Structural and morphological aspects of (fluoro)quinolone delivery by layered double hydroxide nanoparticles

Article abstract of DOI:10.1039/c8nj03072c

Layered double hydroxides (LDHs) have been proposed as delivery systems (DSs) of (fluoro)quinolones (QLNs) to overcome their low bioavailability and to prevent the emergence of resistant bacteria. Both LDH-DS synthesis as nanoparticles (NPs) and QLN interactions with the metal ions that constitute the layers are essential to improve their physicochemical, biopharmaceutical and antimicrobial properties. Here, LDH-DSs containing the basic form of nalidixic acid (Nal), used as a probe, were obtained by coprecipitation at variable and constant pH (LDH-Nal-pHvar and LDH-Nal-pHcte, respectively). For both syntheses, LDH NPs containing Nal anions (LDH-Nal-NPs), with sizes between 30 and 40 nm, were obtained. A coordination compound (Mg(Nal)₂·4H₂O, Mg(Nal)₂) was also concurrent for LDH-Nal-pHcte, which modulated the drug release profile and antimicrobial properties of the LDH-Nal-NPs. Thus, Nal release from LDH-Nal-pHvar was produced mainly by anion exchange. The best fits, obtained for the Higuchi model, showed rate constants dependent on the exchanging anions (k_H = 0.88 and 1.53 for NaCl 0.9% and buffer phosphate 0.05 M, pH = 7.4, respectively). The nanometric size of LDH-Nal-pHvar as well as its faster release rate allowed a minimum inhibitory concentration decrease (MIC = 32 μg mL^?1) compared to the pure drug (MIC = 128 μg mL^?1). Instead, the presence of Mg(Nal)₂ in LDH-Nal-pHcte led to a more sustained and media independent Nal release, but a lower MIC (64 μg mL^?1) than LDH-Nal-pHvar.

Full text of DOI:10.1039/c8nj03072c

NJC
View Article Online
View Journal | View Issue
PAPER
Structural and morphological aspects of
fluoro)quinolone delivery by layered double
(
Cite this: New J. Chem., 2018,
hydroxide nanoparticles†
42, 19144
a
b
c
Yadira Salguero, Monica Cristina Garcıa,
Ricardo Rojas
Giuliana Mosconi and
´
´
a
*
Layered double hydroxides (LDHs) have been proposed as delivery systems (DSs) of (fluoro)quinolones
QLNs) to overcome their low bioavailability and to prevent the emergence of resistant bacteria. Both LDH-
(
DS synthesis as nanoparticles (NPs) and QLN interactions with the metal ions that constitute the layers are
essential to improve their physicochemical, biopharmaceutical and antimicrobial properties. Here, LDH-DSs
containing the basic form of nalidixic acid (Nal), used as a probe, were obtained by coprecipitation at
variable and constant pH (LDH-Nal-pHvar and LDH-Nal-pHcte, respectively). For both syntheses, LDH NPs
containing Nal anions (LDH-Nal-NPs), with sizes between 30 and 40 nm, were obtained. A coordination
compound (Mg(Nal)
2
ꢀ4H
2 2
O, Mg(Nal) ) was also concurrent for LDH-Nal-pHcte, which modulated the drug
release profile and antimicrobial properties of the LDH-Nal-NPs. Thus, Nal release from LDH-Nal-pHvar
was produced mainly by anion exchange. The best fits, obtained for the Higuchi model, showed rate
Received 24th July 2018,
H
constants dependent on the exchanging anions (k = 0.88 and 1.53 for NaCl 0.9% and buffer phosphate
Accepted 31st October 2018
0.05 M, pH = 7.4, respectively). The nanometric size of LDH-Nal-pHvar as well as its faster release rate
ꢁ
1
allowed a minimum inhibitory concentration decrease (MIC = 32 mg mL ) compared to the pure drug
DOI: 10.1039/c8nj03072c
ꢁ1
(
2
MIC = 128 mg mL ). Instead, the presence of Mg(Nal) in LDH-Nal-pHcte led to a more sustained and
ꢁ1
rsc.li/njc
media independent Nal release, but a lower MIC (64 mg mL ) than LDH-Nal-pHvar.
3
developed below it. Organic matrixes (polymeric particles,
1
. Introduction
liposomes or dendrimers) are mainly studied as hosts for
2
,4,5
(
Fluoro)quinolones (QLNs), due to their variety, broad- QLNs
but the inorganic ones (carbon nanotubes, iron
spectrum bactericidal activity and efficacy, are widely used for oxides, clays and layered double hydroxides (LDHs)) are receiving
1
4,6,7
therapeutic and prophylactic purposes. Delivery systems (DSs) increasing attention.
are increasingly studied for QLNs to reduce their fluctuations LDHs are inorganic anion exchangers that have been studied
in plasma concentration, to protect them from degradation and as matrixes in DSs for anionic drugs, including anti-
2
8
9
10,11
to increase their bioavailability. Also, DSs are intended to inflammatories, anticancer agents or antimicrobials.
produce antimicrobial levels above the minimal inhibitory structure can be described as a stack of brucite-like layers with a
Their
II
III
II
2+
concentration (MIC) for longer times, preventing the resistance general formula [M
M
(OH) ]A ꢀmH O (where M = Mg ,
1ꢁx
x
2
n/x
2
2
+
2+
III
3+ 3+
nꢁ
Ca , Zn , . . .; M = Al , Fe , . . . and A is the interlayer
anion). The anionic drugs are placed between the layers and
released by either anion exchange or dissolution of the host
layers. Release profile modification as well as increased drug
solubility and stability
of LDHs.
1
2
a
b
c
INFIQC-CONICET, Departamento de Fisicoqu ´ı mica, Facultad de Ciencias
Qu ´ı micas, Universidad Nacional de C o´ rdoba. Ciudad Universitaria, X5000HUA,
C o´ rdoba, Argentina. E-mail: rrojas@fcq.unc.edu.ar, yadisalgueros@gmail.com
UNITEFA-CONICET, Departamento de Ciencias Farmac ´e uticas. Facultad de
Ciencias Qu ´ı micas, Universidad Nacional de C o´ rdoba. Ciudad Universitaria,
X5000HUA, C o´ rdoba, Argentina. E-mail: mgarcia@fcq.unc.edu.ar
1
3
14
15,16
are among the described advantages
LDHs have been proposed as hosts of anionic QLNs such as
the basic form of nalidixic acid (HNal), enrofloxacin, ciprofloxacin
IPQA-CONICET, Departamento de Qu ´ı mica Org a´ nica, Facultad de Ciencias
Qu ´ı micas, Universidad Nacional de C o´ rdoba. Ciudad Universitaria, X5000HUA,
C o´ rdoba, Argentina. E-mail: mosconigiuliana@gmail.com
17–22
and norfloxacin.
The obtained DSs exhibited modified release
†
Electronic supplementary information (ESI) available: Analysis of the synthesis
kinetics, while the bacteriological studies performed indicated
that the antimicrobial activity against both Gram-positive and
Gram-negative bacteria was maintained.
parameters by titrations of metal salt solutions in presence and absence of HNal.
Study of Nal interaction with LDH layers by Nal sorption isotherm on LDH-Cl.
Bacteria count of lawn cultures of MIC determinations. See DOI: 10.1039/
c8nj03072c
1
8,23
It has also been
proposed that LDH nanoparticles (LDH-NPs) establish electrostatic
19144 | New J. Chem., 2018, 42, 19144--19152 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
2
4
interactions with the membrane of both types of bacteria.
2.2.2. Coprecipitation at constant pH. A 100 mL solution
ꢁ1
ꢁ1
2 3
This interaction leads to membrane disruption, bacterial aggre- containing MgCl (0.3 mol L ) and AlCl (0.1 mol L ) was
gation and membrane charging reduction that would potentially added dropwise to a 200 mL solution containing 0.01 mol
synergize with the action of intercalated QLNs. Nevertheless, the HNal, whose pH was previously set to 9. The pH of the media
ꢁ1
synthesis of LDH-NPs containing QLNs has not yet been was maintained constant to this value by addition of 1 mol L
described to our knowledge. NaOH solution. Once the addition was finished, the DS obtained
Another unexplored feature is the role of the QLNs interac- (LDH-Nal-pHcte) was processed according to that described
tions with the metal ions of the LDH layers. QLNs are able to in Section 2.2.1.
2
+
2+
establish coordination bonds with metal ions such as Mg , Zn
2.2.3. Reference samples. A sodium salt of the anionic
3
+
25
or Al with their 3-carboxyl and 4-oxo groups. This interaction form of nalixidic acid (NaNal) was prepared by neutralization
is dependent on the drug, the metal ions, and conditions such of HNal with NaOH 1 mol L , (final pH = 11), followed by
as pH. HNal, for example, present complexation reactions lyophilization of the obtained solution. Also, a magnesium
with Mg and Al ions, which produce important changes in complex with the anionic form of nalixidic acid (Mg(Nal) )
ꢁ
1
2
6
2
+
3+
2
ꢁ
1
its properties, namely lipophilicity, solubility, acidity constant as was obtained by mixing equivalent volumes of a 0.1 mol L
ꢁ1
well as in its bioavailability, permeation capacity and bactericidal MgCl
2
solution and a 0.2 mol L NaNal solution. The obtained
Consequently, these interactions are not only impor- solid was washed and lyophilized to perform its chemical and
tant to successfully obtain effective LDH-based DSs but also structural characterization. Finally, a chloride intercalated LDH
provide a tool for their customization. (LDH-Cl) was prepared by the procedure described in Section 2.2.2.
2
7–29
action.
In this work, we explored the properties of QLNs-loaded DSs in absence of Nal anions.
based on Mg–Al LDHs, using basic form of HNal (Nal) as a probe,
2
.3. Structural and morphological characterization
with emphasis on the synthesis of NPs and the interactions of Nal
2+
with Mg ions. Nal-loaded DSs (LDH-Nal-DSs) were prepared by Mg/Al ratio was determined by energy dispersive X-ray spectro-
different coprecipitation methods and the composition, structure scopy (EDS) on pellets obtained by compression of 0.2 g of the
and morphology of the obtained samples were determined. corresponding sample at 2 tons in a 11 mm die. C and
Finally, the drug release profiles and the antimicrobial activity N content was determined by Pregl-Dumas method in a CHN
of the DSs were compared to that of the free drug.
2400 Series II Elemental Analyzer, using cysteine as reference.
ꢁ
1
Exchangeable Nal content was determined in 0.3 g L disper-
sions in PBS7.4, equilibrated for 24 hours. After centrifugation,
Nal concentration in the supernatants was determined by
UV-Vis spectrometry (Shimadzu UV1601, Japan), at l = 330 nm.
Thermogravimetric and differential thermal analyses (TG/DTA)
were performed in a Shimadzu DTG 60 instrument, under
2. Materials and methods
2
.1. Materials
MgCl O and AlCl
ꢀ6H
acid (HNal, Sigma Aldrich ), NaOH granules and 37% w w
s
2
2
3
ꢀ6H
2
O (PA grade, Anedra ), nalidixic
s
ꢁ1
ꢁ1
flowing air at a 10 1C min slope. Powder X-ray diffraction
s
2 4 2 4
HCl solution (PA grade, Cicarelli ), KH PO and K HPO (PA
(PXRD) patterns were recorded in a Phillips X’pert Pro instru-
s
s
grade, Anedra ) and NaCl (PA grade, Parafarm ) were used
without further treatment in the experiments. All solutions
were prepared with purified water (18 MO MilliQ, Millipore
System) and all experiments, unless otherwise stated, were
performed at room temperature. The LDH syntheses were done
ment equipped with a Pixcell 1D detector using a Cu Ka
lamp (l = 1.5418 Å) at 40 kV and 40 mA in continuous mode.
Fourier Transform Infrared Spectroscopy (FT-IR) spectra were
measured in a Bruker IFS28 instrument using KBr pellets
(1: 100 sample : KBr ratio).
using decarbonated water by boiling and N
2
bubbling. NaCl
The hydrodynamic apparent diameter (d) and zeta potential
z) of the samples were determined by dynamic light scattering
ꢁ
1
0.9% w v solution (NaCl0.9%) and phosphate buffer solutions
(
at pH 5.8 (PBS5.8) and 7.4 (PBS7.4) were prepared according to
USP specifications.
and electrophoretic light scattering measurements, respectively,
using a Delsa Nano C instrument (Beckman Coulter). The
size distributions were obtained by the CONTIN algorithm,
while electrophoretic mobilities were converted to z using the
2.2. Synthesis of the LDH-Nal-DSs
2
.2.1. Coprecipitation at variable pH. A 100 mL solution Smoluchowski equation. Morphological images of the solids
ꢁ1
ꢁ1
containing MgCl
quickly added to a 200 mL solution containing NaOH (0.45 mol L
and HNal (0.05 mol L ) under vigorous stirring provided by an To obtain SEM images, a drop of a 0.1 g L dispersion of each
Ultraturrax T18 BASIC agitator at 25000 rpm. Once the addition sample was placed on a silicon wafer, dried at 60 1C and covered
was finished, the prepared solid was separated by centrifugation with a Cr layer.
2 3
(0.3 mol L ) and AlCl (0.1 mol L ) was and EDS determinations were obtained in a Field Emission-
ꢁ
1
)
Scanning Electron Microscopy (FE-SEM) Sigma instrument.
ꢁ
1
ꢁ1
and washed twice with water. Subsequently, a suspension of the
DS obtained (LDH-Nal-pHvar) in 400 mL of water was prepared
2.4. Drug release studies
and aged by hydrothermal treatment at 80 1C for 4 hours. A Drug release profiles were determined in a bicompartmental
portion of this dispersion was freeze-dried to perform the diffusion device (Franz cells) mounted with a semisynthetic cellulose
s
chemical and structural characterization of the sample.
membrane (12 kDa, Sigma-Aldrich , USA). 1 mL of the DSs
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19144--19152 | 19145

View Article Online
NJC
Paper
(
or the free drug, used as reference) dispersions was carefully filled with 100 mL of the corresponding sample dispersion
ꢁ1
placed in the donor compartment. It was kept in contact with ([Nal] = 512 mg mL ). The plates were incubated for 24 hours
1
6.0 ꢂ 0.7 mL of receptor medium (water, NaCl0.9%, PBS5.8 at 37 1C. Then, the diameters (mm) of the growth inhibition
and PBS7.4) at 37 1C. 1.0 mL portions of receptor medium were zone were carefully measured with a caliper.
withdrawn at different time intervals and replaced with equivalent
volumes of preheated fresh medium. Drug concentration in
the samples was determined by UV-Vis spectrometry, using
calibration curves constructed for each receptor medium. All
experiments were carried out in triplicate.
3
. Results and discussion
3.1. Synthesis and characterization of LDH-Nal-DSs
The mean release profiles were fitted according to following The structure of the LDH-Nal-DSs were studied by PXRD patterns
3
0,31
equations:
(Fig. 1) and FT-IR spectra (Fig. 2). The PXRD patterns of both
LDH-Nal-pHvar and LDH-Nal-pHcte (Fig. 1e and f, respectively)
exhibited peaks corresponding to a LDH phase with a basal
spacing of 8.1 Å. This basal spacing was similar to that of LDH-Cl
Zero-order model: M = M + k ꢃ t
(1)
(2)
t
0
0
Mt
0:5
Higuchi model:
¼ kH ꢃ t
3
2
M0
(Fig. 1d), and it is typical of chloride intercalated LDHs.
Nevertheless, the chemical analysis of the samples (Table 1)
indicated an important amount of Nal anions were incorporated
to the solids. These results indicated that Nal was intercalated
Mt
n
Peppas model:
¼ kP ꢃ t
(3)
M0
where M
ment, M is the amount of drug permeated at time t, and k
and k
P
are the corresponding kinetic constants. Finally, n diffraction on (0 0 l) plane for the Nal intercalated phase and also
is the initial amount of drug in the donor compart- between LDH layers, which produced a strong structural
0
3
3
34
t
0
,
disorder and/or low layer stacking that completely hindered
k
H
parameter in the Peppas model equation (eqn (3)) characterizes widened the 00l peaks corresponding to the chloride intercalated
the release mechanism. For n = 0.5 and n = 1, the equation is one. The Nal anions were also placed at the particle surface,
3
+
identical to that of zero order (eqn (1)) and Higuchi (eqn (2)) where a portion of the positive charge generated by Al ions are
models, respectively, while n values between 0.5 and 1 indicate
an anomalous process or superposition of both phenomena.
When the drug release was stabilized within the time of the
experiments, the data used for the fittings where those under
80% of the maximum cumulative drug release.
2.5. Antimicrobial studies
The antimicrobial activity of the LDH-Nal-DSs was evaluated
against Staphylococcus aureus ATCC 25923 due to its incidence
5
in topical infections. The bacterial cultures were grown over-
night at 37 1C in nutrient Mueller–Hinton medium and then
8
ꢁ1
diluted to 10 CFU mL
.
MIC assays were performed by the broth dilution method,
where LDH-Nal-pHcte, LDH-Nal-pHvar and the pure drug were
tested at different concentrations, corresponding to [Nal] = 8,
ꢁ
1
1
6, 32, 64, 128, 256 and 512 mg mL in Mueller–Hinton broth
medium placed in individual tubes. In order to do that, 1 mL
5
ꢁ1
of Mueller–Hinton broth (10 CFU mL ) was added to 1 mL of
aliquot of each consecutive dilution. The microbial cultures were
incubated at 37 1C for 24 hours and the MIC was determined by
UV absorption measurements at l = 600 nm of the obtained
dispersions. Also, lawn cultures on Mueller Hinton agar plates
were made with dispersions that contained a [Nal] above the MIC.
These plates were inverted and incubated for 24 hours at 37 1C
and finally, the number of colony forming units was counted.
These experiments were performed in duplicate on separate.
Finally, the inhibition zone assay was used to compare the
antimicrobial activity of LDH-Nal-DSs and the pure drug against
8
ꢁ1
S. aureus. The bacterial cultures (10 CFU mL ) were distributed
over the surface of the Mueller–Hinton agar to obtain a uniform
growth. In each plate, 4 stainless steel cylinders of the same size
2
Fig. 1 PXRD patterns of (a) HNal, (b) NaNal, (c) Mg(Nal) , (d) LDH-Cl,
were placed on the surface of Mueller–Hinton agar and were (e) LDH-Nal-pHvar, (f) LDH-Nal-pHcte.
19146 | New J. Chem., 2018, 42, 19144--19152 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
and also with those of NaNal and Mg(Nal)
2
reference samples
ꢁ
1
(Fig. 2b and c). On the contrary, bands at 1718 cm (nCOOH)
were absent, which confirmed that HNal (Fig. 2a) was not
present in the LDH-Nal-DSs. Also, their spectra showed wide
bands corresponding to hydroxyl anions as well as wide bands
ꢁ
1
in the 900–400 cm range due to nM–O and nM–O–M of LDH
3
5
layers, analogous to those of LDH-Cl (Fig. 2d). Finally, LDH-
ꢁ1
Nal-pHcte presented bands at 542 and 495 cm , similar to
those obtained for Mg(Nal) and to those obtained Nal com-
2
2
+
3+ 36
plexes with Ca or Fe . On the contrary, these bands were
absent in LDH-Nal-pHvar spectrum.
The LDH-Nal-DSs composition was explained by two com-
peting reactions, namely the formation of a Nal-containing
LDH and the precipitation of Mg(Nal) , a coordination com-
2
2
8
2+
pound with a 1 : 2 Mg : Nal ratio. Titration curves of the Mg
3
+
and Al solutions in the presence of HNal (Fig. S1, ESI†)
indicated that Mg(Nal) was formed upon nalixidate dissolu-
tion in the presence of Mg ions. Nevertheless, increasing the
pH of the media produced the dissolution of the complex and
the precipitation of Mg ions as part of LDH layers. This
reaction can be written as:
2
2
+
2
+
2
+
ꢁ(aq)
Al(OH)3(s) + 2.5Mg (aq) + 5OH
Mg Al(OH)
+ 0.5Mg(Nal)
NalꢀnH
2
ꢀ4H
2
O
(s)
-
3
8
2
O
(s)
(4)
2
Fig. 2 FT-IR spectra of (a) HNal, (b) NaNal, (c) Mg(Nal) , (d) LDH-Cl,
Thus, the synthesis of LDH-Nal-pHvar involved the addition of
(e) LDH-Nal-pHvar, (f) LDH-Nal-pHcte. The main bands of Nal anions
ꢁ
the metal salts solution to a solution with a large [OH ] and
are indicated by grey lines.
3
7
[
Nal] and, consequently, the formation of the Nal containing
LDH phase was favored. On the other hand, the synthesis
ꢁ
located. This portion was particularly important for small LDH- method of LDH-Nal-pHcte maintained a low [OH ] (pH = 9)
3
8
NPs such as those observed by SEM (see below, Fig. 4). Instead, during the whole process in a medium with a high [Nal] and,
LDH-Nal-pHcte showed additional peaks, narrow and with high consequently the formation of Mg(Nal) was more favorable
intensity, identical to those obtained in the pattern of the than for LDH-Nal-pHvar. Mg(Nal) was also formed by Nal
reference sample Mg(Nal)
(Fig. 1c). No peaks of HNal interaction with LDHs, as demonstrated by adsorption experi-
Fig. 1a and Joint Committee on Powder Diffraction Standards ments on LDH-Cl (Fig. S2, ESI†). In these experiments, Nal
2
2
2
(
PDF data number 30-1820) or NaNal (Fig. 1b) were observed in anions showed a low affinity for LDH layers and they were
the patterns of the neither of the LDH-Nal-DSs, which indicates incorporated by two main processes: adsorption at low [Nal]
that neither form of the pure drug was present in the samples. and Mg(Nal)
The incorporation of Nal anions and the presence of particle size and stacking of LDH-Cl sample, Nal incorporation
Mg(Nal) in LDH-Nal-pHcte were confirmed by FT-IR spectra at the surface of LDH-Cl particles was quite lower than for
Fig. 2). Both LDH-Nal-DSs (Fig. 2e and f) exhibited bands of the samples obtained by direct synthesis by coprecipitation.
2
precipitation at high ones. Due to the large
2
(
ꢁ
1
ꢁ1
Nal anions at 1570–1590 cm and 1390–1400 cm ranges A complete description of both titration and adsorption experi-
ꢁ
1
(nasym,COO and nsym,COO), 1629 cm
(nCO, ketone group), ments are included with the Fig. S1 and S2 of the ESI.†
ꢁ
1
ꢁ1
1
448 cm (dC–H) and 1262 cm (nCO + nNO coupling). The The incorporation of Nal anions to the LDH phase was con-
positions of these bands were similar between both samples firmed by the TG/DTA curves (Fig. 3). Thus, LDH-Nal-pHvar (Fig. 3e)
Table 1 Chemical analyses and proposed formulae of the LDH-Nal-DSs. Theoretical values according to the proposed formulae are given between
parentheses
a
b
b
c
d
e
Sample
LDH-Cl
Mg : Al % C
%N
%H
2
O
%w750
%Nalexc
Formula
2.8 (2.8) —
—
14.1 (14.0) 55.9 (54.2) —
Mg0.74Al0.26(OH)
2
Cl0.26ꢀ0.62H
2
O
LDH-Nal-pHvar 3.3 (3.3) 18.3 (18.0) 3.3 (3.4) 15.3 (15.3) 39.0 (41.2) 32.0 (27.9) Mg0.77Al0.23(OH)
2
Nal0.13Cl0.10ꢀ0.88H
2
O
2
LDH-Nal-pHcte 2.4 (2.4) 28.4 (26.7) 5.1 (5.1) (14.0) (14.1) 29.5 (31.9) 35.7 (36.1) Mg0.70Al0.30(OH)2.00Nal0.22Cl0.08ꢀ0.99H
O + 0.02Mg(Nal)
2
2
ꢀ4H O
Mg(Nal)
2
—
50.4 (51.6) 10.2 (10.0) 12.7 (12.9) 6.3 (7.2) 5.4 (0)
Mg(Nal)
2
ꢀ4H
2
O
a
ꢁ1
b
ꢁ1
c
mol mol , determined by EDS. w w , as determined by the Pregl–Dumas method. Residual weight at 175 1C as determined by TG analysis
Fig. 3). Residual weight at 750 1C as determined by TG analysis (Fig. 3), theoretical values were calculated considering that the only residues at
this temperature are metal oxides. Exchangeable Nal percentage (w w ), determined by UV-Vis spectrometry in 1 g L dispersions in PBS7.4.
d
(
e
ꢁ1
ꢁ1
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19144--19152 | 19147

View Article Online
NJC
Paper
2
Fig. 3 TG/DTA curves of (a) HNal, (b) NaNal, (c) Mg(Nal) , (d) LDH-Cl, (e) LDH-Nal-pHvar, (f) LDH-Nal-pHcte.
showed the three main processes involved in LDHs thermal the exothermic peak at 490 1C were similar to those of LDH-Nal-
2
decomposition, namely dehydration, dehydroxylation and anion pHvar. Nevertheless, the presence of the complex Mg(Nal) O
2
ꢀ4H
decomposition. Dehydration process extended up to 180 1C in the sample LDH-Nal-pHcte produced a widening of the
approximately with the peak of the associated endothermic effect exothermic peak and its displacement to a temperature inter-
centered at 90 1C, while the weight loss due to dehydroxylation mediate between those of LDH-Nal-pHvar and Mg(Nal) . The
2
and loss of the Nal anion was produced between 250 1C and weight of the residue at 750 1C (%w750) was 29.5% for this sample,
6
5
00 1C, where the main peak was exothermic and registered at which was lower than that of LDH-Nal-pHvar (39.0%) but much
20 1C. The peaks obtained for this sample was quite different higher than that obtained for Mg(Nal) reference sample.
The structural characterization, together with the weight
2
to those obtained for either HNal or NaNal (Fig. 3a and b,
respectively), which confirmed that neither of the pure drug forms losses obtained in the TG curves and the elemental analysis
was precipitated together with the LDH phase. Alternatively, performed to the samples, allowed establishing the chemical
LDH-Nal-pHcte (Fig. 3f) curve showed an intermediate behavior formulae of the LDH-Nal-DSs (Table 1). There was a good
between those of LDH-Nal-pHvar and Mg(Nal) (Fig. 3c). The correlation between the experimental values and those calcu-
2
curve of the latter presented two endothermic effects, centered at lated on the basis of the proposed formula, which indicated
100 and 140 1C, assigned to dehydration, while the drug oxida- that they were a good approximation to the actual composition
tion produced a sudden weight loss that started at 400 1C and of the LDH-Nal-DSs. LDH-Nal-pHvar formula was proposed
presented exothermic peaks at 460, 560 and 670 1C. The %w750 considering the presence of only one LDH phase containing
ꢁ
ꢁ
value for Mg(Nal)
formula Mg(Nal)
2
(6.3%) was concordant with the chemical both Nal and Cl anions. The Mg : Al ratio (3.3 : 1) was similar
O, considering MgO as the only residue to that of the starting solution, while the Nal content corre-
2
ꢀ4H
2
at this temperature. Finally, LDH-Nal-pHcte showed an inter- sponded to 54% of the anion exchange capacity (AEC) of
mediate behavior between LDH-Nal-pHvar and Mg(Nal) . Thus, the LDH layers. On the other hand, LDH-Nal-pHcte formula
2
the dehydration step (up to 200 1C) produced a couple of considered the presence of two phases, one corresponding to a
thermal events similar to those registered for Mg(Nal)
were framed inside a wide band similar to that of LDH-Nal-pHvar. equivalent to 70% of the AEC and another corresponding to
Also, the continuous weight loss in the 250–500 1C range and the Mg(Nal) salt (Mg(Nal) O). According to the chemical
ꢀ4H
2
, but they Nal intercalated LDH with a 2.3 : 1 Mg: Al ratio and a Nal content
2
2
2
19148 | New J. Chem., 2018, 42, 19144--19152 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
formula proposed, approximately 85% of the Nal anions were of large particles due to the larger light dispersion capacity of
3
9
included between the LDH layers, while only 15% were in the large particles.
Mg(Nal) salt. The percentage of exchangeable Nal (%Nalexc),
determined by the drug release produced in PBS7.4, exhibited
2
3.2. Drug release studies
a good fitting for both LDH-Nal-pHvar and LDH-Nal-pHcte. Nal release from LDH-Nal-DSs was studied to evaluate their
Nevertheless, for the latter, it was considered that Nal ions in performance for topical administration (Fig. 5) and the results
Mg(Nal)
2
ꢀ4H
2
O were not released in this medium, in good obtained were compared to those of the free drug. Fitting of
release data with representative kinetic models (zero order,
accord with the %Nalexc (o6%) of this reference sample.
LDH-Nal-pHvar presented a wide size distribution (Fig. 4A) Higuchi and Peppas) allowed to parameterize the kinetic behavior
with a maximum at 342 nm which was in the same order to that and to give a insight on the drug release mechanism (Table 2).
obtained for LDH-Cl (76 nm). Nevertheless, while the SEM Theoretical kinetic profiles obtained with the better fits of each
images of LDH-Cl-NPs (Fig. 4B) showed thick, regularly shaped sample and medium (marked with * in Table 2) were included
particles with sizes within the size range similar to those in Fig. 5 to provide a guide for the eye and an indication of the
obtained by DLS (between 50 and 200 nm according to Fig. 4A), goodness-of-fit of these models.
the LDH-Nal-NPs in Fig. 4D presented particle size between 30
and 40 nm, quite below to the size distribution obtained by DLS NaCl0.9% and PBS5.8, which produced cumulative Nal releases
between 100 nm and 1 mm), which was associated to aggrega- at the end of the experiment (%Nal360) around 13%. Also, the
The free drug showed similar behaviors towards water,
(
tion processes. The morphology of LDH-Nal-pHvar (Fig. 4D) release profiles were linear in all cases, as indicated by the
2
particles (thin, small, and irregularly shaped) was consistent regression coefficient values (R ), above 0.99 for zero order
with the low crystallinity and layer stacking indicated by its model. Then, the determining step for Nal release from the
PXRD patterns. On the other hand, the particle size distribution free drug was dissolution from the particle surface, which was
of LDH-Nal-pHcte was displaced to the micrometric range, limited for all these media due to the low solubility of HNal.
with a wide size distribution. The particle size increase for Instead, %Nal360 reached around 33% for PBS7.4 and, contra-
LDH-Nal-pHcte was not due to aggregation of LDH-NPs but rily to that obtained in the previous media, drug release
2
to the presence of the Mg(Nal) phase. Effectively, the SEM stabilized within the time of the experiment. Nevertheless,
images of LDH-Nal-pHcte (Fig. 4E and F) showed the same the best fit for the profile (up to 80% of the overall release at
small particles observed for LDH-Nal-pHvar and assigned to the the plateau) was obtained for Peppas model with n B 1, which,
LDH-Nal-NPs phase, but also needle-shaped particles with a as previously stated, relates to a zero-order model. Consequently,
length of a few mm assigned to the Mg(Nal) complex. These drug dissolution was still the rate determining step in this
2
particles present a similar shape to those of Mg(Nal)
Fig. 4C) but, in the case of LDH-Nal-pHcte, they were quite ment due to neutral pH of the receptor media.
larger and they were covered by LDH-Nal-NPs. These results On the other hand, the release profiles from LDH-Nal-pHvar
confirmed the presence of a mixture of LDH-Nal-NPs and were quite dependent on the receptor media. Thus, %Nal360
Mg(Nal)
2
sample media, and the only difference was a HNal solubility enhance-
(
2
particles. Although the particles of the former were increased in the order water { PBS5.8 o NaCl0.9% { PBS7.4
more numerous than those of the latter in Fig. 4E and F, the (6, 12, 14 and 22%, respectively). The best fits were obtained for
size distribution by DLS (Fig. 4A) overweighed the presence Higuchi model, which indicated that diffusion was the determining
2
Fig. 4 Intensity size distribution of LDH-Nal-DSs and reference samples (A). SEM images of LDH-Cl (B), Mg(Nal) (C), LDH-Nal-pHvar (D), LDH-Nal-
pHcte (E and F).
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19144--19152 | 19149

View Article Online
NJC
Paper
Fig. 5 In vitro Nal release profile from the free drug, LDH-Nal-pHvar and LDH-Nal-pHcte, placed as dispersions in the donor compartment and released
toward (A) water, (B) NaCl0.9%, (C) PBS5.8 and (D) PBS7.4. Lines are calculated release profiles using the best fitting model for each experiment, marked
with asterisks (*) in Table 2.
2
Table 2 Kinetic data obtained by modeling Nal release profiles with zero order, Higuchi and Peppas model. Fittings with the best R are marked with
asterisks
Kinetic model
Release media
Water
Zero order
Higuchi
Peppas
2
2
2
Sample
%Nal
0
K
Z
R
%Nal
0
K
H
R
ꢁlog(K
P
)
n
R
HNal
LDH-Nal-pHvar
LDH-Nal-pHcte
0.36*
0.71
0.29*
0.0364*
0.0163
0.0154*
0.995*
0.984
0.997*
ꢁ2.6
0.78
0.35
0.33
0.981
0.994
0.971
1.37
0.90*
1.31
0.99
0.66*
0.80
0.993
0.999*
0.995
ꢁ0.66
ꢁ0.95
NaCl 0.9%
HNal
LDH-Nal-pHvar
LDH-Nal-pHcte
0.20*
1.57
0.41
0.036*
0.040
0.014
0.997*
0.960
0.985
ꢁ2.71
ꢁ1.94*
ꢁ0.76
0.77
0.88*
0.30
0.975
0.998*
0.992
1.46
0.88
1.24*
1.02
0.83
0.77*
0.997
0.985
0.999*
PBS pH = 5.8
PBS pH = 7.4
HNal
LDH-Nal-pHvar
LDH-Nal-pHcte
0.29
1.50
0.36
0.038
0.034
0.020
0.997
0.939
0.995
ꢁ2.79
ꢁ1.6*
ꢁ1.25
0.81
0.76*
0.42
0.976
0.997*
0.982
1.28*
1.67
1.25*
0.95*
1.17
0.82*
1.000*
0.861
1.000*
HNal
LDH-Nal-pHvar
LDH-Nal-pHcte
0.42
2.45
0.54
0.30
0.094
0.018
0.997
0.959
0.988
ꢁ6.02
ꢁ2.31*
ꢁ0.95
3.03
1.53*
0.38
0.986
1.000*
0.990
0.48*
0.42
1.06*
0.98*
0.77
0.74*
0.999*
0.980
1.000*
step of the drug release. Such behavior is typical of LDHs at
2
Finally, the presence of Mg(Nal) in LDH-Nal-pHcte produced
pHs above 5, where layers dissolution is minimized and anion relevant changes respect to that obtained for LDH-Nal-pHvar.
4
0,41
exchange is the main release mechanism.
The exception %Nal360 was lower (around 6–7%) for the former and quite
was the release profile for water, whose best fit was obtained independent of the receptor media. Also, the best fits were not
for Peppas model with n = 0.66. Anion exchange was limited in obtained for Higuchi model, which indicated a change in the
this media by the absence of anions, and other processes, such release mechanism. In the case of water, the best fit was
as matrix dissolution, became relevant. Oppositely, the fastest obtained for the zero-order model, which indicated that anion
release was observed for PBS7.4 due to presence of divalent exchange was completely suppressed. For the remaining media,
2
ꢁ
13
HPO
4
anions, which present high affinity for LDHs.
the best fits were obtained for Peppas model with n values
19150 | New J. Chem., 2018, 42, 19144--19152 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
The dispersions of the LDH-Nal-DSs became increasingly
turbid at large [Nal], which was not due only to an increasing
particle concentration, but also to an increasing bacterial
proliferation, as determined by the colony forming units
count (Fig. S3, ESI†). Nevertheless, this is not an uncommon
4
4
behavior for QLNs: i.e., Nal presents its higher activity in the
ꢁ
1
5
0–200 mg mL concentration range for most Gram-negative
4
5
bacteria, while its activity decrease at large concentrations.
This effect was enhanced for LDH-Nal-NPs due to the permea-
tion enhancement proposed above.
These results were complemented with inhibition zone
assays (Fig. 6B). The diameter of the inhibitory zone for pure
incubation, (B) zones of bacterial inhibition at [Nal] = 512 mg mL in Nal was 12 mm, which indicated its poor antimicrobial activity
Fig. 6 Microbiological studies of LDH-Nal-DSs and the pure drug.
(
A) Absorbance at l = 600 nm of cultures (A600)vs. [Nal], after 24 hours
ꢁ1
8
Mueller Hinton agar plates seeded with 10 CFU bacterial cultures.
against S. aureus. LDH-Nal-pHvar presented a slightly larger
protection area (14 mm) than the pure drug, which was in good
agreement with the lower MIC obtained for this sample. Also,
the diameter of the inhibitory zone for LDH-Nal-pHcte was the
same of the pure drug (12 mm), which was consistent with its
higher MIC and its slower release kinetics compared to LDH-
Nal-pHvar.
between 0.74 and 0.82. These results indicated that Mg(Nal)2
dissolution was an important drug release mechanism for LDH-
Nal-pHcte. Due to the protonation reactions of Nal anions
2
(Fig. S1, ESI†), Mg(Nal) increased its solubility at pHs lower
than 8. This mechanism explained that, contrarily to LDH-Nal-
pHvar, the %Nal360 values were less affected by the anions
present in the release medium. Nevertheless, there was an 4. Conclusions
increasing contribution of the anion exchange mechanism with
the affinity of the anions in the release media, as demonstrated
Layered double hydroxide-based delivery systems containing
the basic form of nalidixic acid (LDH-Nal-DSs) were prepared
using different synthesis procedures. Pure Nal-containing LDH
nanoparticles (LDH-Nal-NPs) were obtained by coprecipitation
at variable pH, while an additional phase, corresponding to a
by the decreasing n values for water (no exchanging anions),
ꢁ
ꢁ
0
.9% NaCl (Cl ), PBS5.8 (mainly H
2
PO
4
anions) and PBS7.4
2
ꢁ
(mainly HPO
4
anions).
These results indicated that, although both LDH-Nal-DSs
Mg(Nal) ꢀ4H O complex, was obtained for the one at constant
presented good properties for their application in topical for-
mulations, their characteristics were quite different. Thus, Nal
release from LDH-Nal-pHvar toward both PBS5.8 and PBS7.4, which
simulate intact and damage skin, respectively, indicated a pH-
sensitive behavior that was not present on LDH-Nal-pHcte. This
behavior would lead to a faster release in an injured area, where it is
more needed. Also, due the faster drug release from LDH-Nal-pHvar
than from LDH-Nal-pHcte, the former was expected to achieve
more rapidly MIC values, and the latter to provide a more
sustained action independently of the release medium.
2
2
pH (LDH-Nal-pHcte). Drug release from LDH-Nal-NPs was
produced mainly by an anion exchange mechanism and its
rate was dependent on the exchanging anions concentration.
The presence of Mg(Nal)
diminution, which was explained by the change in the main
release mechanism due to the influence of Mg(Nal) dissolu-
2
in LDH-Nal-pHcte led to a release rate
2
tion in this sample. LDH-Nal-NPs also produced an enhanced
antimicrobial activity, which was related to their positive z and
the consequent interactions with the bacterial cell membrane,
negatively charged. These results indicate that LDH-NPs portray
interesting features as DSs for (fluoro)quinolones, but careful
control of the synthesis method is necessary to control the
interactions of these drugs with the metal ions of LDHs and,
therefore, the properties of the resulting DSs.
3.3. Antimicrobial activity
The microbiological studies (Fig. 6) showed that LDH-Nal-DSs
possessed better activity than the pure drug. Thus, the MIC
ꢁ1
determined for the pure drug was 128 mg mL , while it was 64
ꢁ
1
and 32 mg mL for LDH-Nal-pHcte and LDH-Nal-pHvar, respectively
(
Fig. 6A). The larger activity of LDH-Nal-DSs was related exclusively to Conflicts of interest
an enhancement of the antimicrobial action of Nal, as demonstrated
by the absence of antimicrobial activity for LDH-Cl in the concen-
tration range studied. The MIC diminution was explained by the
mechanism of action of HNal and effects produced by its vehiculiza-
tion as LDH-Nal-NPs. The antimicrobial activity of Nal is based
on the interference with the synthesis of nucleic acids and, Salguero Y thanks to Secretar ´ı a de Educaci o´ n Superior, Ciencia,
consequently, permeation of the bacterial cell membrane is Tecnolog ´ı a e Innovaci ´o n (SENESCYT) de la Rep u´ blica del Ecuador
essential to its antimicrobial action. Due to the positive charge for her master fellowship. Garc ´ı a MC thanks to CONICET for
and size of the LDH-Nal-NPs, they are able to interact with the her postdoctoral fellowship. Economic supports by SeCyT-UNC,
There are no conflicts to declare.
Acknowledgements
4
2
2
4,43
cell membrane, increasing in Nal permeability.
project number 05/C585, FONCyT, project PICT 12/0634, and
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19144--19152 | 19151

View Article Online
NJC
Paper
CONICET, PIP 11220120100575, are gratefully acknowledged. 22 H. Nabipour, M. H. Sadr and N. Thomas, New J. Chem.,
SEM images were obtained at LAMARX.
2016, 40, 238–244.
2
2
3 B. Kutlu, P. Schr o¨ ttner, A. Leuteritz, R. Boldt, E. Jacobs and
G. Heinrich, Mater. Sci. Eng., C, 2014, 41, 8–16.
4 S. Malekkhaiat Haffner, L. Nystrom, R. Nordstrom, Z. P. Xu,
M. Davoudi, A. Schmidtchen and M. Malmsten, Phys. Chem.
Chem. Phys., 2017, 19, 23832–23842.
References
1
2
3
4
5
J. S. Wolfson and D. C. Hooper, Antimicrob. Agents Chemother.,
985, 28, 581–586.
1
M. H. Xiong, Y. Bao, X. Z. Yang, Y. H. Zhu and J. Wang, Adv. 25 I. Turel, Coord. Chem. Rev., 2002, 232, 27–47.
Drug Delivery Rev., 2014, 78, 63–76.
26 B. L. Lomaestro and G. R. Bailie, Drugs, 2001, 25, 411–416.
P. Gao, X. Nie, M. Zou, Y. Shi and G. Cheng, J. Antibiot., 27 M. Nakano, M. Yamamoto and T. Arita, Chem. Pharm. Bull.,
011, 64, 625–634.
1978, 26, 1505–1510.
R. S. Kalhapure, N. Suleman, C. Mocktar, N. Seedat and 28 N. B. Behrens, G. M. Diaz and D. M. L. Goodgame, Inorg.
T. Govender, J. Pharm. Sci., 2015, 104, 872–905.
Chim. Acta, 1986, 125, 21–26.
M. C. Garc ´ı a, J. C. Cuggino, C. I. Rosset, P. L. P ´a ez, M. C. 29 A. J. G. Bailey, A. Cole, J. Goodfield, P. M. May, M. E. Dreyfuss,
2
Strumia, R. H. Manzo, F. L. Alovero, C. I. Alvarez Igarzabal and
J. M. Midgley and D. R. Williams, Int. J. Pharm., 1984, 22,
A. F. Jimenez-Kairuz, Mater. Sci. Eng., C, 2016, 69, 236–246.
283–290.
6
7
8
9
V. Rives, M. del Arco and C. Mart ´ı n, Appl. Clay Sci., 2014, 30 P. Costa and J. M. Sousa Lobo, Eur. J. Pharm. Sci., 2001, 13,
8–89, 239–269.
123–133.
M. Jafarbeglou, M. Abdouss, A. M. Shoushtari and 31 J. Siepmann and N. A. Peppas, Adv. Drug Delivery Rev., 2001,
M. Jafarbeglou, RSC Adv., 2016, 6, 50002–50016.
48, 139–157.
D. Mosangi, L. Moyo, S. Kesavan Pillai and S. S. Ray, RSC 32 V. A. Drits and A. S. Bookin, in Layered Double hydroxides:
8
Adv., 2016, 6, 105862–105870.
Present and Future, ed. V. Rives, Nova Science, New York,
Z.-L. Liu, D.-Y. Tian, S.-P. Li, X.-D. Li and T.-H. Lu, Int.
J. Pharm., 2014, 473, 414–425.
0 M. A. Djebbi, A. Elabed, Z. Bouaziz, M. Sadiki, S. Elabed,
2001, pp. 39–92.
33 D. Mosangi, S. Kesavan Pillai, L. Moyo and S. S. Ray,
RSC Adv., 2016, 6, 77709–77716.
1
1
P. Namour, N. Jaffrezic-Renault and A. B. H. Amara, Int. 34 S. P. Lonkar, A. Leuteritz and G. Heinrich, RSC Adv., 2013, 3,
J. Pharm., 2016, 515, 422–430.
1495–1501.
1 E. P. Komarala, S. Doshi, S. Thiyagarajan, M. Aslam and 35 J. T. Kloprogge and R. L. Frost, in Layered Double Hydroxides:
D. Bahadur, New J. Chem., 2018, 42, 129–136.
Present and Future, ed. V. Rives, Nova Science Publishers,
1
1
2 S. P. Newman and W. Jones, New J. Chem., 1998, 22, 105–115.
New York, 2001, pp. 139–192.
3 R. Rojas, M. C. Palena, A. F. Jimenez-Kairuz, R. H. Manzo 36 F. A. I. Al-Khodir and M. S. Refat, J. Mol. Struct., 2015, 1094,
and C. E. Giacomelli, Appl. Clay Sci., 2012, 62–63, 15–20.
22–35.
4 Z. Gu, A. C. Thomas, Z. P. Xu, J. H. Campbell, G. Qing and 37 J. S. Valente, M. S ´a nchez-Cant u´ , E. Lima and F. Figueras,
M. Lu, Chem. Mater., 2008, 3715–3722.
Chem. Mater., 2009, 21, 5809–5818.
5 L. Perioli, V. Ambrogi, L. di Nauta, M. Nocchetti and 38 E. M. Seftel, E. Popovici, M. Mertens, K. De Witte,
1
1
1
1
1
C. Rossi, Appl. Clay Sci., 2011, 51, 407–413.
G. Van Tendeloo, P. Cool and E. F. Vansant, Microporous
Mesoporous Mater., 2008, 113, 296–304.
39 J. Stetefeld, S. A. McKenna and T. R. Patel, Biophys. Rev.,
2016, 8, 409–427.
6 G. Choi, J.-H. Lee, Y.-J. Oh, Y. Bin Choy, M. C. Park, H. C.
Chang and J.-H. Choy, Int. J. Pharm., 2010, 402, 117–122.
7 W. Zhang, P. Jiang, Y. Chen, P. Luo, G. Li, B. Zheng,
W. Chen, Z. Mao and C. Gao, Nanoscale, 2016, 8, 9572–9582. 40 M. del Arco, A. Fernandez, C. Martin and V. Rives, Appl. Clay
8 D. Hesse, M. Badar, A. Bleich, A. Smoczek, S. Glage, M. Kieke,
Sci., 2009, 42, 538–544.
P. Behrens, P. P. M u¨ ller, K. H. Esser, M. Stieve and N. K. 41 M. L. Parello, R. Rojas and C. E. Giacomelli, J. Colloid
Prenzler, J. Mater. Sci.: Mater. Med., 2013, 24, 129–136.
Interface Sci., 2010, 351, 134–139.
9 Z. Rezvani, M. Shahbaei and M. S. Zolfaghar Rezvani, Polym. 42 F. C. Tenover, Am. J. Infect. Control, 2006, 34, 3–9.
1
Polym. Compos., 2008, 16, 101–113.
43 C. Vasti, V. Pfaffen, E. Ambroggio, M. R. Galiano, R. Rojas
20 F. Duda, M. Kieke, F. Waltz, M. E. Schweinefuss, M. Badar,
and C. E. Giacomelli, Appl. Clay Sci., 2017, 137, 151–159.
P. P. M u¨ ller, K. H. Esser, T. Lenarz, P. Behrens and 44 G. Cheng, H. Hao, M. Dai, Z. Liu and Z. Yuan, Eur. J. Med.
N. K. Prenzler, J. Mater. Sci.: Mater. Med., 2015, 26, 5334.
Chem., 2013, 66, 555–562.
1 Y. Li, H.-Y. Bi, H. Li, X.-M. Mao and Y.-Q. Liang, Mater. Sci. 45 G. C. Crumplin and J. T. Smith, Antimicrob. Agents Chemother.,
Eng., C, 2017, 78, 886–891. 1975, 8, 251–261.
2
19152 | New J. Chem., 2018, 42, 19144--19152 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018