**1. Introduction**

Gene therapy offers highly specific therapeutics without significant side effects. It covers a wide variety of approaches, ranging from the replacement of malfunctioning genetic information, often in the form of a large nonviral plasmid DNA (pDNA), to the currently more investigated oligonucleotide (OND)-based therapy acting on the level of mRNA destabilization or translation [1,2]. As for OND-based therapy, synthetic 20–30-mer

**Citation:** Kubackova, J.; Holas, O.; Zbytovska, J.; Vranikova, B.; Zeng, G.; Pavek, P.; Mullertz, A. Oligonucleotide Delivery across the Caco-2 Monolayer: The Design and Evaluation of Self-Emulsifying Drug Delivery Systems (SEDDS). *Pharmaceutics* **2021**, *13* , 459 . https://doi.org/10.3390/ pharmaceutics13040459

Academic Editor: Ana Catarina Silva

Received: 15 February 2021 Accepted: 24 March 2021 Published: 28 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

nucleotides connected by phosphodiester bonds form a polyanionic backbone that determines the physicochemical properties of this hydrophilic macromolecule independently of the nucleotide sequence [3]. Therefore, there should be versatility in formulating OND for various applications. Nevertheless, the nucleotide structure is unstable in the presence of nucleases. Rapid enzymatic degradation combined with insufficient cellular uptake, often followed by entrapment in endosomes, makes the delivery of OND challenging [4]. Nonviral drug delivery systems, mainly lipid-based nanoformulations, have emerged as a promising option to overcome not only these limitations but, also, to enable oral delivery [5].

Oral delivery is the preferred route of administration—in particular, in the case of local gastrointestinal diseases. Many of these pathologies are defined or accompanied by intestinal inflammation, such as inflammatory bowel disease [6,7]. Novel colloidal drug delivery systems benefit from the enhanced permeation and retention effect of the inflamed tissue, which leads to increased accumulation of nanocarriers (~100–500 nm) within the leaky inflammatory site, which is also accompanied with an altered mucus layer [8,9]. Moreover, the increased accumulation of nanocarriers enables their uptake by phagocytic immune cells invading the area affected by inflammation; their rapid elimination due to diarrhea commonly present is avoided. In addition to the size of carriers, negative surface charge and hydrophilic surface decoration are also strategies utilized for passive targeting to the inflamed intestinal tissue [9–11]. The local delivery of OND has been investigated using many kinds of polymeric nanoparticles [12], lipid nanoparticles [13], microencapsulated nanogel [14], thioketal nanoparticles [15] and nanoparticles-in-microspheres multicompartment systems [16].

Self-emulsifying drug delivery systems (SEDDS) are isotropic mixtures of oils, surfactants, cosurfactants and cosolvents forming oil-in-water nanoemulsions upon gentle dispersion in an aqueous environment, e.g., gastrointestinal fluids [17]. These oral lipidbased drug delivery systems were originally established to improve the bioavailability of small poorly water-soluble molecules [18]. Recently, the delivery of hydrophilic macromolecules in SEDDS has gained increasing attention in order to take advantage of its beneficial properties, such as easy upscaling, nanosize of formed droplets and protection of the loaded substance from chemical and enzymatic degradation [19]. Nevertheless, the hydrophilicity of a potential hydrophilic substance needs to be reduced, usually by hydrophobic ion pairing [19]. The hydrophobic ion pairing of nucleic acids is a method based on replacing the counterions associated with negatively charged phosphate groups with surfactants carrying the positive charge. This method leads to a lipophilicity increase of the resulting ion-paired complexes [20]. For this purpose, positively charged surfactants, such as dimethyldioctadecylammonium bromide (DDAB) and 1,2-dioleoyl-3 trimethylammonium propane (DOTAP), are frequently utilized [21–24]. In addition, these quaternary ammonium salts enhance endosomal escape so the nucleic acid can reach the cytosol [25,26].

ONDs as hydrophilic macromolecules use predominantly paracellular transport through the intestinal monolayer; therefore, their permeability under physiological conditions is very limited [27]. Oral formulations containing intestinal permeation enhancers (PEs) have been widely investigated to enhance oral delivery of hydrophilic macromolecules [28]. PEs based on medium-chain fatty acids (MCFA) have been found to enhance paracellular transport via the opening of tight junctions (TJs) in a reversible manner by interaction with the cytoskeleton [29,30]. In addition, at higher concentrations the action of MCFA might be supported by transcellular permeation as a result of cell membrane alterations [28].

Hauptstein et al. pioneered the delivery of nucleic acids in SEDDS. In their research, several hydrophobic pDNA-cationic lipid complexes were tested. Among the tested cationic lipids, the complexes with cetrimide delivered in SEDDS showed a good transfection efficiency of HEK-293 cells comparable to Lipofectin®, a gold standard transfection reagent [31]. Mahmood et al. improved the transfection efficiency of pDNA-cetrimide in

analogical SEDDS by the incorporation of a cell-penetrating peptide and confirmed an internalization of 50-nm nanoemulsions in Caco-2 cells [32]. Both studies focused on the delivery of large pDNA into cells. In contrast to considerably larger pDNA molecules, this study focuses on the delivery of short OND sequences as emerging, highly potent therapeutics. ery of large pDNA into cells. In contrast to considerably larger pDNA molecules, this study focuses on the delivery of short OND sequences as emerging, highly potent therapeutics. The aim of this study was to prepare and characterize MCFA-based SEDDS loaded with hydrophobized OND and to investigate the ability of this system to deliver OND

efficiency of HEK-293 cells comparable to Lipofectin®, a gold standard transfection reagent [31]. Mahmood et al. improved the transfection efficiency of pDNA-cetrimide in analogical SEDDS by the incorporation of a cell-penetrating peptide and confirmed an internalization of 50-nm nanoemulsions in Caco-2 cells [32]. Both studies focused on the deliv-

The aim of this study was to prepare and characterize MCFA-based SEDDS loaded with hydrophobized OND and to investigate the ability of this system to deliver OND across the intestinal Caco-2 monolayer. This formulation approach has a considerable potential to overcome OND low stability and poor permeability. In this study, firstly, hydrophobized complexes of cationic lipids (DDAB or DOTAP) and a model fluorescently labeled 20-mer OND were thoroughly described and subsequently loaded into SEDDS (Figure 1). Dispersed SEDDS (negatively charged or neutral SEDDS) were characterized in terms of size, zeta potential, lipolysis, protective effect against nucleases and OND permeability. across the intestinal Caco-2 monolayer. This formulation approach has a considerable potential to overcome OND low stability and poor permeability. In this study, firstly, hydrophobized complexes of cationic lipids (DDAB or DOTAP) and a model fluorescently labeled 20-mer OND were thoroughly described and subsequently loaded into SEDDS (Figure 1). Dispersed SEDDS (negatively charged or neutral SEDDS) were characterized in terms of size, zeta potential, lipolysis, protective effect against nucleases and OND permeability.

*Pharmaceutics* **2021**, *13*, x 3 of 27

**Figure 1.** A general scheme showing the formulation of an oligonucleotide (OND) into a self-emulsifying drug delivery system (SEDDS). **Figure 1.** A general scheme showing the formulation of an oligonucleotide (OND) into a self-emulsifying drug delivery system (SEDDS).

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

#### *2.1. Materials 2.1. Materials*

The 20-mer model 5′-6-carboxyfluorescein (FAM) labelled OND composed of random DNA nucleotides (molecular weight 6654.5 Da) was purchased from Generi Biotech (Hradec Kralove, Czech Republic). Lipoid S LPC 80® (LPC) (from soybeans, containing 80.8% monoacyl phosphatidylcholine and 13.2% phosphatidylcholine) and phospholipids Lipoid S PC (from soybeans, containing 98.0%phosphatidylcholine) were provided by Lipoid GmbH (Ludwigshafen am Rhein, Germany). Labrasol® (caprylocaproyl macrogol-8 glycerides), Maisine CC® (glyceryl monolinoleate) and Peceol® (glyceryl monooleate) were provided by Gattefossé (Saint-Priest, France). Captex 300® (glyceryl tricaprylate/tricaprate) was provided by Abitec (Columbus, OH, USA) and Citrem® (glycerides of citric acid) by Danisco-DuPont (Grindsted, Denmark). Dimethyldioctadecylammonium bromide (DDAB), bovine bile extract, maleic acid, porcine pancreatic lipase extract, bovine serum albumin (BSA), 2-(N-morpholino)ethansulfonic acid (MES), N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) and orlistat were purchased from Sigma-Aldrich (now Merck, Darmstadt, Germany). 1,2-dioleoyl-3-trimethylammonium propane (chloride salt) (DOTAP) was purchased from Avanti Polar Lipids (Alabaster, AL, USA). The 20-mer model 50 -6-carboxyfluorescein (FAM) labelled OND composed of random DNA nucleotides (molecular weight 6654.5 Da) was purchased from Generi Biotech (Hradec Kralove, Czech Republic). Lipoid S LPC 80® (LPC) (from soybeans, containing 80.8% monoacyl phosphatidylcholine and 13.2% phosphatidylcholine) and phospholipids Lipoid S PC (from soybeans, containing 98.0%phosphatidylcholine) were provided by Lipoid GmbH (Ludwigshafen am Rhein, Germany). Labrasol® (caprylocaproyl macrogol-8 glycerides), Maisine CC® (glyceryl monolinoleate) and Peceol® (glyceryl monooleate) were provided by Gattefossé (Saint-Priest, France). Captex 300® (glyceryl tricaprylate/tricaprate) was provided by Abitec (Columbus, OH, USA) and Citrem® (glycerides of citric acid) by Danisco-DuPont (Grindsted, Denmark). Dimethyldioctadecylammonium bromide (DDAB), bovine bile extract, maleic acid, porcine pancreatic lipase extract, bovine serum albumin (BSA), 2-(N-morpholino)ethansulfonic acid (MES), N-2-hydroxyethylpiperazine-N0 -2-ethanesulfonic acid (HEPES) and orlistat were purchased from Sigma-Aldrich (now Merck, Darmstadt, Germany). 1,2-dioleoyl-3-trimethylammonium propane (chloride salt) (DOTAP) was purchased from Avanti Polar Lipids (Alabaster, AL, USA). Gibco® Hanks' Balanced Salt Solution (HBSS) was purchased from Thermo Fisher Scientific (Copenhagen, Denmark). S1 nuclease (Cat. No. EN0321) was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Triethylamine 99.5%, glacial acetic acid 99.99%, and acetonitrile were of HPLC grade. Deionized (DI) water was purified by SG Ultraclear water systems (SG Ultra Clear™ 2002, SG Water GmbH, Barsbüttel, Germany). Hoechst 33342 and propidium

iodide (PI) were purchased from Molecular Probes (Eugene, OR, USA). All other chemicals were of analytical grade and commercially available.
