Research Paper
Inter-nanocarrier and nanocarrier-to-cell transfer assays demonstrate the risk of an immediate unloading of dye from labeled lipid
nanocapsules
Carl Simonsson
a,b,1, Guillaume Bastiat
a,b,⇑,1, Marion Pitorre
a,b, Andrey S. Klymchenko
c, Jérôme Béjaud
a,b, Yves Mély
c, Jean-Pierre Benoit
a,baINSERM, U 1066, Micro et Nanomédecines biomimétiques – MINT, Angers F-49933, France
bLUNAM Université, UMR-S1066, Angers F-49933, France
cLaboratoire de biophotonique et de Pharmacologie, UMR 7213 CNRS, Université de Strasbourg, Faculté de Pharmacie, 74 route du Rhin, 67401 Illkirch, France
a r t i c l e i n f o
Article history:
Received 8 July 2015 Revised 10 October 2015
Accepted in revised form 21 October 2015 Available online 30 October 2015
Keywords:
Lipid nanocapsule Fluorescent dye Labeling of nanocarriers Cell culture
Size-exclusion chromatography FRET
a b s t r a c t
Release studies constitute a fundamental part of the nanovector characterization. However, it can be dif- ficult to correctly assess the release of lipophilic compounds from lipid nanocarriers using conventional assays. Previously, we proposed a method including an extraction with oil to measure the loading stabil- ity of lipophilic dyes in lipid nanocapsules (LNCs). The method indicated a rapid release of Nile Red from LNCs, while the loading of lipophilic carbocyanine dyes remained stable. This method, although interest- ing for a rapid screening of the fluorescence labeling stability of nanocarriers, is far from what happens in vivo, where lipid acceptor phases are nanostructured. Here, lipophilic dye loading stability has been assessed, by monitoring dye transfer from LNCs toward stable colloidal lipid nanocompartments,i.e.
non-loaded LNCs, using new methodology based on size exclusion chromatography (SEC) and Förster Resonance Energy Transfer (FRET). Dye transfer between LNCs and THP-1 cells (as model for circulating cells) has also been studied by FACS. The assays reveal an almost instantaneous transfer of Nile Red between LNCs, from LNCs to THP-1 cells, between THP-1 cells, and a reversal transfer from THP-1 cells to LNCs. On the contrary, there was no detectable transfer of the lipophilic carbocyanine dyes. Dye release was also analyzed using dialyses, which only revealed a very slow release of Nile Red from LNCs, demon- strating the weakness of membrane based assays for investigations of the lipophilic compound loading stability in lipid nanocarriers. These results highlight the importance of using relevant release assays, and the potential risk of an immediate unloading of lipophilic fluorescent dyes from lipid nanocarriers, in the presence of a lipid acceptor nanocompartment. Some misinterpretations of cellular trafficking andin vivobiodistribution of fluorescent nanoparticles should be avoided.
Ó2015 Elsevier B.V. All rights reserved.
1. Introduction
Recent advances in nanomedicine have provided significant improvements in drug delivery and delivery of contrast agents for bio-imaging applications. Lipid vesicles, polymeric nanoparti- cles, and lipid nanocapsules (LNCs) are examples of nanovectors used in pharmaceutical applications [1,2]. Better selectivity, increased uptake over biological barriers, sustained release, decreased toxic side effects, and reduced chemical and enzymatic
degradation of active agents, are some of the reported benefits of nanovectors compared to ‘free drugs’[3]. Fluorescent dye-labeled nanocarriers are often used inin vitroandin vivomechanistic stud- ies. For example, cellular trafficking[4–6]andin vivobiodistribu- tion [7–11] studies have been conducted using fluorescent dye labeled nanocarriers. Advantages of fluorescence techniques include simple labeling procedures, high sensitivity, no specific requirement regarding the handling of the labels, and the use of standardized analytical methods which are available in most labo- ratories, e.g. fluorescence spectrophotometry, fluorescence micro- scopy, FACS, andin vivofluorescence imaging.
Dye-labeled nanocarriers are often obtained through non- covalent association between label and carriers. Consequently, one important drawback is the risk of uncontrolled release
http://dx.doi.org/10.1016/j.ejpb.2015.10.011 0939-6411/Ó2015 Elsevier B.V. All rights reserved.
⇑ Corresponding author at: MINT – UMR_S1066, Institut de Biologie en Santé – IRIS, 4 rue Larrey CHU, 49933 Angers Cedex 9, France. Tel.: +33 244 688531; fax:
+33 244 688546.
E-mail address:guillaume.bastiat@univ-angers.fr(G. Bastiat).
1 Equal contribution to the work.
Contents lists available atScienceDirect
European Journal of Pharmaceutics and Biopharmaceutics
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / e j p b
in vitroor duringin vivocirculation. Before being applied, the sta- bility of the label must be studied to ensure that it does not leak from the carrier. However, it can be difficult to correctly assess the encapsulation stability of lipophilic compounds. The nature of the dispersion medium is a key factor controlling the release.
When dispersed in an aqueous phase, lipophilic compounds will obviously remain inside the lipid nanocarriers. Thus, a classical release assay,e.g.measuring release against water or a buffer, will not mimic the release in anin vivoenvironment where multiple lipophilic acceptor compartments are present. Therefore, in our previous work, we proposed a new methodology to investigate the labeling stability of LNCs loaded with lipophilic fluorescent dyes[12]. Briefly, dye release from LNCs toward lipophilic acceptor compartments was assessed by vortexing an aqueous suspension of labeled LNCs with an oil, followed by a separation by centrifugation.
The results showed that two dyes, Nile Red (NR) and Coumarin- 6 (6-Cou), were readily transferred between LNCs and the oil, while three lipophilic carbocyanine dyes (1,10-dioctadecyl-3,3,30,30-tetra methylindocarbocyanine perchlorate: DiI, 1,10-dioctadecyl-3,3,30,3
0-tetramethylindodicarbocyanine 4-chlorobenzenesulfonate: DiD and 3,30-dioctadecyloxacarbocyanine perchlorate: DiO) did not transfer. These two opposite properties could be explained by dif- ferent localizations of the dyes inside the nanocarriers. For NR, a similar rapid release was demonstrated when LNCs were incubated in serum, using a Förster Resonance Energy Transfer (FRET) assay [13]. Assuming that the vortex with the oil has no direct effect on the stability of the label, the results highlight the possibility of a very rapid release of lipophilic compounds from intact lipid nanocarriers.
However, the nature of the oil acceptor phase in the extraction assay is relatively far from the nature of acceptor compartments present underin vivoconditions. Indeed, in systemic circulation, lipid nanoparticles such as lipoproteins can be found, as well as the lipid nanocompartments of cells (phospholipid bilayer of endothelial cells, monocytes, etc.), and not free macroscopic oil phase. New methodologies, in which the acceptor consists of stable colloidal lipophilic nano-compartments, need to be investigated.
The aim of this study was to investigate the loading stability of fluorescent dyes (NR and lipophilic carbocyanine dyes) in LNCs, when the loaded nanoparticle suspension was in contact with non-loaded LNC suspension (as lipid acceptor phase), instead of a single oil phase. We did not primarily want to use a dialysis assay, as it was believed that the membrane could have an influence on the measured release rates, and is far from an in vivosituation.
Instead, we first studied dye transfer between LNCs of different sizes, using a new method based on a separation by size exclusion chromatography (SEC). In a second step, transfer was investigated by monitoring the FRET signals between encapsulated dyes, for LNCs with identical sizes. Thirdly, for a comparison of methods and to test our assumption regarding the dialysis membrane, we also performed a dialysis assay. In a final step, dye transfer was investigated from LNCs toward THP-1 cells (monocyte cell line).
This was already showed with HEI-OC1 cells (auditory cell line) [12]and we wanted to confirm the behavior was not cell depen- dent, and could occur with model for circulating cells. Dye transfer was also investigated between THP-1 cells, and from THP-1 cells back to non-loaded LNCs, as these events could be observed in anin vivosituation.
2. Materials and methods 2.1. Chemicals
LipoïdÒ S75-3 (soybean lecithin: 69% of phosphatidylcholine and 10% of phosphatidylethanolamine) and KolliphorÒ HS15
(mixture of free polyethylene glycol 660 and polyethylene glycol 660 hydroxystearate) were kindly supplied by Lipoïd GmbH (Ludwigshafen, Germany) and BASF (Ludwigshafen, Germany), respectively. LabrafacÒ WL 1349 (caprylic-capric acid triglycerides) was generously provided by Gattefossé S.A.
(Saint-Priest, France). NaCl was purchased from Prolabo (Fontenay-Sous-Bois, France). Disodium hydrogen phosphate and sodium dihydrogen phosphate were purchased from VWR International SAS (Fontenay-Sous-Bois, France). Deionized water was obtained from a Milli-Q plusÒ system (Millipore, Billerica, MA). Nile Red (NR) and Sepharose CL4b were purchased from Sigma–Aldrich (St. Quentin Fallavier, France). DiI, DiD and DiO were provided by Molecular ProbesÒ (Eugene, OR).
40-Dioctylamino-3-octyloxyflavone (F888) was synthesized as described[13].
2.2. Lipid nanocapsule formulation
Non-labeled and fluorescent-labeled LNC formulation processes were based on a phase-inversion property, which has been described elsewhere [2,12,14]. Briefly, the quantities of Labrafac (Lab) (oil phase), water, NaCl (aqueous phase), Kolliphor (Kol) and Lipoid (Lip) (surfactants) for each formulation were precisely weighed. For 30 nm LNC (LNC30), mLab= 0.846, mKol= 1.934, mLip= 0.075, mWater= 2.055 and mNaCl= 0.103 g; for 60 nm LNC (LNC60), mLab= 1.028, mKol= 0.846, mLip= 0.075, mWater= 2.962 and mNaCl= 0.148 g; for 120 nm LNC (LNC120), mLab= 1.209, mKol= 0.484,mLip= 0.075,mWater= 3.143 andmNaCl= 0.157 g. The mixtures were heated to 95°C at a rate of 5°C min1under mag- netic stirring followed by cooling at the same rate to 50°C. This cycle was repeated three times and the LNC suspensions were returned to room temperature at the end of the last cycle. The LNC suspensions were only purified using filtration through a 0.45
l
m syringe filter before use.To obtain fluorescent-labeled LNCs, the various dyes were dis- solved in Lab prior to the formulation procedure described above.
For size-exclusion chromatography, dialysis experiments and in vitrocell assays, NR and/or DiO were added at a concentration of 1 mg g1 (weight ratio dye/Lab), to obtain NR-LNC, DiO-LNC and NR/DiO-LNC. For FRET experiments, NR and/or F888 were added at a concentration of 5 mg g1 (weight ratio dye/Lab), to obtain NR-LNC, F888-LNC and NR/F888-LNC. In addition, DiI and/
or DiD were added at concentrations of 3.75 and 2.90 mg g1 (weight ratio dye/Lab), respectively, to obtain DiI-LNC, DiD-LNC and DiI/DiD-LNC. All the dye-loaded LNCs were prepared at the 3 sizes described for the non-labeled LNCs.
2.3. Hydrodynamic diameter, polydispersity index and derived count rate measurements
The hydrodynamic diameter (Z-ave), polydispersity index (PdI) and derived count rate (DCR) of LNC formulations were determined by quasi-elastic light scattering using a ZetasizerÒNano Series DTS 1060 (Malvern Instrument Ltd., Worcestershire, UK). The instru- ment is equipped with a 4 mW Helium–Neon laser, with an output wavelength of 633 nm, and a scatter angle fixed at 173°. All mea- surements were performed at 25°C on LNC suspensions diluted by a factor of 60 (v/v).
The correlation functions were fitted using an exponential fit (Cumulant approach) forZ-ave and PdI determinations for LNC sus- pensions. DCR measurements were performed to confirm the pres- ence of LNC in suspensions. DCR can be calculated from the measured count rates of the scattered light on the detector and the attenuation factor (linked to the attenuator value) according to the following relation:
DCR¼Measured count rate Attenuation factor
For attenuator values of 1–11, attenuation factors are 8.66106, 8.26105, 1.23104, 6.01104, 1.42103, 3.62103, 1.26102, 0.044, 0.111, 0.281 and 1 (as provided by Malvern Instrument Ltd.).
2.4. Analysis of inter-nanocarrier dye transfer by size exclusion chromatography
Measurement of inter-nanocarrier dye transfer using size exclu- sion chromatography (SEC) was performed in two steps. First, NR- LNC30 or DiO-LNC30 suspension was gently mixed with non- labeled LNC120. In the case of NR-LNC, LNCs were diluted and mixed so that the amount of Lab in the non-labeled LNC120 frac- tion was equal (NR-LNC30/LNC1201:1), twice (NR-LNC30/
LNC1201:2) or ten times (NR-LNC30/LNC1201:10) the amount of Lab in the NR-LNC30 fraction. In the case of DiO, LNC formulations were diluted and mixed so that the amount of Lab in the DiO- LNC30 and the non-labeled LNC120 fractions was equal (DiO- LNC30/LNC1201:1). For equal amounts of Lab in both fractions, the mixture was prepared from a volume of suspension of NR- LNC30 or DiO-LNC30 containing 94 mg/mL Lab mixed with the same volume of suspension of non-labeled LNC120 containing 94 mg/mL Lab. For twice the amount of Lab in the non-labeled LNC120 fraction, the mixture was prepared from a volume of sus- pension of NR-LNC30 containing 94 mg/mL Lab mixed with the same volume of suspension of non-labeled LNC120 containing 188 mg/mL Lab. For ten times the amount of Lab in the non- labeled LNC120 fraction, the mixture was prepared from a suspen- sion of NR-LNC30 containing 94 mg/mL Lab mixed in 11:39 volume ratio with a suspension of non-labeled LNC120 containing 268 mg/
mL Lab.
The NR-LNC30/LNC1201:1mixture was incubated at room tem- perature for 5 min, 2 h, 24 h and 72 h. Incubation for 5 min was performed without further mixing, whereas incubation for 2 h and more was performed on a Heidolph Titramax platform shaker (Schwabach, Germany). The NR-LNC30/LNC1201:2, NR-LNC30/
LNC1201:10 and DiO-LNC30/LNC1201:1 mixtures were incubated at room temperature for 5 min without further mixing.
Secondly, 100
l
L mixtures were separated by SEC. The separa- tion was performed on a Sepharose CL4b gel packed in a 15/40 col- umn. The mobile phase was a 0.050 M sodium phosphate buffer at pH 7 with 0.1 M sodium chloride. The mobile phase was pumped with an Ismatec IPS peristaltic pump (Glattbrugg, Switzerland) at a flow rate of 1 mL/min. 1-mL eluent fractions were collected using an Amersham Scientific Frac-920 fraction collector (GE Healthcare, Uppsala, Sweden). Chromatograms were obtained by analyzing the fluorescence of the 1-mL eluent fractions using a Fluoroskan AscentÒ microplate reader (Labsystems SA, Cergy-Pontoise, France). Filter pairs with adequate excitation and emission wave- length couples (kex–kem) were used, i.e. 515–590 nm and 485–518 nm for NR and DiO, respectively. The retention times of LNC120 and LNC30 on the gel were controlled by injection of unmixed dye-labeled LNC30 or LNC120. Due to the manual injec- tion procedure, chromatograms were adjusted ±1 min, so that the retention time (tr) of either the LNC30 or the LNC120 peaks was equivalent between separations. The concentration of NR trans- ferred from the labeled LNC30 fraction to the non-labeled LNC120 fraction in the assay was calculated from the area under the curve of the LNC120 peak in the chromatograms (tr= 20–29 min). The concentration was calculated from a calibra- tion curve established by injections of loaded LNC120 with differ- ent dye concentrations. The assays (mixing and separations) were performed in duplicate (DiO) or triplicate (NR).
2.5. Analysis of inter-nanocarrier dye transfer by FRET
FRET experiments were performed to study the transfer of NR and DiI or DiD between LNCs of similar sizes. For NR transfer, var- ious LNC suspensions (500
l
L) were mixed: F888/NR-LNC + LNC to study the NR transfer from labeled to non-labeled LNCs and F888- LNC + NR-LNC to study the NR transfer between labeled LNCs.Three controls were performed: F888/NR-LNC + water (total FRET), F888-LNC + LNC (F888 emission spectrum alone), and NR-LNC + LNC (NR emission spectrum alone). For DiD (or DiI) transfer, var- ious LNC suspensions (500
l
L) were mixed: DiI/DiD-LNC + LNC to study the dye transfer from labeled to non-labeled LNCs; and DiI-LNC + DiD-LNC to study the dye transfer between labeled LNCs.Three controls were performed: DiI/DiD-LNC + water (total FRET), DiI-LNC + LNC (DiI emission spectrum alone), and DiD-LNC + LNC (DiD emission spectrum alone). All the mixtures were centrifuged for 30 s. Three LNC sizes were tested;i.e.30, 60 and 120 nm. The fluorescence spectra of the mixtures after dilution in pure water were obtained using a Spectramax M2 (Molecular Devices, Sunny- vale, CA) at 25°C, with a 1 mm width quartz cell (Hellma, Paris, France). The FRET donors, F888 and DiI, were excited at 390 and 549 nm, respectively.
2.6. Analysis of inter-nanocarrier dye transfer by dialysis
Dialysis experiments were performed using a static dialysis cell, built in-house, composed of a donor and an acceptor half-cell chamber separated by the dialysis membrane. The volume of each half-cell chamber was 2 mL and the area of the dialysis membrane was 1.5 cm2. The membrane used was a Spectra PorÒ6 standard regenerated cellulose membrane (Spectrum Laboratories, Inc., Ran- cho Dominguez, CA) with a molecular weight cutoff at 50 kDa.
According to the suppliers, the thickness of the dialysis membrane is approximately 60
l
m. The donor chamber was filled with the sample solution, i.e. NR-LNC60, and the acceptor chamber with the dialysate, i.e. an aqueous suspension of non-labeled LNC60.The concentration of non-labeled LNC60 in the dialysate was the same as the concentration of NR-LNC60 in the donor chamber. To investigate the role of membrane thickness on the release rate, 2 or 4 dialysis membranes were mounted stacked on top of each other between the donor and acceptor chambers. The dialysis was performed at room temperature for 24, 48 or 72 h. Mixing of the sample solution and dialysate was achieved by placing the cells on a Heidolph Titramax platform shaker under the same conditions as for SEC experiments. The concentration of NR in the dialysate was analyzed using a Fluoroskan AscentÒmicroplate reader. Filter pairs with adequate excitation and emission wavelength couples (kex–kem) were used,i.e.515–590 nm for NR.
2.7. Cell culture
THP-1 cells (human monocyte/macrophage cell line obtained by ATCC, LGC Promochem, Molsheim, France) were grown in suspen- sion at 37°C, 90% humidity, and 5% CO2in ATCC medium.
2.8. Analysis of nanocarrier-to-cell dye transfer
THP-1 cell suspension was centrifuged, washed with PBS (Lon- za, Verviers, Belgium), centrifuged once again, and then suspended in PBS (450
l
L) at various concentrations (0.5, 1, 5 and 10106 cells). LNC60, NR-LNC60, DiO-LNC60 and DiO/NR-LNC60 suspensions (50l
L) were added and the final LNC concentrations were 1, 5, 10, 20 and 40 mg/mL. Incubation was performed at 25°C under gentle stirring for 30 s, 1, 15 and 30 min. After cen- trifugation, supernatants (LNC suspensions) were collected and flu- orescence was analyzed using a Fluoroskan AscentÒ microplatereader (see above for wavelengths) and compared to initial LNC suspensions to determine the fluorescence dye transfer from LNC to cells. Size measurements (Z-ave) and LNC concentrations (DCR) were determined using a ZetasizerÒNano Series DTS 1060 as described above. THP-1 cell pellets were washed twice with PBS and analyzed in a FACScan flow cytometer with the CellQuert Software (BD Biosciences, Le Pont-de-Claix, France). All experi- ments were performed in triplicate.
2.9. Analysis of cell-to-cell dye transfer
A NR-labeled THP-1 cell suspension in PBS was obtained after incubation of THP-1 cells (10106cells, 450
l
L) with NR-LNC60 (1 mg/mL, 50l
L) suspensions at 25°C for 30 s. NR-labeled THP-1 cell suspension in PBS (2106cells, 250l
L) was incubated with non-labeled THP-1 cell suspensions in PBS (0.66, 2 and 6106- cells, 250l
L) at 25°C under gentle stirring for 30 s. Labeled THP- 1 cell/non-labeled THP-1 cell ratios were 3/1, 1/1 and 1/3, respec- tively. DiO-labeled THP-1 cell suspension in PBS was obtained after incubation of THP-1 cells (10106cells, 495l
L) with DiO (1 mM, 5l
L) solution in ethanol during 4 h at 37°C, in the dark. DiO- labeled THP-1 cell suspension in PBS (2.5106cells, 250l
L) wasincubated with non-labeled THP-1 cell suspensions in PBS (2.5106cells, 250
l
L) at 25°C under gentle stirring for 30 s.THP-1 cell suspensions were centrifuged and the transfer of NR or DiO from labeled to non-labeled THP-1 cells was analyzed using flow cytometry (10,000 cells analyzed). All experiments were per- formed in triplicate.
2.10. Analysis of cell-to-nanocarrier dye transfer
A NR-labeled THP-1 cell suspension in PBS (10106cells, 450
l
L) was incubated with non-labeled LNC60 suspensions (1, 5 and 10 mg/mL, 50l
L) at 25°C under gentle stirring for 30 s. DiO- labeled THP-1 cell suspension in PBS (2.5106cells, 250l
L) wasincubated with non-labeled LNC60 suspensions (20 mg/mL, 250
l
L) at 25°C under gentle stirring for 30 s. The fluorescence of the supernatants, LNC size and concentration (DCR value), and THP-1 cell pellets were analyzed using the microplate reader Fluo- roskan Ascent, ZetasizerÒNano Series DTS 1060 and flow cytome- try, respectively, as described above, to determine the transfer of NR or DiO from labeled THP-1 cells to non-labeled LNCs. All exper- iments were performed in triplicate.3. Results
3.1. LNC formulations
Non-labeled and dye-labeled (NR, DiO, DiI, DiD and F888) LNCs of three different sizes were prepared, i.e. 30, 60 and 120 nm (LNC30, LNC60 and LNC120, respectively). The LNC characteriza- tion data are reported insupporting information (Table S1). Slight variations in LNC diameters occurred when dyes were encapsu- lated, for the 3 sizes of LNCs. For all LNCs, the polydispersity index (PdI) was lower than 0.2, indicating a monomodal and monodis- persed distribution. The size of DiD and DiD/DiI-loaded LNC could not be determined, as the wavelength of the He–Ne laser at 633 nm in the ZetasizerÒNano Series DTS 1060 apparatus overlaps with the absorption spectrum of DiD.
3.2. Analysis of inter-nanocarrier dye transfer by size exclusion chromatography
An inter-nanoparticle dye transfer assay using SEC was used to measure the dye transfer rate from NR- or DiO-labeled LNCs to non-labeled LNCs. Briefly, NR- or DiO-LNC30 was mixed with
non-labeled LNC120 and incubated at room temperature. The LNC mixtures were then separated by SEC, and the inter- nanocarrier dye-transfer was determined by analyzing the dye concentration in the LNC120 fraction by fluorescence spectroscopy.
Chromatograms of the separations of LNC30 and LNC120 are presented inFig. 1.
The NR-LNC30/LNC1201:1mixture (with equal amounts of Lab in NR-LNC30 and LNC120 fractions) was analyzed after incubation for 5 min, 2 h, 24 h and 72 h. The results showed that the concen- tration of NR in the LNC120 fraction was 16.2 ± 0.6
l
g/mL after5 min of incubation, 18.3 ± 0.2
l
g/mL after 2 h of incubation, 17.6 ± 0.7l
g/mL after 24 h of incubation and 18.0 ± 0.3l
g/mLafter 72 h of incubation (Figs. 2A and 1B). This corresponds to a transfer of 34–39% of the total amount of NR in the assay.
In the NR-LNC30/LNC1201:2 and NR-LNC30/LNC1201:10 mix- tures, the concentration of Lab in the LNC120 fraction was twice or ten times the amount of Lab in the NR-LNC30 fraction. The NR-LNC30/LNC1201:2 and NR-LNC30/LNC1201:10 mixtures were incubated 5 min before separation by SEC. The concentration of NR in the LNC120 fraction corresponded to 57% of the total amount of NR in the NR-LNC30/LNC1201:2mixture, and close to 100% of the total amount of NR in the NR-LNC30/LNC1201:10 mixture (Figs. 2B and1C). The DiO-LNC30/LNC1201:1mixture (with equal amounts of Lab in DiO-LNC30 fraction as in the LNC120) was ana- lyzed after incubation for 5 min. Contrary to the NR-LNC mixtures, DiO could not be detected in the LNC120 fraction (Fig. 1D). Thus, the SEC data indicate a rapid release of NR from LNC and a stable DiO labeling.
3.3. Analysis of inter-nanocarrier dye transfer by FRET
A FRET-based assay was used to monitor the dye transfer between loaded and non-loaded LNCs of similar sizes. The assay was performed using F888/NR and DiI/DiD as FRET donor/acceptor pairs[13,15]. LNC60 F888/NR FRET experiments are presented in Fig. 3A. With an excitation wavelength of 390 nm, the emission maxima were 455 nm for F888 (F888-LNC + LNC mixture) and 620 nm for NR (NR-LNC + LNC mixture), respectively. As expected, the NR fluorescence signal in the NR sample was very low, as the maximum of the NR excitation spectrum is at 525 nm. When NR and F888 were loaded in the same LNC (F888-NR-LNC suspension), FRET between the two dyes was evidenced by the decrease of the F888 peak and the increase of the NR peak in comparison with the spectra of F888-LNC + LNC and NR-LNC + LNC, respectively. The fluorescence intensity ratio between donor and acceptor emission peaks was 0.7. For the F888-LNC + NR-LNC and F888-NR-LNC + LNC mixtures, a decrease in the FRET level was observed, as shown by the increase of the F888 peak and the concomitant decrease of the NR peak. As a result, the ratio of the peaks increased to 2.0 and 1.8, respectively. These results indicate that NR was parti- tioned between LNCs in a similar way for the two mixtures, so that NR was transferred from NR-LNC to F888-LNCs in the F888-LNC + NR-LNC mixture and from F888-NR-LNC to non-labeled LNCs in the F888-NR-LNC + LNC mixture.
LNC60 DiI/DiD FRET experiments are presented inFig. 3B. With an excitation wavelength of 549 nm, the emission maxima were 575 nm for DiI (DiI-LNC + LNC mixture) and 680 nm for DiD (DiD-LNC + LNC). As expected, excitation of DiD-LNC at 549 nm resulted only in a small emission peak, as the excitation maximum of DiD was at 644 nm. When DiI and DiD were loaded in the same LNC (DiI–DiD-LNC suspension), and excited at 549 nm, FRET was observed between the two dyes, as evidenced by the decrease in the DiI peak and the increase in the DiD peak. The ratio between donor and acceptor peaks was 0.2. A mixture of DiI–DiD-LNC + LNC produces the same emission spectrum as the DiI–DiD-LNC suspension, with a ratio between the peaks of 0.2, indicating that
none of the two dyes leaked from the DiI/DiD-loaded LNCs.
Furthermore, no significant FRET was observed with a mixture of DiI-LNC + DiD-LNC. The ratio between the two peaks was 2.8. The emission peaks corresponding to DiD fluorescence for this mixture
and for DiD alone (DiD-LNC + LNC mixture) were identical.
Together, our data indicate that neither DiI nor DiD are transferred between LNCs. Similar results were obtained with LNC30 and LNC120 (Supporting information Fig. S1 and S2).
0 100 200 300
20 30 40 50 60 70
Time (min)
Fluorescence (a.u.)
NR LNC120 NR LNC30
A
0 50 100 150
20 30 40 50 60 70
Time (min)
Fluorescence (a.u.)
1:1 1:2 1:10
C
0 30 60 90 120
20 30 40 50 60 70
Time (min)
Fluorescence (a.u.)
0.1h 2h 24h 72h
B
0 50 100 150
20 30 40 50 60 70
Time (min)
Fluorescence (a.u.)
DiO LNC30 DiO LNC120 DiO LNC30+LNC120
D
Fig. 1.Size-exclusion chromatograms of (A) NR-LNC30 and NR-LNC120 injected separately; (B) NR-LNC30 mixed with LNC120 (1:1 Labrafac (Lab) weight ratio) after incubation for 0.1 h, 2 h, 24 h or 72 h at 25°C; (C) NR-LNC30 mixed with LNC120 (1:1, 1:2 or 1:10 Lab weight ratio) after incubation for 0.1 h at 25°C; (D) DiO-LNC30 and DiO- LNC120 injected separately and DiO-LNC30 mixed with LNC120 (1:1 Lab weight ratio) after incubation for 0.1 h at 25°C. The experiments were repeated in triplicate.
3.4. Analysis of inter-nanocarrier dye transfer by dialysis
Inter-nanocarrier transfer of NR between labeled and non- labeled LNCs was also investigated in a dialysis assay, using static dialysis cell composed of a donor and an acceptor half-cell cham- ber separated by a dialysis membrane. The dialysis was performed at room temperature against an aqueous suspension of non- labeled LNC60 in the acceptor chamber. The concentration of NR in the dialysate was analyzed on a microplate fluorometer after 24, 72 or 168 h.
The dialysis assay demonstrated a slow release of NR toward non-labeled LNCs (Fig. 4A). The concentration of NR in the dialy- sate was 0.05 ± 0.02
l
g/mL (n= 5) after 24 h, 0.24 ± 0.06l
g/mL(n= 7) after 72 h, and 0.61 ± 0.05
l
g/mL (n= 6) after 168 h (Fig. 4A). The initial concentration of NR in the LNC formulation in the donor chamber was 230l
g/mL. Thus, only a minor fraction (0.3%) of the NR in the labeled LNC was released during 168 h ofdialysis against a suspension of non-labeled LNC60. We also compared the transfer of NR using a single membrane (60
l
m)with the transfer when using 2 (120
l
m) or 4 (240l
m) mem-branes mounted stacked on top of each other between the donor and acceptor chamber in the dialysis cell. The result showed a rapid decrease in the release of NR with the thickness of the mem- brane (Fig. 4B). While the release of NR was 0.24 ± 0.06
l
g/mL(n= 7) after 72 h dialysis using a single membrane, it was only 0.086 ± 0.002
l
g/mL (n= 5) with 2 membranes and0.044 ± 0.008
l
g/mL (n= 5) with 4 membranes. The decreased transfer clearly illustrates the effect of the membrane thickness on the transfer of NR in the dialysis assay.3.5. In vitro nanocarrier-to-cell and cell-to-nanocarrier sequential transfer of dyes
Anin vitrocell culture assay was used to investigate whether a similar instantaneous transfer of lipophilic dye can occur between LNC and cells and/or between cells, as had been demonstrated between LNC populations.
0 25 50 75 100
1:1 1:2 1:10
Labrafac Ratio (LNC30:LNC120)
% NR in LNC120
B
0 10 20 30 40
0.1 2 24 72
Incubation time (h)
% NR in LNC120
A
Fig. 2.(A) Percentage of NR transferred from NR-LNC30 to LNC120 (mixture with 1:1 Labrafac (Lab) weight ratio) after incubation for 0.1 h, 2 h, 24 h or 72 h at 25°C;
and (B) percentage of NR transferred from NR-LNC30 to LNC120 (mixtures with 1:1, 1:2 or 1:10 Lab weight ratio) after incubation for 0.1 h at 25°C; measured by the size-exclusion chromatography assay (n= 3, mean ± standard deviation).
0 5000 10000
400 500 600 700
Wavelength (nm)
Fluorescence (a.u.)
F888 NR LNC F888 LNC+LNC F888 LNC+NR LNC F888 NR LNC+LNC NR LNC+LNC
A
0 2500 5000 7500 10000
600 700 800
Wavelength (nm)
Fluorescence (a.u.)
DiI DiD LNC DiI LNC+LNC DiI LNC+DiD LNC DiI DiD LNC+LNC DiD LNC+LNC
B
Fig. 3.Fluorescence emission spectra for (A) F888/NR-LNC and mixtures of F888- LNC and LNC, F888-LNC and NR-LNC, F888/NR-LNC and LNC, NR-LNC and LNC; with an excitation wavelength of 390 nm; and (B) DiI/DiD-LNC and mixtures of DiI-LNC and LNC, DiI-LNC and DiD-LNC, DiI/DiD-LNC and LNC, DiD-LNC and LNC; with an excitation wavelength of 549 nm. LNC size was 60 nm and the Labrafac weight ratio was 1:1 for all mixtures. The experiments were repeated in triplicate.
To investigate dye transfer from LNC to cells, human monocyte cells (THP-1, from 0.1 to 10106cells) were incubated at room temperature for 30 s with either NR-LNC60 or DiO-LNC60 (10 mg/mL LNC final concentration). DiO was chosen because its emission spectrum only has a minor overlap with that of NR [12]. After incubation, cells and LNCs were separated by centrifuga- tion and the dye transfer was analyzed by flow cytometry.
As in the LNC assays reported above, the cell experiments showed that NR transferred rapidly from NR-LNCs to THP-1 cells.
Independently of the number of cells, similar fluorescence inten- sity values were obtained by flow cytometry (Fig. 5). Concurrently, NR concentrations in LNCs decreased by 4 ± 2% and 31 ± 3% when the cell numbers were 0.1 and 10106, respectively (Supporting information Fig. S3). The results indicate that NR is transferred to THP-1 cells and the same fluorescence intensity in the THP-1 cells for all cell numbers can be explained by a dye concentration satu- ration effect in the cell (up to 10106cells). The transfer occurred without uptake of LNCs by the cells as the DCR values of the LNC
suspensions obtained by light scattering, which are proportional to the LNC concentrations[12], were similar before and after incu- bation. As expected, no transfer of DiO was observed between DiO- LNC and THP-1 cells (Fig. 5). The DiO fluorescence intensity values remained almost similar to the control (THP-1 without LNC), inde- pendently of the number of cells. In addition, DiO concentrations in LNCs were unchanged before and after incubation as determined by fluorescence spectroscopy (Supporting information Fig. S3).
NR and DiO co-labeling (Fig. 5 and supporting information Fig. S3), increased incubation times (from 30 s to 30 min) (Support- ing information Fig. S4) or increased concentrations of LNCs (from 1 to 40 mg/mL) (Supporting information Fig. S5) generated the same results as described above.
In a second step, NR and DiO transfer between cells was inves- tigated. NR-labeled THP-1 suspension previously obtained (2106cells) was incubated with non-labeled THP-1 cells (0, 0.66, 2 or 6106cells) at room temperature for 30 s. Similarly, DiO-labeled THP-1 suspension (2.5106cells) (obtained after cell incubation with DiO ethanol solution) was incubated with the same amount of non-labeled THP-1 cells at room temperature for 30 s. NR and DiO transfer from labeled to non-labeled THP-1 cells was then analyzed by FACS. For NR, a single THP-1 population was observed and its relative fluorescence intensity decreased, while the number of non-labeled THP-1 cells increased (Fig. 6A).
Specifically, the relative fluorescence intensity values (presumably equivalent to the average number of NR molecules per cell), i.e.
about 75%, 50% and 25%, reflected the labeled/total THP-1 ratio, i.e.3/4, 1/2 and 1/4, respectively. Two distinct populations, fluores- cent and non-fluorescent, were recovered after incubation of a mixture of DiO-labeled THP-1 cells and non-labeled THP-1 cells with fluorescence intensities close to those for the initial DiO- labeled and non-labeled THP-1 cells, respectively (Fig. 6B), indicat- ing that there were no transfer of DiO between THP-1 cells.
Finally, NR and DiO transfer from cells to LNCs was analyzed.
NR-labeled THP-1 suspensions (10106cells) were incubated with non-labeled LNCs (1, 5 or 10 mg/mL LNC final concentration) at room temperature for 30 s and NR transfer was measured by flow-cytometry. For DiO transfer, DiO-labeled THP-1 suspension obtained previously (2.5106cells) was incubated with non- labeled LNCs (10 mg/mL LNC final concentration) at room temper- ature for 30 s. With NR, flow cytometry analysis showed that the relative fluorescence intensity value of the THP-1 cells decreased by 85–95% when mixed with non-labeled LNCs (1–10 mg/mL, Fig. 6C), and that no fluorescence intensity change was observed using DiO (Fig. 6D). In addition, fluorescence spectroscopy con- firmed the presence of NR on one hand and the absence of DiO on the other hand in the LNC population after incubation (Support- ing information Fig. S6).
Together the cellular studies presented here demonstrate a rapid sequential transfer of NR from LNC to cells, between cells and from cells to LNC, while DiO remains bound to LNCs (when DIO-LNCs are used) or THP-1 cells (when DIO-cell labeling is per- formed) in the same timescale.
4. Discussion
When using a fluorescent dye as a biological positioning system for a nanocarrier, it is vital that the label is stable and does not dis- sociate from the carrier inin vitroorin vivoassays. The risk for a de-localization of the dye with respect to the nanocarrier cannot be neglected, especially if the labeling is achieved through non- covalent interactions. To validate the stability of a dye label, poten- tial risks of leakage should be investigated under appropriate con- ditions. In the case of leakage of lipophilic compounds from lipophilic carriers, acceptors would be various lipid compartments 0.0
0.2 0.4 0.6
24 72 168
Dialysis time (h)
Conc. in dialysate (µg/mL)
A
0.0 0.1 0.2 0.3
60 120 240
Membrane thickness (µm)
Conc. in dialysate (µg/mL)
B
Fig. 4.NR concentration in dialysate as a function of (A) time and (B) membrane thickness, after inter-nanocarrier NR transfer between labeled (in donor cell chamber) and non-labeled LNC60 (in acceptor cell chamber,n= 5–7, mean ± stan- dard deviation).
encountered by the carriers during their diffusion through the biological fluids. Therefore, a conventional release assay, e.g. a membrane based dialysis experiment, could give misleading results regarding labeling stability. Indeed, in a previous study, we demonstrated an almost instantaneous partition-dependent release of Cou-6 and NR dyes from LNC using a new nanoparti cle-to-oil-phase-extraction assay. Simultaneously, the same assay indicated a stable encapsulation of three indocarbocyanine (DiD, DiI and DiO)[12].
Assuming the mixing with the oil has no destabilizing effect on the label, the results highlight a rapid leakage of some lipophilic compounds from LNC, when in direct contact with a lipid acceptor compartment. However, in a biological environment nanocarriers are not expected to encounter any free macroscopic oil acceptor phase, but rather nanostructured lipid compartments. Therefore, to mimic lipophilic acceptors in vivo, we wanted to investigate the labeling stability of NR and lipophilic carbocyanine dyes using release assays, in which the acceptor phases consist of stable col- loidal lipid compartments. For this, the oil-phase was replaced with an aqueous suspension of non-loaded LNC.
Initially dye transfer was studied using a new method based on a separation by SEC. To be able to separate labeled from non- labeled LNCs after incubation, we had to use labeled and non- labeled LNCs of different sizes. Using this technique, an instanta- neous partitioning of NR from NR-LNC30 to LNC120 was observed.
Increasing the incubation time from 5 min up to 72 h did not change the amount of NR transferred to the LNC120 fraction. Inter- estingly, when increasing the amount of LNC120 with respect to the amount of NR-LNC30, the NR was found to further accumulate
1 10 100
0.5 1 5 10 Controls
Number of THP-1 Cells (x106)
Fluorescence Intensity in THP-1 Cells (Geo-Mean, a.u.)
Dye Labeled LNC (Detection Channel):
DiO-LNC (DiO) NR-LNC (NR) NR/DiO-LNC (DiO) NR/DiO-LNC (NR)
Fig. 5.Fluorescence intensity in THP-1 cells after incubation for 30 s at 25°C of non-labeled THP-1 cells (0.5, 1, 5 and 10106cells) with DiO-LNC, NR-LNC, or NR/
DiO-LNC suspension (1 mg/mL, 60 nm), using flow cytometry with NR (FL2 channel) or DiO (FL1 channel) detection. (n= 3, mean ± standard deviation).
Fig. 6.Flow cytometry histograms of cell counts (10,000 cells analyzed) versus fluorescence intensities in THP-1 cells for (A) mixtures of NR-THP-1 and non- labeled THP-1 cells (1:3, 1:1 or 3:1 cell ratio), (B) mixture of DiO-THP-1 and non- labeled THP-1 cells (1:1 cell ratio), (C) mixtures of NR-THP-1 and non-labeled LNC (10106cells; 1, 5 or 10 mg/mL LNC concentration, 60 nm), and (D) mixtures of DiO-THP-1 and non-labeled LNC (2.5106cells; 10 mg/mL LNC concentration, 60 nm). All mixtures were stirred for 30 s at 25°C. Non-labeled (A, B, C, and D) and labeled THP-1 cells (with NR for A and C; and with DiO for B and D) were used as control. The experiments were repeated in triplicate.
in the LNC120 fraction, clearly demonstrating an equilibrium based partitioning effect on the redistribution of dye between the two LNC populations. In contrast to NR, there was no detectable trans- fer of DiO from DiO-LNC30 to LNC120 in the SEC assay. This assay is, to our knowledge, the first direct quantification of the transfer of compounds between two LNC populations in suspension.
Subsequently, a FRET assay was used to study transfer between LNCs of similar sizes to verify that transfer was size independent.
FRET occurs on a scale of a few nanometers and the loss or appear- ance of the FRET signal is an excellent indicator of the transport of fluorescent compounds in and out of nano-compartments. This technique has been used to examine the stability of dye-labeled nanocarriers, both inin vitroandin vivoassays[7,13,16–21]. Using the F888/NR pair that was previously applied to monitor the leak- age of NR from lipid nanocarriers in serum[13], we observed an instantaneous exchange of NR between LNC populations, i.e. a transfer of NR from NR-LNC to F888-LNC, and a transfer of NR from F888/NR-LNCs to non-labeled LNCs. These results confirm the rapid release of NR seen in both the SEC assay and the extraction assay with an oil-acceptor phase[12]. They are also in line with FRET and fluorescence correlation spectroscopy (FCS) experiments showing fast NR release from lipid nanocarriers in serum[13]. In contrast, no transfer of the lipophilic carbocyanine dyes (DiI/DiD) between LNCs could be detected based on the variation in the FRET signal in various LNC mixtures that were analyzed. Similar conclu- sions were obtained with LNCs of three different sizes. These results corroborate the stability of DiO-labeled LNCs shown by the SEC assay, and of DiO-, DiI- and DiD-LNCs in the nanoparti cle-to-oil-phase-extraction assay [12]. Our data are in line with the high stability of DiD/FP730-C18 FRET pair-labeled LNCs in plasma[7], and the stability of DiD/DiO FRET pair-loaded LNC over 24 h at 37°C in PBS or BSA, with or without addition of non-labeled LNCs[16].
Nanocarrier release rates are often evaluated by dialysis[22,23].
However, this method has some important drawbacks, especially when measuring the release rate of lipophilic compounds from lipophilic nanocarriers. The membrane creates an artificial bound- ary between the carrier and the acceptor compartment which is not present in anin vivosituation. Therefore, dialysis was not our first choice in the study of the dye transfer between LNCs. How- ever, for a comparison of methods, a dialysis experiment measur- ing transfer of NR between LNCs was performed. In contrast to the results obtained using the nanoparticle-to-oil-phase-extraction assay, the SEC assay, and the FRET assay, the dialysis method indi- cated a very slow release of NR from labeled to non-labeled LNCs.
Only a minor fraction (0.3%) of NR in LNC was released during 168 h of dialysis against non-labeled LNC suspension. When using dialysis, the appearance of dye on the acceptor side of the dialysis membrane is not necessarily controlled by the release rate from the nanocarriers on the donor side. It can also be controlled by the transport in the dispersion medium and/or the dialysis mem- brane. For example, Zambito et al. recently demonstrated a membrane-controlled transport of diclofenac and ofloxacin from chitosan nanoparticles in a dialysis assay[24]. The thickness of the dialysis membrane used in our study is approximately 60
l
m. This is a considerable distance as compared to the scale of the LNC. Indeed, the membrane thickness was found to play a major role on the amount of NR transferred to the acceptor phase, as illustrated by the strong decrease in the transfer of NR with increasing membrane thickness. In addition to the distance, lipo- philic compounds may adhere to the membrane, which should fur- ther slow down the dye flux[25].Taken together, the SEC and the FRET assays demonstrate an instantaneous transfer of NR between two LNC populations, whereas the dialkylcarbocyanine dyes (DiD, DiI and DiO) remain in the labeled LNC without transfer to non-labeled LNCs. The
results corroborate the release shown using extraction with an oil phase, validating this method as a fast and simple technique to investigate the release of lipophilic compounds from LNCs and the labeling stability of lipid nanocarriers. The SEC assay, which is more time consuming and complex to apply, still remains an good alternative for studies of transfer between colloidal nano- compartments in the absence of an artificial dialysis membrane, in particularly, for dyes for which FRET cannot be applied. The dial- ysis assay displays the limitations of using this technique to mea- sure the release of lipophilic compounds from a LNC formulation.
The measured release of NR using the dialysis assay appears to be membrane-transport related. The rapid equilibrium-based transfer of NR from NR-LNC30 to LNC120 in the SEC assay and between LNCs of similar sizes in the FRET-based membrane-free inter-nanoparticle release assays, compared to the dialysis experi- ments, is probably a result of a combination of parameters includ- ing the shorter distance between donor and acceptor compartments when omitting the membrane (diffusion length in the aqueous dispersion medium), and the absence of undefined associations of dye molecules with the membrane.
The inter-nanoparticle release assays suggest that NR is released in a rapid equilibrium-based manner if there is a lipophilic acceptor in close proximity to the LNCs. Thus, a rapid release is also expected to occur toward biological lipid reservoirs, e.g. the plas- mic membrane of cells or intracellular lipid compartments. Indeed, NR (encapsulated in the LNC core) was rapidly transferred from labeled LNCs to human monocytes (THP-1 cell line), used as model for circulating cells. A similar release has previously been reported for NR in both HEI-OC1 (auditory cells derived from the organ of Corti), and HeLa (derived from cervical cancer) cell lines[12,13], and for Coumarin-6, in an A30 lung cell line[26]. Thus, the cell nat- ure does not seem to play a key role for the transfer; only the pres- ence of lipophilic compartments is necessary[12,27]. In addition to the rapid release of NR from LNCs to THP-1 cells, this study also demonstrates a rapid sequential transfer of NR from both labeled to non-labeled THP-1 cells, and from labeled THP-1 cells to non- labeled LNCs. The almost instantaneous partitioning of NR between loaded and non-loaded cells could explain the rapid diffusion of the dye both to the endothelial cell layers of the vessels and to cell layers in the underlying tissue, after intravenous administration in lipid nanocarriers[13]. Lipophilic carbocyanine dyes are generally considered stable cell membrane labels used, for example, in cell–
cell fusion [28], cellular adhesion [29], cell migration [30], and intracellular compartment characterization [31] assays. Releases have been reported only after longer incubation times. For exam- ple, Lassailly et al. observed DiI and DiR transfer (10–30%) between labeled human promyelocytic leukemia cells and non-labeled fibroblastic cells, after 24 h incubation at 37°C[32]. In this study, no transfer of DiO could be detected within the time limits (30 s) of our cell assays, either from labeled LNCs to non-labeled THP-1 cells, from labeled to non-labeled THP-1 cells, or from labeled THP-1 cells to non-labeled LNCs. The results corroborate the stabil- ity of this label shown in the SEC and FRET assays, and are sup- ported by the previously reported absence of DiO transfer from lipid nanocarriers to HEI-OC1 cells[12].
5. Conclusion
New methods were investigated to study the loading stability of fluorescence dyes in lipid nanocapsules, in the presence of lipid nanostructured acceptor compartments. The results corresponded to our previous observations using a nanoparticle-to-oil phase- extraction assay. The SEC and FRET assays used in this study demonstrate an instantaneous transfer of NR between two LNC populations, whereas the lipophilic carbocyanine dyes (DiD, DiI
and DiO) remain in the labeled LNCs without transfer to non- labeled LNCs. Similar release behavior was found in cell studies, revealing a rapid sequential transfer of NR from LNCs to cells, between cells, and from cells to LNCs, whereas no transfer was observed with DiO. Taken together, our results suggest that NR is released in a rapid equilibrium-based manner if there are lipophilic acceptor compartments in close proximity. This is an important phenomenon as the experiments reported here can be considered close toin vivobehavior. The result of the dialysis assays demon- strates the drawback of this method to study the release of lipophi- lic compounds from lipid nanocarriers, and highlight the importance of relevant assays to determine the stability of dye labeling in LNC formulations. Membrane-free release assays, such as the nanoparticle-to-oil phase extraction assay presented in our previous study or the inter-nanocarrier transfer assays by SEC and FRET used here appear as appropriate tools to analyze the release of fluorescent dyes from lipid nanocarriers, and to confirm the fluorescent dye labeling stability of nanocarriers.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgments
The authors wish to acknowledge the help of the European Community (LYMPHOTARG and NICHE programs: EuroNanoMed ERA-NET 09 and EuroNanoMed 2, respectively).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the online version, athttp://dx.doi.org/10.1016/j.ejpb.2015.10.011.
References
[1]G. Bastiat, P. Oliger, G. Karlsson, K. Edwards, M. Lafleur, Development of non- phospholipid liposomes containing a high cholesterol concentration, Langmuir 23 (2007) 7695–7699.
[2]B. Heurtault, P. Saulnier, B. Pech, J.E. Proust, J.-P. Benoit, A novel phase inversion-based process for the preparation of lipid nanocarriers, Pharm. Res.
19 (2002) 875–880.
[3]T.M. Allen, P.R. Cullis, Drug delivery systems: entering the mainstream, Science 303 (2004) 1818–1822.
[4]D. Pozzi, C. Marchini, F. Cardarelli, F. Salomone, S. Coppola, M. Montani, M.E.
Zabaleta, M.A. Digman, E. Gratton, V. Colapicchioni, G. Caracciolo, Mechanistic evaluation of the transfection barriers involved in lipid-mediated gene delivery: interplay between nanostructure and composition, Biochim.
Biophys. Acta 2014 (1838) 957–967.
[5]C. Zhou, Y. Zhang, B. Yu, M.A. Phelps, L.J. Lee, R.J. Lee, Comparative cellular pharmacokinetics and pharmacodynamics of siRNA delivery by SPANosomes and by cationic liposomes, Nanomedicine 9 (2013) 504–513.
[6]F. Huang, E. Watson, C. Dempsey, J. Suh, Real-time particle tracking for studying intracellular trafficking of pharmaceutical nanocarriers, Methods Mol. Biol. 991 (2013) 211–223.
[7]A.-L. Laine, J. Gravier, M. Henry, L. Sancey, J. Béjaud, E. Pancani, M. Wiber, I.
Texier, J.-L. Coll, J.-P. Benoit, C. Passirani, Conventional versus stealth lipid nanoparticles: formulation and in vivo fate prediction through FRET monitoring, J. Control. Release 188 (2014) 1–8.
[8]J.R. McDaniel, S.R. MacEwan, X. Li, D.C. Radford, C.D. Landon, M. Dewhirst, A.
Chilkoti, Rational design of ‘‘heat seeking” drug loaded polypeptide nanoparticles that thermally target solid tumors, Nano Lett. 14 (2014) 2890–
2895.
[9]T. Thambi, D.G. You, H.S. Han, V.G. Deepagan, S.M. Jeon, Y.D. Suh, K.Y. Choi, K.
Kim, I.C. Kwon, G.-R. Yi, J.Y. Lee, D.S. Lee, J.H. Park, Bioreducible carboxymethyl dextran nanoparticles for tumor-targeted drug delivery, Adv. Healthc. Mater. 3 (2014) 1829–1838.
[10]S. Hirsjärvi, S. Dufort, G. Bastiat, P. Saulnier, C. Passirani, J.-L. Coll, J.-P. Benoit, Surface modification of lipid nanocapsules with polysaccharides: from physicochemical characteristics to in vivo aspects, Acta Biomater. 9 (2013) 6686–6693.
[11]J. Balzeau, M. Pinier, R. Berges, P. Saulnier, J.-P. Benoit, J. Eyer, The effect of functionalizing lipid nanocapsules with NFL-TBS.40-63 peptide on their uptake by glioblastoma cells, Biomaterials 34 (2013) 3381–3389.
[12]G. Bastiat, C.O. Pritz, C. Roider, F. Fouchet, E. Lignières, A. Jesacher, R. Glueckert, M. Ritsch-Marte, A. Schrott-Fischer, P. Saulnier, J.-P. Benoit, A new tool to ensure the fluorescent dye labeling stability of nanocarriers: a real challenge for fluorescence imaging, J. Control. Release 170 (2013) 334–342.
[13]A.S. Klymchenko, E. Roger, N. Anton, H. Anton, I. Shulov, J. Vermot, Y. Mely, T.F.
Vandamme, Highly lipophilic fluorescent dyes in nano-emulsions: towards bright non-leaking nano-droplets, RSC Adv. 2 (2012) 11876–11886.
[14] B. Heurtault, P. Saulnier, B. Pech, J.E. Proust, J. Richard, J.-P. Benoit, World Pat., 2001 064 328 A1, 2001.
[15]T. Yoon, B. Okumus, F. Zhang, Y. Shin, T. Ha, Multiple intermediates in SNARE- induced membrane fusion, Proc. Natl. Acad. Sci. 103 (2006) 19731–19736.
[16]J. Gravier, L. Sancey, S. Hirsjärvi, E. Rustique, C. Passirani, J. Benoit, J.L. Coll, I.
Texier, FRET imaging approaches for in vitro and in vivo characterization of synthetic lipid nanoparticles, Mol. Pharm. 11 (2014) 3133–3144.
[17]A.-C. Groo, P. Saulnier, J.-C. Gimel, J. Gravier, C. Ailhas, J.-P. Benoit, F. Lagarce, Fate of paclitaxel lipid nanocapsules in intestinal mucus in view of their oral delivery, Int. J. Nanomed. 8 (2013) 4291–4302.
[18]J. Lu, S.C. Owen, M.S. Shoichet, Stability of self-assembled polymeric micelles in serum, Macromolecules 44 (2011) 6002–6008.
[19]A. Wagh, S.Y. Qian, B. Law, Development of biocompatible polymeric nanoparticles for in vivo NIR and FRET imaging, Bioconjug. Chem. 23 (2012) 981–992.
[20]P. Zou, H. Chen, H.J. Paholak, D. Sun, Noninvasive fluorescence resonance energy transfer imaging of in vivo premature drug release from polymeric nanoparticles, Mol. Pharm. 10 (2013) 4185–4194.
[21]H. Chen, S. Kim, L. Li, S. Wang, K. Park, J. Cheng, Release of hydrophobic molecules from polymer micelles into cell membranes revealed by Förster resonance energy transfer imaging, Proc. Natl. Acad. Sci. USA 105 (2008) 6596–
6601.
[22]V. Bali, M. Ali, J. Ali, Nanocarrier for the enhanced bioavailability of a cardiovascular agent: in vitro, pharmacodynamic, pharmacokinetic and stability assessment, Int. J. Pharm. 403 (2011) 46–56.
[23]M. Rawat, A. Jain, A. Mishra, M. Muthu, S. Singh, Development of repaglinide loaded solid lipid nanocarrier: selection of fabrication, Method Curr. Drug Deliv. 7 (2012) 44–50.
[24]Y. Zambito, E. Pedreschi, G. Di Colo, Is dialysis a reliable method for studying drug release from nanoparticulate systems? a case study, Int. J. Pharm. 434 (2012) 28–34.
[25]S. Modi, B.D. Anderson, Determination of drug release kinetics from nanoparticles: overcoming pitfalls of the dynamic dialysis method, Mol.
Pharm. 10 (2013) 3076–3089.
[26]I. Rivolta, A. Panariti, B. Lettiero, S. Sesana, P. Gasco, M.R. Gasco, M. Masserini, G. Miserocchi, Cellular uptake of coumarin-6 as a model drug loaded in solid lipid nanoparticle, J. Physiol. Pharmacol. 62 (2011) 45–53.
[27]B.-C. Tian, W.-J. Zhang, H.-M. Xu, M.-X. Hao, Y.-B. Liu, X.-G. Yang, W.-S. Pan, X.- H. Liu, Further investigation of nanostructured lipid carriers as an ocular delivery system: in vivo transcorneal mechanism and in vitro release study, Colloids Surf. B. Biointerfaces 102 (2013) 251–256.
[28]L. Spotl, A. Sarti, M.P. Dierich, J. Most, L. Franzens, U. Innsbruck, Cell membrane labeling with fluorescent dyes for the demonstration of cytokine-induced fusion between monocytes and tumor cells, Cytometry 21 (1995) 160–169.
[29]J.D. Malhotra, P. Tsiotra, D. Karagogeos, M. Hortsch, Cis-activation of L1- mediated ankyrin recruitment by TAG-1 homophilic cell adhesion, J. Biol.
Chem. 273 (1998) 33354–33359.
[30]S. Kuriyama, M. Yamazaki, A. Mitoro, T. Tsujimoto, M. Kikukawa, H. Okuda, H.
Tsujinoue, T. Nakatani, H. Yoshiji, Y. Toyokawa, Analysis of intrahepatic invasion of hepatocellular carcinoma using fluorescent dye-labeled cells in mice, Anticancer Res. 18 (1998) 4181–4188.
[31]S. Mukherjee, T.T. Soe, F.R. Maxfield, Endocytic sorting of lipid analogues differing solely in the chemistry of their hydrophobic tails, J. Cell Biol. 144 (1999) 1271–1284.
[32]F. Lassailly, E. Griessinger, D. Bonnet, ‘‘Microenvironmental contaminations ” induced by fluorescent lipophilic dyes used for noninvasive in vitro and in vivo cell tracking, Blood 115 (2010) 5347–5354.
