**1. Introduction**

Therapeutics based on oligonucleotides have significant potential for the treatment of a wide variety of diseases [1,2]. In principle, any disease with a known genetic origin can be treated by modifying genetic functions with oligonucleotide-based drugs. Compared to traditional pharmaceuticals, this approach has several advantages, including specificity, potency, and possibility for a rapid and rational drug design. The major limitation for the clinical translation of these therapeutics is the difficulty of in vivo delivery; as large, anionic macromolecules that are prone to degradation by nucleases, oligonucleotides require sophisticated carrier systems to enable delivery into the target cells [3,4]. An optimal carrier protects oligonucleotides from enzymatic degradation and clearance, and transfers them selectively into the cytoplasm of the target cells with minimal toxicity.

The most investigated synthetic vectors for oligonucleotides are lipid-based nanoparticles [5–7]. Commonly, these carriers contain cationic lipids that enable high loading capacity by complexing with negatively charged oligonucleotides and efficient intracellular delivery by interacting with the negatively charged cell membranes. However, the utility of cationic liposomes in vivo is limited,

as excess positive charge results in toxicity, innate immune activation, and poor pharmacokinetic properties [8–10]. Neutral or anionic liposomes show less interaction with serum proteins and complement components and, consequently, are less toxic and have better pharmacokinetic profiles. Yet, as non-cationic lipids do not interact with cellular membranes as efficiently as cationic ones, cellular uptake and intracellular release of entrapped oligonucleotide cargo with neutral liposomes is usually poor. Currently, the most promising lipid formulations contain ionizable lipids [11–13]. These lipids switch charge pH-dependently, enabling neutral particles at physiological pH in the blood circulation and in the extracellular space of tissues, and positive charge in the acidic environment of endosomes after cellular internalization.

In general, liposomal drug delivery is associated with poorly controlled and insufficient cytosolic oligonucleotide release. To improve the control and effectiveness of cytosolic delivery and drug release at the target site, systems that are activated by external or internal signals, such as temperature, pH, ultrasound, specific enzymes, magnetic field and light, have been developed [14,15]. In the present study, we applied a previously developed light-triggered liposomal system for oligonucleotide delivery [16–18]. The system consists of thermosensitive liposomes with indocyanine green (ICG) as the photosensitizing agent. The light sensitivity is based on the photothermal ability of ICG to absorb light energy and convert it to heat [18,19]; when the temperature-sensitive liposomes containing ICG are exposed to light, the released heat creates a localized temperature increase, leading to fluidization of the thermosensitive lipid membranes and release of the encapsulated drug.

ICG injections have been approved by the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) for fluorescence-based clinical imaging [20,21]. Compared to the most commonly used photothermal agents, gold and carbon nanomaterials, ICG has certain advantages: (1) ICG has absorption maximum at the near infrared (NIR) range, enabling excitation at a safe wavelength of 800 nm that penetrates into tissues [22,23]. (2) As an organic molecule, ICG can be conveniently incorporated into delivery systems. Also, processes for particle size control, such as extrusion and microfluidization, can be used, as the presence of ICG does not limit the size of the carrier [17]. (3) Since ICG is a fluorescent compound, it enables imaging-guided drug delivery in certain tissues [24–26]. (4) The safety profile of ICG is well-documented, while the long-term toxicity of non-biodegradable inorganic nanoparticles is unknown and they have not been approved for clinical use [27,28].

We have previously shown that the ICG-containing liposomes are functional in light-triggered release of small and large fluorescently labeled model compounds [16]. In the present work, we extended the concept to the delivery of oligonucleotides. Our aim was to investigate the effects of light induction and ICG on cellular delivery of oligonucleotides and liposomal oligonucleotides.

#### **2. Materials and Methods**

#### *2.1. Materials*

1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-*N*-[methoxy(polyethylene glycol)-2000] (DSPE-PEG) were bought from Lipoid (Ludwigshafen, Germany). 1-stearoyl-2-hydroxy-sn-glycero-3-phosphocholine (Lyso PC) was from Avanti Polar Lipids, Inc. (Alabaster, AL, USA). The oligonucleotides used in this study were a splice switching antisense oligonucleotide (SSO) restoring correct splicing of EGFP [29] and an siRNA against luciferase. The sequences of the oligonucleotides are the following:

```
SSO: 5
     -GCT ATT ACC TTA ACC CAG-3
siRNA: sense 5
           -CUUACGCUGAGUACUUCGAdTdT-3
anti-sense 5
         -UCGAAGUACUCAGCGUAAGdTdT-3
```
Underlined bases indicate a 2 -*O*-methyl modification. dT indicates deoxyribonucleic acid bases with phosphorothioate (PS) bonds. The SSO consists completely of PS bonds. The SSO was purchased from Biosearch Technologies (Petaluma, CA, USA) and the siRNA from Integrated DNA technologies (Leuven, Belgium). Cell medium and supplements were from GibcoBRL, Thermo Fisher Scientific (Naarden, The Netherlands). Indocyanine green purchased from Sigma-Aldrich (St. Louis, MO, USA) was the United States Pharmacopeia (USP) Reference Standard (mw. 775 g/mol). All other compounds were bought from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise mentioned. Fluorescence measurements of ICG and the Ribogreen assay were performed with a Jasco FP8300 Spectrofluorometer with micro-well plate reader (JASCO Benelux BV., De Meern, The Netherlands). An 808N10W laser system with a circular beam of 7 mm in diameter was used for the light triggering studies (Changchun Dragon Lasers Co., Ltd., Changchun, China). The output light intensities with different power settings were measured using a P-9710-1 optometer with RCH-102-2 custom-made detector head (Te Lintelo Systems BV, Zevenaar, The Netherlands). Light intensities (mW/cm2) corresponding to the power settings of 1–10 W are shown in the supporting information (Table S1).
