Self-Assembly of Lipoaminoacids-DNA Based on Thermodynamic and Aggregation Properties

Article abstract of DOI:10.1002/jsde.12391

Lipoamino acids (LAA) are biocompatible and biodegradable biosurfactants, a promising alternative to viral vectors in gene delivery. LAA are constituted by a polar head, the amino acid, and a hydrocarbon (alkyl) chain usually from a fatty acid or fatty ac

Full text of DOI:10.1002/jsde.12391

J Surfact Deterg (2020)
DOI 10.1002/jsde.12391
ORIGINAL ARTICLE
Self-Assembly of Lipoaminoacids-DNA Based on
Thermodynamic and Aggregation Properties
Célia Faustino1 · Tiago Martins1 · Noélia Duarte1 · Maria H. Ribeiro1
Received: 7 April 2019 / Revised: 4 January 2020 / Accepted: 7 January 2020
© 2020 AOCS
Abstract Lipoamino acids (LAA) are biocompatible and
biodegradable biosurfactants, a promising alternative to viral
vectors in gene delivery. LAA are constituted by a polar head,
the amino acid, and a hydrocarbon (alkyl) chain usually from a
fatty acid or fatty acid derivative, such as a fatty amine or a fatty
alcohol. In this work, dodecyl LAA was produced from dode-
cylamine and natural L-amino acids cystine (Cys), lysine (Lys),
and phenylalanine (Phe) using an enzyme-based approach with
porcine pancreatic lipase. The self-assembly behavior of LAA
solutions, in the absence or presence of DNA, was studied by
conductivity and fluorescence regarding the application as
transfection agents. Conductivity measurements yielded impor-
tant system parameters, including critical micelle concentration
with electrostatic interactions. The CACapp values decreased
with increasing LAA hydrophobicity, reflecting the relevance
of hydrophobic interactions in complex coacervation.
Keywords Lipoamino acid ꢁ DNA ꢁ Cationic surfactant ꢁ
Lipase ꢁ Gene delivery vectors
J Surfact Deterg (2020).
Introduction
Gene therapy has emerged as a promising approach for the
treatment or prevention of both genetic and acquired diseases
(Ilarduya et al., 2010; Wettig et al., 2008). However, there are
several hurdles to overcome in developing effective gene-based
therapeutics, namely cellular uptake, endosomal escape
avoiding DNA degradation, and nuclear localization (Ilarduya
et al., 2010; Singh et al., 2011; Singh et al., 2012; Wettig et al.,
2008). Successful gene therapy depends crucially on the devel-
opment of effective vectors, especially the introduction of the
selected gene into living cells toward the target site. Some
drawbacks may be associated with viral vectors, namely resid-
ual infectivity and immunogenic and inflammatory responses
(Ilarduya et al., 2010; Wettig et al., 2008). Cationic lipids or cat-
ionic surfactants that bind and condense the negatively charged
phosphate backbone of DNA into nanosized cationic lipid-
DNA complexes (lipoplexes) are attractive synthetic alterna-
tives to viral vectors due to easy preparation, higher loading
capacity, lower cytotoxicity, and safer immunogenic profile
(Ilarduya et al., 2010; Wettig et al., 2008).
(CMC) and standard Gibbs energy of micellization (ΔGꢀ
)
mic
for pure LAA systems, and apparent critical aggregation con-
centration (CACapp) and apparent standard Gibbs energy of
aggregation (ΔGꢀagg) for the mixed LAA-DNA systems. The
CMC increased in the order of decreasing lipophilicity:
(C12Cys)2 < C12Phe < C12Lys, CMC values were higher in the
presence of DNA, suggesting the formation of a LAA-DNA
complex responsible for hindering the micellization process.
Binding of the LAA with DNA was confirmed from fluores-
cence measurements for the ethidium bromide exclusion assay.
Results suggest a weak interaction of the LAA with DNA
which can be attributed to their relatively short dodecyl chains
and/or the ionic strength of the buffer solution, supporting the
role of hydrophobic interactions in complex formation between
DNA and the oppositely charged surfactant in combination
* *Maria H.L. Ribeiro
mhribeiro@ff.ul.pt
Lipoamino acids (LAA), which are condensation products of
amino acids with fatty acids or their derivatives, usually have
good biocompatibility and biodegradability (Bordes and
Holmberg, 2015; Pérez et al., 2014). The polar headgroup of
1
Faculty of Pharmacy, Research Institute for Medicines (iMed.
ULisboa), Universidade de Lisboa, Av. Prof. Gama Pinto,
1649-003 Lisbon, Portugal
J Surfact Deterg (2020)

J Surfact Deterg
LAA, which contains both donor and acceptor hydrogen bond-
ing groups capable of intra- and intermolecular interactions, adds
further complexity to their self-assembly behavior. LAA have
been employed as drug delivery agents for hydrophobic drugs
(Ménard et al., 2012; Serafim et al., 2016), enhancing drug solu-
bilization and bioavailability. Moreover, LAA are obtained from
naturally renewable sources and can be produced by green-
chemistry approaches that include biotechnological procedures,
such as fermentation or enzymatic catalysis (Ménard et al., 2012;
Pérez et al., 2014; Pinazo et al., 2011; Serafim et al., 2016). Cat-
ionic LAA are thus promising transfection agents that meet the
requirements of both physiological and ecological compatibility.
Dodecyl LAA varying in the nature of the headgroup
and number of dodecyl chains has been obtained from
renewable raw materials by enzyme-catalyzed coupling of
dodecylamine (DDA) with lysine, phenylalanine, and cys-
tine (Fig. 1).
membrane phospholipids (Faustino et al., 2011), serum albumin
proteins (Branco et al., 2015) and DNA (Faustino et al., 2015).
The presence of pH-sensitive amino acid headgroups in
the LAA structure allow for changes in the aggregation mor-
phology with variations in the protonation state of the amino
acid residues that facilitate membrane fusion and DNA
escape from the endosome often avoiding the need for addi-
tion of helper lipids (Ilarduya et al., 2010; Wettig et al.,
2008). In this work, the supramolecular behavior of the dode-
cyl LAA obtained and their interactions with DNA were
studied regarding further application as transfection agents.
Conductivity measurements yielded important system param-
eters, including critical micelle concentration (CMC) and
standard Gibbs energy of micellization (ΔGꢀmic) for pure
LAA systems as well as apparent critical aggregation concen-
tration (CACapp) and apparent standard Gibbs energy of
aggregation (ΔGꢀagg) for the mixed LAA-DNA systems.
The LAA from cystine (the dimer of cysteine) with its
dimeric structure comprising two cysteine residues linked by a
disulphide bond, is expected to promote the release of the com-
plexed DNA and enhance transfection efficiency (Ilarduya
et al., 2010; Wettig et al., 2008), through reductive cleavage of
the disulphide spacer by endogenous glutathione. Furthermore,
dimeric or gemini LAA are versatile molecules able to interact
with biological interfaces and biomacromolecules, including
Materials and Methods
Materials
Double-stranded deoxyribonucleic acid (dsDNA) in the
form of sodium salt from salmon testes was supplied by
O
NH2
H
S
N
H N
S
NH2
O
(C12Cys)2
O
O
H2 N
NH
NH
NH2
NH2
C12Phe
C
12Lys
Fig. 1 Lipoamino acids produced by biocatalysis
J Surfact Deterg (2020)

J Surfact Deterg
Sigma-Aldrich (St. Louis, MO, USA) and used as
received. Ethidium bromide (EB; 3,8-diamino-5-ethyl-
6-phenylphenanthridinium bromide) solution (0.5 mg mL−1
in water), tris(hydroxymethyl) aminomethane (Tris), DDA
L-cystine (Cys), L-lysine (Lys), L-phenylalanine (Phe) and
porcine pancreatic lipase (PPL), type II crude, were also
obtained from Sigma-Aldrich (St. Louis, MO, USA).
Quattromicro® API (Waters®, Dublin, Ireland), triple quad-
rupole type. Compound ionization was performed by an
electrospray source in positive mode (ESI+). For acquisi-
tion and processing of HPLC-MS/MS data MassLynx®
version 4.1 was used.
The separation was performed on a normal-phase col-
umn (Luna HILIC 1000 × 3.00 mm) at 35 ꢀC using an
injection volume of 10 mL. Photodiode Array Detector was
used to scan wavelength absorption from 210 to 600 nm.
The mobile phase consisted of Milli-Q (Millipore, Bedford,
MA, USA) water containing 0.5 formic acid (i): acetonitrile
(ii) at a flow rate of 0.30 mL min−1 and the following elut-
ing conditions were used: 0 min—5% (A) and 95% (B);
5.0 min—100.0% (A) and 0.0% (B); 7.0 min—100.0%
(A) and 0.0% (B); 10.0 min—5.0% (A) and 95.0% (B);
15.0 min—5.0% (A) and 95.0% (B).
Biosynthesis
The dodecylLAAs were prepared from natural L-amino acids
cystine, lysine, and phenylalanine by condensation with DDA
using the biocatalyst, lipase PPL (EC number 232–619-9)
encapsulated in sol–gel, in a reaction media of eutectic mix-
ture, Dowtherm©A (Dow Chemicals, Midland, MI, USA).
PPL was immobilized using a developed in-house pro-
cess. The immobilization of PPL was carried out in sol–gel,
following the method described by Vila-Real et al. (2010).
Briefly, in an eppendorf were added 96 mg of glycerol
(98%), 70 μL of distilled water, 15 μL of HCl (80 mM),
and 300 μL of TMOS. The mixturꢀe was sonicated for
20 min at a temperature from 0 to 4 C. To the formation
of the hydrogel a ratio of 1:1 was used, adding to 25 μL of
sol solution, 25 μL PPL 2.5 mg mL−1 in 10 mM Tris buffer
pH 8. The lens shape was created by using a 96 rounded
well microplate as a recipient for the sol–gel with the PPL.
Gentle tapping on the microplate was used to unify the
mixture.
The lenses (126 mg) were used in the bioconversion
assays at 45 ꢀC, with the substrate, DDA, which is liquid at
this temperature. One gram of DDA was weighted into
each reaction flask, which was stabilized at 45 ꢀC, to com-
plete melting of the DDA. After, 300 mg of cystine, or
300 mg lysine, or 300 mg phenylalanine was added to each
flask, along with 500 μL of the eutectic mixture Dowt-
herm®A. The bioconversion assays were performed for
96 h. All the assays were carried out in triplicate.
NMR spectra were recorded on
a
Fourier
300 spectrometer (300 MHz) from Bruker (Billerica, MA,
USA) using CDCl3 as solvent, based on the studies of
1
Faustino et al., (2010). H chemical shift is expressed in
parts per million, δ (ppm), referenced to the solvent used.
1
(C12Cys)2. H NMR (CDCl3): δ (ppm) = 0.88 (t, 6H,
2 × CH3), 1.25 (m, 36H, 2 × (CH2)9CH3), 1.53 (m, 4H,
2 × CH2CH2NHCO), 2.78 (m, 4H, 2 × CH2NHCO), 3.15
(m, 4H, 2 × CH2S), 3.56 (d, 2H, 2 × CH). M+ (m/z): 575.
C12Lys. 1H NMR (CDCl3): δ (ppm) = 0.88 (t, 3H, CH3),
1.25 (m, 18H, (CH2)9CH3), 1.45 (m, 4H, CH2CH2NHCO
+ H2N(CH2)2CH2), 2.10 (m, 4H, H2NCH2CH2CH2CH2),
2.69 (m, 2H, CH2NHCO), 3.71 (m, 1H, CH). M+ (m/z): 314.
C12Phe. 1H NMR (CDCl3): δ (ppm) = 0.88 (t, 3H, CH3),
1.26 (m, 18H, (CH2)9CH3), 1.45 (m, 2H, CH2CH2NHCO),
2.21 (br s, 2H, CH2Ph), 2.70 (m, 2H, CH2NHCO), 3.75 (d,
1H, CH), 7.10–7.60 (m, 5H, C6H5). M+ (m/z): 333.
Sample Preparation
DNA stock solutions at a concentration of 0.01 g L−1 were
prepared by accurately weighing the appropriate amount of
lyophilized material and dissolving in 0.09 mol L−1 Tris–
HCl aqueous buffer solution (pH 8.0) under gentle mag-
netic stirring for at least 12 h prior to use. Surfactant solu-
tions, both in the absence and in the presence of 0.01 g L−1
DNA, were prepared by accurately weighing the desired
amount of surfactant and dissolving in Tris–HCl buffer
solution (pH 8.0) or DNA stock solution, respectively. Pre-
viously, the solubilization of the new surfactants was tried
in water but solubility was lower than that in the Tris–HCl
buffer solution, presumably due to the presence of the
uncharged form of the surfactant, with low water solubility,
eventually resulting from partial deprotonation of the LAA
in the aqueous media. LAA solutions were prepared in the
concentration range 0.1–2 mg mL−1 chosen due to the low
solubility of LAA at concentrations above 2 mg mL−1 and
The products were obtained in a solid state form, precipi-
tated and separated from the reaction media. A yield of
approximately 85% was attained.
The final products obtained in a solid form were identi-
fied by high-performance liquid chromatography–tandem
mass spectrometry (HPLC-MS/MS) and characterized by
nuclear magnetic resonance (NMR).
The samples analyzed were prepared by collecting 2 mg
of the reaction and adding 2 mL of dimethylsulfoxide for a
final concentration of 1 mg mL−1
.
The HPLC analyses were performed on a Waters Alli-
ance 2695 (Waters®, Dublin, Ireland) equipped with a qua-
ternary pump, solvent degasser, auto sampler, and column
oven, coupled to a Photodiode Array Detector Waters
996 PDA (Waters®, Dublin, Ireland). The tandem mass
spectrometer (MS/MS) used was
a
MicroMass
J Surfact Deterg (2020)

J Surfact Deterg
conductivity masking by the buffer at LAA concentrations
below 0.1 mg mL−1
DNA) was studied by conductometry. The conductivity
profiles for the studied LAA solutions (in the absence of
DNA) are shown in Fig. 2.
.
The CMC of the pure LAA in buffer solution was deter-
mined from the breakpoint in the specific conductivity
against LAA concentration plots (Fig. 2), which show two
linear regions due to the LAA self-assembly process. At
concentrations above the CMC, the slope decreases due to
lower conductivity of micelles compared to free surfactant
monomers since mobility of condensed counterions at the
micellar interface is restricted while the larger micellar size
leads to increased friction with the solvent and thus
decreased diffusion coefficient (Das et al., 2016; Ali et al.,
2009; Faustino et al., 2009a, 2009b).
Conductance Measurements
Conductance of LAA and LAA-DNA solutions was mea-
sured at 20 ꢀC with an uncertainty of less than ꢂ0.1% using
a digital conductivity meter (XS Instruments, Carpi, Italy)
equipped with a conductivity cell with platinum electrodes
and a cell constant of 1.11 cm−1 and calibrated with stan-
dard KCl solutions (Crison Instruments, Allela, Spain) in
the appropriate concentration range.
EB Exclusion Assay
The average degree of micellar ionization (α) was taken as
the ratio of the slopes above and below the CMC (Ali et al.,
2009; Das et al., 2016; Faustino et al., 2009a, 2009b), which
allowed determination of the degree of counterion binding to
micelles (β), according to Equation (1):
To 20 μL of a DNA solution (0.01 mg mL−1) in 0.09 mol L−1
Tris–HCl buffer (pH 8.0) was added 10 μL of EB solution
(0.5 mg mL−1). Different amounts of each LAA stock solu-
tion (2 mg mL−1) in Tris–HCl buffer were added to the
DNA/EB solution in a 96-well microplate and diluted with
Tris–HCl buffer to a final volume of 100 μL. Fluorescence
intensity was measured in a microplate reader (Zenyth 3100
from Anthos, Salzburg, Austria). The excitation and emission
wavelengths were set at 480 and 600 nm, respectively.
β = 1−α:
ð1Þ
The standard ΔGꢀmic for the LAA systems in the absence
of DNA was calculated from Equation (2) based on the
phase separation model for ionic surfactants with monova-
lent counterions in the presence of excess electrolyte (Zana,
1996, 2002):
ΔGꢀmic = RT ð1 + βÞ lnCMC
ð2Þ
Results and Discussion
where R is the ideal gas law constant and T the absolute
temperature. The thermodynamic parameters obtained for
the studied pure LAA solutions are summarized in Table 1.
The hydrophobic effect serves as the driving force for
micelle formation, which is thermodynamically favored
Self-Assembly Behavior of Pure LAA in Aqueous
Media
The self-assembly behavior of the biosynthesized LAA in
aqueous medium (in the absence and in the presence of
Fig. 2 Conductivity profile for the studied LAA in 0.09 Mol L−1 Tris–HCl buffer (pH 8.0) at 293 K. The insert represents the conductivity pro-
file for C12Phe
J Surfact Deterg (2020)

J Surfact Deterg
Table 1 Critical micelle concentration (CMC), degree of micelle ionization (α), degree of counterion binding (β), and standard Gibbs energy of
micellization (ΔGꢀmic) for the lipoamino acids studied in 0.09 Mol L−1 Tris buffer (pH 8.0) at 293 K
Surfactant
CMC/mg mL−1 (mmol L−1
)
α
β
ΔGꢀmic/kJ mol−1
(C12Cys)2
0.15 (0.25)
0.85 (2.72)
0.43 (1.31)
0.63
0.80
0.66
0.37
0.20
0.34
−27.7
−17.7
−22.1
C
C
12Lys
12Phe
Note: Values in brackets refer to CMC in Mmol L−1
.
(Zana, 1996, 2002; He et al., 2011). Accordingly, the
the surfactant structure. A similar effect has been reported
for quaternary ammonium ester surfactants (esterquats) of
the same alkyl chain length, where the ester bond orienta-
tion as O(C O)(CH2)10CH3 in regular esterquats or as
(C O)O(CH2)11CH3 in betaine esters shifted the CMC
from 0.7 to 5 mmol L−1, respectively (Para et al., 2016).
Gemini amide quats with dodecyl tails and a tetramethylene
spacer, whose structure is very similar to (C12Cys)2, show
an almost identical CMC value of 0.23 mmol L−1 (Hoque
et al., 2012).
On the other hand, the additional amino group at the
lysine side chain increases the hydrophilicity of C12Lys,
which improves water solubility and lowers the tendency
of the surfactant to form micelles in aqueous solution, thus
contributing to a higher CMC (Colomer et al., 2011; Col-
omer et al., 2012).
The solubility of the cationic surfactants in 20 mM Tris
buffer pH 8 is higher than in (more acidic) water which can
be attributed to a decrease in Krafft temperature (TK)
resulting from the increase in ionic strength due to the
buffer. Solubility of monomeric ionic surfactants usually
increases with temperature up to the critical micelle temper-
ature which for most ionic detergents is identical to TK
(Pérez et al., 2014; Pinazo et al., 2011; Zana, 2002). Elec-
trolyte addition is known to modulate both the CMC and
the TK of ionic surfactants in aqueous solution (Islam et al.,
2015). Islam et al., (2015) studied the effect of inorganic
sodium salts on the TK and CMC of anionic surfactant
benzyldimethylhexadecylammonium chloride in aqueous
media. The authors found that kosmotropic anions such as
SO42−, F−, and CO32−, which are extensively hydrated in
the bulk, lower TK with increasing concentration of the
electrolytes while chaotropic ions like Br− and I− tend to
form ion pairs with the surfactant, resulting in decreased
monomer solubility and increased TK (Islam et al., 2015).
Conductivity measurements and isothermal titration calo-
rimetry for Gemini surfactants also showed that the nature
of the counterion strongly influenced CMC and micellar
ionization degree, which were correlated with the
Hofmeister (lyotropic) series of counteranions (Jiang et al.,
2004). The size of the counterion has also been shown to
influence micellar growth in N-dodecyl alaninate surfac-
tants, which is also promoted in the presence of buffer due
to the electrolyte screening effect (Takeuchi et al., 2002).
values of ΔGꢀ for the self-assembly process of the stud-
mic
ied LAA are negative, as shown in Table 1. As expected,
the Gemini surfactant, with two hydrocarbon chains,
showed higher |ΔGꢀmic| as a result of increased hydropho-
bic interactions. The more hydrophobic Gemini surfactant
also possessed the highest β while the more hydrophilic
C12Lys with a lysine headgroup had the lowest β (Table 1).
More hydrophobic surfactants tend to form larger micelles,
which reduce the surface area per headgroup, increasing
surface charge density and therefore the degree of counter-
ion binding (Faustino et al., 2009a, 2009b, 2010; Pérez
et al., 2014; Pinazo et al., 2011). The additional amino
group of C12Lys at the side chain of the amino acid residue
contributes to increased electrostatic repulsion among the
headgroups and thus is more likely to form smaller aggre-
gates with higher surface area per head group, resulting in
lower counterion binding (Faustino et al., 2009a, 2009b,
2010; Pérez et al., 2014; Pinazo et al., 2011). A decrease in
β has been associated to an increase in the binding ability
of surfactants toward DNA (Bhadani and Singh, 2009);
thus, the relatively low β values obtained for the studied
LAA suggest formation of small micelles with low inter-
face charge density, which can be attributed to the large
hydrophilic amino acid headgroup.
The CMC values for the studied LAA increased in the
order
(C12Cys)2 < C12Phe < C12Lys
due
to
the
corresponding decrease in hydrophobicity of the LAA
(Table 1). The micellization process is mainly governed by
the number and length of the alkyl chains, and also by the
nature of the polar headgroup. Thus, among the studied
LAA, the Gemini surfactant (C12Cys)2 produced lower
CMC than the other single-tail surfactants of the same alkyl
chain length, as expected, due to enhanced hydrophobic
interactions that result from the presence of the additional
dodecyl chain (Faustino et al., 2010; Pérez et al., 2014; Pin-
azo et al., 2011). The CMC of the studied Gemini LAA is
slightly higher than the ones reported for cationic Gemini
surfactants of the same alkyl chain length that also contain
a disulphide bond in the spacer, obtained by reaction of
dodecyldimethylglycine betaine with cystine or cystamine,
0.065 and 0.120 mmol L−1, respectively (Pinazo et al.,
2011). This can be attributed to their quaternary ammonium
groups and to the different position of the amide bond in
J Surfact Deterg (2020)

J Surfact Deterg
The nature and size of the amino acid residue can also
affect the CMC and TK of LAA. Haldar and Maji studied
N-n-hexadecanoyl amino acid amphiphiles with hydropho-
bic Cα-side chains in Tris buffer (pH 9.3) and showed from
differential scanning calorimetry measurements that TK and
enthalpy of solution were affected by the type of the amino
acid side chain (Haldar and Maji, 2013). The authors elimi-
nated the effect of the counterion on self-assembly, solubil-
ity, and ionization behavior of the amphiphiles by using a
large excess of Tris buffer compared to the CMC of the sur-
factants (Haldar and Maji, 2013). Ohta et al., (2003) stud-
ied N-hexadecanoyl amino acid surfactants and observed
that TK increased with decreasing size of the amino acid
residue.
Increasing the concentration of the chloride counterions
by addition of Tris–HCl buffer also contributes to decrease
the CMC of the studied cationic LAA. The electrostatic
repulsion between the charged head groups is reduced due
to screening of the surface charge of micelles by the associ-
ated counterions resulting in a decrease in CMC and TK.
Similar results have been found by Colomer et al., (2012)
for pH-sensitive dodecyl cationic surfactants from lysine
and also by our group when comparing the CMC of anionic
Gemini surfactants derived from cysteine with octyl chain
length in water (CMC = 8.2 mmol L−1; Faustino et al.,
2010) and in 0.020 M phosphate buffered-saline at pH 7.4
micellization and thus the CMC (Chatterjee et al., 2002).
Despite the bulky phenyl group in C12Phe, which could
hinder micellization, additional micelle stabilization by π-π
stacking interactions between the aromatic rings has been
described for phenylalanine-derived surfactants (Chatterjee
et al., 2002). CMC values of 0.126 and 0.154 mmol L−1
(Joondan et al., 2014) have been reported from surface ten-
sion and conductivity measurements, respectively, for the
dodecyl ester of phenylalanine (hydrochloride salt), which
are lower than the one obtained for the corresponding dode-
cyl amide studied. However, since a similar CMC of
0.212 mmol L−1 has been obtained for the hexadecyl amide
of phenylalanine quaternized at the α-amino group (Jadhav
et al., 2008), a higher value is expected for C12Phe due to
the shorter dodecyl chain and thus less hydrophobic
character.
The CMC values obtained for the studied LAA are also
lower than the one reported for the commercial quaternary
ammonium surfactant of the same alkyl chain length, dode-
cyltrimethylammonium bromide (DTAB), which has a
CMC of 16 mmol L−1 in aqueous solution (Pérez et al.,
2014). Contrary to quaternary ammonium surfactants,
whose charge does not depend on the pH of the solution,
amino acid-based surfactants are pH-sensitive. Thus,
increasing the pH above the pKa of the surfactant increases
the percentage of uncharged species and can lead to
decrease in solubility (Chatterjee et al., 2002; Colomer
et al., 2012; Pinazo et al., 2011) and/or CMC due to forma-
tion of mixed micelles with reduced electrostatic repulsions
between the headgroups due to the presence of the
uncharged species (Colomer et al., 2012; Mezei
et al., 2012).
The α-amino groups of cysteine, phenylalanine, and
lysine have pKa values of 9.1, 9.3, and 10.7, respectively,
while the ε-amino group at the lysine side chain has a pKa
value of 10.5 (Colomer et al., 2012, 2013; Pérez et al.,
2014; Pinazo et al., 2011). Thus, the studied LAA are
expected to be mainly in their protonated form in Tris–HCl
buffer (pH 8.0) medium; however, significant pKa shifts
have been reported for amino acid–based surfactants upon
micellization as aggregation can induce changes in surfac-
tant protonation (Colomer et al., 2011, 2012; Mezei et al.,
2012; Pinazo et al., 2011). Formation of micelles from cat-
ionic LAA has been reported to shift the acid-base equilib-
rium toward the acid species corresponding to the
protonated surfactant, since micellar counterion binding
stabilizes electrostatic repulsions among the charged
headgroups and reduces their proton releasing ability
(Colomer et al., 2012, 2013).
containing 0.15 mol L−1 NaCl (CMC
=
0.612 ꢂ
0.014 mmol L−1) (Faustino et al., 2015), both values
obtained at 25 ꢀC. Moreover, Colomer et al., (2012) also
found that pKa of the amino group for their cationic lysine-
based surfactants usually increased with the increase in
ionic strength, which suggests that the studied amino acid–
based surfactants are mainly in their protonated form at the
pH = 8 of the Tris buffer employed.
Regarding the lysine surfactants, it has been shown that
the position of the cationic charge either at the α- or
ε-amino group of the side-chain does not significantly
affect the CMC. Nε-dodecanoyl lysine methyl ester hydro-
chloride has a CMC value of 5.5 mmol L−1 while Nα-
dodecanoyl lysine methyl ester hydrochloride has a CMC
of 7.2 mmol L−1 (Colomer et al., 2011, 2012). The studied
lysine dodecyl amide C12Lys has a slightly lower value,
close to the CMC value of 1.8 mmol L−1 found for arginine
dodecyl amide dihydrochloride (Pinazo et al., 2011), which
also has an amide bond (instead of ester) at the same posi-
tion of the studied LAA and unsubstituted nitrogen basic
functions.
The phenylalanine-derived surfactant C12Phe has a lower
CMC than the corresponding lysine surfactant C12Lys due
to the hydrophobic phenyl group at the side chain of the
amino acid residue. Thermodynamic studies have revealed
that the hydrophobicity of the Cα-substituents in amino
acid-based surfactants lower the free energy of
Furthermore, pKa values of weak acids can be influenced
by the ionic strength of the medium since increasing the
ionic strength stabilizes charge separation and thus
increases the acid dissociation constant. For weak acids like
J Surfact Deterg (2020)

J Surfact Deterg
alkylammonium cations, pKa values have been reported to
increase with ionic strength (Colomer et al., 2012). The
influence of the ionic strength on the pKa of lysine- and
arginine-derived surfactants has been studied and no signif-
icant differences have been observed between apparent pKa
values measured in water and those measured in aqueous
saline solutions up to 0.5 mol L−1 NaCl added salt
(Colomer et al., 2013). Thus, ionic strength is expected to
have a minor effect on the apparent pKa of the studied
LAA, suggesting that they remain mainly in their proton-
ated form at pH 8.0 and hence behave like cationic surfac-
tants. On the other hand, high ionic strength of buffer
solutions have been reported to lower the CMC due to the
electrolyte screening effect, which reduces the electrostatic
repulsions between the surfactant headgroups in the micel-
lar aggregates thus promoting micelle formation at lower
surfactant concentrations (Faustino et al., 2009a, 2009b;
Zhu and Evans, 2006).
Fig. 3 Conductivity profile for the studied LAA in the presence of
0.01 g L−1 DNA in 0.09 Mol L−1 Tris–HCl buffer (pH 8.0) at 293 K.
curved lines are just guides, not models. Straight lines result from
regression analysis in the concentration ranges where linearity
between conductivity and surfactant concentration occurred
Except for the C12Phe-DNA system, deviation from linear-
ity occurred within a narrow LAA concentration range near
CACapp, where the increase in conductivity with LAA con-
centration is interrupted by a slow rise immediately
followed by a sharp and linear increase up to the CMC*
(Fig. 3). Nonlinearity between CAC and CMC has been
observed in polymer-surfactant systems and has been inter-
preted by assuming that more than one type of aggregate is
developing (Chatterjee et al., 2002), as the result of confor-
mational changes (Mitra et al., 2008) or self-assembly of
the polymer-surfactant complex is involved (Chatterjee
et al., 2002). Transformation of dsDNA from an extended
coil state to a compact globule condition has been demon-
strated by fluorescence microscopy studies and the onset of
DNA compaction are in agreement with the CAC value
obtained from conductivity measurements, with a narrow
concentration range of coexistence of coils and globules
(Dias et al., 2008).
LAA–DNA Interactions
Self-Assembly Behavior of LAA in the Presence of DNA
Electrical conductivity has been used to provide valuable
information on the association process between ionic sur-
factants and water-soluble polymers or polyelectrolytes,
such as polyvinyl pyrrolidone (Ali et al., 2009), inulin (Dan
et al., 2009), bovine serum albumin (BSA; Branco et al.,
2015; Faustino et al., 2009a, 2009b), lysozyme (Chatterjee
et al., 2002), gelatine (Chatterjee et al., 2002; Mitra et al.,
2008), sodium hyaluronate (Zhu and Evans, 2006), and
sodium carboxymethyl cellulose (NaCMC; Ansari et al.,
2013; Das et al., 2016; Wang et al., 2005). Plots of conduc-
tivity against surfactant concentration in the presence of
polymers and proteins usually show three different linear
regions with two characteristic transition points (Ali et al.,
2009; Branco et al., 2015; Chatterjee et al., 2002; Faustino
et al., 2009a, 2009b). The break point at lowest concentra-
tion corresponds to the critical aggregation concentration
(CAC), representing the onset of cooperative binding,
while the second transition point corresponds to CMC in
the presence of the polymer (CMC*) due to bulk micelle
formation after polymer saturation has been reached.
Although this behavior has also been found in
polyelectrolyte-oppositely charged surfactant systems
(Bracˇicˇ et al., 2015; Chatterjee et al., 2002), conductivity
dependence on cationic surfactant concentration in the pres-
ence of DNA is often more complex due to the tertiary
structure of the biopolymer and strong electrostatic interac-
tions. The conductivity profiles for the studied LAA in the
presence of 0.01 g L−1 DNA are shown in Fig. 3.
The CACapp and CMC* values obtained from the electri-
cal conductivity profiles for the studied LAA-DNA mixed
systems are summarized in Table 2, along with the values
found for the standard ΔGꢀagg for the surfactant-DNA com-
plexation process, which were calculated according to
Equation (3):
ΔGꢀagg = RTð1 + β0lnðCACÞ
ð3Þ
where β0 is the degree of counterion binding to DNA-
surfactant complexes. This value can be obtained from the
ratio of the slopes of the conductivity plots against LAA
concentration above and below CACapp (while the ratio of
the slopes above CMC* and below CAC give the degree of
counterion binding to free micelles) whenever there is a
decrease in the slopes from the CAC to the CMC* concen-
tration regions. For the studied LAA-DNA systems, the
All conductivity profiles exhibit two discontinuities,
corresponding to apparent CAC (CACapp) and CMC*.
J Surfact Deterg (2020)

J Surfact Deterg
slope below CAC is lower than the slope above CAC (and
also lower than the slope above CMC*); thus, β0 could not
be determined by this method. Therefore, the β value
obtained previously, corresponding to the degree of coun-
terion binding to micelles in the absence of DNA, was used
instead of β0 in Equation (3). Although lower counterion
binding is usually found for polymer-bound micelles due to
their smaller size (Ali et al., 2009; Bracˇicˇ et al., 2015;
Faustino et al., 2009a, 2009b), the β values obtained for the
free micelles are low and thus are not expected to differ sig-
nificantly from β0.
For the studied LAA-DNA systems, initial binding of
LAA to the polyelectrolyte is expected to take place
through electrostatic interactions between the cationic
headgroups of the surfactant and the anionic phosphate
groups of the DNA backbone along with hydrophobic
interactions involving DNA-bound LAA molecules (Zhou
et al., 2012). For aromatic LAA, such as C12Phe, additional
π-π interactions between the phenyl substituent of the phe-
nylalanine headgroup and the DNA nucleobases can occur
(Zhou et al., 2012).
Cooperative binding of cationic surfactants to DNA usu-
ally occurs at a critical surfactant concentration (i.e., CAC)
much lower than CMC due to the strong electrostatic inter-
actions, similar to other polyelectrolyte-oppositely charged
surfactant systems (Ansari et al., 2013; Chatterjee et al.,
2002; Das et al., 2016; Wang et al., 2005). The CAC of
DDA and DTAB was found to shift to lower values in the
presence of dsDNA and synthetic polyribonucleotides in
buffer solution (1 mmol L−1 Tris–HCl, pH 7.2,
10 mmol L−1 NaCl) as compared to CMC of pure DDA
and DTAB buffer solutions (Petrov et al., 2002). From
DNA does occur. Moreover, surfactant concentration
corresponding to formation of free micelles in solution
(CMC) was higher in the presence of DNA, suggesting for-
mation of a LAA-DNA complex responsible for hindering
the micellization process (Ali et al., 2009; Branco et al.,
2015; Chatterjee et al., 2002; Faustino et al., 2009a,
2009b). DNA complexation by the studied LAA was fur-
ther confirmed by the EB exclusion assay, as will be dis-
cussed below.
The high values obtained for CACapp may reflect com-
plex coacervation due to self-assembly of a LAA-DNA
complex formed at lower LAA concentrations. These LAA
concentrations were not detected by conductivity measure-
ments under the experimental conditions. The buffer used
had a high ionic strength, which hides the conductivity of
highly diluted ionic solutions, precluding further studies
of more diluted LAA-surfactant solutions. Often, CAC
values that occur at very low surfactant concentration cannot
be detected by conductometry. This was the case for inulin-
alkyltrimethylammonium surfactants (Takeuchi et al., 2002),
gelatine-DTAB (Mitra et al., 2008), NaCMC-alkyltrimethy-
lammonium surfactants (Takeuchi et al., 2002), and NaCMC-
cationic gemini surfactant tetramethylene-1,4-bis(dimethy-
ltetradecylammonium bromide) (14-4-14) systems (Das et al.,
2016). For the octadecyltrimethylammonium (OTAB)-inulin
system, CAC values, although determined by tensiometry,
could not be detected by either conductivity, turbidity, or vis-
cometry (Dan et al., 2009). Surface tension measurements
are sensitive mainly to the concentration of the mono-
meric form of the surfactant while conductometric
methods depend on the electrical conductivity of all the
ionic species present in solution. Regarding polymer-
surfactant systems, tensiometry can reveal the nature of
the interaction of the surfactant with the polymer both at
the interface and in the bulk, while conductivity is
restricted to bulk phase properties, being less sensitive
for detection of self-assembly processes at low surfactant
concentrations. Surface tension measurements were per-
formed for the studied LAA solutions; however, the long
equilibration times did not allow for a constant surface
tension value to be reached due to separation of a minor
amount of coacervate from the solution upon standing for
a long time. It is known that surfactants with high surface
activity (i.e., with equilibrium surface tension
≤30 mN m−1) and low CMC can lead to long equilibra-
tion times due to low adsorption dynamics and may
require several hours of measurement for surface tension
to reach a constant value (Bhadani and Singh, 2009;
Branco et al., 2015; Faustino et al., 2010; Mitra et al.,
2008; Zana, 2002).
Table 2, it is evident that the CAC values (CACapp
)
obtained are all higher than the CMC for the corresponding
LAA (Table 1), which usually means that formation of free
micelles is favored over polymer-bound ones, and DNA
complexation does not occur. However, comparison of the
conductivity profiles for the studied LAA in the absence
and in the presence of DNA (Figs. 2 and 3, respectively)
show striking differences, suggesting that interaction with
Table 2 Apparent critical aggregation concentration (CACapp), criti-
cal micelle concentration in the presence of biopolymer (CMC*), and
standard Gibbs energy of aggregation (ΔGꢀagg) for the lipoamino
acids studied in the presence of 0.01 g L−1 DNA in Tris buffer
(0.09 Mol L−1, pH 8.0) at 293 K
CACapp/mg mL−1
CMC*/mg mL−1
(mmol L−1
ΔGꢀ
/
agg
Surfactant (mmol L−1
)
)
kJ mol−1
(C12Cys)2
12Lys
C12Phe
0.63 (1.09)
1.12 (3.57)
0.90 (2.71)
1.76 (3.06)
1.74 (5.56)
1.60 (4.82)
−24.7
−20.4
−23.4
C
Mixtures of polyelectrolytes and surfactants of opposite
charges often show strong tendency to form insoluble com-
plexes, leading to phase separation either by coacervation
Note: Values in brackets refer to critical concentrations in mmol L−1
.
J Surfact Deterg (2020)

J Surfact Deterg
or by precipitation due to reduced electrostatic repulsions
between the polymeric chains upon charge neutralization
(Das et al., 2016; Dias et al., 2008). Compensation of the
electric charges of the polyelectrolyte by increasing the
concentration of oppositely charged surfactant eventually
leads to the formation of electrically neutral complexes of
increased hydrophobicity that can separate out from the
solution (Das et al., 2016; Dias et al., 2008). Further surfac-
tant addition leading to charge overcompensation may
result in redissolution of coacervates or formation of turbid
colloidal dispersions (Dan et al., 2009; Das et al., 2016;
Mitra et al., 2008). Free micelle formation in the bulk phase
upon reaching CMC* can also result in coacervate
(or precipitate) solubilization with depletion of visible tur-
bidity (Mitra et al., 2008). Coacervation is mainly depen-
dent on the balance between electrostatic, hydrophobic, and
solvent interactions (Das et al., 2016). Several factors are
known to influence complex coacervation, such as poly-
electrolyte composition and charge density, chain length
and flexibility, polymer molecular weight, micelle size,
shape and surface charge density, temperature, pH, and
ionic strength of the medium (Das et al., 2016).
The behavior of the studied LAA-DNA systems was
similar to the one found for cationic Gemini surfactant
14-4-14 in the presence of anionic polymer NaCMC (Das
et al., 2016). The tensiometric profile in the cited study sys-
tem showed that 14-4-14 monomers interacted with the
polymer at very low surfactant concentration with a CAC
value in the micromolar range (Das et al., 2016). For sur-
factant concentrations above CAC, the surfactant-NaCMC
complex self-assembled to form a turbid coacervate phase
once polymer saturation concentration, Csat, was reached.
Above Csat, the system became a turbid colloidal solution
and free surfactant micelles were formed above CMC* that
coexisted with the coacervates, even at higher surfactant
concentrations (Das et al., 2016). On the other hand, the
conductivity profile for this system showed two distinct
transitions at surfactant concentrations corresponding to the
Csat and CMC* values. No breakpoint was observed in the
concentration range near tensiometric CAC in the experi-
ments performed by tensiometry (Das et al., 2016). Thus,
the two break points in the conductivity plots against sur-
factant concentration for the studied LAA-DNA systems,
CACapp and CMC*, likely correspond to the onset of coac-
ervation and to free micelle formation in the bulk phase,
respectively.
surfactant micelles were formed in solution (Dan et al.,
2009). For the OTAB-inulin system, the conductometric
profile showed no transitions close to the CAC, similar to
the gelatine-cetyl trimethylammonium (CTAB) system, but
the breakpoints corresponding to Csat and CMC* were
observed and agreed with the values obtained via tensiome-
try (Dan et al., 2009). An increase of turbidity for OTAB-
inulin solutions appearing above Csat, due to coacervation,
was not significantly reduced as the surfactant concentra-
tion was further increased, even above CMC*, suggesting
the coexistence of coacervates and free OTAB micelles.
This was also the case for the Gemini 14-4-14-NaCMC
system and for the studied LAA-DNA systems, where tur-
bidity occurred after CACapp and did not disappear in the
post-CMC* region upon micelle formation in the bulk,
being more pronounced at higher surfactant concentrations.
Moreover, for the studied LAA-DNA systems, as well as
for the 14-4-14-NaCMC (Das et al., 2016) and OTAB-
inulin (Dan et al., 2009) systems, a minor amount of
surfactant-polymer complex separated from the solution
after standing for a long time. On the other hand, gelatin-
CTAB complexes also formed turbid colloidal dispersions
due to coacervation but solubilization of the coacervates by
free surfactant micelles led to depletion of visible turbidity
at surfactant concentrations above CMC* (Mitra
et al., 2008).
The conductivity profiles for the studied cationic LAA-
DNA systems show a linear increase of electrical conduc-
tivity with surfactant concentration in the low surfactant
region up to the first breakpoint (CACapp),attributable to
the release of DNA counterions (Na+) into the bulk phase
upon surfactant binding, which also contributes to an
increase in the entropy of the solution that drives coopera-
tive binding (Matulis et al., 2002). After the first break
point, which corresponds to the onset of coacervation, a
sharp increase in electrical conductivity with surfactant
concentration was observed. In addition, an increase in the
slope between the first and the second break points
occurred, reflecting the contribution of surface conduction
(Stern layer conduction). This contribution is typical of col-
loidal dispersions to the overall conductance of the
coacervate-containing colloidal solution (Dan et al., 2009;
Das et al., 2016). Above the second break point (CMC*),
the slope decreased due to condensation of counterions
from the bulk phase to the interface of DNA-free micelles
(Das et al., 2016). The same behavior was observed in the
conductometric profiles for the cationic Gemini surfactant
14-4-14 in the presence of NaCMC (Das et al., 2016) and
for the OTAB-inulin system (Dan et al., 2009). On the
other hand, short-chain alkyltrimethylammonium surfac-
tants, such as DTAB and tetradecyl trimethylammonium
bromide (TTAB), showed a very weak interaction with inu-
lin and only a single break point was detected in the
Similar behavior has also been found for single-chain
cationic surfactant OTAB in the presence of inulin, with
formation of a surfactant-polysaccharide complex at low
OTAB concentrations, as revealed by tensiometry, which
self-aggregated above the CAC, leading to coacervate for-
mation at the polymer saturation concentration Csat (Dan
et al., 2009). At concentrations above CMC*, free
J Surfact Deterg (2020)

J Surfact Deterg
conductivity profile, which corresponded to CMC* (Dan
et al., 2009). The strength of the interaction of the quater-
nary ammonium surfactants with inulin increased with
alkyl chain length, supporting the significant role played by
surfactant hydrophobicity in the complexation process
(Dan et al., 2009).
The CMC* values for the studied LAA-DNA systems
(Table 2) are all higher than CMC values obtained in the
absence of the polyelectrolyte (Table 1), which was
expected since formation of the surfactant-DNA complex
hinders the micellization process. This behavior is an indi-
cation of effective interaction between surfactant and DNA
and it is typical of other surfactant-polymer systems,
including mixed systems formed by the same Gemini LAA
studied (but prepared by chemical synthesis) and BSA
(Branco et al., 2015; Faustino et al., 2009a, 2009b).
exclusion assay. EB is a cationic dye commonly used as a
probe for native DNA that shows enhanced fluorescence
emission upon intercalation between the DNA base pairs in
comparison to residual fluorescence emission in water
(Bhadani and Singh, 2009; Dasgupta et al., 2007).
The extent of LAA binding to DNA can be deter-
mined by the ability of the surfactant to displace interca-
lated EB from the DNA-EB complex, which results in
quenching of the fluorescence signal due to the formation
of the surfactant-DNA complex (Dan et al., 2009; Zhao
et al., 2008). The percentage of quenching (Q) observed
from the displacement of EB from the intercalation com-
plex with DNA upon addition of increasing amounts of
cationic Gemini LAA was determined according to
Equation (4):
Qð%Þ = ð1 – F=F0Þ × 100
ð4Þ
Both CMC* and CACapp for the studied LAA-DNA sys-
tems decreased with increasing LAA hydrophobicity
(Table 2). CMC* has a direct correlation with surfactant
hydrophobicity, as discussed above, for the pure LAA sys-
tems. Thus, (C12Cys)2, a Gemini surfactant, i.e., more
hydrophobic than the surfactants investigated herein, has a
lower CMC* and CACapp, supporting the role of hydropho-
bic interactions in complex coacervation (Ansari et al.,
2013; Matulis et al., 2002). Observing that the values
where F0 and F are the fluorescence intensities at 600 nm
of the DNA-EB complex in the absence and presence of
LAA, respectively.
Although fluorescence quenching was observed at all
surfactant concentrations tested, saturation of the fluores-
cence quenching was not achieved in the LAA concentra-
tion range studied. However, more concentrated solutions
were not analyzed due to solubility problems. Thus, the
accessibility of DNA to EB is preserved to a significant
extent in the presence of the studied LAA, particularly for
the more hydrophilic surfactant C12Lys, as shown in Fig. 4.
The differences in the EB displacement shown by the
different surfactants can be interpreted in terms of the
strength of interaction between LAA and DNA. On the
other hand, these differences can also be due to the type of
self-assembly structure that the surfactant adopts in the
vicinity of the DNA molecules. The fact that 50% fluores-
cence quenching was not achieved even at the highest LAA
concentrations tested suggests a weak interaction of the
LAA with DNA, which can be due to their relatively short
dodecyl chain(s). Surfactant-DNA interaction is usually
enhanced with increasing alkyl chain length, supporting the
role of hydrophobic interactions in the process, in combina-
tion with the electrostatic interactions (Han and Wang, 2011;
He et al., 2011; Husale et al., 2008). The CACapp values
decreased with increasing LAA hydrophobicity, reflecting
the relevance of hydrophobic interactions in complex coacer-
vation (Fig. 4). The same trend has been observed in other
polyelectrolyte oppositely charged surfactant systems.
Among the quaternary ammonium surfactants DTAB,
TTAB, and CTAB, the one with the shorter (dodecyl) chain,
DTAB, underwent very weak interaction with NaCMC
(Ansari et al., 2013). Alkyltrimethylammonium surfactants
with higher alkyl chain lengths were also more interactive
with gelatin than their short-chain homologues
(Mitra et al., 2008).
obtained for ΔGꢀ were all negative (Table 2), coacerva-
agg
tion was a spontaneous process favored over free micelle
formation in the bulk phase in the biopolymer environment
since |ΔGꢀagg| is slightly higher than |ΔGꢀmic| for
micellization in pure LAA solutions (Table 1).
Formation of the LAA-DNA Complex
The binding of the studied cationic LAA with dsDNA was
confirmed from fluorescence measurements in the EB
Fig. 4 Quencher efficiency (%) of the studied LAA at different con-
centrations (below CACapp, above CACapp and below CMC*, and
above CMC*) in the ethidium bromide exclusion assay in
0.09 Mol L−1 Tris–HCl buffer solution (pH 8.0) and 0.01 g L−1 DNA
J Surfact Deterg (2020)

J Surfact Deterg
The high ionic strength of the buffer solution used can
also weaken the LAA-DNA interaction due to the screening
of electrostatic interactions between DNA and the oppo-
sitely charged LAA (Wang et al., 2013; Zhu and Evans,
2006). Addition of salt has been shown to markedly influ-
ence the formation of NaCMC/DTAB complexes, which is
prevented at high ionic strength due to complete salt
screening of the electrostatic attractions between DTAB
and NaCMC (Wang et al., 2005). The (C12Cys)2-BSA sys-
tem has been previously studied in a high ionic strength
medium and a relatively weak interaction was observed
(CAC/CMC ratio of 0.63 in phosphate buffer at pH 7.4
containing 0.15 mol L−1 NaCl; Branco et al., 2015).
complex by π-π interactions between the surfactant phe-
nyl group and the nucleotide bases that impart conforma-
tional restrictions and block the accessibility to EB
(Bhadani and Singh, 2009).
Conclusions
Cationic single-chain and dimeric surfactants derived from
natural amino acids (LAA) were prepared by a green bio-
technological approach using a lipase encapsulated in sol–
gel and their supramolecular behavior in aqueous media
(Tris buffer, pH 8.0) was characterized both in the absence
and in the presence of dsDNA. Conductivity measurements
showed that all LAA self-assembled to form micelles in
solution which occurred at a higher LAA concentration in
the presence of DNA, under conditions where coacervation
occurs. The hydrophobicity of the LAA, which was mainly
determined by the number of dodecyl chains and nature of
the amino acid group, strongly influenced the self-assembly
behavior and interaction with DNA. Formation of a
surfactant-DNA complex was demonstrated from fluores-
cence measurements of the EB exclusion assay, although
results suggest a weak interaction. The phenomenon can be
attributed to the relatively short dodecyl chains, thereby
supporting the major role played by hydrophobic interac-
tions in the LAA-DNA association process in addition to
electrostatic interactions between the negatively charged
DNA phosphate groups and the positively charged amino
groups of the LAA. Moreover, the high ionic strength of
the buffer solution employed may have also contributed to
screening of electrostatic interactions.
The more hydrophobic LAA, (C12Cys)2 and C12Phe,
were able to quench 35.6% and 44.2% of the initial fluores-
cence intensity, respectively, while C12Lys only quenched
15.5% (Fig. 4). Thus, the efficiency of EB exclusion for the
studied LAA decreased in the order C12Phe >
(C12Cys)2 > C12Lys. Although enhanced EB exclusion
from DNA has been associated with increased hydrophobic
interactions due to increase in the alkyl chain length
(Bhadani and Singh, 2009), the studied LAA all have dode-
cyl chains; therefore, the differences observed in the EB
exclusion assay can be attributed to the polarity of the
amino acid group and thus increased hydration resulting in
larger effective size (Dasgupta et al., 2007). A similar trend
has been observed for cationic surfactants with hexadecyl
chains upon introduction of additional hydroxyl substitu-
ents in the head group (Dasgupta et al., 2007). Surfactants
with larger head groups form smaller micellar aggregates
and their higher curvature is associated with less efficient
surface coverage of DNA, leaving parts of the polynucleo-
tide structure accessible to EB binding (Singh et al., 2011).
Among the studied LAA, C12Lys with a headgroup bearing
two polar amine functions (Fig. 1) is the most hydrophilic,
and it has been shown that surfactants with more hydro-
philic and/or more hydrated headgroups are less efficient in
expelling EB due to packing constraints (Dasgupta et al.,
2007). In fact, Degusa et al. … showed that the accessibil-
ity of DNA to EB was preserved to a significantly larger
extent for the more hydrophilic surfactants, in terms of sur-
factant packing, reflecting that the surfactants with more
substituents have a larger headgroup and therefore form
smaller micellar aggregates. Moreover, restrictive confor-
mational freedom has also been related to better binding
capability and enhanced quenching, e.g., cationic
pyridinium surfactants have shown higher efficiency in
EB displacement over quaternary ammonium surfactants
of the same alkyl chain length (Bhadani and Singh,
2009). This fact can explain the higher quenching effi-
ciency of C12Phe over Gemini LAA (C12Cys)2 due to the
presence of the phenyl substituent in the former, which
can provide additional stabilization of the C12Phe-DNA
Acknowledgements This work was supported by Portuguese Sci-
ence Foundation (FCT) strategic project [PEst-OE/SAU/
UI4013/2013] and for funding the project REDE/1518/REM/2005
that allowed the HPLC–MS analysis.
Conflict of interest The authors declare that they have no conflict
of interest.
References
Ali, M. S., Suhail, M., Ghosh, G., & Kamil, M. (2009) Kabir-ud-din,
interactions between cationic gemini/conventional surfactants with
polyvinylpyrrolidone: Specific conductivity and dynamic light scat-
tering studies. Colloids Surfaces A Physicochemical and Engineer-
ing Aspects, 350:51–56.
Ansari, A. A., Kamil, M., & Kabir-ud-din (2013) Polymer-surfactant
interactions and the effect of tail size variation on micellization pro-
cess of cationic ATAB surfactants in aqueous medium. Journal of
Dispersion Science and Technology, 34:722–730.
Bhadani, A., & Singh, S. (2009) Novel gemini pyridinium surfactants:
Synthesis and study of their surface activity, DNA binding, and
cytotoxicity. Langmuir, 25:11703–11712.
J Surfact Deterg (2020)

J Surfact Deterg
Bordes, R., & Holmberg, K. (2015) Amino acid-based surfactants—
Do they deserve more attention? Advances in Colloid and Interface
Science, 222:79–91.
Bracˇicˇ, M., Hansson, P., Pérez, L., Zemljic, L. F., & Kogej, K. (2015)
Interaction of sodium hyaluronate with a biocompatible cationic
Haldar, S., & Maji, S. K. (2013) Role of non-covalent interactions in
the molecular organization of N-n-hexadecanoyl amino acid amphi-
philes with hydrophobic Cα-side chains in Tris buffer (pH 9.3).
Colloids and Surfaces A: Physicochemical and Engineering
Aspects, 420:10–21.
surfactant from lysine:
12043–12053.
A
binding study. Langmuir, 31:
Han, Y., & Wang, Y. (2011) Aggregation behavior of gemini surfac-
tants and their interaction with macromolecules in aqueous solu-
tion. Physical Chemistry Chemical Physics, 13:1939–1956.
He, Y., Shang, Y., Shao, S., Liu, H., & Hu, Y. (2011) Micellization
of cationic gemini surfactant and its interaction with DNA in dilute
brine. Journal of Colloid and Interface Science, 358:513–520.
Hoque, J., Akkapeddi, P., Yarlagadda, V., Uppu, D. S. S. M.,
Kumar, P., & Haldar, J. (2012) Cleavable cationic antibacterial
amphiphiles: Synthesis, mechanism of action, and cytotoxicities.
Langmuir, 28:12225–12234.
Husale, S., Grange, W., Karle, M., Burgi, S., & Hegner, M. (2008)
Interaction of cationic surfactants with DNA: A single-molecule
study. Nucleic Acids Research, 36:1443–1449.
Ilarduya, C. T., Sun, Y., & Duzgunes, N. (2010) Gene delivery by
lipoplexes and polyplexes. European Journal of Pharmaceutical
Sciences, 40:159–170.
Islam, M. N., Sharker, K. K., & Sarker, K. C. (2015) Salt-induced
modulation of the Kraft temperature and critical micelle concentra-
tion of benzyldimethylhexadecylammonium chloride. Journal of
Surfaces and Detergents, 18:651–659.
Jadhav, V., Maiti, S., Dasgupta, A., Das, P. K., Dias, R. S.,
Miguel, M. G., & Lindman, B. (2008) Effect of the headgroup
geometry of amino acid-based cationic surfactants on interaction
with plasmid DNA. Biomacromolecules, 9:1852–1859.
Branco, M. A., Pinheiro, L., & Faustino, C. (2015) Amino acid-based
cationic gemini surfactant-protein interactions. Colloids Surfaces A
Physicochemical and Engineering Aspects, 480:105–112.
Chatterjee, A., Moulik, S. P., Majhi, P. R., & Sanyal, S. K. (2002)
Studies on surfactant–biopolymer interaction. I. Microcalorimetric
investigation on the interaction of cetyltrimetylammonium bromide
(CTAB) and sodium dodecylsulfate (SDS) with gelatin (Gn), lyso-
zyme (Lz) and deoxyribonucleic acid (DNA). Biophysical Chemis-
try, 98:313–327.
Colomer, A., Pérez, L., Pons, R., Infante, M. R., Pérez-Clos, D.,
Manresa, A., … Pinazo, A. (2013) Mixed monolayer of DPPC and
lysine-based cationic surfactants: An investigation into the antimi-
crobial activity. Langmuir, 29:7912–7921.
Colomer, A., Pinazo, A., Garcia, M. T., Mitjans, M., Vinardell, M. P.,
Infante, M. R., … Pérez, L. (2012) pH-sensitive surfactants from
lysine: Assessment of their cytotoxicity and environmental behav-
iour. Langmuir, 28:5900–5912.
Colomer, A., Pinazo, A., Manresa, M. A., Vinardell, M. P.,
Mitjans, M., Infante, M. R., & Pérez, L. (2011) Cationic surfactants
derived from lysine: Effects of their structure and charge type on
antimicrobial and hemolytic activities. Journal of Medicinal Chem-
istry, 54:989–1002.
Dan, A., Ghosh, S., & Moulik, S. P. (2009) Physicochemistry of the
interaction between inulin and alkyltrimethylammonium bromides
in aqueous medium and the formed coacervates. The Journal of
Physical Chemistry. B, 113:8505–8513.
Das, S., Mondal, S., & Ghosh, S. (2016) Interaction of cationic gem-
ini surfactant tetramethylene-1,4-bis(dimethyltetradecylammonium
bromide) with anionic polyelectrolyte sodium carboxymethyl cellu-
lose, with two different molar masses, in aqueous and aquo-organic
(isopropanol) media. RSC Advances, 6:30795–30803.
Jiang, N., Li, P., Wang, Y., Wang, J., Yan, H., & Thomas, R. K.
(2004) Micellization of cationic gemini surfactants with various
counterions and their interaction with DNA in aqueous solution.
The Journal of Physical Chemistry. B, 108:15385–15391.
Joondan, N., Jhaumeer-Laulloo, S., & Caumul, P. (2014) A study of
the antibacterial activity of L-phenylalanine and L-tyrosine esters in
relation to their CMCs and their interactions with 1,2-dipalmitoyl-
sn-glycero-3-phosphocholine, DPPC as model membrane. Microbi-
ological Research, 169:675–685.
Dasgupta, A., Das, P. K., Dias, R. S., Miguel, M. G., Lindman, B.,
Jadhav, V. M., … Maiti, S. (2007) Effect of headgroup on DNA-
cationic surfactant interactions. The Journal of Physical Chemistry.
B, 111:8502–8508.
Matulis, D., Rouzina, I., & Bloomfield, V. A. (2002) Thermodynam-
ics of cationic lipid binding to DNA and DNA condensation: Roles
of electrostatics and hydrophobicity. Journal of the American
Chemical Society, 124:7331–7342.
Dias, R. S., Magno, L. M., Valente, A. J., Das, D., Das, P. K.,
Maiti, S., … Lindman, B. (2008) Interaction between DNA and cat-
ionic surfactants: Effect of DNA conformation and surfactant
headgroup. The Journal of Physical Chemistry. B, 112:
14446–14452.
Faustino, C., Serafim, C., Ferreira, I., Pinheiro, L., & Calado, A.
(2015) Solubilization power of an amino acid-based gemini surfac-
tant towards the hydrophobic drug amphotericin B. Colloids and
Surfaces A: Physicochemical and Engineering Aspects, 480:
426–432.
Faustino, C. M. C., Calado, A. R. T., & Garcia-Rio, L. (2009a) Gem-
ini surfactant-protein interactions: Effect of pH, temperature and
surfactant stereochemistry. Biomacromolecules, 10:2508–2514.
Faustino, C. M. C., Calado, A. R. T., & Garcia-Rio, L. (2009b) New
urea-based surfactants derived from α,ω-amino acids. The Journal
of Physical Chemistry. B, 113:977–982.
Faustino, C. M. C., Calado, A. R. T., & Garcia-Rio, L. (2010)
Dimeric and monomeric surfactants derived from sulfur-containing
amino acids. Journal of Colloid and Interface Science, 351:
472–477.
Ménard, N., Tsapis, N., Poirier, C., Arnauld, T., Moine, L.,
Lefoulon, F., … Fattal, E. (2012) Drug solubilisation and in vitro
toxicity evaluation of lipoamino acid surfactants. International
Journal of Pharmaceutics, 423:312–320.
Mezei, A., Pérez, L., Pinazo, A., Comelles, F., Infante, M. R., &
Pons, R. (2012) Self-assembly of pH-sensitive cationic lysine based
surfactants. Langmuir, 28:16761–16771.
Mitra, D., Bhattacharya, S. C., & Moulik, S. P. (2008) Physicochemi-
cal studies on the interaction of gelatine with cationic surfactants
alkyltrimethylammonium bromides (ATABs) with special focus on
the behaviour of the hexadecyl homologue. The Journal of Physical
Chemistry. B, 112:6609–6619.
Ohta, A., Ozawa, N., Nakashima, S., Asakawa, T., & Miyagishi, S.
(2003) Kraft temperature and enthalpy of solution of N-acyl amino
acid surfactants and their racemic modifications: Effect of the
amino acid residue. Colloid & Polymer Science, 281:363–369.
Para, G., Luczynski, J., Palus, J., Jarek, E., Wilk, K. A., &
Warszynski, P. (2016) Hydrolysis driven surface activity of
esterquat surfactants. Journal of Colloid and Interface Science,
465:174–182.
Faustino, C. M. C., Calado, A. R. T., & Garcia-Rio, L. (2011) Mixed
micelle formation between amino acid-based surfactants and phos-
pholipids. Journal of Colloid and Interface Science, 359:493–498.
Pérez, L., Pinazo, A., Pons, R., & Infante, M. R. (2014) Gemini sur-
factants from natural amino acids. Advances in Colloids and Inter-
face Science, 205:134–155.
J Surfact Deterg (2020)

J Surfact Deterg
Petrov, A. I., Khalil, D. N., Kazaryan, R. L., Savintsev, I. V., &
Sukhorukov, B. I. (2002) Structural and thermodynamic features of
complexes formed by DNA and synthetic polynucleotides with
dodecylamine and dodecyltrimethylammonium bromide. Bio-
electrochemistry, 58:75–85.
Pinazo, A., Pons, R., Pérez, L., & Infante, M. R. (2011) Amino acids
as raw material for biocompatible surfactants. Industrial and Engi-
neering Chemistry Research, 50:4805–4817.
bioencapsulation process for glycosides hydrolysis. Process Bio-
chemistry, 45:841–850.
Wang, X., Li, Y., Li, J., Wang, J., Wang, Y., Guo, Z., & Yan, H.
(2005) Salt effect on the complex formation between polyelectro-
lyte and oppositely charged surfactant in aqueous solution. The
Journal of Physical Chemistry. B, 109:10807–10812.
Wang, Y., Jiang, X., Zhou, L., Yang, L., Xia, G., Chen, Z., &
Duan, M. (2013) Synthesis and binding with BSA of a new gemini
surfactant. Colloids Surfaces A Physicochemical and Engineering
Aspects, 436:1159–1169.
Wettig, S. D., Verrall, R. E., & Foldvari, M. (2008) Gemini surfac-
tants: A new family of building blocks for non-viral gene delivery
systems. Current Gene Therapy, 8:9–23.
Zana, R. (1996) Critical micellization concentration of surfactants in
aqueous solution and free energy of micellization. Langmuir, 12:
1208–1211.
Zana, R. (2002) Dimeric and oligomeric surfactants. Behaviour at
interfaces and in aqueous solution: A review. Advances in Colloid
and Interface Science, 97:205–253.
Zhao, X., Shang, Y., Hu, J., Liu, H., & Hu, Y. (2008) Biophysical
characterization of complexation of DNA with oppositely charged
Gemini surfactant 12-3-12. Biophysical Chemistry, 138:144–149.
Zhou, T., Xu, G., Ao, M., Yang, Y., & Wang, C. (2012) DNA com-
paction and multi-molecular DNA condensation induced by cat-
Serafim, C., Ferreira, I., Rijo, P., Pinheiro, L., Faustino, C.,
Calado, A.,
& Garcia-Rio, L. (2016) Lipoamino acid-based
micelles as promising delivery vehicles for monomeric
amphotericin B. International Journal of Pharmaceutics, 497:
23–35.
Singh, J., Michel, D., Chitanda, J. M., Verrall, R. E., & Badea, I.
(2012) Evaluation of cellular uptake and intracellular trafficking as
determining factors of gene expression for amino acid-substituted
gemini surfactant-based DNA nanoparticles. Journal of
Nanobiotechnology, 10:7.
Singh, J., Yang, P., Michel, D., Verrall, R. E., Foldvari, M., &
Baeda, I. (2011) Amino acid-substituted gemini surfactant-based
nanoparticles as safety and versatile gene delivery agents. Current
Drug Delivery, 8:299–306.
Takeuchi, N., Asakawa, T., Inoh, M., & Miyagishi, S. (2002) Micellar
growth of N-dodecanoyl-alaninates with different counterions and
its quantitative relation with some factors. Colloids Surfaces A
Physicochemical and Engineering Aspects, 197:125–132.
ionic imidazolium gemini surfactants. Colloids Surfaces
Physicochemical and Engineering Aspects, 414:33–40.
A
Zhu, D. M., & Evans, R. K. (2006) Molecular mechanism and ther-
modynamics study of plasmid DNA and cationic surfactants inter-
actions. Langmuir, 22:3735–3743.
Vila-Real, H., Alfaia, A. J., Rosa, M. E., Calado, A. R., &
Ribeiro, M. H. L. (2010) An innovative sol–gel naringinase
J Surfact Deterg (2020)