**About the Editors**

#### **Sofia Lima**

Sofia Lima is an assistant researcher at REQUIMTE, University of Porto, with a great interest in polymers as drug delivery systems. She completed her PhD in chemistry at the University of Porto, and her research is focused on the study of cell–nanomaterial interactions using biochemical and biophysical techniques to unravel related mechanisms of action. Improving the knowledge on nanomaterial safety is crucial to ensure their emerging introduction in consumer goods (food and health). During the last 10 years, she has particularly been dedicated to the development of drug delivery systems able to overcome the barriers (skin and gastrointestinal) leading to the exploitation of natural compounds as tools to enhance permeation and to thus modulate biological processes, including the immune system.

#### **Salette Reis**

Salette Reis is the coordinator of the NanoPlatforms Research Group of LAQV, REQUIMTE (https://www.requimte.pt/laqv/); is the leader of the Molecular Biophysics and Biotechnology Unit of REQUIMTE at the University of Porto; and is a Cathedratic Professor at the Faculty of Pharmacy of the University of Porto. Her research activities are focused on the study of the complex interplay between drugs and lipid membranes. In this field, she is interested in the study of membrane biophysical changes related to the mechanism of action and toxicity of drugs and in the study of the effect of drugs on the activity of membrane enzymes involved in inflammation. Salette Reis is also involved in the study of the role of the membrane biophysical properties on membrane peroxidation and on the mechanism of action of antioxidants. Recently, her research activities have also been based on the development of drug nanocarrier systems to overcome the disadvantages of classical therapies.

## *Editorial* **Polymeric Carriers for Biomedical and Nanomedicine Application**

**Sofia A. Costa Lima \* and Salette Reis \***

LAQV, REQUIMTE, Departamento de Ciências Químicas, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira, 228, 4050-313 Porto, Portugal

**\*** Correspondence: slima@ff.up.pt (S.A.C.L.); shreis@ff.up.pt (S.R.)

Polymeric carriers play a key role in modern biomedical and nanomedicine applications. Polymers can be obtained from natural or synthetic sources and have been exploited given their chemistry to achieve interaction with living tissues and cells. Different types of carriers can be produced for drug delivery, namely, micelles, nanoparticles, dendrimers, sponges, hydrogels, and microneedles. With different coatings, appropriate adhesion and targeting features can be designed. Polymeric carriers allow the incorporation or conjugation of both hydrophilic and hydrophobic molecules and have tunable chemical and physical features that allow effective drug protection from degradation or denaturation. Other features are noteworthy in polymeric carriers, like their generally good biocompatibility and the ability to exhibit a slow and controlled dug release, allowing for the use in biomedical applications.

This Special Issue provides an encompassing view on the state of the art of polymeric carriers, showing how current research is dealing with new stimuli-responsive systems for cancer therapies and biomedical challenges, namely, overcoming the skin barrier. The published papers cover topics ranging from novel production methods and insights on hybrid polymers to applications as diverse as nanoparticles, hydrogels, and microneedles to antifungal skin therapy, peptide and siRNA delivery, enhanced skin absorption of bioactive molecules, and anticancer therapy. This Special Issue contains one review paper on modulation of macrophage polarization mediated by carbohydrate-functionalized polymeric nanoparticles [1]. A couple of polymeric carriers targeting macrophages have been reviewed in terms of production methods and conjugation approaches. The role of mannose receptor in the polarization of macrophages is highlighted as strategies for infectious diseases and cancer therapies as well as prevention actions.

Taking advantage of polysaccharides' physicochemical features, Pontillo et al. designed new biocompatible and cost-effective carriers for tyrosol, a bioactive natural product present in olive oil and white wine [2]. A chitosan based nanosystem was obtained using the ionic gelation method, while for β-cyclodextrin (βCD), the kneading method was employed. Additionally, coating of the tyrosol–βCD inclusion complex with chitosan led to a sustained release of tyrosol and slowed down the initial burst effect observed from the inclusion complex. The nanosystems were extensively characterized after optimized production based on a two- or three-factor, three-level Box–Behnken experimental design. Moreover, the interaction of tyrosol and the corresponding nanosystems with ctDNA was evaluated. Data suggest that tyrosol is a ctDNA groove binder, which was confirmed by molecular modeling studies. The same mode of binding was found only for the tyrosol/βCD and tyrosol/βCD/chitosan nanosystems. Nanocomposites of chitosan and alginate were exploited by Sabbagh et al. to deliver metronidazole [3]. Optimization of the formulation was obtained using a full factorial design to study the effect of chitosan and alginate polymer concentrations and calcium chloride concentration on drug loading efficiency, particle size, and zeta potential. These dependent variables were affected by the chitosan, alginate, and calcium chloride concentrations, while zeta potential depended only on the alginate and calcium chloride concentrations. The applied mathematical models revealed that the devel-

**Citation:** Lima, S.A.C.; Reis, S. Polymeric Carriers for Biomedical and Nanomedicine Application. *Polymers* **2021**, *13*, 1261. https:// doi.org/10.3390/polym13081261

Received: 25 March 2021 Accepted: 7 April 2021 Published: 13 April 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/).

oped response surface methodology models were statistically significant and adequate for all conditions. High correlation values were determined between the experimental data and predicted ones. The optimized nanocomposites were physiochemically characterized by X-ray diffraction, Fourier-transform infrared spectroscopy, thermal gravimetric analysis, scanning electron microscopy, and in vitro drug release studies. Overall, the optimized nanocomposites could be effective in sustaining the metronidazole release for a prolonged period. Hybrid nanosystems have been studied by Duskey et al. to increase the applicability of poly(lactic-co-glycolic acid) (PLGA) in drug delivery [4]. A series of unique PLGA–chitosan hybrid polymers with tailored and tunable physicochemical characteristics were obtained with two different synthetic methods: solid-phase synthesis on a film or in solution chemical reaction with polycaprolactone as intermediate. The hybrid polymers were physiochemically characterized using nuclear magnetic resonance, Fourier-transform infrared spectroscopy, and dynamic scanning calorimetry. A sodium dodecyl sulfate (SDS) salting-out reaction led to a chitosan SDS intermediate that is soluble in organic solvents, and consequently, a new series of PLGA–chitosan copolymers with different molar ratios were produced. The unique series of PLGA–chitosan hybrids with various molar rapports and solubilities represent the expansion of the PLGA delivery system for the protection and delivery of a wide range of previously noncompatible drugs either as nanoparticles formed through chitosan self-assembly techniques (for those still soluble in acidic solutions) or for the encapsulation in stable and nontoxic films for long-term controlled release (for those insoluble in biological solutions).

Shin et al. developed PLGA nanoparticles as siRNA carriers to overcome ROS/oxidative stress-induced chondrocyte damage in osteoarthritis [5]. A double emulsion technique allowed the successful incorporation of siRNA p47phox within PLGA nanoparticles. The nanosystem was physicochemically characterized and evaluated in chondrocytes and in an osteoarthritis in vivo model. The formulated PLGA nanoparticles provided a sustained release of siRNA, which could reduce dosing frequency to a weekly regimen. Inhibition of p47phox by nanoparticles delivered siRNA-attenuated pain behavior, cartilage damage, and ROS production in knee joints with induced osteoarthritis. The developed polymeric nanosystem may represent a promising novel therapeutic avenue for the treatment of osteoarthritis. PLGA nanoparticles were explored for peptide delivery by Lima et al. [6]. A peptide from the myeloid proteolipid protein (PLP) was encapsulated in PLGA nanoparticles and further incorporated within polymeric microneedle patches for an effective skin delivery. Trehalose was included to preserve the nanoparticles during the freeze-drying process. Polydimethylsiloxane molds were used to obtain poly(vinyl alcohol)–poly(vinyl pyrrolidone) microneedles to carry the freeze-dried PLP-loaded PLGA nanoparticles. Microneedle patches with 550 μm height and 180 μm diameter allowed the peptide release in physiological media. The achieved outcomes motivate the exploitation of this strategy as a new antigen-specific therapy, providing minimally invasive administration of PLP-loaded nanoparticles into the skin. Struggling with skin drug delivery, Zhang et al. studied the effect of poly(ethylene glycol) (PEG) 400 and PEG-6-caprylic/capric glycerides on the dermal absorption of niacinamide [7]. Binary and ternary systems composed of PEGs or PEG derivatives combined with other solvents were studied for skin delivery of niacinamide. Porcine skin permeation assays over 24 hours revealed improved performance of all designed vehicles in relation to PEG 400. High skin retention was observed for these vehicles when compared with the neat solvents investigated. Hence, these results indicate PEG 400 as a useful tool to deliver the bioactive agents to the skin, instead of through the skin. According to the bioactive agent, skin retention may be more interesting than skin permeation. Skin retention of terbinafine was investigated by Ghose et al. through the design of polymeric nanosponge hydrogel. The antifungal agent was incorporated in Box–Behnken-design-optimized nanosponge formulations [8]. In vitro drug release from the nanosponge incorporated into the hydrogel was higher than the drug suspension or the marked formulation. Antifungal activity, nonirritancy, and no erythema or edema

confirmed the promising application of the developed nanosponge hydrogel for efficient topical delivery of terbinafine hydrochloride.

Stimuli-responsive nanosystems have been designed to control the release of active molecules into the intended site of action. Van Gheluwe et al. applied a three-step synthesis of a redox-responsive blend of poly(ethylene glycol)–*block*–poly(lactide) (PEG–*block*–PLA) and poly(lactide) (PLA) to deliver retinol in the skin [9]. The selection of short-length polymers to incorporate the lipophilic active molecule allowed for achieving a high loading and rapid release of retinol. Stimuli responsiveness of the nanosystem was confirmed in vitro in the presence of L-glutathione. Good biocompatibility of black nanocarriers was observed in human keratinocytes, and low toxicity was detected in the presence of retinol. The redox-responsive blend of PEG–*block*–PLA and PLA were assembled by nanoprecipitation in smart nanocarriers able deliver other retinoid molecules for the treatment of skin diseases, like acne, photoaging, psoriasis vulgaris, melisma, and skin cancers. Cano-Cortes et al. investigated the drug covalent conjugation to the polymeric nanosystem based on PEGylated polystyrene pH-responsive polymer [10]. Doxorubicin was selected to evaluate this pH-responsive approach to cancer therapy. An efficient loading was achieved upon covalent conjugation of doxorubicin to cross-linked polystyrene nanoparticles, allowing selective drug release under acidic pH values. Breast and lung cancer cell lines were studied to determine the efficiency of cellular uptake, therapeutic activity, and genotoxicity effect. The pH-responsive polymeric nanosystems exhibited better antitumor activity in relation to free doxorubicin. The implemented chemical strategy could be further applied to other molecules and types of cancer.

**Data Availability Statement:** No new data were created or analyzed in this study. Data sharing is not applicable to this article.

**Acknowledgments:** The guest editors would like to thank all contributors of this Special Issue in the Polymers journal (MDPI). Special thanks to all reviewers who help us to ensure the quality of each published article in this Special Issue; special thanks to the editor in chief and assistant editorial team of Polymers for helping us to complete this work. The guest editors are thankful for the support from FEDER funds through the COMPETE 2020 Operational Programme for Competitiveness and Internationalisation (POCI), Portugal 2020, and national funds through FCT/MCTES in the framework of the project POCI-01-0145-FEDER-030834, and Base Funding UIDB/50006/2020. Sofia Lima thanks the Portuguese Foundation for Science and Technology (FCT) for the financial support for her work contract through the Scientific Employment Stimulus-Individual Call (CEECIND/01620/2017).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Review* **Modulation of Macrophages M1/M2 Polarization Using Carbohydrate-Functionalized Polymeric Nanoparticles**

**Raquel G. D. Andrade 1, Bruno Reis 2,3, Benjamin Costas 1,\* Sofia A. Costa Lima and Salette Reis <sup>2</sup>**


**Abstract:** Exploiting surface endocytosis receptors using carbohydrate-conjugated nanocarriers brings outstanding approaches to an efficient delivery towards a specific target. Macrophages are cells of innate immunity found throughout the body. Plasticity of macrophages is evidenced by alterations in phenotypic polarization in response to stimuli, and is associated with changes in effector molecules, receptor expression, and cytokine profile. M1-polarized macrophages are involved in pro-inflammatory responses while M2 macrophages are capable of anti-inflammatory response and tissue repair. Modulation of macrophages' activation state is an effective approach for several disease therapies, mediated by carbohydrate-coated nanocarriers. In this review, polymeric nanocarriers targeting macrophages are described in terms of production methods and conjugation strategies, highlighting the role of mannose receptor in the polarization of macrophages, and targeting approaches for infectious diseases, cancer immunotherapy, and prevention. Translation of this nanomedicine approach still requires further elucidation of the interaction mechanism between nanocarriers and macrophages towards clinical applications.

**Keywords:** glyconanoparticles; immunotherapy; infectious diseases; mannose receptors; nutraceuticals

#### **1. Introduction**

Nanomedicine aims to improve health and life welfare with nanosized materials. Nanoparticles can be designed for drug delivery by modulating surface properties and composition to improve therapeutic effect and targeting specificity. Active targeting can be obtained with surface functionalization of the nanoparticles using specific ligands to reach the target of interest [1]. Taking advantage of this receptor-mediated specificity will reduce toxicity and side-effects to healthy tissues.

Macrophages are innate immune cells widely present in the body acting to maintain homeostasis and to resist pathogen invasion [2]. Macrophages are distributed according to their functions, surface-expressed markers, and secreted cytokines in M1/M2-polarized phenotypes. However, the simplicity of M1/M2 dichotomy of macrophage activation is too broad to explain all the actual states of the macrophages as a response to several stimuli. To have a proper description of the macrophage activation, it is generally accepted to include information on macrophage source, type of activators, and markers [3]. An imbalance in the M1/M2 ratio weakens the immune response and leads to inflammation. Hence, macrophages constitute an important player in the therapeutic strategies against infections, inflammatory conditions, and cancer. Receptors frequently expressed on the surface of macrophages constitute a potential target for nanomedicine-based approaches. Macrophage scavenger receptor, Toll-like receptors, glucan receptor, folate receptor, and mannose receptor are among the most used surface receptors of macrophages [4].

**Citation:** Andrade, R.G.D.; Reis, B.; Costas, B.; Lima, S.A.C.; Reis, S. Modulation of Macrophages M1/M2 Polarization Using Carbohydrate- Functionalized Polymeric Nanoparticles . *Polymers* **2020**, *13*, 88. https:// doi.org/10.3390/polym13010088

Received: 9 December 2020 Accepted: 23 December 2020 Published: 28 December 2020

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

**Copyright:** © 2020 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/).

Mannose receptor (MR) is composed by several domains that allows recognition to various molecules of the carbohydrate family and contributes to receptor-mediated endocytosis. Upon internalization, nanocarriers can elicit macrophage polarization in vivo. Different types of polymeric based carriers (e.g., nanoparticles, micelles, dendrimers) are emerging as macrophage-targeted delivery systems [5]. Nanocarriers can also reset the macrophage activation state, as it is the case of the conversion of M2 phenotype to M1 in tumor-associated macrophages [6]. Understanding the interaction mechanisms between nanoparticles and macrophages is essential to a successful and effective nanocarrier's design towards a therapeutic or prevention strategy.

#### **2. Polymeric Nanoparticles as Biomedical Delivery Devices**

Over the past few decades, the development of new strategies that surpasses the problems associated with conventional diagnosis and therapies have gained great importance on the scope of nanomedicine. One of the main goals in this field is to design nanoparticles capable of a targeted delivery and controlled release of bioactive compounds to a specific site, increasing its therapeutic effect while minimizing its side effects [7,8]. Several types of nanoparticles can be prepared from different building blocks like lipids, proteins, metals, and polymers [7–9]. Polymeric nanoparticles have gained great importance as biocompatible drug delivery systems given their simplicity and low-cost production [10]. The use of polymeric nanoparticles in drug delivery has many advantages over the use of other types of nanocarriers: a growing choice of biodegradable and biocompatible polymers, higher encapsulation efficiencies, higher stability in physiological conditions, improved drug bioavailability, and simpler preparation (for more detailed information on the synthesis methods see ref. [11]).

The design of drug delivery systems needs to consider several characteristics, namely, hydrophobicity, size, surface charge, biological interactions/toxicity, and biodegradability. A wide variety of natural or synthetic polymers are available for the preparation of the nanoparticles [12,13]. To produce nanoparticles, the most commonly used natural polymers include chitosan, a linear polysaccharide extracted from the exoskeletons of marine crustaceans [14], alginate that is isolated from brown algae [15], and gelatin obtained by hydrolyzed collagen [16]. Natural polymers have the advantage of combining biological properties like mimicking the extracellular matrix, allowing to sustain cell growth in tissue engineering applications, and tunable mechanical properties like stimuli-responsiveness, degradation, swelling, and crosslinking capabilities [13–15]. However, the application of natural polymers is often hampered by contaminants and batch-to-batch variability. Other constraints involve low hydrophobicity that compromises lipophilic drugs encapsulation, and a rapid drug release from the matrix [17,18].

Limitations of natural polymers can be overcome with the use of synthetic polymers, which are more reproducible in manufacture and more stable. Polymeric nanoparticles obtained with synthetic polymers allow drug-controlled release for a period of days up to several weeks [18]. Drawbacks associated with these type of nanoparticles involve their limited aqueous solubility and the need of surfactants to form stable suspensions [19]. The outcome of the nanoparticle as a drug delivery system can be modulated in the composition not only in the nature of the polymer, but also in the molecular weight, copolymer composition, and selected surfactant. To produce a targeted drug release within the body, the nanoparticles can be considered with additional properties to respond to external or internal stimuli such as redox state or pH [20]. Polylactic acid (PLA), poly(glycolic acid) (PGA), and their copolymers (PLGA) represent the most extensively used and studied synthetic polymers for drug delivery [21–23]. The presence of ester linkages in their backbones make these polyesters biodegradable. In fact, in a living organism, these polymers suffer a hydrolysis, and the resulting products are easily metabolized in the Krebs cycle and eliminated as carbon dioxide and water [17]. Also widely applied in the production of nanoparticles is poly-E-caprolactone (PCL) that allows a slow degradation rate in comparison with PLA and PLGA, and thus is more adequate for long-term drug delivery. Poly(alkylcyanoacrylate) (PACA) is an interesting polymer whose properties can be controlled by the side of the introduced chains, being that the longer the side-chains the longer the half-life of the nanoparticles [17].

Depending on the preparation method, used polymers and desired application, different polymeric nanocarriers can be obtained such as polymer-drug conjugates, polymeric micelles, polymeric nanogels, and dendrimers [24]. Two types of polymer nanoparticles can be obtained for drug delivery: nanocapsules, composed of a liquid or semisolid core covered by a polymer membrane; or nanospheres that consist in a solid polymer matrix [11,12,23–25]. The drugs can be either entrapped in nanoparticles or adsorbed at the surface. In nanocapsules, the drug can be encapsulated in the inner core, while in nanospheres it is uniformly dispersed in the polymer matrix (Figure 1). These represent versatile tools for surface modification, as well as shape, size, and even optical characteristics. In nanomedicine, core–shell polymeric nanoparticles are also interesting, as the polymeric platform allows a second shell, usually a solid, which may confer smart properties to the nanoparticle (e.g., pH sensitive, thermo- and enzyme-responsive) [26–28].

**Figure 1.** Schematic representation of the two types of polymeric nanoparticles: nanocapsules (**A**) and nanospheres (**B**). Nanocapsules comprise an inner cavity, composed of water or a semi solid (oil), and covered with a polymer membrane, while in nanospheres the entire mass is a polymer matrix. Drug molecules can be entrapped in both types of nanoparticles.

#### **3. Production Methods for Polymeric Nanoparticles and Surface Properties Modifications**

Currently, there are several methods developed and well-implemented for the preparation of polymeric nanoparticles. At first, one needs to ponder on (i) the physicochemical properties of the bioactive compound to be delivered, (ii) the nature and type of polymer, (iii) the target and biological environment, and (iv) the administration route. Based on this information it is possible to select the most adequate production method among emulsification-solvent evaporation, nanoprecipitation, emulsification reverse salting-out, and emulsification solvent diffusion. These polymerization processes allow production of nanoparticles with control of physicochemical and biological properties of the nanoparticles that are formed (Figure 2). At least two steps are involved in these conventional production methods: (i) polymer dissolution in an organic solvent followed by emulsification in an aqueous phase, and (ii) solvent evaporation to obtain the nanoparticles [13,29]. Polymeric nanoparticles can also be produced using monomers in an emulsion or as a micellar suspension by interfacial poly-condensation [13,17].

**Figure 2.** Diagram representing the options of production methods to obtain polymeric nanoparticles. Abbreviations: NMP (nitroxide-mediated polymerization); ATRP (atom transfer radical polymerization); RAFT (reversible addition and fragmentation transfer chain polymerization).

Hydrophilicity of the drug delivery systems represents an important feature to be considered for biological application. In fact, upon intravenous administration, hydrophobic nanoparticles are taken as foreign and the organism removes them from circulation to the excretion organs (liver, spleen, and lymph nodes) using the mononuclear phagocytic system [30]. If the intended treatment targets one of these organs, hydrophobic nanoparticles are the best solution. When aiming different targets, systemic circulation needs to occur, so the delivery systems reaches the diseased site. In this case, the surface of the nanoparticles must be modified with hydrophilic polymers to prevent the action of the mononuclear phagocytic system and phagocytosis. Hydrophilic nanoparticles will have long circulation times and reduced nonspecific distribution [31,32]. The list of hydrophilic polymers is long and include polyethylene glycol (PEG), poly-vinyl pyrrolidone (PVP), pluronics (poly-ethylene oxides), poloxamers, vitamin E TPGS, polysorbate 20, polysorbate 80, and polysaccharides (e.g., dextran) [33]. A protective layer can be obtained at the surface of the nanoparticles with these hydrophilic compounds, by adsorption or grafting shield groups. In some cases, PEG can be incorporated as copolymer [30,34,35]. The most used hydrophilic polymer for nanoparticles' surface modification is PEG. The nature (flexible chains) and the physicochemical (hydrophilicity) features of this polymer as well as the presence of functional groups able to prevent plasma proteins binding are the reasons for this success [36]. A significant decrease in the opsonization and macrophage internalization of nanoparticles was observed with PEG coating, which lead to an enhanced long-term blood circulation. PEGylated nanoparticles promote a higher drug uptake by target tissues when compared to non-PEGylated ones [37–39].

In sum, a crucial feature of polymeric nanoparticles is their surface modification in order to improve drug delivery. On the one hand, the addition of a stealth layer (PEG, PVA, polysorbate) at the surface of nanoparticles allows an increased blood circulation time, avoiding the binding of opsonins and the rapid clearance from the mononuclear phagocytic system, and on the other hand, the functionalization at the surface with targeting ligands (proteins, peptides, antibodies [40–42]) improves the specificity of the treatment [43,44].

#### **4. Carbohydrate-Functionalized Polymeric Nanoparticles**

As stated before, polymeric nanoparticles have excellent features that make them promising delivery systems for therapeutic applications. A higher specificity of drug delivery to a certain site of action is achieved when targeting ligands are incorporated in these nanocarriers. The functionalization of nanoparticles with carbohydrates, also known as glyconanoparticles, plays a key role in receptor-mediated delivery, as it allows to establish specific interactions with carbohydrate-binding proteins (lectins) [45,46]. Besides molecular recognition, sugars can act as colloidal stabilizers [47], reduce toxicity [48] and immunogenicity [49] and unlike PEG, increase circulation time in the bloodstream without compromising cellular uptake [50].

Glycopolymers can be prepared either by post-polymerization modification, which consists in the functionalization of a preformed polymeric backbone, or in the polymerization of glycosylated monomers [51,52] that can be performed by several synthetic routes that provide controllable architectures, stereochemistry, and molecular weights, such as free radical polymerization (ring-opening polymerization (ROP)), ionic polymerization, controlled radical polymerization (nitroxide-mediated polymerization (NMP), atom transfer radical polymerization (ATRP), reversible addition fragmentation chain transfer (RAFT), and enzyme-mediated polymerization [51,53–55]. Here, we will focus on the post-functionalization of polymeric nanoparticles with carbohydrates, as it allows the attachment of pendant carbohydrate moieties (Figure 3), making it ideal for targeted delivery.

**Figure 3.** Chemical structure of some carbohydrates commonly used to produce glyconanoparticles. (**A**) Galactose; (**B**) mannose; (**C**) mannan.

The coupling of a ligand to a nanoparticle can be achieved either by electrostatic interactions or by covalent conjugation strategies [56,57]. The last requires the presence of reactive functional groups (amine, carboxyl, sulfhydryl, hydroxyl, azide-reactive groups) at the surface of nanoparticle that enable conjugation with ligands [58]. A very popular method used for chemical conjugation is the carbodiimide method, which consists of the activation of carboxylate functional groups that react with primary amines to form amide bonds [58,59]. In this case, a direct conjugation is performed, but sometimes linkers are used. For instance, Kim and collaborators used *N,N* -dicyclohexyl carbodiimide (DCC) for a two-step coupling reaction of a galactose moiety to polymeric nanoparticles composed of cholic acid and diamine-terminated poly(ethylene glycol) as a linker [60]. Palmioli and co-workers also described the functionalization of PLGA with sugar entities bearing a 2-(2 aminoethoxy)ethanol linker through amide bond using *N,N* -diisopropylcarbodiimide and NHS [61].

Crucho and colleagues produced a polymeric conjugate composed of PLGA modified with sucrose and cholic acid moieties [62]. The functionalization of the PLGA backbone was made through esterification using DCC/NHS reactions, and then sucrose and cholic acid-functionalized PLGA nanoparticles were obtained by nanoprecipitation. Sucrose addition provided colloidal stability to the nanoparticles, demonstrated by the decrease of the negative surface charge. Rieger and collaborators reported a simple method for the preparation of mannose-functionalized PLA NPs [63]. The synthesis approach consisted in the co-nanoprecipitation evaporation of a mannosylated poly(ethylene oxide)-*b*-poly( caprolactone) (PEO-*b*-PCL) diblock copolymer with PLA. The amphiphilic copolymers bearing the mannose moieties worked as surface modifiers and were able to specifically bind to MR.

Freichels and co-workers prepared crosslinked hydroxyethyl starch (HES) nanocapsules, which is a hydroxyethylated glucose polymer, functionalized with (oligo)mannose [64]. The preparation of the nanocapsules consisted in interfacial addition of HES with 2,4 toluene diisocyanate (TDI) in inverse miniemulsion. This procedure leaves an amount of non-reacted amine groups that were used to perform the functionalization with three types of mannose molecules: a-D-mannopyranosylphenyl isothiocyanate, 3-O-(a-D-mannopyranosyl) -D-mannose (di-mannose), and α3,α6-mannotriose (tri-mannose). The amine groups on the surface of nanocapsules were used to react directly with mannose isothiocyanate while di- and tri-mannose were coupled through reductive amination. The obtained delivery systems exhibit a specific binding to agglutinin and the presence of a PEG linker showed to increase the interaction to the receptor, due to a higher accessibility of the sugar molecule.

Kim et al. developed a siRNA delivery system composed of PEI, PEG, and mannose [65]. PEI molecules were used to form the polymer/siRNA polyplex, PEG was used as a stabilizer, and mannose as a targeting ligand for macrophages. Here, two different functionalization methods were performed: one in which PEG and mannose molecules were directly linked to PEI backbone (mannose-PEI-PEG), and another in which mannose chains were conjugated to PEI using a PEG spacer, i.e., mannose was linked to PEG before reaction of mannose-PEG chains to PEI backbone. In these reactions, like the ones described before, α-D-mannopyranosylphenyl isothiocyanate was used for mannosylation and PEG was conjugated to PEI via glutaraldehyde linkage. The researchers also found that the location in which mannose ligands are conjugated affect the cytotoxicity of nanocarriers. Table 1 resumes examples of glycoproteins produced with electrostatic interactions and covalent conjugation strategies identifying the ligand and the target defined for the nanocarriers.


**Table 1.** List of developed carbohydrate-functionalized polymeric nanoparticles.


**Table 1.** *Cont.*

#### **5. Macrophages**

#### *5.1. Functions and Polarization State*

The mononuclear phagocytic system, also designated as the reticuloendothelial system, is composed of monocytes in the blood and macrophages in the tissues and is part of the innate immune system. During the hematopoiesis process, mature monocytes circulate for about 8 h, grow, and end up in specific tissues, as macrophages [68].

Macrophages are present throughout the body resident in tissues and also motile, known as free or wandering macrophages. They can originate from circulating monocytes, but also from embryonic hematopoietic stem cells or yolk sac [69]. Macrophages play relevant roles in the immune response, as they act in tissue development, inflammation related to pathogens, cancer, and organ transplantation. During phagocytosis, macrophages engulf pathogens, mediated by receptors on macrophage surface that bind to the fragment crystallizable (Fc) region of molecule from the pathogen. This process leads to the formation of a phagosome that merges with the lysosome where the target is digested. Macrophages act as antigen presenting cells, when displaying foreign material or parts of antigens on its surface in association with class II major histocompatibility complex (MHC) molecules. This triggers T-cells, and consequently, the adaptive immunity. Likewise, macrophages can secrete several cytokines involved in the immune response, homeostasis, and inflammation, which modulate their function and surface marker expression [70].

Macrophages are polarized to respond to alterations in their environment, being classified as M1 macrophages and M2 macrophages [71]. Contact with pathogen-associated molecular patterns (PAMPs), such as bacterial lipopolysaccharide (LPS) from *Escherichia coli* (Gram-negative) or peptidoglycan (PGN) from *Staphylococcus aureus* (Gram-positive) drives macrophages polarization towards M1 phenotype, with the ability to elicit proinflammatory response and production of interleukin (IL) 6 (IL-6), IL-12, and tumor necrosis factoralpha (TNF-α), all pro-inflammatory cytokines. Alternatively, activated macrophages are produced in the presence of the Th2 cytokines IL-4 and/or IL-13, which can lead macrophage polarization to M2, characterized by anti-inflammatory responses and tissue repair abilities [72].

Regulation of macrophage polarization phenotype is reversible and modulates their immune function. An important feature in this mechanism is the expression of the cell surface markers. M1 macrophages overexpress CD80, CD86, and CD16/32, while M2 exhibits more arginase-1 and mannose receptor (CD206).

#### *5.2. Macrophage Polarization Mediated by Nanocarriers*

To date, several nanocarriers were able to induce inflammatory and immune responses in vitro and in vivo [73–75]. Nanocarriers can be internalized by macrophages inducing changes at the cell surface as well as secretion of cytokines and chemokines [76]. Understanding the mechanism of interaction between nanocarriers and macrophages will contribute to an effective design of nanocarriers for a specific therapeutic strategy. Macrophagemediated therapies are emerging as a promising and effective approach towards the treatment of several diseases. In particular, uptake of nanocarriers by macrophages implies interaction between nanocarriers' surface and macrophage cell membrane. Therefore, the formed membrane-bound vesicle will have a size, composition, and internal environment according to the internalization, resulting in endosomes, phagosomes, or macropinosomes. In fact, the uptake mechanisms can be described as phagocytosis, micropinocytosis, endocytosis mediated by clathrin or by caveolin, or independent from both [77]. Passive and active targeting approaches can be designed to achieve the intended effect. Size and surface of the nanocarrier govern passive targeting, while for an active targeting the surface of the nanocarrier requires functionalization with a specific ligand towards a particular surface cell receptor. Carbohydrate-coated nanocarriers have been exploited to target mannose receptors expressed in macrophages and dendritic cells (antigen presenting cells, APCs) [51].

Conjugation of ligands at the surface of nanocarriers may modulate the immune system. The use of targeted nanocarriers elicit the maturation of APCs, with alterations at the surface expression of co-stimulatory molecules and in the secretion of cytokines that activate T-cell responses [78–80]. Active targeting of nanocarriers towards endocytic receptors present on macrophage surface can be achieved using C-type lectin receptors (CLR) or the mannose receptor CD206.

#### *5.3. Mannose Receptor*

CD206 or mannose receptor (MR) has the ability to recognize mannosylated or fucosylated glycoproteins and engulf them [81]. This 175 kDa endocytic receptor was first identified in rabbit alveolar macrophages and is a type I transmembrane receptor composed by an extracellular region containing a cysteine-rich (CR) domain that acts as second lectin domain, and a fibronectin type II (FNII) domain that is involved in collagen binding, and multiple C-type lectin-like domains (CTLDs) within a single polypeptide backbone where the binding of sugars terminated in D-mannose, L-fucose, or N-acetyl glucosamine occurs [82]. Based on their structure, CLR are grouped as transmembrane CLRs and soluble CLRs (collectins). Type I transmembrane CLRs include MR and ENDO180 (mannose receptor C type 2), while type II transmembrane CLRs include dendritic cell-specific intracellular adhesin molecule 3 grabbing non-integrin (DC-SIGN), langerin, and macrophage galactose type lectin (MGL) receptors [83].

MR expression is not restricted to resident macrophages and dendritic cells. It was also found on immature monocyte-derived dendritic cells [84], hepatic endothelial cells [85], tracheal smooth muscle cells [86], and kidney mesangial cells, among others [84]. Expression of this receptor is modulated by cytokines, immunoglobulin receptors, and pathogens [87]. MR synthesis is more rapid in the presence of immunoglobulins IgG2a and IgG2b [88]. Cytokines regulate MR expression as IL-4 [89], IL-13 [90], and IL-10 [91] enhance macrophages receptors expression, while interferon-G (IFN-G) [92] down-regulate MR expression and increase macrophage's activation.

Macrophages cell surface express about 10–30% of MR at steady state and the remaining 70–90% have an intracellular location. Early endosomes contain MR internalized and are able to send these receptors back to the cell surface through the interaction with the clathrin-mediated endocytic machinery [80]. This mechanism is mediated by small intracellular vesicles (below 0.2 μm) and drive the MR to be recycled to the macrophage membrane or delivered into late endosomes, filled with lysosomal hydrolases. Here, under acidic pH and hydrolase-rich environment, the final degradation of the internalized cargo

happens. The ability of nanoparticles to modulate the macrophage state through MR was described for several authors (Table 1). For example, chitin and mannose-coated beads improved the production of tumor necrosis factor-alfa (TNF-α), IFN-G, and IL-12 by murine spleen cells in relation to non-coated beads [93].

MR is also expressed in DCs and actively contributes to antigen recognition and processing. Evidences confer MR an important part in the antigen-internalization mechanism in DCs. For example, bovine serum albumin coated with mannose enhanced the uptake and presentation of this antigen to T cells [94,95].

Macrophages are responsible for the internalization and degradation of pathogens, acting as pattern recognition receptors, given the highly conserved C-type lectin receptors, in a calcium dependent manner. Thus, this first line of defense binds to carbohydrate molecules (e.g., mannose, fucose, and *N*-acetyl glucosamine) present on the surface of a wide variety of pathogens, including *Candida albicans* [81], *Leishmania donovani* [96], and *Mycobacterium tuberculosis* [97].

#### *5.4. Mannose Receptor-Targeted Nanocarriers Interactions with Macrophages*

Targeting MR in macrophages using polysaccharides or glycoproteins containing mannose or fucose residues has been exploited to develop nanocarrier-based macrophagemediated therapies [98]. Mannose-based glycopolymers exhibited an increased internalization by macrophages in comparison to galactose-containing glycoprolymers [99]. Given the variability of ligand–target interaction according to the activation and differentiation state of macrophages, studies should consider various types of carbohydrate moieties. The design of the nanocarriers should also pay attention to their surface charge, as it affects macrophage binding affinity. Anionic sialic acid is present on macrophages surface and enhances phagocytosis of positively charged nanocarriers [100]. Nanocarriers coated with albumin, folic acid, or cholesterol are easily internalized by caveolin-mediated endocytosis which prevents lysosomal degradation. However, mechanisms of uptake are interchangeable and blocking a path may "open" another endocytic path, which poses a challenge in the design of a nanocarrier (Figure 4) [5,101].

**Figure 4.** Glyconanoparticle interaction with macrophages through receptor mediated endocytosis mechanism.

#### 5.4.1. Mannose Receptor-Targeting Nanocarriers towards Infection Resolution

Macrophages are host cells of many intra-cellular pathogens (bacteria, parasites, and virus) causing infectious diseases that could be managed with nanocarriers targeting MR. Recent examples of carbohydrate-based polymeric nanocarriers towards macrophages are described and shown in Table 2.

Tuberculosis is the bacterial infection responsible for more deaths worldwide. The treatment regimen involves oral administration of rifampicin, isoniazid, pyrazinamide, and ethambutol for long periods, usually over six months. The completion rate is highly dependent of patient compliance, but interruptions may occur due to adverse side-effects. Hence, new therapeutic approaches which are more efficient, with less side-effects and shorter duration of treatment are envisaged [102]. Aminoglycoside antibiotics are used against mycobacterial infections, but usually are not highly membrane permeable eliciting adverse side effects. Chitosan nanoparticles loaded with aminoglycoside were produced with dextran sulphate as counter ion to shield the positive charge of the antibiotic. In vivo results showed effective killing of intracellular *M. tuberculosis* upon oral administration of antibiotic-loaded nanocarriers [103]. Isoniazid, an anti-tuberculostatic agent, was incorporated in mannosylated gelatin nanoparticles. Macrophages were effectively targeted by these nanoparticles, as assessed by flow cytometry [104]. For rifampicin, several examples of nanocarriers have been described. Rifampicin was loaded in dendrimers able to enhance alveolar macrophage uptake and drug release at pH 5 [105], and also in flower-like polymeric micelles which surface was modified with hydrolyzed galactomannan [106]. The latter combined mannose and galactose were both recognized by CLRs. A complex nanocarrier based on poly(epsilon-caprolactone)-*b*-poly(ethylene-glycol)-*b*-poly(epsilon-caprolactone) flower-like polymeric micelles (PMs) coated with chitosan or GalM-h/chitosan was produced allowing higher intracellular levels of rifampicin in murine macrophages, relative to its free and chitosan-loaded forms.

The protozoa Leishmania is the causative agent of several infectious diseases upon invading macrophages in the liver and spleen (visceral leishmaniasis) or in the skin (cutaneous leishmaniasis). Leishmaniasis remains endemic in developing countries and without proper treatment leads to death. Pentavalent antimonials were the first anti-leishmanial agents used, but given their toxicity, treatment evolved to amphotericin B, miltefosine, pentamidine, primaquine, paromomycin, and even natural compounds (e.g., amarogentin and andrographolide) [107]. Treatment is hampered by the intracellular localization of the protozoa inside the phagolysosome. The US FDA-approved poly(d,l-lactide-coglycolide) (PLGA) polymer was functionalized with carbohydrate moieties (mannose, mannan, and mannosamine) to identity the most effective in targeting macrophages infected with *Leishmania*. In vitro data obtained with murine primary macrophages evidenced the immunemodulatory properties of the nanocarriers, with activation of macrophages and production of pro-inflammatory cytokines, upon clathrin-mediated endocytosis. Amphotericin B-loaded on mannan-functionalized PLGA nanocarriers confirmed in vivo efficacy in relation to Fungizone© alone, in a visceral leishmaniasis model [66]. MR was targeted by coating polyanhydride nanoparticles with carbohydrates (galactose and di-mannose) by Chavez-Santoscoy and co-workers [108]. The designed nanocarriers increased surface expression of markers in alveolar macrophages, enhanced the expression of MR, and promoted production of pro-inflammatory cytokines (IL-1b, IL-6, and TNF-a). Curcuminloaded mannosylated chitosan nanoparticles improved the drug mean residence time within infected macrophages [109]. Effective endocytosis mediated by MR lead to better pharmacokinetic parameters.


**Table 2.** Mannose receptor-targeting nanocarriers towards infection resolution.


**Table 2.** *Cont.*

Targeted mannose-coated gelatin nanoparticles were produced to enhance therapeutic efficacy of didanosine towards human immunodeficiency virus [110]. Higher uptake by alveolar macrophages was observed with the mannose coating, and in vivo biodistribution studies revealed the presence of the nanocarriers in the spleen, lymph nodes, and lungs. Lamiduvine delivery towards HIV was improved with the incorporation in stearate-gchitosan oligosaccharide polymeric micelles. The nanocarrier led to high internalization and low cytotoxicity in viral transfected cells [111]. Another antiretroviral drug, zidovudine, was incorporated within sialic acid and mannose dual-coated poly(propyleneimine) dendrimer [112]. This nanocarrier produced less cell toxicity and hemolysis, most probably related to the zidovudine-sustained release and enhanced internalization by macrophages. In vivo biodistribution revealed targeting to sialo-adhesin and carbohydrate receptors in the lymph nodes.

#### 5.4.2. Mannose Receptor-Targeting Nanocarriers towards Tumor-Associated Macrophages

Macrophages accumulate in the tumor microenvironment, being designated as tumorassociated macrophages (TAM). These represent the major contribution of tumor immune escape, angiogenesis, growth, and metastasis [113]. Mannosylated nanocarriers can modulate macrophage polarization from M2 phenotype to the M1 phenotype enhancing antitumor immunity. Delivery of Toll-like receptor (TLR) agonists reset TAM polarization towards an antitumor M1 phenotype. Rodell and co-workers produced b-cyclodextrin nanoparticles containing a TLR7/8 agonist that reprogrammed TAM and, as a consequence, efficiently controlled tumor growth [114].

MR targeting can also contribute to improve gene delivery efficiency, by improving transfection and tissue specificity. Reeducation of TAM can be accomplished with delivery of siRNA, miRNA, or mRNA using mannosylated nanoparticles [115,116]. Likewise, chitosan nanoparticles allowed to deliver therapeutic DNA by MR-mediated endocytosis [117]. Experimental data highlights less cytotoxicity, improved gene transfection, and induction of IFN-γ production upon IL-12 gene delivery, in comparison to plain chitosan nanocarriers. IL-12-based gene delivery can be applied for cancer immunotherapy, as it elicits a Th1-type immunity and also cell-mediated immunity.

Instead of only modulating TAM polarization to control cancer progression, it is also possible to completely neutralize or kill them, with the delivery of cytotoxic compounds using TAM-targeted nanoparticles [118].

Nanoparticle-based immunotherapies represent a promising approach to target tumor environment, in particular TAM, instead of aiming for the tumor cells, preventing immunemediated adverse-effects. Another application could be cancer vaccination by targeting immune cells in the lymph node.

#### 5.4.3. Mannose Receptor-Targeting Nanocarriers towards Prevention Approaches

Oral delivery is the preferred route for drug/bioactive compounds administration, due to effects both at a local and systemic level, minimal invasiveness, and cost-effectiveness [119, 120]. However, a question of bioavailability and efficacy emerges when these immunomodulatory compounds are orally administered in its free form. This can be attributed to compound degradation due to pH variation and enzymatic activity in the gastrointestinal (GI) tract or poor permeability across intestinal biological membranes [119,121,122]. Delivery systems such as carbohydrate-functionalized polymeric nanoparticles are able to provide protection from degradation in the GI tract, increase absorption by the intestinal epithelium due to its mucoadhesive properties (e.g., PLGA, chitosan, and alginate) and cell or tissue-targeted delivery and sustained release [121,123–126]. Gentamicin (GM) is an antibiotic that can only be administered in parenteral form or in topical formulations, and it cannot be orally administered due to enzymatic degradation and poor bioavailability. However, when GM was encapsulated in chitosan-functionalized PLGA nanoparticles and orally given to healthy rabbits, it not only reached the GI tract, as it was able to cross the membrane entering the blood stream [127]. Based upon these findings the authors concluded that biodegradable chitosan-functionalized PLGA nanoparticles are potential candidates for GM oral delivery. Furthermore, these polysaccharide polymersbased nanoparticles show unique physicochemical properties, namely, biocompatibility, biodegradability, non-toxicity, and low cost [128,129].

Immunomodulators targeting myeloid cells, particularly macrophages, are a proven strategy to improve the host immunological status and immune response. Several studies show that oral immunostimulation with bioactive compounds can be an effective prophylactic strategy to prevent infectious disease or curtail its effects [130–132]. As already mentioned, macrophages perform critical roles in innate immune response, including inflammation and tissue repair, pathogen elimination, and coordination of the adaptive immune response. Cell surface receptors that recognize polysaccharide residues such as mannose, galactose, or *N*-acetylglucosamine residues are paramount for macrophage activation and response. Carriers comprising a matrix of polysaccharide moieties, or surface ligands composed of carbohydrates, are suitable candidates for macrophage targeting or stimulation. A chitosan nanoparticle functionalized with a high molecular weight ulvan polysaccharide, activated Senegalese sole (*Solea senegalensis*) macrophages and triggered a stronger immune response than the ulvan extract free form. Ulvan is a complex polysaccharide composed of glucuronic acid and sulphated rhamnose, known to activate and induce a potent stimulating effect on macrophage oxidative burst [133]. It was hypothesized that ulvan stimulating properties improved in the chitosan/ulvan nanoparticles possibly due to particle endocytic uptake by macrophages [134]. Particle size is an important feature for cell uptake: when comparing microparticles to nanoparticles, the latter is generally having higher cell internalization rates, and thus can be utilized to target cellular and intracellular receptors due to their smaller size and mobility [135]. Furthermore, several studies explored mannose-functionalized nanoparticles recognition by the macrophage MR as a way to stimulate macrophages [67,136].

The potential to use orally delivered carbohydrate-functionalized polymeric nanoparticles to target macrophages is recognized, mostly because of the unique structural features of polysaccharides referred above. As research progresses in the field of nutraceuticals, these glyconanoparticles seem to be a highly suitable delivery system for biologically active compounds targeting macrophages.

#### **6. Future Perspectives**

Further application of carbohydrate-functionalized polymeric nanoparticles depends on more efficient production methods and improved selectivity towards macrophages or other defined targets (Table 3). The design should consider drug release rate to assure rapid release of the cargo at the target site. The amount of loaded cargo is also crucial, since a balance needs to be achieved between high capacity and safety of the total administered dose. Altogether, the product should be scalable and cost-effective to attract investors and industries. However, not all these requirements are currently met. In fact, the production methods are hardly reproducible, as the molecular weight, functional groups, and purity of polymers depends on the source and batch. More knowledge on the mechanism of interaction between glyconanoparticles and targeted macrophages will certainly allow to optimize these parameters and obtain a product for further translation. In fact, the potential of the carbohydrate-functionalized nanoparticles is highlighted by the increasing number of patents found on the World Intellectual Property Organization and recently discussed by Patil and Deshpande [73].

**Table 3.** A resume of the advantages and limitations of mannose receptor-targeting polymeric nanocarriers.


#### **7. Conclusions**

Carbohydrates play a fundamental role in many aspects of receptor-mediated delivery and therapies. The insertion of carbohydrates in biodegradable polymeric nanoparticles enhances their biocompatibility and favors their use for biomedical applications. In this review, we focused on the preparation methods and use of carbohydrate-functionalized polymeric nanoparticles for macrophage targeting. The sugar moieties present in these nanocarriers are able of specifically interacting with receptors at the surface of macrophage cells and trigger immune responses. The study of this interaction makes the development of new macrophage-mediated therapies possible, with the mannose receptor binding being the most exploited, due to its abundant expression in dendritic cells and increased internalization. Mannose-targeting nanocarriers have shown to be effective in increasing the production of pro-inflammatory cytokines, in infection resolution, modulate tumorassociated macrophages' polarization, and improving nutraceuticals oral administration.

**Author Contributions:** Conceptualization, S.A.C.L.; resources, S.A.C.L., B.C., and S.R.; writing original draft preparation, R.G.D.A., B.R., and S.A.C.L.; writing—review and editing, S.A.C.L., B.C., and S.R.; supervision, S.A.C.L. and B.C.; project administration, S.R. funding acquisition, S.A.C.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was partially supported by PT national funds provided by FCT–Foundation for Science and Technology through COMPETE POCI-01-0145-FEDER-030834 and National Funds (FCT) through project PTDC/QUI-COL/30834/2017.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Acknowledgments:** The authors acknowledge the support obtained within the scope of UIDB/04423/2020 and UIDP/04423/2020. SCL, BR, and BC are grateful for the funding from FCT/MEC (CEECIND/01620/2017, PD/BDE/129262/2017, and IF/00197/2015, respectively) financed by national funds. To all financing sources, the authors are greatly indebted.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Encapsulation of the Natural Product Tyrosol in Carbohydrate Nanosystems and Study of Their Binding with ctDNA**

**Antonella Rozaria Nefeli Pontillo 1, Evangelia Konstanteli 1,2 , Maria M. Bairaktari <sup>1</sup> and Anastasia Detsi 1,\***


**Abstract:** Tyrosol, a natural product present in olive oil and white wine, possesses a wide range of bioactivity. The aim of this study was to optimize the preparation of nanosystems encapsulating tyrosol in carbohydrate matrices and the investigation of their ability to bind with DNA. The first encapsulation matrix of choice was chitosan using the ionic gelation method. The second matrix was β-cyclodextrin (βCD) using the kneading method. Coating of the tyrosol-βCD ICs with chitosan resulted in a third nanosystem with very interesting properties. Optimal preparation parameters of each nanosystem were obtained through two three-factor, three-level Box-Behnken experimental designs and statistical analysis of the results. Thereafter, the nanoparticles were evaluated for their physical and thermal characteristics using several techniques (DLS, NMR, FT-IR, DSC, TGA). The study was completed with the investigation of the impact of the encapsulation on the ability of tyrosol to bind to calf thymus DNA. The results revealed that tyrosol and all the studied systems bind to the minor groove of ctDNA. Tyrosol interacts with ctDNA via hydrogen bond formation, as predicted via molecular modeling studies and corroborated by the experiments. The tyrosolchitosan nanosystem does not show any binding to ctDNA whereas the βCD inclusion complex shows analogous interaction with that of free tyrosol.

**Keywords:** tyrosol; nanoparticles; Design of Experiment (DoE); chitosan; β cyclodextrin; DNA binding

#### **1. Introduction**

Tyrosol (2-(4-Hydroxyphenyl)ethanol) is a biophenol that is found in olive oil, white wine, beer and vermouth (Figure 1) [1]. Even though tyrosol does not exhibit strong antioxidant activity, it contributes to the cellular defences due to intracellular accumulation [2,3]. Moreover, numerous studies affirm that tyrosol offers neuroprotective and cardioprotective effect and enhances the regulation of the human LDL levels [4,5]. However, its hydrophilic nature impedes its incorporation in lipid substrates and limits its absorption and bioavailability [6].

**Figure 1.** Structure of tyrosol.

Nanoencapsulation of bioactive compounds and pharmaceutical agents in suitable carriers is a very promising technology, as it offers protection and stabilisation of the

**Citation:** Pontillo, A.R.N.; Konstanteli, E.; Bairaktari, M.M.; Detsi, A. Encapsulation of the Natural Product Tyrosol in Carbohydrate Nanosystems and Study of Their Binding with ctDNA. *Polymers* **2021**, *13*, 87. https://doi.org/10.3390/ polym13010087

Received: 22 November 2020 Accepted: 24 December 2020 Published: 28 December 2020

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

**Copyright:** © 2020 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/).

encapsulated compound. Furthermore, the encapsulation of a compound may lead to a controllable and sustained release, thus enhancing its activity. Therefore, this technology is incorporated in a broad range of applications in different fields, such as in medicinal and pharmaceutical science, cosmetics, agrochemical and food industry [7–10].

β-cyclodextrin (βCD) is a truncated cone-shaped oligosaccharide, with a hydrophobic inner cavity and a hydrophilic outer surface [11]. Small, hydrophobic molecules can be entrapped in the cavity forming an inclusion complex (IC), increasing their solubility, while more hydrophilic compounds can be bound on the external surface [12–14].

Chitosan (CS) is a naturally occurring polymer widely used as a nanocarrier. It is nontoxic, biocompatible and biodegradable and is recognised as Generally Recognised as Safe (GRAS) by the Food and Drug Administration (FDA) [15,16]. The process of encapsulation in chitosan nanoparticles (NPs) has been extensively studied and various techniques have been reported. The nature of the polymer permits the encapsulation of small or larger molecules, natural products like plant extracts and essential oils, or even other nanosystems [17–19].

The properties of the particulate system are defined by the selected carrier and preparation process. Therefore, the ability to design and engineer the experimental process in order to obtain desirable results is an asset for any application. To that end, experimental design and statistical analysis are implemented. Box-Behnken design (BBD) is a Response Surface Methodology (RSM) that enables the multivariate optimisation of a quadratic model [20–22].

Intercalators and groove binders are a class of compounds that interact with the doublestranded DNA. Many anticancer drugs, such as anthracyclines, interact with the DNA through intercalation between adjacent base pairs perpendicularly to the axis of the helix. Many substituents in the intercalator molecule can greatly influence the binding mechanism, the geometry of the ligand–DNA complex and the selectivity of the sequence [23,24].

The interactions between the various cyclodextrins and DNA have yet to be completely identified; however, they are of utmost importance as there are many marketed formulations that contain cyclodextrins. Modified cationic cyclodextrins are known to interact with DNA for gene therapy applications while a strong interaction of a βCD complex was proved to be formed with DNA as the ribose and phosphate groups of the DNA exert a stabilizing effect by forming H-bonds with the outer surface of CD [25,26].

The aim of this study was to develop and optimize the encapsulation process of tyrosol in nanosystems using different matrices namely: chitosan (**TYR/CS**), βCD (**TYR-**β**CD**) as well as in the combined system of βCD/CS (**TYR-**β**CD/CS**). The kneading method was used for the preparation of the inclusion complex of tyrosol with βCD (**TYR-**β**CD**) and ionic gelation for the synthesis of the chitosan nanoparticles. The process optimisation was performed in both cases using a three-factor three-level BBD. The independent variables were set as the initial concentration of the polymer, the loading capacity of **TYR** or the **TYR-**β**CD** inclusion complex and the amount of the cross-linking agent. The examined range was elicited from literature data and data from preliminary experiments.

Complete characterisation of the systems was performed by various methods and techniques, such as Nuclear Magnetic Resonance Spectroscopy (NMR), Infrared Spectroscopy (FT-IR), antioxidant activity determination by the DPPH method, Dynamic Light Scattering (DLS) for the measurement of size, polydispersity index and ζ-potential, Thermogravimetric Analysis (TGA) and Scanning Electron Microscopy (SEM). Finally, the effect of the encapsulation matrix on the ability of tyrosol to interact with Calf-thymus DNA (ctDNA) was investigated.

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

#### *2.1. Materials*

Tyrosol was purchased from Fluorochem (Hadfield, Derbyshire, UK), β-Cyclodextrin in an assay ≥99% (HPLC) was obtained from Fluka (Buchs, Switzerland) and Chitosan (5–20 mPa·s, 0.5% in 0.5% Acetic Acid at 20 ◦C) from TCI (TCI (Shanghai, China). Tween 80, Tris Base, rhodamine B and deoxyribonucleic acid sodium salt, calf thymus were purchased from Alfa Aesar (Ward Hill, MA, USA) and Sodium Tripolyphosphate from Acros Organics (Morris Plains, NJ, USA). For all the experiments double-deionised water was used.

#### *2.2. Synthesis of Nanoparticles*

#### 2.2.1. Preparation of Inclusion Complexes of Tyrosol with βCD (**TYR-**β**CD**)

The kneading method was used for the preparation of the inclusion complex. Briefly, equimolar quantities of βCD (569 mg) and TYR (70 mg) were mixed in an agate mortar and pestle with minimum amount of the solvent (H2O:EtOH 7:3 v/v) for 30 min. The paste was dried to constant weight in a high vacuum pump [27].

#### 2.2.2. Encapsulation of Tyrosol in Chitosan Nanoparticles (**TYR/CS**)

In a 1% aqueous acetic acid solution, 0.05%, 0.2% or 0.35% CS was fully dissolved. Then, an equal to CS amount of emulsifier Tween 80 is added and left to stir at room temperature overnight until complete dissolution. In 5 mL of the occurring solution 2.5, 10 or 17.5 mg of tyrosol was added and the solution was left to stir until total dissolution. Then 1 mL of TPP solution of concentration 0.42, 1.67 or 0.35 mg/mL was added dropwise. The sample was left under magnetic stirring at 700 rpm, for 10 min at 25 ◦C. The nanoparticles were centrifuged at 30,000 rpm for 45 min. The sediment was dispersed and washed with two more consecutive centrifugations. Finally, the nanoparticles were freeze dried and stored in a desiccator.

#### 2.2.3. Coating of the Tyrosol-βCD Inclusion Complexes with Chitosan (**TYR-**β**CD/CS**)

For the synthesis of **TYR-**β**CD/CS** nanoparticles, the procedure described in "Section 2.2.2" was followed, adding **TYR-**β**CD** instead of tyrosol. For preparation of blank chitosan nanoparticles, in 5 mL of 0.2% chitosan, 1 mL of TPP solution 1.67 mg/mL was added and the solution was left to stir for 10 min.

#### *2.3. Design of Experiments (DoE)*

Two experimental designs were conducted to optimize the processes for preparation of **TYR/CS** and **TYR-**β**CD/CS** nanoparticles. Design-Expert® in trial version (Version 12, Stat-Ease, Inc., Minneapolis, MN, USA) was used.

Three factors at three levels were selected to control the size (response R1) and ζpotential (response R2) of the nanosystem. Factor A was the concentration of the chitosan solution, factor B was the mg of TPP and factor **C** the amount of tyrosol for the **TYR/CS** nanosystem or of the inclusion complex for the **TYR-**β**CD/CS** system.

Factors' levels and their normalised values are shown in Table 1. Central Point (0, 0, 0) was repeated three times, and a total of 15 runs were performed for each set.


**Table 1.** Factors, level and responses of the DoE studies.

The data obtained were analysed with Analysis of Variance (ANOVA). Linear, secondorder and quadratic models were evaluated for all responses and in terms of statistical significance, R2 values and the deviation of the predicted to the experimentally obtained results.

The confidence level was set at 95% and *p*-values ≤ 0.05 to determine statistically significant factors.

#### *2.4. Characterisation of the Nanoparticles*

#### 2.4.1. Dynamic Light Scattering (DLS)

The measurements for size, polydispersity index (PDI) and ζ-potential were performed in a Zetasizer Nano ZS, Malvern Instruments Ltd. (Malvern, UK) using a cuvette DTS1070. The results were analysed with the Zetasizer 7.12, Malvern Instruments Ltd. Software.

For **TYR-**β**CD**, 1 mg of the dried sample was dissolved in 4 mL of water and was fully dispersed with 2 min ultrasound at a 2210 Ultrasonic Bath, Branson. DLS measurements of the **TYR/CS and TYR-**β**CD/CS** samples were conducted by diluting 1 mL of freshly made sample in 1 mL of water.

#### 2.4.2. Encapsulation Efficiency (EE%) and Loading Capacity (LC%)

After ultracentrifugation, the quantification of free tyrosol in the supernatant was performed using UV-Vis spectroscopy.

$$EE\% = \frac{\text{Total TYR (mg)} - \text{TYR in supernantant (mg)}}{\text{Total TYR (mg)}} \times 100\tag{1}$$

*LC% = [Total encapsulated (mg)/total nanoparticles weight (mg)]* × *100* (2)

#### 2.4.3. Release Study

The release profile of tyrosol was investigated by determining the quantity of tyrosol released form the nanosystem at given time intervals. For that reason, 50 mg of the each nanosystem were dissolved in a 12 mL solution of aqueous 1% acetic acid (6 mL) and 6 mL DMSO and were stirred at 37 ◦C at 100 rpm. At specific time intervals, 0.5 mL of sample was obtained and filtered through a 0.45 μm pore syringe filter and analysed by UV-Vis spectroscopy. Each time 0.5 mL of fresh solvent was added to the solution

#### 2.4.4. Fourier Transform Infrared Spectroscopy (FTIR)

KBr pellets containing the dried sample were prepared with hydraulic pellet press. The FT-IR spectra were recorded with a JASCO FT/IR-4200 spectrometer (Japan Spectroscopic Company, Tokyo, Japan).

#### 2.4.5. Nuclear Magnetic Resonance Spectroscopy (NMR)

1H NMR spectrum of TYR-βCD IC was obtained to determine the host–guest interactions. The spectra were obtained on a Varian V 600 MHz instrument (National Hellenic Research Foundation, Institute of Chemical Biology, Athens, Greece). The inclusion complex was dissolved in deuterium oxide (D2O).

#### 2.4.6. Thermal Characterisation

Thermal characterisation of the dried samples was performed with Differential Scanning Calorimetry (DSC) with a DSC 1 STARe System device (Mettler Toledo, Columbus, OH, USA) at temperature range from 20 ◦C to 350 ◦C with heating rate 10 ◦C/min, under nitrogen gas flow 20 mL/min and Thermogravimetric Analysis (TGA) the TGA/DSC 1 STARe System Thermobalance (Mettler Toledo, Columbus, OH, USA) at 25 ◦C–600 ◦C, with heating rate 10 ◦C/min, under nitrogen gas flow 10 mL/min.

#### *2.5. Molecular Docking*

The study of the interaction mode and binding affinity docking studies has been performed with the crystal structure of the DNA (PDB ID: 1bna), was obtained from the RSCB protein Data Bank. The optimisation of the docking parameters was performed using AutoDock Vina software (The Scripps Research Institute, La Jolla, CA, USA) and implemented empirical free energy function. Only polar hydrogens were added to the DNA in AutoDock Tools [28]. Finally, the image has been generated using PyMol software. The name and the number of nucleotides were designed according to PyMol software.

#### *2.6. DNA Binding Studies Using UV-Vis Spectroscopy*

Lyophilised Calf-thymus DNA (ctDNA) was dissolved in Tris-HCl buffer solution of concentration 10 mM and pH 7.4, and left overnight at 4 ◦C. Then, 1 mg ctDNA was dissolved in 1 mL buffer and the concentration was determined from the absorbance at 260 nm using an extinction coefficient of 6600 M−<sup>1</sup> cm−<sup>1</sup> [29]. Tyrosol, βCD, CS, TYR-βCD, TYR-βCD/CS, TYR/CS and Rhodamine B were dissolved in the buffer to a concentration of 10 mM for tyrosol or Rhodamine B or 2 mg/mL for the two carriers and all the nanosystems, which were then used as the stock solution for the preparation of the concentration of 100μM. Afterwards, various concentrations (0–100 μM) of ctDNA were added to the prepared solutions which were incubated for 5 min and 30 min at 37 ◦C. Absorption spectra were measured using a JASCO double beam V-770 UV-Vis/NIR spectrophotometer in range of 230–400 nm.

#### **3. Results**

#### *3.1. DoE for the TYR/CS Preparation Process*

The measured responses for the **TYR/CS** nanoparticles are presented in Table 2.

**Table 2.** Experimental data of **TYR/CS** nanosystem and obtained results.


From the table above, it can be observed that the size of the occurring nanoparticles ranged from 115.3 nm at central point to 679.6 nm except for points (−1, 1, 0), (−1, 0, 1) and (−1, 0, −1) in which the particles were over 3 μm and precipitated.

ζ-potential was positive in all cases and ranged from 4.3 to 46.4 mV. As chitosan is a cationic polymer at acidic environments, highly positive ζ-potential is expected, and is indicative of its stability in dispersion. However, the presence of the polyanion TPP manages to reduce the value, by interacting with the protonated amino groups.

Particles in the micro scale were observed only in three points in all of which the concentration of chitosan was at the low level while the concentration of TPP was on its medium or high level. Hence, it can be deduced that when the ratio of chitosan to TPP is low, there is accumulation of the cross-linker in the particles surface, which can also be confirmed by the low ζ-potential. On the other hand, increase in the particles' size is also observed when the ratio of chitosan to TPP is high, as it occurs from points (1, −1, 0), (0, −1, −1) and (0, −1, 1). This could be attributed to an excess of chitosan in the particles surface resulting in insufficient cross-linking of the polymer and low crosslinking density between the polymer and the cross-linking agent. The high ζ-potential of those points confirms the existence of protonated amino groups on the surface of the NPs. Moreover, increased concentration of chitosan leads in decreased intermolecular distance and increased intermolecular hydrogen bonding between the polymeric chains [30–34].

3D surface plots of R1 were designed (Figure 2), and statistical analysis of the results was performed. A Reduced Quadratic Model better described the results (Equation (3)) and was found to be statistically significant (*p*-value 0.0012). Factors A, B, AB, A2 were found to be statistically significant, verifying the observation that the size of the nanoparticles depends both on the amount of chitosan and the interaction with the crosslinking agent TPP. The Model F-value was calculated to be 9.43, indicating a significant model. The coded equation that describes the size response is:

R1= <sup>−</sup>85.74 <sup>−</sup> 2666.41 A + 1450.25 B + 162.06 C <sup>−</sup> 3301.28 AB + 2674.75 A2 + 729.22 B<sup>2</sup> (3)

**Figure 2.** 3D surface plot of response R1 of TYR/CS system (**a**) CS Vs TPP (tyrosol: 10 mg) (**b**) CS Vs TYR (TPP: 1.67 mg) (**c**) TPP Vs TYR (CS: 0.2%).

For the ζ-potential response, the linear model (Equation (4)) best described the relation between the experiment data of R2 and the model F-value was 9.64. The 3D surface plots of R2 are presented in Figure 3.

**Figure 3.** 3D surface plot of response R2 of TYR/CS system (**a**) CS Vs TPP (tyrosol: 10 mg) (**b**) CS Vs TYR (TPP: 1.67 mg) (**c**) TPP Vs TYR (CS: 0.2%).

The coded equation that describes the ζ-potential-response is:

$$R\_2 = 30.64 + 14.25 \text{ A} - 5.29 \text{ B} + 0.74 \text{ C} \tag{4}$$

Table 3 summarises the significance of each factor for the responses R1 and R2 of the TYR/CS nanosystem.

Optimal preparation conditions for TYR/CS nanoparticles are found to be close to the Central Point (CP) values of factors B and C and in the range 0.2–0.35% for the CS concentration.

Those results are in accordance with literature. Shah et al. [31], prepared CS nanoparticles loaded with quetiapine fumarate sized between 140 and 487 nm. The optimal preparation conditions were CS concentration 0.1% and CS:TPP ratio 4.8:1, with stirring time 15 min at 700 rpm and resulted in nanoparticles of size 131.08 nm with ζ-potential 34.4 mV. Delan et al. [30] used BBD for the optimisation of the synthesis of chitosan nanoparticles loaded with the anionic and lipophilic drug simvastatin. It was found that the best CS concentration was 0.34% and the CS:TPP ratio was 3:1, leading to nanoparticles of size

106 nm and ζ-potential 43.3 mv. Sharma et al. [34] ran a four-factor, three-level BBD to assess the process of synthesis of Carvedilol loaded CS nanoparticles. In their research, optimum CS concentration was found to be 0.262% and the nanoparticle size was measured with TEM 102.12 nm.


**Table 3.** Significance of each factor equation model terms of the TYR/CS system.

#### *3.2. DoE for TYR-βCD/CS*

The aqueous dispersion of the inclusion complex of tyrosol with βCD is formed by nanoparticles of size 478.1 nm with a negative, almost neutral ζ-potential of −7.18 mV. The entrapment of the inclusion complex into the chains of chitosan reversed the ζ-potential, due to the presence of chitosan in the outer layer of the inclusion complexes. Furthermore, in most of the 15 runs of the experimental design, the obtained particles were significantly smaller than the inclusion complex. This could be attributed to the strong electrostatic interaction between the oppositely charged carriers, which could separate agglomerations.

From the data in Table 4, it can be observed that the measured sizes ranged from 132.6 nm to over 3 μm, which precipitate, while the ζ-potential ranged from 4.8 to 46.5 mV.

The lowest result for R1 is at the point (0, +1, −1), but at the CP, the size is also very close (average is 190.5 nm). In this design, it can also be observed that the chitosan-to-TPP ratio has a strong impact on the particles' size and ζ-potential. Therefore, for points (−1, 1, 0), (−1, 0, −1) and (−1, 0, 1), the particles precipitate, and the ζ-potential is very low. Moreover, from the 3D surface plot of the response R1 (Figure 4), the size tends to decrease when the concentration of CS increases.

According to the mathematical analysis, the Reduced Quadratic Model was the most suitable for describing this response (F = 12.08). The coded equation (Equation (5)) that describes the system is:

R1 <sup>=</sup> <sup>−</sup>185.97 <sup>−</sup> 3350.14 A + 1093.59 B <sup>−</sup> 226.53 C <sup>−</sup> 2506.92 AB + 3392.22 A<sup>2</sup> + 906.57 C2, (5)

The ζ-potential values range from 4.8 to 46.5 mV for the TYR-βCD/CS system.

Response R2 is best described by the quadratic model, with F-value = 24.02. The coded equation (Equation (6)) of the model is as follows:

R2 = 25.04 + 13.11 A <sup>−</sup> 4.30 B + 2.54 C + 2.18 AB + 1.50 AC <sup>−</sup> 1.03 BC <sup>−</sup> 6.72 A<sup>2</sup> + 7.21 B2 + 5.98 C2, (6)

The 3D surface plots of R2 are presented in Figure 5.

Table 5 summarises the significance of each factor for the responses R1 and R2 of the **TYR-**β**CD/CS** nanosystem.


**Table 4.** Experimental data of TYR-βCD/CS nanosystem and obtained results.

**Table 5.** Significance of each factor equation model terms of the TYR-βCD/CS system.


For this system, the optimal preparation formula was found to be the CP, giving the smallest nanoparticles and a ζ-potential of 25.9 mV. This point was chosen for further experiments.

**Figure 4.** 3D surface plot of response R1 of TYR-βCD/CS system (**a**) CS Vs TPP (TYR-βCD: 10 mg) (**b**) CS Vs TYR-βCD (TPP: 1.67 mg) (**c**) TPP Vs TYR-βCD (CS: 0.2%).

Comparing the experimental results, the lowest size values were given at the point (0, +1, −1) and the CP. In both cases, the ζ-potential was highly positive; hence, the central point was chosen for comparative reasons. This result is in accordance with the software's predictions of the optimal points.

Comparing the two nanosystems, many similarities can be observed. First, the initial concentration of chitosan in the nanoparticle-forming solution plays a significant role in the properties of the particles. Increased concentration of chitosan results in the agglomeration of particles and a very high ζ-potential, attributed to the presence of many protonated amino groups. On the other hand, chitosan to TPP ratio is also important and strongly affects both responses. Moreover, the range of both responses does not differ significantly for the two systems, suggesting that the polymeric chains form a matrix entrapping the molecule or the inclusion complex and form the nanoparticles.

The main difference between the two systems is the impact of the **TYR-**β**CD** to the ζ-potential. The reason for this difference could be that the inclusion complex is negatively charged and that a strong electrostatic interaction between the oligosaccharide and chitosan exists.

**Figure 5.** 3D surface plot of response R2 of the TYR-βCD/CS system (**a**) CS Vs TPP (TYR-βCD: 10 mg) (**b**) CS Vs TYR-βCD (TPP: 1.67 mg) (**c**) TPP Vs TYR-βCD (CS: 0.2%).

#### *3.3. Encapsulation Efficiency and Loading Capacity Calculation*

After identifying the optimal conditions for obtaining nanoparticles, the encapsulation efficiency (EE%) and loading capacities (LC%) were determined for the three nanosystems. For the inclusion complex of tyrosol with the βCD, it was found that the EE% was 98%.

The high encapsulation efficiency is expected for this system, as the kneading technique was implemented, and no washing of the tyrosol was performed. The structure of βCd is presented in Figure 6.

**Figure 6.** The glucose monomer in βCD.

For the **TYR/CS** nanosystem, the EE% was found to be 46% and the LC 12%, while for the **TYR-**β**CD/CS** nanosystem, the corresponding values were 12 and 4.2%, respectively.

#### *3.4. Structural Identification of TYR-*β*CD Using 1H NMR Spectroscopy*

The analysis of the 1H NMR spectrum of the inclusion complexes with βCD provides important evidence regarding the host–guest interactions. The 1H NMR spectrum of **TYR-**β**CD,** as well as those of tyrosol and βCD, are presented in Figure 7. In Table 6, the chemical shift changes of 1H-NMR signals of the protons of βCD before and after the formation of the **TYR-**β**CD** inclusion complexes are shown.

**Table 6.** Chemical shift changes of 1H-NMR signals of βCD before and after the formation of the TYR-βCD IC.


**Figure 7.** *Cont*.

**Figure 7.** 1H NMR spectrum (600MHz, D2O) of (**a**) **TYR-**β**CD** and expansion of the region 3.4–4.1 ppm, (**b**) βCD, and (**c**) tyrosol.

The peaks of the H-3 and H-5 protons of βCD (which are located inside the cavity) present significant upfield shift (−0.030 and −0.100 ppm, respectively), indicative of strong hydrophobic interactions between tyrosol and βCD inside the cavity. The upfield shift of the H-6 (−0.039 ppm), which lie at the primary face of the cyclodextrin cone, implies that there is also a strong interaction with tyrosol, probably between the aliphatic OH of the tyrosol molecule and the 6-OH of βCD. These results lead us to believe that the tyrosol molecule enters the cyclodextrin cone in such a way that the aromatic ring lies well inside the hydrophobic cavity whereas the aliphatic hydroxyethyl moiety points towards the primary face of the cone and strong hydrogen bonds are formed between the OH

groups. These observations are in accordance with the results of Rescifina et al. [35] and Lopez-Garcia et al. [36], who studied the structure of the inclusion complexes of tyrosol and hydroxytyrosol with βCD.

#### *3.5. FT-IR Analysis of the TYR-β-CD IC*

The analysis of the FT-IR spectra of pure βCD, tyrosol and the inclusion complex can serve as further proof of the formation of the inclusion complex. The most characteristic peaks in the FT-IR spectra of the above-mentioned compounds are shown in Table 7 and the spectra in Figure 8.

**Table 7.** Characteristic FT-IR absorption bands of β-CD, tyrosol and the tyrosol-β-CD inclusion complex.


**Figure 8.** FT-IR spectra of β-CD (red), tyrosol (green) and tyrosol-β-CD inclusion complex (blue).

The FT-IR spectrum of the inclusion complex of tyrosol with βCD (TYR-βCD) shows a broad absorption band at 3376 cm−<sup>1</sup> owed to the OH stretching vibration. This band is shifted compared to the corresponding band at the spectra of pure βCD (3382 cm−1) and tyrosol (3389 cm−1). This shift is indicative of strong interactive forces between the host and guest in the inclusion complex. The band at 1543 cm−<sup>1</sup> present in the FT-IR spectrum of the inclusion complex is attributed to the C=C stretching vibration in aromatic compounds and is shifted by 29 cm−<sup>1</sup> from the value of the same band at the spectrum of pure tyrosol. This large shift is further evidence of the efficient inclusion of tyrosol in the cyclodextrin cone with the aromatic part of the molecule lying inside the cone, as concluded also by the NMR spectra. The absorption band of the OH bending vibration appears at 1423 cm−<sup>1</sup> in the spectrum of the inclusion complex and is shifted from the value of the same absorbance at the spectra of β-CD and tyrosol by 9 cm−<sup>1</sup> and 29 cm<sup>−</sup>1, respectively. Again, these shifts corroborate the strong interaction between tyrosol and β-CD.

#### *3.6. Differential Scanning Calorimetry Analysis (DSC)*

The DSC thermograms of tyrosol, the two carbohydrate carriers and the corresponding nanosystems are presented in Figure 9.

**Figure 9.** DSC thermograms of tyrosol (black), β-cyclodextrin (cyan), chitosan (blue), TYR-βCD inclusion complex (magenta), TYR/CS (red), TYR-βCD/CS (green).

In the DSC thermogram of tyrosol, the sharp endothermic process at 85–113 ◦C with peak at 95 ◦C, corresponds to the melting point of the compound [6]. In the chitosan DSC thermogram, in the studied range, only the loss of water is observed at the temperature range 85–107 ◦C, with a peak at 94 ◦C corresponding to the loss of water. The loss of water from cyclodextrin occurs in the range 98–155 ◦C with an endothermic peak at 131 ◦C [37–39].

The DSC thermogram of the inclusion complex undergoes an endothermic thermal transition from 87 to 133 ◦C with peak observed at 116 ◦C, ascribed to the water loss. The melting of tyrosol cannot be observed in this curve, yet the decrease of the temperature of water loss, compared to the one of the pure βCD, may be evidence of the protection that the carrier offers to the molecule. This is in accordance with what has previously been reported when encapsulating a molecule in βCD [37,39].

In the **TYR/CS** nanosystem, water loss occurs at 60 ◦C. Similarly, in the **TYR-**β**CD/CS** nanosystem, the water loss is observed with an endothermic peak at 61 ◦C. Moreover, in this system a second endothermic process takes place at the temperature range 75–87 ◦C with peak at 79 ◦C and could correspond to the loss of the water bound inside the cavity of the βCD.

The endothermic peak owed to the melting of tyrosol is not present in any of the nanosystems studied, and can be considered as further evidence of the successful encapsulation of the compound in the different matrices.

#### *3.7. Thermogravimetric Analysis (TGA)*

The thermal stability of the samples can be determined using TGA. As can be seen in Figure 10, the degradation temperature (Td) of TYR is 220 ◦C.

**Figure 10.** The TGA thermogram of tyrosol.

The thermal degradation of chitosan (Figure 11) occurs in two stages: water loss occurs at 92 ◦C, resulting in an 8% mass loss; whereas the decomposition temperature is 300 ◦C. The residue of the polymer at 500 ◦C is approximately 44%.

**Figure 11.** TGA thermograms of (**a**) β-cyclodextrin and TYR-βCD and (**b**) chitosan, TYR/CS and TYR-βCD.

For the **TYR/CS** nanosystem, the TGA profile presents three stages of degradation ((Figure 11). First, there is the water loss of at 67 ◦C then there is a mass loss of 5.0% at 187 ◦C that could be ascribed to tyrosol and, finally, the decomposition of chitosan occurs at 270 ◦C, leaving a residue of 52% at 500 ◦C. The slight decrease in the Td of chitosan is attributed to the formation of nanoparticles through anionic gelation and, hence, the synthesis of a new material [40].

In the TG curve of βCD, first there is the loss of water molecules, externally and internally bound, at 103 ◦C, resulting in a decrease of 11% of the total mass [41]. The decomposition of βCD happens at 323.8 ◦C and the mass loss is 72.4%. The degradation profile of **TYR-**β**CD** presents three stages: the water loss at 105 ◦C with 4.7% mass loss, then the decomposition of tyrosol at 266 ◦C, resulting in further mass loss of 9.5%, and finally, at 316 ◦C, the decomposition of the βCD. The decomposition of tyrosol happens at a significantly higher temperature compared to the decomposition of the free molecule, confirming that the formation of the inclusion complex protects it from thermal degradation. The residue of the inclusion complex at 500 ◦C is 11%.

For the double-encapsulation system **TYR-**β**CD/CS**, TGA reveals an improved thermal stability of the system. This system degrades in two stages: at 64 ◦C, the dehydration of the sample happens, losing approximately 14% of the mass, and at the temperature range 161–438 ◦C, there is another mass loss of 37%, which corresponds to the degradation of the nanosystem. The residue of the sample at 500 ◦C is 44%.

Therefore, it seems that tyrosol is better protected in the double encapsulated system than in the chitosan matrix.

#### *3.8. Molecular Docking*

In Figure 12, the binding architecture of tyrosol in the crystal structure of DNA (source: PDB:1bna) is presented, depicting its stabilisation in the binding cavity of minor groove of DNA. The docked complex between tyrosol and DNA is illustrated as cartoon (Figure 12a,c,d) and in the form of spheres (Figure 12b) showing the interaction of tyrosol in the binding cavity of minor groove of DNA. The minor groove is smaller in size than the major groove and has the benefit that it is available for attack from small molecules such as tyrosol. Most of the anticancer and antibiotic drugs that have been reported are small molecules, so the minor groove is important as their main binding site.

The stabilisation of the complex is achieved by the formation of hydrogen bond (Figure 12), polar and hydrophobic interactions. From the five hydrogen bonds between DC-11, DG-10, DG-14 and DG-16 nucleotides, three hydrogen bonds are formed between the aliphatic hydroxyl group of tyrosol and the purines of DG-10 and DG-16 base pairs. One more hydrogen bond is formed between the hydroxyl group and the pentose of DG-11 nucleotide and another hydrogen bond between the phenolic hydroxyl group of tyrosol and the purines of DG-14 base pair.

In addition, Table 8 illustrates the nucleotides, the number of hydrogen bonds, and the binding energy that are formed with tyrosol.

**Table 8.** Binding scores of the docked tyrosol on the active site of DNA.


The above results are in accordance with the results obtained from the DNA-binding studies using UV spectroscopy, which indicated an external interaction between TYR and ctDNA.

**Figure 12.** Binding architecture of tyrosol in the crystal structure of CT DNA (PDB:1bna) depicting its stabilisation in the binding cavity of minor groove of DNA. (**a**) DNA structure and tyrosol are illustrated as cartoons, (**b**) DNA structure and tyrosol are formed as spheres, (**c**) The docking pose from a view above the axis of the helix, (**d**) Nucleotides are rendered in line mode and the yellow dotted lines indicate hydrogen bonds between the docked molecule and the nucleotides of the binding pocket in the minor groove of DNA.

#### *3.9. DNA Binding Studies with ctDNA Using UV Spectroscopy*

The interaction of the compounds and nanosystems with ctDNA was studied by UV spectroscopy to obtain information on the existence of any interaction and to further calculate the DNA-binding constants of the compounds (kb). An interaction between a chemical entity and DNA can disrupt the ctDNA band located at 260–280 nm in the presence of increasing amounts of ctDNA.

In absorption spectroscopy, hyperchromism and hypochromism are significant spectral features to study the changes of the double helical structure of DNA. Due to the strong interactions between a molecule and DNA bases, a change in absorption is observed, showing the proximity of the molecule to the DNA bases. On the basis of the interaction of compounds with DNA, the binding constant kb for ligand–DNA binding was determined in the present work, using the Benesi–Hildebrand plot [42]. For the sake of comparison, the binding of a well-known xanthene dye, namely Rhodamine B, which has been shown to have a nonintercalative ctDNA binding in the DNA minor groove [43,44], was also studied.

It is worth noting that, in the present work, no ctDNA binding was observed for any of the nanosystems at 5 min; thus, the measurements were repeated after 30 min. This phenomenon is attributed to the nature of the nanosystems: at 5 min, no significant amount of tyrosol was released from the matrix of the nanosystem; whereas after 30 min, the released tyrosol was able to bind to ctDNA. On the other hand, the binding of rhodamine B and tyrosol to ctDNA at 5 min and 30 min did not show specific change. The UV-Vis data are summarised in Table 9.


**Table 9.** UV-Vis absorption data for rhodamine B, tyrosol and nanosystems in the absence and presence of ctDNA.

\* Hyperchromicity for complexes formed by compounds and nanosystems and 100 μM of ctDNA in comparison to free ligands.

As shown in Figure 13a, TYR (100 μM, pH = 7.4) showed absorption maxima at 275.8 nm. With incremental addition of ctDNA to the solution of TYR, an increase in the absorption intensity at 275.8 nm was observed with concomitant blue shift of the λmax at 274 nm. This hyperchromism suggests that TYR binds to ctDNA by groove binding mode. The binding constant (k) of TYR was calculated from the ratio of the intercept to the slope and was found to be k = 2.09 ± 0.07 × 104 <sup>M</sup><sup>−</sup>1. The results indicate the binding of TYR in the minor groove of ctDNA, as predicted by the molecular modeling studies.

**Figure 13.** Absorption spectra of (**a**) TYR and (**b**) βCD (100 μM, pH 7.4) in Tris-HCl buffer with increasing concentrations of ctDNA (0–100 μM). Arrows ( ) and ( ) refer to hyperchromic and hypochromic (blue shift) effects, respectively.

CDs are known to interact with nucleic acids. Recently, it was reported that when a complex with βCD is formed, the deoxyribose or ribose and the phosphate groups stabilize the docked complex by hydrogen bonds with the outer rim of the CD molecules [26,45]. In this context, the interaction of β-CD and the **TYR-**β**-CD** inclusion complex with ctDNA was investigated.

As shown in Figure 13b, βCD showed absorption maxima at 261 nm. With incremental addition of ctDNA to the βCD solution, an increase in the absorption intensity at 261 nm was observed with concomitant shift of the λmax at 259.2 nm (blue shift). The binding constant (k) of βCD was calculated from the ratio of the intercept to the slope and was found to be k = 0.78 ± 0.05 × 104 <sup>M</sup>−1. Therefore, it seems that <sup>β</sup>CD binds to ctDNA in a non-intercalative mode, yet much less strongly than tyrosol.

As shown in Figure 14 the complex **TYR-**β**CD** showed absorption maxima at 276.4 nm. With incremental addition of ctDNA to the solution of **TYR-**β**CD**, an increase in the absorption intensity at 276.4 nm was observed with a blue shift at λmax 272.6 nm. The binding constant was found to be k = 2.40 ± 0.16 × 104 <sup>M</sup>−1. The complex **TYR-**β**CD** showed enhanced interaction with ctDNA in comparison to the free tyrosol, which could be explained by the enhanced aqueous solubility of the inclusion complex and by the mode of insertion of tyrosol in the βCD cavity. As previously mentioned, if the aromatic ring lies well inside the hydrophobic cavity, while the aliphatic hydroxyethyl group points towards the primary face of the cone, it means that hydrogen bonds can be formed and stabilize the interaction between the complex and ctDNA.

**Figure 14.** Absorption spectra of (**a**) **TYR-**β**CD** and (**b**) **TYR-**β**CD/CS** (100 μM, pH 7.4) in Tris-HCl buffer with increasing concentrations of ctDNA (0–100 μM). Arrows ( ) and ( ) refer to hyperchromic and hypsochromic (blue shift) effects respectively.

The complex **TYR-**β**CD/CS** showed absorption maxima at 260.2 nm (Figure 14). With incremental addition of ctDNA to the solution of **TYR-**β**CD/CS**, an increase in the absorption intensity at 260.2 nm was observed, with a blue shift of the λmax 258.8 nm. The binding constant was found to be k = 1.30 ± 0.08 × 104 <sup>M</sup><sup>−</sup>1.

Rhodamine B showed a hypochromic behaviour and the binding constant was determined to be 10.92 ± 0.20 × 104 <sup>M</sup>−<sup>1</sup> (Figure 15). This is consistent with the minor groove binding of Rhodamine B in ctDNA, in accordance with the work of Islam et al. [44].

All the tested compounds and nanosystems, with the exception of Rhodamine B, showed hyperchromism and a blue-shift upon increasing DNA concentration, indicating that they all interact with the DNA helix. However, as no significant changes in the spectra could be observed, the results indicate a nonintercalative mode of binding. CS and **TYR/CS** were also studied, but no binding with ctDNA was observed.

**Figure 15.** UV-Vis spectra of Rhodamine B with increasing concentrations of ctDNA (0–100 μM). Arrow refers to hypochromic effect.

#### *3.10. Release Profiles*

The release profile of tyrosol was investigated for all nanosystems prepared, and the results are presented in Figure 16. For all systems, the release study was carried out for 45 h in a solution of pH 3.4.

**Figure 16.** The release profile of tyrosol from **TYR-**β**CD** (blue line), **TYR/CS** (black line) and **TYR-**β**CD/CS** (red line).

All nanosystems showed a "burst release" of tyrosol during the first hour. For the **TYR-**β**CD** system, approximately 77% of the encapsulated tyrosol was released during that time. Thereafter, the release rate significantly slowed down, reaching 93% after 18 h.

On the other hand, the release profile of tyrosol from the **TYR/CS** system was significantly different. During the first hour of the study, 13% of tyrosol was released reaching the maximum release after 23 h (26%).

For the **TYR-**β**CD/CS** system, 37% of the encapsulated tyrosol was released during the first hour, reaching the 99% after 45 h. The slower rate presented by this system as compared to the **TYR-**β**CD** system can be attributed to the existence of chitosan as a coating and provides proof that this "double" encapsulation system can offer a more sustained release than the TYR/βCD by slowing down the initial burst effect observed in the case of the inclusion complex.

#### **4. Discussion**

The comparative study of the optimisation of the encapsulation parameters of tyrosol and the inclusion complex of β-cyclodextrin:tyrosol in chitosan revealed that the concentration of chitosan plays a significant role on the properties of the particles that are formed. Moreover, the ratio between the polymer and the cross-linking agent or the molecule to be encapsulated also affects the responses of the particulate system. In both systems studied, it was observed that the smaller sized particles presented a good ζ-potential, which is indicative of a very stable formulation.

The interaction of tyrosol, the inclusion complex of tyrosol with βCD and the inclusion complex coated with chitosan with ctDNA was studied in an effort to elucidate the potential of the prepared nanosystems to be exploited as pharmacologically active agents. Tyrosol and all the examined nanosystems showed a nonintercalative mode of binding to ctDNA. The complexation of tyrosol with cyclodextrin resulted in slightly better interaction with ctDNA as compared to the interaction exhibited by the **TYR-**β**CD/CS** system. This difference is attributed to the small, yet existent, interaction of the oligosaccharide with ctDNA whereas the polysaccharide chitosan did not interact with ctDNA in the performed experiments. Moreover, from the NMR analysis of the inclusion complex, it was deduced that the aromatic ring of the molecule lies inside the cavity, while the aliphatic hydroxyethyl moiety points towards the primary face of the cone. Therefore, it could be suggested that βCD carries the molecule close to ctDNA and an interaction can be observed. On the other hand, no interaction was observed when tyrosol was encapsulated in chitosan nanoparticles, presumably due to the strong positive charge on the surface of the **TYR/CS** nanosystem which impedes it from approaching DNA.

Finally, the release profile of tyrosol is different for each nanosystem. A burst effect is observed for all systems. The release of tyrosol from the inclusion complex formed with βCD is completed in approximately 20 h hours, but when chitosan coating is added, the release rate is delayed significantly. On the other hand, when tyrosol is encapsulated into chitosan nanoparticles, the release of only the 26% of encapsulated tyrosol occurs during the same time.

#### **5. Conclusions**

In the present study, the impact of two different carriers for the encapsulation of tyrosol is investigated: the oligosaccharide β-cyclodextrin and the polysaccharide chitosan. The effect of coating the tyrosol-βCD inclusion complex with chitosan was also investigated as a potential tool to modify the release profile of the bioactive compound. We were gratified to find that this was true for the systems studied in this work: the coating resulted in a sustained release of tyrosol and slowed down the initial burst effect observed from the inclusion complex.

For the formation of the tyrosol-βCD inclusion complex a well-known methodology was implemented, whereas for the two chitosan-containing systems, the formation of the

nanoparticles was succeeded via the ionic gelation method with sodium tripolyphosphate. The latter processes were optimised using experimental design.

Moreover, the interaction of tyrosol and the corresponding nanosystems with ctDNA was investigated. The results show that tyrosol is a ctDNA groove binder and this was confirmed by molecular modeling studies. The same mode of binding was found for the **TYR/**β**CD** and **TYR/**β**CD/CS** nanosystems.

**Author Contributions:** Conceptualisation, A.D. and A.R.N.P.; Methodology, A.R.N.P., E.K., M.M.B.; Software, A.R.N.P., E.K., M.M.B.; Resources, A.D.; Data Curation, A.R.N.P., E.K., M.M.B.; Writing— Original Draft Preparation, A.D., A.R.N.P., E.K., M.M.B.; Writing—Review & Editing, A.D., A.R.N.P., E.K., M.M.B.; Supervision, A.D.; Project Administration, A.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** A R N Pontillo would like to thank the General Secretariat for Research and Technology (GSRT) and the Hellenic Foundation for Research and Innovation (HFRI) for funding her PhD research. M Bairaktari gratefully acknowledges financial support from State Scholarships Foundation (IKY): This research is co-financed by Greece and the European Union (European Social Fund- ESF) through the Operational Programme "Human Resources Development, Education and Lifelong Learning" in the context of the project "Strengthening Human Resources Research Potential via Doctorate Research" (MIS-5000432), implemented by the State Scholarships Foundation (IKΥ).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data sharing not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


*Article*
