understanding the physicochemical properties of the
cationic surfactants in association with their biological properties can lead to a rational
design of safe cationic lipids.
ACKNOWLEDGEMENTS
The authors would like to thank the Spanish Plan National I+D+I MAT2012-38047-
C02-02,
CTQ2014-59632-R and the “Grupo de Tensioactivos con Membranas Celulares”
Universitat de Barcelona), Unidad asociada CSIC.
MAT2012-38047-C02-01,
MIMECO-CTQ2013-41514-P,
MIMECO-
(
References

[
1] (a)L. Pérez, A. Pinazo, R. Pons, M.R. Infante, Advances in Colloid and Interface Science, 205 (2014)
34. (b)A. Manosroi, K. Thathang, R.G. Werner, R. Schubert, R. Peschka-Süss, J. Manosroi, Gene
Therapeutics, 58 (2008) 485.
2] (a)M.A. Llies, W.A. Seitz, B.H. Johnson, E.L. Ezell, A.L. Miller, E.B. Thompson, A.T. Balaban, J.
1
[
Med. Chem., 49 (2006) 3872. (b)S.M. Vyas, J. Turánek, P. Knötigová, A. Kasná, V. Kvardova, E.
Rankin, L. Knutson, H.J. Lehmler, New J. Chem., 30 (2006) 944. (c) J.A. Heyes, D. Nicolescu-Duvaz,
R.G. Cooper, C.J. Springer, J. Med. Chem., 45 (2002) 99.
[
3] a) JPN Silva, ACN Oliveira, M Lucio, AC Gomes, PJG Coutinho, MECDR Oliveira, Colloids
and Surfaces b-Biointerfaces, 121 (2014) 371 b) CL Chan, KK Ewert, RN Majzoub, YK Hwu, KS
Liang, C Leal, CR Safinya, Journal of Gene Medicine, 16 (2014) 84.
[
4] Z. Drulis-Kawa, A. Dorotkiewicz-Jach, J. Gubernator, ; G. Gula, ; T. Bocer, ; W. Doroszkiewicz, ,
International journal of pharmaceutics, 367( 2009) 211.
5] (a) G. Nalecz-Jawecki, E. Granbinska-Sota and P. Narkiewicz, Ecotoxicology and Environmental
[
Safety, 54 (2003) 87. (b) M.T. Garcia, I. Ribosa, T. Guindulian, J. Sanchez-Leal , J. Vives-Rego,
Envirom. Pollut., 111 (2001) 169.
[
6] N.D. Henderson, A review of the environmental impact and toxic effects of DDAC, Environmental
Protection Division, BC Environment, Ministry of Environment, Land and Parks, Victoria, British
Columbia, 1992.
[
7] F. Ferk, M. Misik, C. Hoelzl, M. Uhl, M. Fuerhacker, B. Grillitsch, W. Parzefall, A. Nersesyan, K.
Micieta, T. Grummt, V. Ehrlich, S. Knasmüller, Mutagenesis, 22 (2007) 363.
[
[
8] H. Lv, S. Zhang, B. Wang, S. Cui, J. Yan, J. Controlled Release, 114 (2006) 100.
9] M.C. Moran, A. Pinazo, L. Perez, P. Clapés, M. Angelet, M.T. García, M.P. Vinardell, M.R. Infante,
Green Chemistry, 6 (2004) 233.
[
[
[
[
[
[
[
10] A. Sorrenti, O. Illa, R. Pons, R. M. Ortuño, Langmuir, 31 (2015), 9608.
11] A.S Lawrie, A. Albanyan, R.A. Cardigan, J. Mackie and P. Harrison, Vox Sanguinis, 96 (2009) 206.
12] S.W. Provencher, Comput. Phys. Commun., 27 (1982) 229.
13] J.M. Andrews, Journal of Antimicrobial Chemotherapy, 48 (2001) Suppl. S1, 5-16.
14] T. Mosmann, J. Immunol. Methods, 65 (1983) 55.
15] W.J. Pape, U. Pfannenbecker and U. Hoppe, Mol. Toxicol., 1 (1987) 525.
16] M. R. Infante, L. Pérez,; A. Pinazo, Novel Cationic Surfactants from Arginine. In Novel Surfactants,
Preparation, Applications and Biodegradabilily ; Holmberg, K. Second Ed.; Surfactant Science Series;
Marcel Dekker: New York, 2003; Vol. 76
[
17] F. Goursaud, M. Berchel, J. Guilbot, N. Legros, L. Lemiègre, J. Marcilloux, D. Plusquellec, T.
Benvegnu, Green Chemistry, 10 (2008) 310.
[
[
[
[
(
(
[
18] L. Pérez, A. Pinazo, M. R. Infante, R. Pons, Journal Physical Chemistry B, 111 (2007) 11379.
19] J.N. Israelachvili, D.J. Mitchell, B.W. Ninham, J. Chem. Soc. Faraday Trans. 2, 72 (1976) 1525.
20] V. Vijai, A. Patnaik, Langmuir, 23 (2007) 3523.
21] (a) A. Pinazo, M. Angelet, R. Pons, M. Lozano, M.R. Infante, L. Pérez, Langmuir, 25(2009), 7803.
b) L. Pérez, A. Pinazo, P. Vinardell, P. Clapés, M. Angelet, M.R. Infante, New Journal of Chemistry, 26
2002) 1221.
22] (a) C. Treiner , A. Makayssi, Langmuir, 8 (1992) 794. (b) M.I. Viseu, M.M. Velazquez, C.S.
Campos, I. Garcia-Mateos and S.M..B Costa, Langmuir, 16 (2000) 4882.
[
[
[
23] A. Caria, O. Regev, A. Khan, J. Colloid. Interface Sci., 200 (1998) 19.
24] I. Grillo, J. Penfold, I. Tucker, F. Cousin, Langmuir, 25 (2009) 3932.
25] G. Para, A. Hamerska-Dubra, K. Wilk, P. Warszynski, Colloids and Surfaces A. Physicochem. Eng.
Aspects, 383 (2011) 67.
26] M.J. Rosen, J. T. Kunjappu, Surfactants and Interfacial Phenomena, Fourth Edition, Wiley, New
Jersey, 2012.
27] M. Alvarez Alcalde, A. Jover, F. Meijide, L. Galantini, N. Viorel Pavel, A. Antelo, J. Vazquez Tato,
Langmuir, 24 (2008) 6060.
[
[
[
[
28] L. Perez, M.R. Infate, M. Angelet, P. Clapes, A. Pinazo, Prog.Colloid.Polym. Sci., 123 (2004) 210.
29] P.W.M. Van Dijk, B. De Kruijiff, P.A.M. Aarts, A.J. Verkleij, J. De Gier, Biochim. Biophys. Acta,
5
[
06 (1978) 18.
30] Y. Kondo, M. Abe, K. Ogino, H. Uchiyama, E.D. Tucker, J.F. Scamerhorn, S.D.Christian, Colloids
and surfaces B Biointerfaces, 1 (1993) 51.
[
[
(
[
31] E. Feitosa, J. Jansson and B. Lindman, Chemistry and Physics of lipids, 142 (2006) 128.
32] E.F. Marques, O. Regev, A. Khan, B. Lindman, Advances in Colloidand Interface Science, 100-102
1999) 83.
33] E. Feitosa, G. Karlsson and K. Edwards, Chemistry and Physics of Lipids, 140 (2006) 66.

[
[
34] E. Feitosa and W. Brown, Langmuir, 13 (1997) 4810.
35] I.M. Cuccovia, E. Feitosa, H. Chaimovich and L. Sepulveda, W. Reed, J. Phys. Chem., 94 (1990)
3
722.
36] L. Tavano, M.R. Infante, M. Abo Riya, A. Pinazo, M.P. Vinardell, M. Mitjans, M.A. Manresa, L.
Pérez, Soft Matter, 9 (2013) 306.
[
[
[
37] D.V. Vieira and A.M. Carmona-Ribeiro, J. Antimicrob. Chemother, 58 (2006) 760.
38] C. Bombelli, F. Bordi, S. Ferro, L. Giansanti, G. Jori, G. Mancini, C. Mazzuca, D. Monti, F.
Ricchelli, S. Sennato and M. Venanzi, Molecular Pharmaceutics, 5 (2008) 672.
39] Y. Barenholz, C. Bombelli, M.G. Bonicelli, P.D. Profio, L. Giansanti, G. Mancini and F. Pascale,
Journal of Colloid and Interface Science, 356 (2011) 46.
40] E.L. Wu, P.J. Fleming, M.S. Yeom, G. Widmalm, J.B. Klauda, K.G. Fleming, Biophysical Journal,
06 (2014) 2493.
[
[
1
[
[
41] H. Pockel, C. Lenz and M.L. Barjas-Castro, Rev. Bras. Hematol. Hemoter., 33 (2011) 297.
42] A. Pinazo, L. Pérez, M.R. Infante, M.P. Vinardell, M. Mitjans, M.C. Morán, V. Martinez, in N. L.
Jacobs (Ed.), Chemical structure and toxicity in Arginine-based surfactants, Nova Science Publishers,
New York, 2011, Chapter 6.
[
43] A. Colomer, A. Pinazo, M.A. Manresa, M.P. Vinardell, M. Mitjans, M.R. Infante, L. Pérez, J. Med.
Chem., 54 (2011) 989.
[
[
44] S. Kitagawa, M. Kasamaki and F. Hiyama, Chem. Pharm. Bull., 49 (2001) 1155.
45] M. Rasia, M.I. Spengler, S. Palma, R. Manzo, P. Lo Nostro, D. Allemandi, Clinical Hemorheology
and Microcirculation, 36 (2007) 133.
46] T. Benavides, M. Mitjans, V. Martinez, P. Clapés, M.R. Infante, R.H. Clothier, P. Vinardell,
Toxicology, 197 (2004) 229.
[
[
[
[
[
47] S. Roy and P.K. Das, Biotechnol. Bioeng., 100 (2008) 756.
48] A. Fogt, H. Hagerstrand, B. Isoma, Chem. Biol. Interact., 94 (1995) 147.
49] F. Tang and J.A. Hughes, J. Controlled Release, 62 (1999) 345.
50] M. Dubnickova, M. Bobrowska-Hagerstrand, T. Soderstrom, A. Iglic, H. Hagerstrand, Acta
Biochim. Pol., 47 (2000) 651.
51] H. Nagamune, T. Maeda, K. Ohkura, K. Yamamoto, M. Nakajima, H. Kourai, Toxicology in vitro,
4 (2000) 139.
[
1
[
[
52] D.R. Nogueira, M. Mitjans, M.R. Infante, M.P. Vinardell, Int. J. Pharm., 420 (2011) 51.
53] A. Colomer, A. Pinazo, M.T. García, M. Mitjans, M.P. Vinardell, M.R. Infante, V. Martínez, L.
Pérez, Langmuir, 28 (2012) 5900.
54] D.R. Nogueira, M.C. Morán, M. Mitjans, V. Martínez, L. Pérez, M.P. Vinardell, European Journal of
Pharmaceutics and Biopharmaceutics, 83 (2013) 33.
[
[
[
[
[
55] T. Xia, M. Kovochich, M. Liong, J.I. Zink, A.E. Nel, ACS Nano, 2 (2008) 85.
56] A.M. Carmona-Ribeiro, F. Ortis, R.I. Schumacher, M.C.S. Armelin, Langmuir, 13 (1997) 2215.
57] K.Y. Win and S.S. Feng, Biomaterials, 26 (2005) 2713.
58] N. Scholer, C. Olbrich, K. Tabatt, R.H. Müller, H. Hahn, O. Liesenfeld, Int. J. Pharm., 221 (2001)
5
7.
[
[
[
59] R. Cortesi, E. Esposito, E. Menegatti, R. Gambari, and C. Nastruzzi, Int. J. of Pharm., 139 (1996) 69.
60] C-H. Liang, T-H. Chou, Chemistry and Physics of Lipids, 158 (2009) 81.
61] N. Lozano, L. Pérez, R. Pons, A. Pinazo, Amino Acids, 40 (2011) 721.