Supramolecular assistance between cyclodextrins and didecyldimethylammonium chloride against enveloped viruses: Toward eco-biocidal formulations

Article abstract of DOI:10.1016/j.ijpharm.2016.08.057

Nosocomial infections have emerged as important causes of morbidity and mortality in immunocompromised individuals. In this respect, biocides are widely used in hospitals leading to resistant microorganisms. We show here that cyclodextrins can remarkably boost the virucidal activity of di-n-decyldimethylammonium chloride. These oligosaccharides synergistically work with the biocide affording a noticeable reduction of the active virucide concentration between 40 and 85%. Partial replacement of a significant amount of the biocide by eco- and bio-compatible cyclodextrins whilst maintaining the same activity is of great interest as it allows the reduction of the toxicological drawbacks of classical biocide mixtures.

Full text of DOI:10.1016/j.ijpharm.2016.08.057

Accepted Manuscript
Title: Supramolecular Assistance between Cyclodextrins and
Didecyldimethylammonium Chloride against Enveloped
Viruses: Toward Eco-biocidal Formulations
¨
Author: Loıc Leclercq Anny Dewilde Jean-Marie Aubry
´
Veronique Nardello-Rataj
PII:
S0378-5173(16)30807-9
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.ijpharm.2016.08.057
IJP 16033
To appear in:
International Journal of Pharmaceutics
Received date:
Revised date:
Accepted date:
25-7-2016
23-8-2016
26-8-2016
¨
Please cite this article as: Leclercq, Loıc, Dewilde, Anny, Aubry, Jean-
´
Marie, Nardello-Rataj, Veronique, Supramolecular Assistance between
Cyclodextrins and Didecyldimethylammonium Chloride against Enveloped
Viruses: Toward Eco-biocidal Formulations.International Journal of Pharmaceutics
http://dx.doi.org/10.1016/j.ijpharm.2016.08.057
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.

Supramolecular Assistance between Cyclodextrins and
Didecyldimethylammonium Chloride against Enveloped
Viruses: Toward Eco-biocidal Formulations
Loïc Leclercq,*,a Anny Dewilde,b Jean-Marie Aubry,a and Véronique Nardello-Rataj*,a
a Univ. Lille, CNRS, ENSCL, UMR 8181 – UCCS - Equipe CÏSCO, F-59000 Lille, France.
Phone: (+33) 320 434 448, Fax: (+33) 320 434 260, E-mail: veronique.rataj@univ-lille1.fr
and loic.leclercq@univ-lille1.fr
b
Univ. Lille, Faculté de Médecine, UPRES EA 3610, Institut de Microbiologie, Laboratoire
de Virologie, CHRU Lille, F-59037 Lille Cedex, France.
1

Graphical abstract.
Abstract. Nosocomial infections have emerged as important causes of morbidity and
mortality in immunocompromised individuals. In this respect, biocides are widely used in
hospitals leading to resistant microorganisms. We show here that cyclodextrins can
remarkably boost the virucidal activity of di-n-decyldimethylammonium chloride. These
oligosaccharides synergistically work with the biocide affording a noticeable reduction of the
active virucide concentration between 40 and 85%. Partial replacement of a significant
amount of the biocide by eco- and bio-compatible cyclodextrins whilst maintaining the same
activity is of great interest as it allows the reduction of the toxicological drawbacks of
classical biocide mixtures.
Keywords. Di-n-decyldimethylammonium chloride, Cyclodextrins, Host-Guest Chemistry,
Lipids Extraction, Virus inactivation, Synergistic Formulation, Eco-biocide.
2

1. Introduction
Amphiphilic Quaternary Ammonium Compounds (QACs) are organic cations that contain
one, two or three long alkyl chain(s) (typically from C8 to C18) and methyl or benzyl groups
covalently attached to a central positively charged nitrogen atom. Due to their property of
adsorption especially on negatively charged surfaces, they are mainly used as disinfectants,
fabric softeners and antistatic agents but they can also be found as additives for various
technical applications. They are also widely used in catalysis, better known under the name of
Phase Transfer Catalysts (PTCs) like the commercially important benzyltrimethylammonium
or methyltrioctylammonium chlorides. In 2004, their world-wide annual consumption was
reported as 500,000 tons (Hauthal, 2004). Benzalkonium, [BA], benzethonium, [BZ], and di-
n-decyldimethylammonium, [DiC10], chlorides are widely used in hospitals to prevent and
control pathogen infections (Nardello-Rataj & Leclercq, 2014). Indeed, QACs are membrane-
active agents interacting with lipids and proteins prior to the plasma membrane disruption
(Denyer, 1995; Nardello-Rataj & Leclercq, 2014). Unfortunately with the long-term used of
biocides, a certain tolerance among some types of pathogens occurs (Gerba, 2015).
Consequently, some pathogens were found to have reduced susceptibility to QACs and the
disinfectant can no longer be used against this particular strain (Moore et al., 2008). In some
cases, the resistance against biocides can also lead to resistance to antibiotics (Gilbert &
McBain, 2003) Furthermore, the use of biocides is regulated in both the United States of
America and the European Union. In order to improve efficacy and decrease resistance effect,
mixtures of biocides are often encountered in formulations but they may also lead to an
increase of ecotoxicity and/or sensitivity due to cumulative damages. Moreover, nosocomial
infections are not limited to bacteria or fungi. Indeed, viral infections can be acquired in
healthcare settings by person to person or by fomites transmission. Viruses represent 32% of
pediatric nosocomial infections (Valenti et al., 1980). Respiratory syncytial virus (RSV) is the
3

major cause of lower respiratory tract infections and hospital visits during infancy (Hall,
1977). Another common viral disease is caused by herpes simplex (HSV) leading to several
distinct medical disorders (orofacial and genital herpes, whitlow, keratitis or encephalitis,
Aitken & Jeffries, 2001). People with immature or suppressed immune systems (newborns,
transplant recipients, AIDS) are prone to severe complications from HSV infections. More
recently, between 2013 and 2016, Ebola virus epidemic has resulted in at least 28 646
suspected cases and 11 323 confirmed deaths in West Africa (Ebola situation report, 2016).
To avoid Ebola transmission, infection-control programs have been developed with the use of
disinfectants including QACs (Chepurnov, 2003) but as pointed out by Moore and Payne
(2004): “viruses are more resistant than bacteria or fungi to the QACs”.
In this context, natural cyclodextrins (CDs) have been shown to be effective for the
inactivation of pathogens including viruses. CDs constitute a family of macrocyclic
compounds built up from six, seven or eight -D-glucopyranoside units linked 14 (-, -
and -CD, respectively, see Figure 1). The structure of CDs, involving a hydrophobic cavity
and a hydrophilic exterior, enables them to form water-soluble inclusion complexes with the
hydrophobic part of numerous molecules (Szejtli, 1998). They are very useful in various areas
such as catalysis (Leclercq, 2005), textile (Leclercq, 2016), pharmaceutical (Jambhekar &
Breen, 2016a,b, Leclercq & Nardello-Rataj, 2016; Kurkov & Loftsson, 2013; Del Valle,
2004) and food industries (Astray et al., 2009). Their biocidal mechanism is based on the
lipids extraction from the cell membrane driven by complexation (Fauvelle et al., 1997;
Debouzy et al. 1998). Although native CDs are biocompatible (Stella & He, 2008) and
derived from biomass feedstocks, they unfortunately cannot be used directly to kill pathogens
due to the high CD concentration needs to observe an effective biocidal activity (> 5 mM;
Wallace et al., 2003, Leclercq et al., 2010a, Leclercq et al., 2012). Based on these
considerations and on the fact that developing and marketing new biocides has become almost
4

impossible for industrials, one alternative is to take advantage of synergy effects, as in many
domains nowadays, especially in formulation. We show indeed hereafter that combination of
CDs and QACs can lead to huge synergistic effects on viruses. Some CDs associated with
organic biocides have already been investigated. Two behaviors can be highlighted in respect
with the aqueous solubility of organic biocides: poorly soluble biocides are significantly more
active than the biocides alone due to the solubility improvement (Schmidt et al., 1996)
whereas soluble ones are inactive depending on the degree of complexation (Lehner et al.,
1993, 1994).
Insert Figure 1.
In order to overcome this limitation, Beck et al. have described systems based on chlorite
and sorbic acid biocides which present reduced binding to CDs but this method is very
restrictive (2002). As CD host-guest chemistry enables the implementation of self-assembly
processes in aqueous environment, a simple way to avoid complexation is the dilution. For
infinitely diluted solutions, free CD and biocide species prevail due to the Le Châtelier’s
principle (Leclercq et al., 2012). However, such a system requires highly effective biocides
for which the minimum biocidal concentration (MBC) is very weak. If this is not the case
(e.g. for [BZ] and [BA]), as QACs are relatively well soluble in water, their complexation by
CDs reduces the biocidal activities (Simpson, 1992; Loftsson et al., 1992). In this context, the
use of the [DiC10][Cl] salt is very promising since it achieves biocide effectiveness at a
relatively reduced concentration (< 1 mM). For instance, HSV-1 is deactivated within less
than 1 min of contact at a dose of 0.8 mM (Argy et al., 1999). Therefore, the combination of
[DiC10][Cl] and CDs (active at weak and high concentrations, respectively), is expected to
lead to synergies in terms of antiviral activity since both of them have the same target (i.e.
5

viral membrane, see Figure 1). This effect could be highly helpful in the fight against
nosocomial infections.
2. Materials and methods
2.1. General information
All chemicals were purchased from Sigma-Aldrich Chemical Company at the highest purity
available. NMR spectra were recorded in the indicated solvents. Chemical shifts are given in
ppm () and measured relative to the CDCl3 (Sigma-Aldrich Chemical Company). The
following abbreviations are used to explain the multiplicities: s = singlet, d = doublet, t =
triplet, q = quartet, p = pentet, m = multiplet. All products were dried in a Freeze Dryer
(Alpha 1-2 LD plus). Sterile water for injection was used in all experiments (Fresenius-Kabi,
France,  = 72.0 mN.m-1 at 20.0 °C).
2.2. Synthesis of di-n-decyldimethylammonium chloride
A dry flask (1 L), equipped with a magnetic stir bar, was charged with a solution of 1-
decylbromide (33.17 g, 150 mmol) in acetonitrile (300 mL) and dimethyldecylamine (27.08 g,
150 mmol). The reaction mixture was stirred for 12 hours at 90 °C. The resulting solution was
concentrated under vacuum to give the product. The product was dissolved in water (51 g/100
mL) and ethanol (10 mL) was added to this solution. The aqueous layer was washed with 3 ×
50 mL of hexane (eliminated fractions) and 3 × 25 mL of chloroform. The chloroform layers
were concentrated under vacuum to give the [DiC10][Br]. An aqueous solution of [DiC10][Br]
was eluted on an hydroxide ion exchange resin. The aqueous solution of [DiC10][OH] was
neutralized with an aqueous HCl under dry argon. The [DiC10][Cl] was dried and store under
dry nitrogen. 1H NMR (300 MHz, CDCl3, 25 °C, TMS): (ppm) = 0.85 (t, 6H; CH3, 3J = 6.9
Hz), 1.23-1.32 (m, 28H; CH2), 1.66 (m, 4H; N-CH2-CH2), 3.38 (s, 6H; NCH3), 3.45 (m, 4H;
NCH2). 13C NMR (75 MHz, CDCl3, 25 °C): (ppm) = 14.1, 22.6, 22.7, 26.2, 29.2, 29.4, 31.8,
6

51.3, 63.5. Anal. Calc. for (C22H48NCl): C, 72.98%; H, 13.36%; N, 3.87%; Cl, 9.79%. Found:
C, 72.88%; H, 13.33%; N, 3.96%; Cl, 9.83%. m.p. = 88 °C. (Yield: 95%).
2.3. Surface tension
Surface tensions were measured with the tensiometer K11 (Krüss) using the Wilhelmy plate
method. A concentrated solution was prepared extemporaneously, and equilibrium surface
tension was measured by dilution method in ultrapure water ( = 72.0 mN/m at 25 °C). After
each addition, the solution was gently stirred for 60s. Surface tension was recorded after
equilibration for each mixture. All equilibrium surface tension values were mean quantities of
at least three measurements. The precision of the force transducer of the surface tension
apparatus was 0.1 mN m-1 and before each experiment, the platinum plate was cleaned in
red/orange color flame. The temperature was stabilized at 25 ± 0.05 °C with a thermo-
regulated bath Lauda RC6. The standard deviation was estimated at ± 10% of the CMC
values.
2.4. [DiC10]-selective electrode
The sensitive membrane was prepared using a cross-linked polymer made of a mixing of
silicone gel (Rhodorsil, Rhodia) and polymethylhydromethylcyanopropylsiloxane (Petrarch
system). The selectivity of the membrane is due to a “carrier” of the cationic amphiphile, di-n-
decyldimethylammonium tetraphenylborate [DiC10][BPh4], which was prepared by mixing
equimolar (1 mM) aqueous solutions of sodium tetraphenylborate (NaBPh4, Aldrich) and
[DiC10][Cl]. The insoluble salt obtained was extracted three times by liquid–liquid extractions
in dichloromethane. After evaporation of the solvent, the residual salt was recrystallized twice
from methanol. The salt was then dissolved at 10 mM in THF. Before polymerization, the
composition of the membrane solution is as follows: about 0.7 g of silicone gel, rhodorsil
CAF3 (Rhodia), about 0.3 g of a copolymer, polymethylhydromethylcyanopropylsiloxane
(Petrarch System), about 0.03 g of Aerosil300 (Degussa), 1 mL of the carrier ([DiC10][BPh4])
7

solution into THF and 1 mL of THF. This mix is homogenized few minutes. After
evaporation of THF, the cross-linked polymer forms quickly in contact with air. To construct
the selective electrode, a commercial pH-glass electrode was dipped in the membrane solution
two times, allowing 15 min air-drying between each immersion. For all measurements, the
sensitive electrode was used in potentiometric cells in conjunction with a reference electrode
of KCl-saturated calomel protected from amphiphile diffusion by a saline agar-agar gel made
from 2 M KCl solution contained in a Teflon capillary tube. Emfs were measured with a
Meterlab PHM250 Ion Analyzer (Radiometer Analytical). All the measurements were carried
out at constant temperature (25 ± 0.1 °C).
2.5. Viruses and cells
HSV-1 (Strain Kos) and VACV (Strain Elstree) were propagated in Vero cells (ATCC®
CCL-81™) in Minimum essential Medium (Gibco, Life Technologies) supplemented with 2
mM l-glutamine (Gibco, Life Technologies), 1% non-essential amino-acids (Gibco, Life
Technologies) and 2% inactivated fetal calf serum (Gibco, Life Technologies). Cell-free viral
suspensions of the viruses were obtained by freezing-thawing cycles followed by a low speed
centrifugation to remove cell debris. RSV (local laboratory strain) was propagated in Hep-2
cells (ATCC® CCL-23™) in the same medium. Viruses titers were assayed by the cytopathic
effect of serial dilutions (1:10) of virus-containing samples on Vero or Hep-2 cells. A sample
(100 µL) for each dilution was used to infect four replicate wells in 96-well microtiter plates
(NunclonTM Delta Surface, Thermo ScientificTM NuncTM). Virus-induced cytopathic
effects were scored after 5 days of incubation at 37 °C ± 0.1 °C in a humidified 5% CO2
atmosphere. Titers were expressed as the quantity of viruses infecting 50% of the tissue
culture wells (Tissue Culture Infectious Doses, TCID50) according to Spearman (1908) and
Kärber (1931). The detection limit was 5.62 TCID50 mL-1.Virus stocks were 2 × 107 TCID50
mL-1 for HSV-1, 2 × 106 TCID50 mL-1 for VACV, 3 × 106 TCID50 mL-1 for RSV.
8

2.6. Virus inactivation
200 L of the viral stock solution was added to 200 L of [DiC10][Cl] or 200 L of the
different aqueous binary mixtures of CDs and [DiC10][Cl] (in equimolar proportions) After
incubation for 15 min (30 or 60 min) at room temperature (25 ± 0.1 °C), the mixtures were
immediately filtered on MicroSpin S-400 HR columns (GE Healthcare) to separate viruses
from the other components of the mixtures. Residual viruses were then titrated as described
above. Each experiment was performed at least twice.
3. Results and Discussion
3.1. Virucidal activities of [DiC10][Cl] alone or in combination with native CDs
As control experiments, we evaluated the virucidal activities of [DiC10][Cl] alone against
lipid-containing deoxyribonucleic and ribonucleic acid viruses, such as HSV-1, RSV and
vaccinia viruses (VACV, as model of Variola vera). As expected, the initial viral population
decreases with increasing [DiC10][Cl] concentration after 15 min of contact time (Figure 2,
blue bars). It is noteworthy that a “6-log reduction” is equivalent to 99.9999% reduction of
viral titer. In this case, it appears that it is unlikely that intact viral particles are still present
after the use of [DiC10][Cl]. In other words, the sterilization process is very efficient.
However, the [DiC10][Cl] concentration needed to obtain > 6-log reduction clearly depends on
the virus type: this cation is more active against HSV-1 and RSV (125 µM) than against
VACV (500 µM, see Figure 2, blue bars).
Insert Figure 2.
This differential susceptibility can be correlated with the virus size. Indeed, the diameter of
HSV-1 and RSV is around 200 nm (Plomp et al., 2002; Utley et al., 2008) while that of
VACV is usually 320-380 by 260-340 nm (Malkin et al., 2003). As [DiC10][Cl] interacts and
penetrates the viral membrane, making it fragile, and leads to virus inactivation, we can
9

suppose that the required amount of [DiC10][Cl] in order to obtain a complete membrane
disorganization prior to virus inactivation is higher for VACV than HSV-1 or RSV because of
the viral envelope size. Another plausible explanation can be the difference in term of
cholesterol/phospholipids ratio of viral envelopes. Indeed, the viral envelopes rich in
cholesterol are more stable and less flexible than those poor in cholesterol.
Before considering the mixtures, a second batch of control experiments were undertaken in
the presence of natural CDs without [DiC10][Cl]. In a concentration range from 20 to 1,000
μM, no activity was found after 15 min or 2 h of contact time against HSV-1, RSV and
VACV. It is logical to not observe the virucidal activities of natural CDs with concentrations
six times lower than those found by Wallace et al. (2003). Indeed, exposure of HSV-1 to -
CD (6,000 µM) resulted in “5-log reduction” of the virus titer after 24h. In addition, native
CDs are well known to induce morphological changes in erythrocytes followed by hemolysis
at higher concentrations. As this hemolytic activity is ascribed to the extraction of
phospholipids and/or cholesterol depending on the cavity size, this information is very
important because it is reliable measure for the estimation of CD-induced cytoplasmic
membrane damage. Our results are also in perfect agreement with the hemolytic activities of
native CDs. Indeed, hemolysis of human erythrocytes was initiated at 6,000, 3,000 and 16,000
µM for -, -, and -CD, respectively, whereas complete hemolysis was observed at 20,000,
10,000 and 41,000 µM, respectively (Irie et al., 1982). Based on this, it is reasonable that
native CDs are not able to induce a virucidal activity due to their low concentrations (< 1,000
µM).
Focusing on the virucidal activity of [DiC10][Cl]/CD mixtures, the minimum
concentrations to achieve a “6-log reduction” have been estimated for several diluted aqueous
solutions prepared from a equimolar stock solution of [DiC10][Cl] and CD. In the presence of
-CD, the biocidal activity of [DiC10] is clearly significantly enhanced against the three
10

viruses investigated: the [DiC10][Cl] concentration needed to obtain “> 6-log reduction” is
reduced to 72, 40 and 85% against HSV-1, RSV and VACV, respectively (Figure 2, green
bars). As the efficacy of this mixture is clearly higher than the sum effects of the single
substances, these observations can be directly related to a molecular synergism between the
[DiC10][Cl] and the -CD. Indeed, their combination in an equimolar mixture increases the
overall effectiveness of the formulation. Surprisingly, the mixtures based on - and -CD
show no synergy (Figure 2, purple and orange bars). The differential behavior of the
formulation with -CD compared to those with -CD and -CD on viruses are clearly not
directly linked to the hemolytic activity induced by natives CDs which increases in the order
-CD < -CD < -CD due to the extraction of phospholipids and/or cholesterol depending on
the cavity size (Ohtani et al., 1989). Indeed, - and -CDs are excellently suited to solubilize
phospholipids and cholesterol, respectively, whereas -CD is less lipid selective. Based only
on the potencies of these CDs for solubilizing components of viral envelope, the most
virucidal mixtures should be the - or -CD/[DiC10][Cl] mixtures depending on the
phospholipids/cholesterol ratio in the membranes: -CD for phospholipids-rich envelopes and
-CD for cholesterol-rich membranes. In our case, the efficiency of native CDs in term of
synergy with [DiC10][Cl] (i.e. in term of virus inactivation) is clearly inconsistent with the
order of magnitude of hemolytic activity. This difference could be ascribed to the presence of
[DiC10][Cl]. Indeed, this virucide is also membrane-active agent interacting with lipids or
proteins of the viral envelope prior to its complete disorganization (Denyer, 1995; Nardello-
Rataj & Leclercq, 2014). In the literature, it is reported that CDs, at low concentrations, can
alleviate hemolysis induced by drugs (e.g. imiparine, chlorpromazine hydrochloride,
propantheline bromide, etc.) through the formation of inclusion complexes (Funasaki et al.,
2001; Funasaki et al., 1999). Nevertheless, this behavior is not just limited to drugs but also to
surfactants with hemolytic activity. Indeed, the attenuation of dodecytrimethylammonium
11

bromide-induced hemolysis increases in the order -CD <-CD ≈ -CD (Funasaki et al.,
2000). This effect is directly correlated with the magnitude of the binding constants between
CDs and cationic surfactant that increases in the order -CD <-CD ≈ -CD. Therefore, the
formation of complexes between the dodecytrimethylammonium cation and - or -CD
suppresses the hemolysis-induced by the surfactant whereas this effect is kept with -CD due
to the weak binding constant (Funasaki et al., 2000). It could be argued that CDs in some
manner have a similar effect in our case.
3.2. Thermodynamics of [DiC10][Cl]/[CD] mixtures
In the case of a binary mixture based on a single surfactant (S) and CD, it is assumed that
two inclusion complexes can be observed depending on the cavity size (Leclercq et al.,
2010a). Furthermore, according to its concentration, the surfactants can self-assemble to form
micelles. Additional equilibria occur in the presence of viruses such as lipids extraction from
the viral envelope by CD or the insertion of the surfactant in the membrane. Figure 3 shows a
schematic model of the equilibria occurring in the [DiC10][Cl]/CD mixtures.
Insert Figure 3.
As it is well known that addition of CD to a micellar surfactant solution leads to the
dissociation of the micelle, we have determined the apparent CMC of [DiC10][Cl]/[CD]
mixtures. Consequently, the effect of the natural CDs on the micellization process of the
[DiC10][Cl] itself has been studied by measuring the surface tension, , at 25 °C by the
dilution method. In this method, as the dilution occurs for the surfactant as well as for the CD,
the results are directly comparable with the virucidal tests (see experimental section). In the
absence of CDs, the surface tension isotherm is typical of true surfactants and the critical
micelle concentration (CMC) values is in good agreement with the data previously reported
(1 200 µM, Leclercq et al., 2013). As expected, the presence of CDs leads to significant
12

changes in surface tension isotherms: the surface tension deviates from that observed for the
[DiC10][Cl] alone reflecting the formation of inclusion complexes (Figure 4).
Insert Figure 4.
Unfortunately, the complete rationalization of the isotherms obtained with CDs is very
difficult because of the ability of inclusion complexes to participate in the surface monolayer
stabilization. Actually, our previous results supports that the - or -CD/[DiC10][Cl] 1:1
inclusion complexes participate in the surface-adsorbed monolayer (Leclercq et al., 2013).
However, in the presence of CD, the apparent CMCs are in the order: -CD (5,140 µM) ≈ -
CD (5,550 µM) ⪡ -CD (9,710 µM). Therefore, the micellization equilibrium is not relevant
in this case: the virucidal activities are obtained in the premicellar region for all mixtures (see
discussion above).
Next, in order to explicit the equilibria involved in the initial steady state, we have
experimentally determined the free [DiC10] concentration, [DiC10]free, in the [DiC10][Cl]/CD
mixtures using a [DiC10]-selective electrode. In the absence of CD and viruses and below its
CMC (1,200 µM), the electromotive force (emf) of an aqueous solution of [DiC10][Cl]
presents a pseudo-Nernstian behavior which is a unique function of [DiC10]free (eq. 1, Figure
5, Leclercq et al., 2010b).
emf  59.2  55.1log[DiC10]free
(1)
In contrast, for [DiC10][Cl]/CD mixtures, the emf curves, recorded by the dilution method,
deviates from the Nernstian behavior in the negative direction. These deviations are assigned
to the formation of inclusion complexes which reduce the free [DiC10] concentration. The
extent of deviation is in the order -CD ≈ -CD <-CD.
Insert Figure 5.
13

The free [DiC10] concentrations can be calculated from the emf values since equation 1
provides accurate and reproducible measurements of these concentrations (Rauwel et al.,
2012). The evolution of [DiC10]free concentration as a function of the total concentration of
[DiC10][Cl] in the binary [DiC10][Cl]/CD mixture is presented in Figure 6.
Insert Figure 6.
As the binding constants (K1 and K2) have been previously estimated from the [DiC10]-
selective electrodes by the titration method applied to a fixed concentration of [DiC10][Cl]
with varying concentrations of native CDs, we can easily estimate the initial composition of
our systems (Leclercq et al., 2010a). Indeed, below the CMC of the surfactant (S) and in the
presence of CD, it is assumed that the molarity of 1:1 and 1:2 complexes is expressed as
[SCD]  K1 [S][CD]
(2)
(3)
[SCD2 ]  K2K1 [S][CD]2
where [S] and [CD] are the free concentrations of S and CD, respectively. In eqs 2 and 3,
[SCD] and [SCD2] denote the concentrations of the complexes with 1:1 and 1:2
stoichiometries respectively. The mass balance equations are written as
CCD [CD] K1 [S][CD] 2K1K2[S][CD]2
CS [S] K1[S][CD] K1K2[S][CD]2
(4)
(5)
Eliminating [S] from eqs. 4 and 5 gives the cubic equation:
K1K2[CD]3  K1 [CD]2 

1 K2

CCD  2CS


1 K1CS  K1CCD

[CD]CCD  0
(6)
The free CD concentration ([CD]) can be calculated from eq. 6 and substituting in eq. 5 to
obtain the free surfactant concentration ([S]). Finally, the complexes concentrations ([CDS]
and [CD2S]) can be calculated with eqs. 2 and 3. All results are presented in Figure 6. It is
14

clear that the association constants, previously determined, describe very well the free
[DiC10][Cl] concentrations (Figure 6, compare the continuous lines and the experimental
values). The goodness of fits is given in Table 1. As these values tend to 1, the binding
constants values, previously determined from titration method, fits very well with our present
results obtained with the dilution method. Therefore, it is not necessary to fit values again.
Insert Table 1.
As depicted in Figure 6, infinitely diluted solutions ([DiC10][Cl]T ≤ 30 µM) show the
prevalence of free species ([DiC10] and CD). When [DiC10][Cl]T ≥ 30 µM, [DiC10]CD and
[DiC10]CD2 inclusion complexes are gradually formed leading to a rapid divergence of the
free CD concentration compare to the free [DiC10][Cl]. As this behavior is directly related to
the formation of 1:2 inclusion complexes, the evolution of free species ([DiC10] and CD) are
similar for -CD. For [DiC10][Cl]T = 35 µM, it is clear from these results that the relative ratio
of inclusion complexes are marginal and that the free [DiC10] and CD) concentrations are
quite similar for all native CDs (see Figure 6). Moreover, at this concentration, negligible
fractions of lipids (phospholipids or cholesterol) bound to native CDs in steady state can be
also invoked because the magnitude of binding constants with lipids are quite similar to the
constants observed with [DiC10][Cl] (e.g. ~1.7×104 for -CD/cholesterol; Frijlink et al., 1991;
Shiotani et al., 1995). However, we can reasonably suppose that the inclusion complexes can
be seen as a reservoir of [DiC10] and CD molecules which are readily available for interaction
with the viral envelope (i.e. complexation and insertion, for CDs and [DiC10], respectively).
Consequently, the thermodynamic in initial steady state fails to explain the differential effects
of CDs especially for [DiC10][Cl]/CD mixtures (35 µM each) against HSV-1 (see Figure 2).
15

3.3. Kinetics of the virus inactivation
It is known that virus inactivation is a kinetic process wherein the “viability” of viruses
varies as a function of time (Weavers & Wickramanayake, 2001). As the apparent differential
effect of native CDs, observed after 15 min of exposure, can be related to dissimilarities in
term of inactivation rates, we have determined the virus titers after 15, 30 and 60 min of
exposure to [DiC10][Cl]/CD mixtures (35 µM each) against HSV-1 (Figure 7).
Insert Figure 7.
As expected, virus inactivation varies as a function of time. The time needed to obtain “> 6-
log reduction” (i.e. 99.9999% reduction of viral titer) for [DiC10][Cl]/CD mixtures are -CD
much greater than -CD greater than -CD. It is noteworthy that this time is always smaller
than the [DiC10][Cl] alone. Therefore, the synergism occurs for all native CDs in term of virus
inactivation rates. However, the fast kinetic process is obtained with the [DiC10][Cl]/ -CD
mixtures. The analysis and interpretation of inactivation kinetics are extremely complex.
Based on the hemolytic activity of native CDs, we can suppose that the mechanism of action
is quite different for -CD on the one hand and - and-CDs on the other. Indeed, the -CD
cavity is perfect for the extraction of phospholipids whereas -CD and -CD extract
preferentially cholesterol although -CD is less active than -CD (Ohtani et al., 1989).
However, it is known that the rate of cholesterol transfer between vesicles is considerably
accelerated in the presence of submillimolar concentrations of - and-CD (64- and 63-fold,
respectively) compared to that observed in the absence of CD (Leventis & Silvius, 2001). In
contrast, -CD is able to accelerate the transfer of phospholipids (Leventis & Silvius, 2001).
Therefore, even if the fractions of cholesterol bound to -CD are initially marginal, we can
reasonably suppose that the insertion of [DiC10][Cl] is faster in the presence of -CD because
the concentrations of free [DiC10][Cl] and -CD are identical due to the presence of 1:1
complex (see Figure 6). In these conditions, the insertion of [DiC10][Cl] associated to the
complexation of cholesterol by -CD allows an easy displacement of equilibria and accelerate
the apparent rate of virus inactivation. In contrast, with - and -CD, due to the presence of
1:2 inclusion complexes, the free CD concentration is weaker than [DiC10][Cl] leading to a
slower [DiC10][Cl] insertion than with -CD. In line with the experimental contact time, the
16

- and -CD/[DiC10][Cl] mixtures act slower than the -CD/[DiC10][Cl] mixture resulting in
an apparent loss of activity at given time and concentration. It is noteworthy that similar
results are obtained with the other enveloped viruses investigated here.
4. Conclusion
According to the old adage of strength through unity, we have demonstrated that [DiC10]
cations associated with CDs act as a much more virucidal agent than each compound alone
due to the lipid and/or cholesterol extraction from the viral envelope. As a consequence, the
[DiC10][Cl] concentration is reduced up to 85% for the same virucidal activity with the -CD.
As depicted in Table 2, the best compromise in terms of toxicity (hemolysis, cytotoxicity and
degradability) and virucidal activity (concentration and time to obtain sterilization) is obtained
with the equimolar [DiC10][Cl]/-CD mixture.
Insert Table 2.
Moreover, such antiviral mixtures taking advantage of the synergy between native CDs and
classical virucides allow the partial replacement of a significant amount of biocides by eco-
and bio-compatible CDs. This result is particularly interesting both for the formulator and the
user as it allows to decrease the toxicological and ecotoxicological drawbacks of classical
biocides while complying with the regulation. The eco-compatible and biocide terms, largely
antithetical, may be combined for the first time under the name of eco-biocide. These results
support the idea that the supramolecular assistance by the CDs is a phenomenon of general
importance. This synergistic effect provides elements valuable for further developments and
will probably become the new Holy Grail in the coming years particularly in the fight against
viruses and other nosocomial infections.
Acknowledgements
Chevreul Institute (FR 2638), Ministère de l’Enseignement Supérieur et de la Recherche,
Région Nord - Pas de Calais and Fonds Européen de Développement Régional (FEDER) are
acknowledged for supporting and funding partially this work. We are grateful to Pr. Didier
Hober (Université de Lille) for supplying us the vaccinia virus and Dr. Gaétan Rauwel
(Laboratoires Anios, France) for fruitful discussions.
17

References
Aitken, C., Jeffries, D. J., 2001. Nosocomial spread of viral disease. Clin Microbiol Rev. 14,
528-546.
Antlsperger, G., Schmid, G., Toxicological comparison of cyclodextrins. In: Szejtli, J.,
Szente, L. (Eds.) Proceedings of the Eighth International Symposium on Cyclodextrins,
Springer-Science+Business Media, B.V., Dordrecht, Netherland, 1996, pp. 149-155.
Argy, G., Bricout, F., d’Hermies, F., Cheymol, A., 1999. Study of prophylaxis by didecyl
dimethyl ammonium chloride against herpes simplex virus infection in nude mice. C. R.
Acad. Sci. III, Sci. Vie, 322, 863-870.
Arima, H., Motoyama, K., Irie, T. Recent findings on safety profiles of cyclodextrins,
cyclodextrin conjugates, and polypseudorotaxanes. In: Bilensoy, E. (Ed.) Cyclodextrins in
pharmaceutics, cosmetics, and biomedicine: current and future industrial applications, Wiley,
Hoboken, NJ, U.S.A., 2011, pp. 91-122.
Astray, G., Gonzalez-Barreiro, C., Mejuto, J. C., Rial-Otero, R., Simal-Gándara, J., 2009. A
review on the use of cyclodextrins in foods. Food Hydrocolloid, 23, 1631–1640.
Beck, G. J., Kerslake, E. D. S., Olejnik, O., 2002. Preserved cyclodextrin-containing
compositions. U.S. Patent 0,076,449.
Chepurnov, A. A., Bakulina, L. F., Dadaeva, A. A., Ustinova, E. N., Chepurnova, T. S.,
Baker, J. R. Jr., 2003, Inactivation of Ebola virus with a surfactant nanoemulsion. Acta Trop.
87, 315-320.
Debouzy, J.C., Fauvelle, F., Crouzy, S., Girault, L., Chapron, Y., Göschl, M., Gadelle, A.,
1998. Mechanism of -cyclodextrin induced hemolysis: a study of the factors controlling the
association with serine-, ethanolamine-, and choline-phospholipids. J Pharm Sci. 87, 59-66.
18

Del Valle, E. M. M., 2004. Cyclodextrins and their uses: a review. Process Biochem. 39,
1033-1046.
Denyer, S. P., 1995. Mechanisms of action of antibacterial biocides. Int. Biodeter. Biodegr.
36, 227-245.
Ebola situation report. World Health Organisation. [March 30, 2016].
Fauvelle, F., Debouzy, J. C., Crouzy, S., Göschl, M., Chapron, Y., 1997. Mechanism of
cyclodextrin–induced hemolysis: the twostep extraction of phosphatidylinositol from
themembrane. J. Pharm. Sci. 86, 935-943.
Frijlink, H. W., Eissens, A. C., Hefting, N. R., Poelstra, K., Lerk, C. F., Meijer, D. K. F.,
1991. The effect of parenterally administrated cyclodextrins on cholesterol levels in the rat.
Pharmaceut. Res. 8, 9-16.
Funasaki, N. Ohigashi, M., Hada, S., Neya, S., 2000. Surface tensiometric study of multiple
complexation and hemolysis by mixed surfactants and cyclodextrins. Langmuir, 16, 283-388.
Funasaki, N., Ohigashi, M., Hada, S., Neya, S., 1999. Quantitative Prediction of the
Suppression of Drug-Induced Hemolysis by Cyclodextrins from Surface Tension Data.
Langmuir, 15, 594–599.
Funasaki, N., Okuda, T., Neya S., 2001. Mechanisms and surface chemical prediction of
imipramine-induced hemolysis suppressed by modified cyclodextrins. J. Pharm. Sci. 90,
1056-1065.
Gerba, C. P., 2015. Quaternary ammonium biocides: efficacy in application. Appl. Environ.
Microbiol. 81, 464-469.
19

Gilbert, P., McBain, A. J., 2003. Potential impact of increased use of biocides in consumer
products on prevalence of antibiotic resistance. Clin. Microbiol. Rev. 16, 189-208.
Hall, C. B., 1977. The shedding and spreading of respiratory syncytial virus. Pediatr. Res. 11,
236-239.
Hauthal, H. G., 2004. Dynamic surfactants and nanostructured surfaces for an innovative
industry. SÖFW-Journal, 130, 3-17.
Irie, T., Otagiri, M., Sunada, M., Uekama, K., Ohtani, Y., Yamada, Y., Sugiyama, Y. 1982.
Cyclodextrin-induced hemolysis and shape changes of human erythrocytes in vitro; J. Pharm.
Dyn. 5, 741-744.
Jambhekar, S. S., Breen, P., 2016a. Cyclodextrins in pharmaceutical formulations: structure
and physicochemical properties, formation of complexes, and types of complex. Drug Discov.
Today, 21, 356-362.
Jambhekar, S. S., Breen, P., 2016b. Cyclodextrins in pharmaceutical formulations:
solubilization, binding constant, and complexation efficiency. Drug Discov. Today, 21, 363-
368.
Kärber, G., 1931. Beitrag zur kollectiven Behandlung pharmakologischer Reihenversuche.
Arch. Exp. Path. Pharmak. 162, 480-487.
Kurkov, S. V., Loftsson, T., 2013. Cyclodextrins. Int. J. Pharm. 453, 167-180.
Leclercq, L., Lubart, Q., Aubry, J.-M., Nardello-Rataj, V., 2013. Modeling of multiple
equilibria in the self-aggregation of di-n-decyldimethylammonium chloride/octaethylene
glycol monododecyl ether/cyclodextrin ternary systems. Langmuir, 29, 6242-6252.
20

Leclercq, L., Lubart, Q., Dewilde, A., Aubry, J.-M., Nardello-Rataj, V., 2012. Supramolecular
effects on the antifungal activity of cyclodextrin/di-n-decyldimethylammonium chloride
mixtures. Eur. J. Pharm. Sci. 46, 336-345.
Leclercq, L., Nardello-Rataj, V., 2016. Pickering emulsions based on cyclodextrins: A smart
solution for antifungal azole derivatives topical delivery. Eur. J. Pharm. Sci. 82, 126-137.
Leclercq, L., Nardello-Rataj, V., Rauwel, G., Aubry, J.-M., 2010a. Structure-activity
relationship of cyclodextrin/biocidal double-tailed ammonium surfactant host-guest
complexes: towards a delivery molecular mechanism? Eur. J. Pharm. Sci. 41, 265-75.
Leclercq, L., Nardello-Rataj, V., Turmine, M., Azaroual, N., Aubry, J.-M., 2010b. Stepwise
aggregation of dimethyl-di-n-octylammonium chloride in aqueous solutions: from dimers to
vesicles. Langmuir 26, 1716-1723.
Leclercq, L., Sauthier, M., Castenet, Y., Mortreux, A., Bricout, H., Monflier, E., 2005. Two-
phase hydroformylation of higher olefins using randomly methylated -cyclodextrin as mass
transfer promoter: a smart solution for preserving the intrinsic properties of the
rhodium/trisulfonated triphenylphosphine catalytic system. Adv. Synth. Catal. 347, 55-59.
Leclercq, L., Smart medical textiles based on cyclodextrins for curative or preventive patient
care. In: Hu, J. (Ed.) Active coatings for smart textiles, Woodhead Publishing, Duxford, U.K.
2016, pp. 391-427.
Lehner, S. J., Müller, B. W., Seydel, J. K., 1993. Interactions between p-hydroxybenzoic acid
esters and hydroxypropyl--cyclodextrin and their antimicrobial effect against Candida
albicans. Int. J. Pharm. 93, 201-208.
Lehner, S. J., Müller, B. W., Seydel, J. K., 1994. Effect of hydroxylpropyl--cyclodextrin on
the antimicrobial action of preservatives. J. Pharm. Phamacol. 46, 186-191.
21

Leventis, R., Silvius, J. R. 2001. Use of cyclodextrins to monitor transbilayer movement and
differential lipid affinities of cholesterol. Biophys. J. 81, 2257-2267.
Loftsson, T., Stefánsdóttir, Ó., Friòriksdóttir, H., Guòmundsson, Ō., 1992. Interactions
between preservatives and 2-hydroxypropyl--cyclodextrin. Drug Dev. Ind. Pharm. 18, 1477-
1484.
Malkin, A. J., McPherson, A., Gershon, P. D., 2003. Structure of intracellular mature vaccinia
virus visualized by in situ atomic force microscopy. J. Virol. 77, 6332-6340.
Moore, L. E., Ledder, R. G., Gilbert, P., McBain, A. J., 2008. In vitro study oft he effect of
cationic biocides on bacterial population dynamics and susceptibility. Appl. Environ.
Microbiol. 74, 4825-4834.
Moore, S. L., Payne, D. N. Types of antimicrobial agents. In: Fraise, A. P. Lambert, P. A.
Maillard J.-Y. (Eds.) Russell, Hugo & Ayliffe’s Principles and Practice of Disinfection,
Preservation and Sterilization (4th Ed.), Blackwell Science, Oxford, U.K. 2004, pp. 8-97.
Nardello-Rataj, V., Leclercq, L., 2014. Encapsulation of biocides by cyclodextrins: toward
synergistic effects against pathogens. Beilstein J. Org. Chem. 10, 2603-2622.
Ohtani, M., Irie, T., Uekama, K., Fukunaga, K. Pitha, J. 1989. Differential effects of -, -
and -cyclodextrins on human erythrocytes. Eur. J. Biochem. 186, 17-22.
Plomp, M., Rice, M. K., Wagner, E. K., McPherson, A., Malkin, A. J., 2002. Rapid
visualization at high resolution of pathogens by atomic force microscopy. Am. J. Pathol. 160,
1959-1966.
Rauwel, G., Leclercq, L., Criquelion, J., Aubry, J.-M., Nardello-Rataj, V., 2012. Aqueous
mixtures of di-n-decyldimethylammonium chloride/polyoxyethylene alkyl ether: dramatic
22

influence of tail/tail and head/head interactions on co-micellization and biocidal activity. J.
Colloid Interface Sci. 374, 176-186.
Schmidt, A., von der Eltz, H., Kaluza, K., 1996. Cyclodextrin-biocide complex. U.S. Patent
5,506,216.
Shiotani, K., Uehata, K., Irie, T., Uekama, K., Thompson, D. O., Stella, V., 1995. Differential
effects of sulfate and sulfobutyl ether of beta-cyclodextrin on erythrocyte membranes in vitro.
J. Pharm. Res. 12, 78−84.
Simpson, W. J., 1992. Neutralisation of the antibacterial action of quaternary ammonium
compounds with cyclodextrins. Microbiol. Lett. 90, 197-200.
Spearman, C., 1908. The method of 'right and wrong cases' ('constant stimuli') without Gauss'
formulae. Br. J. Psychol. 2, 227-242.
Stella, V. J., He, Q., 2008. Cyclodextrins. Toxicol. Pathol. 36, 30-42.
Szejtli, J., 1998. Introduction and general overview of cyclodextrin chemistry. Chem Rev. 98,
1743-1754.
Utley, T. J., Ducharme, N. A., Varthakavi, V., Shepherd, B. E., Santangelo, P. J., Lindquist,
M. E., Goldenring, J. R., Crowe, Jr. J. E., 2008. Respiratory syncytial virus uses a Vps4-
independent budding mechanism controlled by Rab11-FIP2. Proc. Natl. Acad. Sci. USA 105,
10209-10214.
Valenti, W. M., Menegus, M. A., Hall, C. B., Pincus, P. H., Douglas, R. G. Jr., 1980.
Nosocomial viral infections: epidemiology and significance. Infect Control., 1, 33-7.
Wallace, K. B., Khan, M. A., Carlson, R. M., Rice, S., Froberg, M. K., 2003. Cyclodextrin
compositions and methods of treating viral infections. U.S. Patent 0,220,294.
Weavers, L. K., Wickramanayake, L. K. Kinetics of inactivation of microorganisms. In:
Block, S. S. (Ed.) Disinfection, Sterilization and Preservation (5th Ed.), Lippincott Williams
& Wilkins, Philadelphia, U.S.A. 2001, pp. 65-78.
23

Figure 1. Effects of di-n-decyldimethylammonium cation, [DiC10], and natural cyclodextrins,
CDs, on enveloped viruses.
24

Figure 2. Dose-dependent virucidal action of [DiC10][Cl] with or without natural CDs from
herpes simplex (HSV-1), respiratory syncytial (RSV) and vaccinia virus (VACV) (NB. the
control experiments correspond to the viral titer in the absence of CDs and [DiC10][Cl]; the
standard deviation of virus titer is ± 10%).
25

Figure 3. Schematic interrelation among the equilibria of complexation (K1 and K2),
micellization (Kmic), CD-induced lipids extraction (Kext), and [DiC10] insertion (Kins) for a
binary mixture of [DiC10][Cl] and CD in the presence of viruses (1. protein, 2. cholesterol, 3.
phospholipid, 4. envelope, and 5. nucleocapsid).
26

Figure 4. Surface tension () plotted against total [DiC10][Cl] concentration for equimolar
[DiC10][Cl]/CD binary mixtures at 25.0 °C compare to [DiC10][Cl] alone.
27

Figure 5. Electromotive force (emf) plotted against total [DiC10][Cl] concentration for
equimolar [DiC10][Cl]/CD binary mixtures at 25.0 °C compare to [DiC10][Cl] alone.
28

Figure 6. Evolution of free [DiC10], free CD, [DiC10]CD, and [DiC10]CD2 concentrations as
a function of the total [DiC10][Cl] concentration for equimolar [DiC10][Cl]/CD binary
mixtures at 25.0 °C in steady state. Circles correspond to experimental values whereas
continuous lines are calculated from the binding constants (see equations. 2-6).
29