Synthesis and Intracellular Uptake of Rhodamine-Nucleolipid Conjugates into a Nanoemulsion Vehicle.

Neurodegenerative diseases represent some of the greatest challenges for both basic science and clinical medicine. Due to their prevalence and the lack of known biochemical-based treatments, these complex pathologies result in an increasing societal cost. Increasing genetic and neuropathological evidence indicates that lysosomal impairment may be a common factor linking these diseases, demanding the development of therapeutic strategies aimed at restoring the lysosomal function. Here, we propose the design and synthesis of a nucleolipid conjugate as a nonviral chemical nanovector to specifically target neuronal cells and intracellular organelles. Herein, thymidine, appropriately substituted to increase its lipophilicity, was used as a model nucleoside and a fluorophore moiety, covalently bound to the nucleoside, allowed the monitoring of nucleolipid internalization in vitro. To improve nucleolipid protection and cellular uptake, these conjugates were formulated in nanoemulsions. In vitro biological assays demonstrated cell uptake- and internalization-associated colocalization with lysosomal markers. Overall, this nucleolipid-nanoemulsion-based formulation represents a promising drug-delivery tool to target the central nervous system, able to deliver drugs to restore the impaired lysosomal function.

Introduction

The selective filter of the blood–brain barrier (BBB) is one of the major obstacles to the achievement of efficient drug delivery[1,2] and therapeutic effect to the brain,[3,4] which hampers further treatments for brain-related diseases. Nowadays, innovation in the field of nanosystems allows a better crossing of biological membranes, thus paving the way for new therapeutic approaches.[1,3,5] Nucleolipids (NLs) are bifunctional hybrid molecules in which a lipid moiety and a nucleic acid moiety (nucleoside, nucleotide, nucleobase, or oligonucleotide) are covalently linked.[6] Exhibiting a wide structural variety, these molecules can be natural such as algelasine F and tunicamycin[7,8] or of synthetic origin like the DOTAU.[9] NLs, either natural or synthetic, are interesting not only for their potential biological activities, including antimicrobial, antifungal, antiviral, and antitumor properties,[10−12] but also for their remarkable ability to self-assemble. Indeed, amphiphilic molecules, such as the DOTAU, can form supramolecular objects like micelles or liposomes, which can be used to deliver DNA, antisense oligonucleotides, or siRNA directly into the cells.[9] Because of their similarity to the lipid bilayer of cell membranes, these molecules are expected to cross the plasma membrane without the need for membrane transporters. Considering the benefit of nanoemulsions (NEs) as vehicles for therapeutic agents to target the brain,[13−15] their combination with NLs, as a potential absorption promoter, seemed to be a tantalizing approach to improve the passage of the BBB.

Oil-in-water (O/W) NEs, made of submicrometric oily droplets stabilized by a corona of amphiphilic surfactants, present enhanced stability compared to other nanosystems and high loading capacity of hydrophobic drugs or imaging probes.[13,16−18] Very recently it has also been reported that some NLs, such as an NL radiotracer, were successfully able to permeate the BBB,[19] suggesting that NLs could be promising absorption promoters. In this work, original NEs associated with nucleolipids were developed to enhance the membrane-crossing properties for therapeutic purposes and particularly in the context of neurodegenerative diseases. Indeed, previous studies have shown that both nanoparticles and NEs loaded with an acidic cargo made of poly(dl-lactide-co-glycolide) (PLGA) were able to induce neuroprotective effects, and following a systemic injection, NEs were able to cross the BBB.[20,21]

Herein, we report the design and synthesis of original NLs and their formulation into NEs to cross the plasma membrane. We show the contributing role of NEs in the uptake and internalization of NLs into neuronal cells in vitro, and colocalization of NLs with lysosome was observed. These results suggest that lysosomes can be efficiently targeted by the use of these nanotechnology-based systems for drug delivery to treat brain diseases, in particular, neurodegenerative diseases, where lysosomal impairment occurs, among others.[22,23]

Results and Discussion

Design and Synthesis of Nucleolipidic Platforms

Six compounds, of which three were original NLs (Figure A), were then designed, synthesized, and biologically evaluated. Commercially available thymidine was used as a nucleosidic platform, and key positions were conveniently functionalized step-by-step (Figure B). A benzyl group was introduced at the N-3 position of thymidine to enhance the lipophilicity of the compound and allow solubilization in the oily phase to provide an NE formulation later on. The first NL, compound A, was substituted with a palmitic chain at the 5′ position and a rhodamine B fluoroprobe at position 3′. In NL compound B, the lipid chain and rhodamine B positions were switched. Thereby, these two lipophilic NLs were formulated into NEs, to carry out in vitro biological assays and evaluate the role of NLs, as well as the positions of the different substituents and their influence on the uptake into cells. The third NL (compound C) has been similarly designed by removing the benzyl group to make it water-soluble and investigate the role of NE itself. The last three compounds were used as a control. Compound D was rhodamine B, one formulated in an NE and solubilized in water, to determine the role of the NEs, and compound E was a rhodamine–lipid conjugate (Figure B). Compound E was designed to compare its internalization into neuronal cells with that of compound A and investigate the benefits of using an NL rather than a simple lipid chain.

Figure 1

(A) General structure of nucleolipids. (B) Chemical structure of the different compounds synthetized for this study. (C) Synthetic sequence leading to NL-A.

Nucleolipids A and B were prepared in seven- and five-step sequences, respectively. The first step was selective protection of the nucleoside to functionalize, thereafter, either position 3′ or 5′. A benzyl group was added on the thymidine N-3 position by a simple alkylation following a procedure, using microwave activation, previously developed in the laboratory.[24] An esterification reaction between palmitic acid and thymidine was performed at the free hydroxyl positions (position 5′ for compound A and position 3′ for compound B). The protective group was then cleaved before a subsequent esterification step with the rhodamine B moiety (Figure C). Compound C was obtained through a six-step sequence, while compound E required a unique esterification step.

Nanoemulsion Preparation and Characterization

The lipidic nature and plasticity of O/W NEs make them suitable intravenous (iv) systems for drug delivery. NEs exhibit good encapsulation ability for hydrophobic drugs and efficiently improve their stability, IV tolerance, and bioavailability.[25] Five nanoemulsions were formulated with medium chain triglyceride oil Miglyol 812N as the oily phase and two surfactants polysorbate 80 (Tween 80) and egg phospholipids (Lecithin E80) (Table ). NE-A, NE-B, NE-D, and NE-E were loaded, respectively, with nucleolipids A and B, rhodamine (D), and the rhodamine–lipid conjugate (E). An unloaded NE named NE-O was used as a control.

Table 1

Composition of Loaded NEs with Compound A, B, D, or E and the Unloaded NE (NE-O)a

NEs’ composition	NE-O	NE-A	NE-B	NE-D	NE-E
miglyol 812	20%	20%	20%	20%	20%
lecithin E80	1.2%	1.2%	1.2%	1.2%	1.2%
tween 80	2.5%	2.5%	2.5%	2.5%	2.5%
compound’s mass (mg)	0	10	10	5	7,5
quantity of compounds (mmol)	0	9.7 × 10–3	9.7 × 10–3	1.0 × 10–2	1.1 × 10–2
water q.s.	100	100	100	100	100

The amounts of miglyol, lecithin, and tween are expressed in % (w/w).

Physicochemical analysis of the formulations by dynamic light scattering (DLS) showed that all five formulations had a submicrometric size range (Table ). All formulations displayed a monodispersed distribution with a dispersity below 0.2. Zetameter measurements showed a negative ζ-potential for the control formulation (NE-O). Indeed, this negative charge is linked with the lecithin E80 surfactant (a mixture of phospholipids including negatively charged phospholipids). Interestingly, NE-A and NE-B were found to exhibit a positive ζ-potential. The same phenomenon was observed for NE-E, the rhodamine–lipid-based NE, which is also naturally positively charged. Contrariwise, a negative ζ-potential was found with the rhodamine B-based formulation (NE-D) because non-lipid-conjugated rhodamine was solubilized in the oil phase at a final concentration of 25 mg/mL, suggesting that in this case, rhodamine B was placed inside the oily droplet and not at its surface (Figure ).

Table 2

Physicochemical Characteristics of Unloaded and Loaded NEs

NEs’ characteristics	NE-O	NE-A	NE-B	NE-D	NE-E
diameter (nm)	180.9	178.6	146.2	186.7	181.3
polydispersity index	0.167	0.177	0.140	0.181	0.183
ζ-potential (mV)	–32.3	+56.1	+35.6	–37.1	+48.8

Figure 2

Physicochemical characteristics of unloaded and loaded NEs. (A) Schematic structure of the different NEs. (B) Evolution of the diameter, polydispersity index (PdI), and ζ-potential of NE-A, NE-D, and NE-E over time.

Finally, the colloidal stability of three of these NEs was monitored in dynamic light scattering (DLS) to assess the size of the globules and the index of polydispersity.[26] Moreover, the ζ-potential was measured using a zetameter (Figure B). Satisfactorily, all formulations proved to remain monodispersed with a mean diameter under 200 nm for at least 14 days, and their ζ-potential was consistent over the same time. Moreover, NE-A, NE-D, and NE-E when assessed for longer times were found to be stable over 30 days.

In Vitro Cytotoxicity Evaluation of Nucleolipid-Loaded Nanoemulsions

Six formulations have been used in vitro to evaluate their respective cytotoxicity on human neuroblastoma cell lines (BE (2)-M17 cells) for either 24 or 48 h. These are the four NEs prepared (NE-A, NE-B, NE-D, and NE-E) and two aqueous formulations of compounds C and D (C: NL without the benzyl group and D: rhodamine B). After 24 or 48 h of exposure, no significant toxicity was observed, except for formulations Aq-D and NE-E (in the range of 20–36%) (Figure ).

Figure 3

Cell viability evolution over time. BE (2)-M17 cells treated with NE-A, NE-B, NE-D, and NE-E and compounds C and D solubilized in water compared to the control condition (UT) after 24 and 48 h. *p < 0.05 compared with untreated cells.

Role of NL-Loaded NEs for Internalization into Lysosomes in Vitro

As previously mentioned, lysosomal impairment is a common factor in neurodegenerative diseases. Lysosomes are intracellular acidic compartments that contain hydrolytic enzymes involved in the degradation of intracellular components through several degradation pathways, including endocytosis, phagocytosis, and autophagy.[26,27] Therefore, to determine if these NE–NL nanovectors can be used as a drug-delivery system for neuronal cells, cellular uptake and lysosomal colocalization of the different compounds, labeled with rhodamine B, were investigated (Figure ). Foremost, it has been observed that compounds/formulations NE-A, NE-B, NE-E, and Aq-C were well internalized into cells, while only 12 and 6% cellular uptake was observed for formulations NE-D and Aq-D, respectively. One possible explanation regarding the low rate of internalization of NE-D and Aq-D might be the leakage of free rhodamine outside the oily droplet of the NE, preventing the NE from fully serving its carrier function. Nevertheless, an uptake of 100% was obtained for the two NLs and the rhodamine–lipid conjugate loaded into NEs (A, B, or E), and it was found that the rate of NL internalized into cells was greatly enhanced with the use of NE. Indeed, we observed that only 50% of Aq-C, solubilized in water, get through plasma membranes (Figures and 5). Furthermore, each of these compounds labeled with rhodamine (red) colocalized into lysosomes (LAMP2, green).

Figure 4

In vitro biological assessment. Epifluorescence microscopy pictures showing uptake and colocalization into BE (2)-M17 cells treated with compounds A, B, D, and E formulated into NEs and compounds C and D solubilized in water. Nuclei were stained with Hoechst (blue), lysosomes were stained with LAMP2 (green), and compounds A–E are highlighted by the presence of rhodamine B (red). Scale bar: 10 μm.

Figure 5

Quantification of internalized rhodamine in treated cells to evaluate uptake into BE (2)-M17 cells of compounds A, B, D, and E formulated into NEs and compounds C and D solubilized in water.

Consequently, the combination of nanoemulsions with properly substituted nucleolipids A and B provided efficient and nontoxic nanovectors for neuronal cells’ internalization and subsequent lysosome colocalization.

Conclusions

We proposed the synthesis of original thymidine-derived NLs bearing the fluorophore rhodamine B to study their biocompatibility and their role as a promoter of absorption to cross biological membranes. First, we designed lipophilic NLs to afford solubilization into the oily phase of O/W NE. Then, the lipidic chains on the sugar moiety and the benzyl group were substituted on the thymine amino group, allowing the formulation of NLs into NEs. The colloidal stability of this NL–NE system was monitored over time by DLS and ζ-potential analysis before carrying out in vitro experiments. Cytotoxicity evaluation of NL–NE complexes on human neuronal cells showed that NLs are fully biocompatible after 24 and 48 h of exposure, while other amphiphilic compounds, such as rhodamine–lipid conjugates, display cellular toxicity. Evaluations of uptake into cells indicate that NL–NE systems are successfully internalized, highlighting the contribution of NEs as drug-delivery vehicles. More importantly, these original NLs bearing a fluorophore moiety are colocalized with lysosomes, suggesting that these nanotechnology-based systems can also be efficient tools to target lysosomes, whose impairment has been involved in neurodegenerative diseases. Therefore, these results were promising and pave the way to test, in future studies, and improve these new NL–NE systems for drug delivery in vivo, offering the possibilities for specific therapeutic solutions targeting pathologies associated with lysosomal impairment.

Experimental Section

General Information

All reactions were carried out under an argon atmosphere. Yields refer to chromatographically and spectroscopically (1H NMR) homogeneous materials, unless otherwise stated. All reagent-grade chemicals were obtained from commercial suppliers and were used as received, unless otherwise stated.

1H NMR and 13C NMR were recorded on a Bruker Avance 300 (1H: 300 MHz, 13C: 75.46 MHz) spectrometer using residual CHCl3 as an internal reference (7.26 ppm) and at 293 K unless otherwise indicated. The chemical shifts (δ) and coupling constants (J) are expressed in ppm and Hz, respectively. The following abbreviations were used to explain the multiplicities: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, and b = broad. Fourier transform infrared (FT-IR) spectra were recorded on a PerkinElmer FT spectrometer Spectrum two (UATR two). For electrospray ionization (ESI) high-resolution mass spectrometry (HRMS) analyses, a Waters Micromass ZQ instrument equipped with an electrospray source was used in the positive and/or negative mode. Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometric analyses were performed on a PerSeptive Biosystems Voyager-De Pro MALDI mass spectrometer in the linear mode using 3,4-dihydroxybenzoic acid as the matrix. Analytical thin-layer chromatography was performed using silica gel 60 F254 precoated plates (Merck) with visualization by ultraviolet light, potassium permanganate, or sulfuric acid. Flash chromatography was performed on a silica gel (0.043–0.063 mm).

Procedure for Compound (A) Synthesis

Synthesis of Compound 1

To a solution of thymidine (1 equiv, 6 g, 24.77 mmol) in pyridine (118 mL) and under argon are sequentially added tert-butyldimethylsilyl chloride (TBDMSCl) (1.2 equiv, 4.48 g, 29.72 mmol) and 4-dimethylaminopyridine (DMAP) (spatula tip). The mixture is stirred overnight at room temperature until completion of reaction. The reaction is then quenched by the addition of NaHCO3 (10 mL, aq.) and diluted with water (10 mL). The aqueous phase was extracted three times with dichloromethane (DCM) (3 × 30 mL), and the combined organic phases are dried over Na2SO4 before concentration to dryness under vacuum. The resulting crude is purified by flash chromatography on a silica gel (pentane/EtOAc, 70/30) to obtain the expected compound 1 as a white powder (7.68 g, 21.55 mmol, 87%). NMR data are consistent with the literature. R: 0.24 (pentane/EtOAc, 70/30). IR (ATR) ν (cm–1): 3549, 3167, 2929, 2855, 1696, 1471, 1257, 1120, 832, 779.

Synthesis of Compound 2

First, to a solution of 1 (1 equiv, 14.7 g, 41.21 mmol) in dimethylformamide (DMF) (88 mL) and under argon are sequentially added imidazole (1.2 equiv, 3.37 g, 49.54 mmol) tert-butylchlorodiphenylsilane (TBDPSCl) (1.2 equiv, 12.88 mL, 49.54 mmol). The mixture is stirred overnight at room temperature until completion of reaction. The mixture is then diluted with toluene (100 mL), and DMF is coevaporated under vacuum to obtain a yellow oil. The oil is diluted in diethyl ether (10 mL), and the white precipitate formed is filtered and washed with an aqueous solution of NaCl. The organic phase is dried over Na2SO4 before concentration to dryness under vacuum. 1′ was obtained as a yellow oil. Second, to a solution of 1′ in methanol (80 mL) and under argon is added para-toluenesulfonic acid monohydrate (0.471 g, 2.477 mmol). The mixture is stirred overnight at room temperature until completion of reaction. The methanol is then evaporated under vacuum to obtain a yellow oil. The crude product is diluted with EtOAc (30 mL) and the organic phase is washed three times with NaHCO3 (10 mL, aq.) and brine (3 × 20 mL). The organic phase is dried over Na2SO4 before concentration to dryness under vacuo. The resulting crude is purified by flash chromatography on a silica gel (petroleum ether/EtOAc, 50/50) to obtain the expected compound (2) as a white powder (6.90 g, 14.36 mmol, 58% over two steps). R: 0.29 (petroleum ether/EtOAc, 50/50). 1H NMR (300 MHz, CDCl3) δ ppm: 9.18 (bs, 1H, NH), 7.59–7.69 (m, 4H, H Ar), 7.34–7.50 (m, 6H, H Ar), 7.30 (d, J = 1.2 Hz 1H, H6), 6.26 (dd, J = 6.0 Hz, 7.8 Hz, 1H, H1′), 4.41–4.48 (m, 1H, H3′), 3.94–4.00 (m, 1H, H4′), 3.62 (dd, J = 2.4 Hz, 12.0 Hz, 1H, H5b′), 3.24 (dd, J = 3.0 Hz, 12.3 Hz, 1H, H5a′), 2.43 (bs, 1H, OH), 2.26 (ddd, J = 3.0 Hz, 6.0 Hz, 13.2 Hz, 1H, H2b′), 2.04–2.22 (m, 1H, H2a′), 1.82 (d, J = 1.2 Hz, 3H, H7), 1.08 (s, 9H, Bu). 13C NMR (75.46 MHz, CDCl3) δ ppm: 164.0 (C4), 150.5 (C2), 136.9 (C6), 135.8 (CH Ar), 133.4 (Cq Ar), 133.3 (Cq Ar), 130.2 (CH Ar), 130.2 (CH Ar), 128.0 (CH Ar), 111.0 (C5), 87.8 (C4′), 86.6 (C1′), 73.1 (C3′), 62.1 (C5′), 40.4 (C2′), 27.0 (CH3Bu), 19.1 (Cq Bu), 12.5 (C7). IR (ATR) ν (cm–1): 2932, 1682, 1471, 1274, 1104, 1031, 740, 702.

Synthesis of Compound 3

To a solution of 2 (1 equiv, 1 g, 2.08 mmol) in anhydrous DMF (10 mL) and under an argon atmosphere, NaH (1.2 equiv, 0.135 g, 3.37 mmol) is added in a vial before activation in a microwave oven (2 min, 40 °C, 200 W). Then, benzyl bromide (1.2 equiv, 0.4 mL, 3.37 mmol) is added, and the reaction mixture is placed in a microwave oven using a temperature control mode (4 min, 40 °C, 200 W). DMF is then coevaporated with toluene, and the resulting crude is purified by flash chromatography on a silica gel (petroleum ether/AcOEt, 70/30) to obtain a white powder 3 (0.846 g, 1.48 mmol, 71%). R: 0.26 (petroleum ether/EtOAc, 70/30). 1H NMR (300 MHz, CDCl3) δ ppm: 7.61–7.69 (m, 4H, H Ar), 7.36–7.51 (m, 8H, H Ar), 7.21–7.35 (m, 4H, H6 and H Ar), 6.28 (appearing t, J = 6.6 Hz, 5.2 Hz, 1H, H1′), 5.11 (m, 2H, H8), 4.41–4.49 (m, 1H, H3′), 3.94–4.00 (m, 1H, H4′), 3.62 (dd, J = 2.1 Hz, 11.7 Hz, 1H, H5b′), 3.23 (dd, J = 2.9 Hz, 12.0 Hz, 1H, H5a′), 2.29 (ddd, J = 2.3 Hz, 4.9 Hz, 13.1 Hz, 1H, H2b′), 2.08–2.33 (m, 1H, H2a′), 1.89 (d, J = 0.9 Hz, 3H, H7), 1.48–1.70 (bs, 1H, OH), 1.10 (s, 9H, Bu). 13C NMR (75.46 MHz, CDCl3) δ ppm: 163.5 (C4), 151.1 (C2), 136.9 (C6), 135.9 (CH Ar), 134.9 (CH Ar), 133.4 (CH Ar), 133.2 (CH Ar), 130.2 (CH Ar), 129.3 (CH Ar), 128.5 (CH Ar), 128.0 (CH Ar), 127.7 (CH Ar), 110.5 (C5), 87.7 (C1′ or C4′), 87.5 (C1′ or C4′), 73.1 (C3′), 62.3 (C5′), 44.6 (C8), 40.4 (C2′), 27.0 (CH3Bu), 19.2 (Cq Bu), 13.4 (C7). IR (ATR) ν (cm–1): 3456, 3071, 2932, 2859, 1666, 1634, 1428, 1240, 1103, 1024, 741, 700.

Synthesis of Compound 4

To a solution of 3 (1 equiv, 0.160 g, 2.80 × 10–1 mmol) in DCM (1.6 mL) and under argon are sequentially added palmitic acid (1.2 equiv, 0.086 g, 3.36 × 10–1 mmol), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC)·HCl (1 equiv, 0.052 g, 2.71 × 10–1 mmol), and DMAP (0.3 equiv, 0.010 g, 8.40 × 10–2 mmol). The mixture is stirred for 48 h at room temperature until completion of the reaction. The reaction is then quenched by the addition of NH4Cl (5 mL, aq.) and diluted with water (10 mL). The aqueous phase is extracted three times with DCM (3 × 10 mL), and the combined organic phases are dried over Na2SO4 before concentration to dryness under vacuum. The resulting crude is purified by flash chromatography on a silica gel (pentane/AcOEt, 90/10) to obtain compound 4 (0.138 g, 1.70 × 10–1 mmol, 63%) as a white foam. R: 0.20 (pentane/EtOAc, 90/10). 1H NMR (300 MHz, CDCl3) δ ppm: 7.62–7.75 (m, 4H, H Ar), 7.23–7.54 (m, 11H, H Ar), 7.19 (d, J = 0.9 Hz, 1H, H6), 6.50 (dd, J = 6.0 Hz, 7.5 Hz, 1H, H1′), 5.16 (q, J = 13.8 Hz, 2H, H8), 4.24–4.33 (m, 1H, H3′), 4.08–4.15 (m, 1H, H4′), 3.94 (dd, J = 3.0 Hz, 12.0 Hz, 1H, H5b′), 3.80 (dd, J = 4.2 Hz, 12.0 Hz, 1H, H5a′), 2.43 (ddd, J = 2.5 Hz, 5.5 Hz, 13.2 Hz, 1H, H2b′), 2.23 (t, J = 7.5 Hz, 2H, O=C–CH lipid), 1.94 (s, 3H, H7), 1.77–1.90 (m, 1H, H2a′), 1.48–1.61 (m, 2H, CH2 lipid), 1.26–1.28 (m, 24H, CH2 lipid), 1.14 (s, 9H, Bu), 0.93 (m, 3H, CH3 lipid). 13C NMR (75.46 MHz, CDCl3) δ ppm: 172.9 (C=O), 163.2 (C2), 150.8 (C4), 136.9 (Cq Ar), 135.7 (CH Ar), 135.7 (CH Ar), 133.0 (CH Ar), 132.9 (CH Ar), 132.8 (C6), 130.2 (Cq Ar), 130.2 (CH Ar), 129.1 (CH Ar), 128.3 (CH Ar), 128.0 (CH Ar), 127.9 (CH Ar), 127.5 (Cq Ar), 110.3 (C5), 85.6 (C1′), 84.7 (C4′), 73.0 (C3′), 63.2 (C5′), 44.5 (C8), 40.8 (C2′), 34.0 (CH2 lipid), 31.9 (CH2 lipid), 29.7 (CH2 lipid), 29.7 (CH2 lipid), 29.6 (CH2 lipid), 29.6 (CH2 lipid), 29.4 (CH2 lipid), 29.4 (CH2 lipid), 29.2 (CH2 lipid), 29.1 (CH2 lipid), 26.9 (CH3Bu), 24.7 (CH2 lipid), 22.7 (CH2 lipid), 19.0 (Cq Bu), 14.1(CH3 lipid), 13.4 (C7). IR (ATR) ν (cm–1): 3333, 2925, 2854, 1743, 1704, 1668, 1648, 1463, 1271, 1106, 1026, 700, 613, 506. HRMS (ESI) [M + H+]: calcd = 809.49194, found = 809.49210.

Synthesis of Compound 5

To a solution of 4 (1 equiv, 0.326 g, 4.03 × 10–1 mmol) in tetrahydrofuran (THF) (4.54 mL) and under argon is added tetra-n-butylammonium fluoride (TBAF) (1 M in THF, 0.403 mL) at 0 °C. The mixture is stirred for 2 h at 0 °C. The reaction is then quenched by the addition of water (5 mL), and the mixture is warmed to room temperature before the addition of DCM (10 mL). The aqueous phase is extracted three times with DCM (3 × 10 mL), and the combined organic phases are dried over Na2SO4 before concentration to dryness under vacuum. The resulting crude is purified by flash chromatography on a silica gel (pentane/AcOEt, gradient 90/10 to 70/30) to obtain the expected compound 5 (0.190 g, 3.33 × 10–1 mmol, 83%). R: 0.31 (pentane/EtOAc, 92/8). 1H NMR (300 MHz, CDCl3) δ ppm: 7.39–7.48 (m, 2H, H Ar), 7.18–7.35 (m, 4H, H Ar and H6), 6.32 (appearing t, J = 6.3 Hz, 1H, H1′), 5.11 (s, 2H, H8), 4.21–4.41 (m, 3H, H3′, H5b′ and H5a′), 4.13 (bs, 1H, H4′), 3.56–3.81 (bs, 1H, OH), 2.29–2.46 (m, 3H, O=C–CH lipid and H2b′), 2.00–2.13 (m, 1H, H2a′), 1.96 (s, 3H, H7), 1.56–1.71 (m, 2H, CH2 lipid), 1.27 (m, 24H, CH2 lipid), 0.85–0.93 (m, 3H, CH3 lipid). 13C NMR (75.46 MHz, CDCl3) δ ppm: 173.6 (C=O), 163.4 (C2), 150.9 (C4), 136.7 (Cq Ar), 133.4 (C6), 128.8 (CH Ar), 128.4(CH Ar), 127.6 (CH Ar), 110.3 (C5), 85.8 (C1′), 84.3 (C4′), 71.3 (C3′), 63.7 (C5′), 44.5 (C8), 40.5 (C2′), 34.2 (CH2 lipid), 31.9 (CH2 lipid), 29.7 (CH2 lipid), 29.6 (CH2 lipid), 29.5 (CH2 lipid), 29.4 (CH2 lipid), 29.2 (CH2 lipid), 29.1 (CH2 lipid), 24.9 (CH2 lipid), 22.7 (CH2 lipid), 14.1 (CH2 lipid), 13.4 (CH3 lipid). IR (ATR) ν (cm–1): 3363, 2918, 2850, 1697, 1621, 1470, 1178, 1092, 711. HRMS (ESI) [M + H+]: calcd = 571.37416, found = 571.37405.

Synthesis of Compound A

To a solution of 5 (1 equiv, 0.190 g, 3.33 × 10–1 mmol) in DCM (1.9 mL) and under argon are sequentially added rhodamine B (1.2 equiv, 0.191 g, 3.99 × 10–1 mmol), EDC·HCl (1 equiv, 0.062 g, 3.23 × 10–1 mmol), and DMAP (0.3 equiv, 0.012 g, 9.99 × 10–2 mmol) were sequentially added. The mixture is stirred for 48 h at room temperature. The reaction is then quenched by the addition of NH4Cl (5 mL, aq.) and diluted with water (10 mL). The aqueous phase is extracted three times with DCM (3 × 10 mL), and the combined organic phases are dried over Na2SO4 before concentration to dryness under vacuum. The resulting crude is purified by flash chromatography on a silica gel (DCM/MeOH, 95/5) to obtain the expected compound A as a purple powder (0.132 g, 1.28 × 10–1 mmol, 40%). R: 0.24 (DCM/MeOH, 95/5). 1H NMR (300 MHz, CDCl3) δ ppm: 8.24–8.34 (d, J = 30.9 Hz, 1H, H Ar), 7.80 (dt, J = 7.5 Hz, 14.7 Hz, 30.3 Hz 2H, H Ar), 7.31–7.41 (m, 2H, H Ar), 7.13–7.30 (m, 5H, H6 and H Ar), 7.09–6.86 (m, 4H, H Ar), 6.74 (bs, 2H, H Ar), 5.08–6.08 (m, 1H, H1′), 5.09–5.15 (m, 1H, H3′), 5.01–5.02 (d, J = 3.0, 2H, H8), 4.08–4.24 (m, 2H, H5′), 3.90–3.98 (m, 1H, H4′), 3.50–3.66 (m, 8H, CH–CH3 rhodamine), 2.04–2.28 (m, 4H, H2′ and O=C–CH lipid), 1.86 (s, 3H, H7), 1.40–1.52 (m, 2H, CH2 lipid), 1.07–1.32 (m, 36H, CH2 lipid and CH2–CH rhodamine), 0.79 (t, J = 6.3 Hz, 3H, CH3 lipid). 13C NMR (75.46 MHz, CDCl3) δ ppm: 172.8 (C=O), 164.1 (CH Ar), 163.0 (C2), 157.5 (Cq Ar or CH Ar), 155.4 (Cq Ar or CH Ar), 155.5 (CH Ar), 155.5 (CH Ar), 150.7 (C4), 136.6 (C6), 133.6 (CH Ar), 133.4 (CH Ar), 131.4 (CH Ar), 130.9 (CH Ar), 130.9 (CH Ar), 130.4 (CH Ar), 130.3 (CH Ar), 128.7 (CH Ar), 128.2 (CH Ar), 127.4 (CH Ar), 114.5 (Cq Ar), 113.2 (C5), 110.5 (Cq Ar), 96.3 (CH Ar), 85.7 (C1′), 81.8 (C4′), 75.5 (C3′), 63.5 (C5′), 46.1 (CH–CH3 rhodamine), 44.3 (CH2 lipid), 31.7 (C2′), 29.5 (CH2 lipid), 29.5 (CH2 lipid), 29.4 (CH2 lipid), 29.2 (CH2 lipid), 29.2 (CH2 lipid), 29.1 (CH2 lipid), 28.9 (CH2 lipid), 24.6 (CH2–CH rhodamine), 22.5 (CH2 lipid), 13.9 (CH2 lipid), 13.2 (CH2 lipid), 12.6 (C7). IR (ATR) ν (cm–1): 3333, 2923, 2852, 1716, 1704, 1647, 1586, 1465, 1412, 1336, 1272, 1246, 1178, 1131, 1072, 923, 682. HRMS (ESI) [M+]: calcd = 995.58924, found = 995.58830.

General Procedure for NE Preparation

Lipoid E80 (Lipoid GmbH, Ludwigshafen, Germany) was dispersed in heated (i.e., 70 °C) oil phase Miglyol 812N (IOI Oleo GmbH, Hamburg, Germany) using ultrasound delivered by a sonication bath. Once the dispersion was complete, 1 × 10–5 mol of the compound (A, B, D, or F) was solubilized in oil and surfactant mixture. Hydrophilic surfactant Tween 80 (SEPPIC, Paris, France) was dispersed in the heated (i.e., 70°) water phase (Milli-Q water). Emulsification was observed on adding the water phase to the oil phase thanks to the phase-inversion method. Homogenization was performed using a sonication probe (Sonic Vibra Cell-VC 250) for 10 min to obtain submicron size range oil droplets. Prior to in vitro experiments, formulation osmolality was adjusted with the addition of glycerol (Coopération pharmaceutique française) at 2% and pH was adjusted close to the physiological value using sodium hydroxide 0.1 N.

NE Characterization and Stability

Granulometric profiles of NE were characterized by dynamic light scattering (DLS) using Malvern Instruments (Zetasizer Nano ZS). NEs were diluted 1:1000 (v/v), and the average size and the polydispersity index were determined by three independent measurements performed at 25 °C. To analyze the ζ-potential, NEs were diluted 1:1000 (v/v) and measurements were performed using Zetasizer Nano ZS coupled with a folded capillary cell (DTS1060) from Malvern Instruments. Short-term stability assessment of the NE was performed for 1 month by checking the lack of visual creaming or phase separation and by checking the granulometric profiles and ζ-potential.

Cell Culture and Cell Viability Assay

Cell lines BE (2)-M17 (human neuroblastoma) were obtained from ATCC (CRL-2267) and grown in OPTIMEM (Life Technologies, 31985-047) by adding 10% fetal bovine serum (Sigma-Aldrich) and 1% penicillin/streptomycin. For all experiments, freshly prepared NE solutions were used and cells were grown at 70–80% confluence. Cells were treated for 24 or 48 h with 0.5 μL of loaded NE. Each experiment was reproduced at least three times. Cell viability was estimated by the MTS assay (ATCC/LGC Promochem) following the manufacturer’s instructions.

Immunostaining and Imaging

For all experiments, NE solutions were freshly prepared. For loaded-NE internalization and colocalization imaging assays, cells were trypsinized and replated in 6-well plates (NUNC) with coverslips. BE (2)-M17 cells were grown at 70–80% confluence per well. Once cells were attached, they were exposed for 24 h at 37 °C to the different loaded nanoemulsions. All nanoemulsions were added at a final concentration of ∼50 μg/mL. Cells were then fixed with 4% paraformaldehyde for 20 min at 4 °C and then washed with phosphate-buffered saline (PBS). Cell permeabilization and blocking steps were realized by the addition of a mixture composed of 5 mL of PBS, 450 μL of normal goat serum, and 225 μL of triton. Lysosomes were marked with the LAMP2 (Mx, H4B4) antibody overnight at 4 °C, and staining was revealed with the appropriate secondary antibody conjugated with GAM 488 (Life Technologies). For cytoplasm and nucleus staining, the cells were incubated for 8 min at room temperature with 8 μM Hoechst dye (ThermoFisher Scientific, #3342) prior to mounting. Slices were air-dried and mounted on #1.5 coverslips with Dako fluorescent mounting medium and left to dry overnight in darkness.

Image stacks (pixel size ∼100 nm, z-step 0.3 μm) were acquired on a laser scanning confocal microscope (Leica TCS SP8 microscope, Leica Microsystems) using a 63X Plan Apo CS oil immersion objective. Rhodamine B was detected using an excitation wavelength of 568 nm (DPSS laser) and with a detection window between 570 and 590 nm. Lysosomes were detected using an excitation wavelength of 488 nm (argon laser) and with a detection window between 545 and 605 nm. Cells without loaded-nanoemulsion treatments were used in parallel as autofluorescence controls using the corresponding excitation and detection wavelengths. Images were analyzed with LAS AF v2.6 acquisition software equipped with an HCS-A module (Leica Microsystems). Rhodamine B intensity relative to the cell surface was quantified with Definiens XD Developer v2.5 software (Definiens). Colocalization, fluorescence profiles, orthogonal projections, and maximal intensity projections were obtained with the ImageJ (NIH) distribution Fiji.

Statistical Analysis

Statistical analysis was performed in GraphPad Prism 6 software. For functional assays, statistical significance of the data was evaluated after one-way analysis of variance (ANOVA), followed by Tukey’s multiple comparison test. The level of significance was set at p < 0.05.