Targeted Delivery of Protein Drugs by Nanocarriers

Recent advances in biotechnology allow the selection and the preparation of novel macromolecular compounds such as peptides, proteins and DNA analogs to be used as drugs (e.g., hormones, monoclonal antibodies, vaccines) for therapeutic purposes. Such compounds show powerful and selective therapeutic activity, but unfortunately they must often be dropped at some development stage, because of their high enzymatic susceptibility, short shelf life or unsuitable efficacy after the administration to the patient, owing to immunogenic reactions or poor bioavailability [ 1 ]. In some cases, from a physicochemical point of view, they cannot reach or enter target cells. Moreover, the drug must cross several biological barriers to reach the site of action, and along its path it can be inactivated or produce undesired side effects. Several approaches have been evaluated to overcome these issues. The first one simply consists of applying the drug to the affected area, since this method minimizes undesired side effects following systemic administration. However, direct application cannot overcome problems connected to the nature of protein drugs, which are hydrophilic and have a large size and an intrinsic instability due to denaturation processes. They also are rapidly filtered by the kidneys or intercepted by the immune system. Drug targeting is a promising tool to solve most of the aforementioned problems. This approach consists of designing a system able to selectively deliver the drug to the area of interest. Transport systems can be designed to control the dispatch of the loaded drug to target areas, increasing its local concentration and bioavailability, while prolonging its retention, half-life and effectiveness. This strategy can avoid diffusion of the drug into normal organs, thus avoiding negative side effects. It is outstanding how favorable this method can be: it can improve pharmacokinetics [ 2 ]; it works independently of the administration method; it minimizes the required amount of drug and hence the cost of the therapy. Over a hundred years ago, Paul Ehrlich was the first one to theorize the use of a “magic bullet” to deliver drugs within the body [ 3 ]. His idea consisted of the use of an entity able to selectively recognize the pathological agent (cells, bacteria or other microorganisms) and to destroy it. The targeting process should assure that the pharmacological effect could only be expressed in the targeted area. Nowadays, these systems have evolved in a structure composed of three major blocks: the pharmacologically active substance, a carrier used to increase the number of active molecules per system (frequently a nanosized carrier) and a targeting moiety able to lead the whole system to the selected site of action. In a branch of his extensive work, Ehrlich identified antibodies as the best targeting moieties, owing to their high affinity and specificity for the relevant antigen. Since then, several different targeting techniques have been investigated, and in parallel many different kinds of carrier have been developed, according to novel information about toxicity, tolerability, biocompatibility and acceptability of properly designed materials by living organisms. This review gives a survey of the different nanocarriers and targeting strategies employed for the specific delivery of pharmaceuticals, with a special focus on peptide and protein based drugs.

The term “nanoparticle” is broadly applied in the description of almost every pharmaceutical carrier or imaging agent system, so further classification is needed for clarity [ 16 ]. One group of nanocarriers includes single-chain polymer–drug conjugates, polymer colloids prepared by techniques such as emulsion polymerization, crosslinked nanogel matrices, dendrimers, and carbon nanotubes. For this group, the carrier is a single synthetic molecule with covalent bonds and a relatively large molar mass. Other types of nanocarriers, often termed nanoparticles, comprise self-assemblies of smaller molecules, which are aggregated through intermolecular forces. Liposomes and polyplexes are the most studied members of this class of particles, but this class of carriers also includes aggregates such as polymersomes and other assemblies of block copolymers, colloidosomal aggregates of latex particles, and protein or peptide assemblies. The dynamic nature of these types of systems depends upon the intermolecular forces in play and the biological conditions. Finally, nanocarriers include also complexes based upon fullerenes, silica, colloidal gold, gold nanoshells, quantum dots, and superparamagnetic particles. The use of a properly designed carrier for the sustained and targeted delivery of pharmaceuticals offers several advantages compared with classic administration: it can increase the amount of drug that reaches the targeted area, improve the transportation mechanism and protect the drug against inactivation, degradation and metabolization phenomena. The main characteristics that the carrier must show are: ■ The ability to encapsulate the drug without deactivating it; ■ The possibility for releasing the drug under proper conditions and according to proper kinetics; ■ A high stability and long circulation time after administration; ■ The capability to actively or passively deliver the drug to a target area. Above all, the use of nanosized carriers offers a way to cross biological barriers that would otherwise forbid the drug to accede to the site of interest, as it often happens in the central nervous system or in the gastrointestinal tract. Nanocarriers have a high surface area to volume ratio, thus providing improved pharmacokinetics and biodistribution of drugs while minimizing toxicity, thanks to specific targeted transport [ 17 ]. Moreover, they can improve the solubility of many drugs and prolong the shelf-life and in vivo stability of peptides, proteins and oligonucleotides [ 18 , 19 ]. In particular, the use of biodegradable materials, which has already been reviewed [ 20 , 21 , 22 ], minimizes the risk for hypersensitivity reactions and ensures good tissue compatibility [ 23 ]. Among the potential nanocarriers, colloidal systems such as liposomes [ 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 ] and nanoparticles [ 29 , 30 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ] have aroused considerable interest and have been extensively reviewed. Complex drug delivery systems are thus a potential alternative to the conventional formulations of proteins, in which the protein is usually either lyophilized, in suspension, or in an aqueous solution. The optimal release pattern may vary between proteins and between indications, and adaptable formulations are therefore required. Some proteins require sustained release, while others require controlled, immediate or pulsed release. Release can be obtained with different particulate drug delivery systems [ 5 ]. Liposomes, solid-lipid nanoparticles, polymeric nanoparticles and virosomes are the most commonly used nanocarriers for protein delivery. In many cases, targeted or untargeted liposomes and nanoparticles are rapidly cleared from the blood stream by the RES; although this event is usually considered a disadvantage, it can lead to the aim of activating macrophages if required by certain therapies. Since macrophages are mostly located in the spleen and liver, their ability to catch particles can be used to selectively deliver substances to these organs. Otherwise, specific chemical modification of the carrier can be performed in order to make the system able to avoid the RES. Surface modification of nanocarriers is commonly performed to give them suitable biological properties, to prolong their life in the blood stream, to limit the uptake by macrophages, and to make them able to target specific organs or tissues. Nanoparticles, such as liposomes, polymeric micelles, lipoplexes and polyplexes have been extensively studied as targeted drug carrier systems over the past three decades. A wide variety of active agents can be incorporated into or complexed with these particles, varying from low molecular weight drug molecules to macromolecules such as proteins and nucleic acids. An important requirement to the systemic intravenous use of this targeted nanomedicine approach is the ability of the nanoparticles to circulate in the bloodstream for a prolonged period of time. To achieve this,

Solid lipid nanoparticles (SLNs) were first described in the nineties [ 93 ]. They are made of solid lipids well tolerated by the body (e.g., glycerides composed of fatty acids, which are commonly used in emulsions for parenteral nutrition, cholesterol [ 94 ], glycerol behenate (Compritol ® 888 ATO) [ 94 ], glyceryl palmitostearate (Precirol ® ATO 5) [ 95 ], glyceryl monostearates (Imwitor ® 900) [ 96 ], tripalmitin [ 97 ] and other triglycerides such as tristearin, trilaurin, hard fats such as Witepsol series, cetyl palmitate, lipid acids such as stearic acid [ 98 ], palmitic acid [ 99 ]), thus minimizing the risk of acute and chronic toxicity [ 100 ]. SLNs are solid at room temperature, thus allowing reduced mobility for incorporated drugs, which is a desirable feature for controlled drug release. Their diameter usually varies between 50 nm and 1 µm, and they can be stabilized using non-toxic surfactants, polymers or both. Large-scale production can be performed in a cost-effective and relatively simple way using hot or cold high-pressure homogenization (HPH) or microemulsion techniques [ 98 ]. Other possible preparation methods, such as emulsification-solvent evaporation [ 101 ], solvent injection [ 102 ], solvent emulsification-diffusion [ 103 , 104 ] and ultrasonication [ 105 ], require the use of organic solvents and do not allow for easy scale up. Among particulate formulations, solid lipid nanoparticles have been successfully explored for drug delivery because they combine the benefits of liquid lipid-based colloidal systems (e.g., emulsions and liposomes) and solid systems [ 106 ]. These products possess excellent tissue biocompatibility, biodegradability, composition flexibility and small size, making them suitable for a variety of applications. Furthermore, they have been found to enhance the drug bioavailability after oral or local administration. On the other hand, solid lipid particle manufacturing techniques are not easily adaptable to protein processing as they operate under high temperature, pressure, and shear stress, which are detrimental to protein stability. To overcome these issues, techniques based on supercritical fluids have been developed to process polymer and lipid materials and produce particulate pharmaceuticals [ 107 ]. These techniques can be properly adapted to produce pharmaceutical grade protein delivery system formulations as they can avoid denaturation and degradation phenomena [ 108 , 109 , 110 ]. Recently, Salmaso et al. described a novel supercritical fluid gas micro-atomization process for the preparation of protein-loaded lipid particles [ 111 ]. They demonstrated that the gas micro-atomization process was suitable for the fabrication of lipid nanoparticles loaded with insulin and recombinant human growth hormone (rh-GH), two proteins of relevant pharmaceutical interest with significantly different physicochemical properties. When using hot [ 112 ] or cold [ 113 ] HPH, the lipid is heated to approximately 5–10 °C above its melting point, then the drug is dissolved in the melt. For the hot homogenization technique, the drug-containing molten lipid is placed into a hot aqueous surfactant solution and stirred to obtain a good dispersion. The pre-emulsion is homogenized using a piston-gap homogenizer and the hot O/W nanoemulsion is then cooled down to room temperature, so that the lipid can crystallize again forming solid lipid nanoparticles. Crystallization can also be initiated at lower temperatures or by lyophilization. Cold homogenization technique is employed in the case of highly temperature-sensitive drugs or hydrophilic drugs. Both hot and cold HPH exclude the use of organic solvents, which could deactivate the drug or produce undesired effects in the body. The HPH equipment can affect the particle characteristics [ 114 ]. To prepare SLNs by the microemulsion technique [ 115 ], a mixture of water, surfactant (e.g., phospholipids) and co-surfactant (e.g., short-chain fatty acids) is heated to the lipid melting temperature and added under gentle stirring to the lipid melt. The compounds must be mixed in the correct ratio to provide a clear stable system for microemulsion formation: nanodrop diameter should be less than 150 nm. The microemulsion is then dispersed in a cold aqueous medium (2–3 °C) under mild mechanical mixing; the precipitated spherical particles have diameters of 70-200 nm. Because of their lipid nature, SLNs are particularly well suited to load synthetic lipophilic drugs. Investigation of drug release kinetics and mechanism performed with etracaine, etomidate and prednisolone model drugs showed how this kind of carrier can be useful in the prolonged release of lipophilic drugs [ 116 ], while hydrophilic drugs would be partially lost during the hot homogenization process because of partitioning between the molten lipid and the water phase. SLNs prepared by hot HPH technique were loaded with tamoxifen, a nonsteroidal antiestrogen used in hormone-pos

Despite the good results obtained using nanocarriers for drug delivery, it is also clear that in some cases even the small size of nanocarriers can limit the efficacy of cell specific receptors that should drive the drug to the desired cell. Since it is not possible to indefinitely reduce the size of the particles, a possible approach to avoid this drawback consists of eliminating the carrier. In this case, the protein drug must be conjugated with the targeting moiety able to selectively deliver the drug to the appropriate tissue [ 191 ]. Strictly speaking this approach is outside the scope of the present review. Nonetheless, protein conjugates cannot be completely ignored if taking into account the huge number of relevant publications. Accordingly, this section will only present a quick overview of their potential application. Protein conjugates can be especially useful to specifically deliver proteins, enzymes, small molecules or DNA into mitochondria to treat mitochondrial diseases. Since mitochondria play an important role in apoptosis, the ability to deliver proteins such as superoxide dismutase (to protect mitochondrial DNA and nuclear DNA from reactive oxygen species), the apoptosis-inducing protein (for cancer therapy) and the anti-apoptosis protein (therapy for cardiomyopathy induced from excess apoptosis) into mitochondria can be of great importance. The most important strategies to deliver proteins into mitochondria consist of conjugating the protein to mitochondrial targeting signal peptides (MTS) or to protein transduction domains (PTD) [ 192 ]. MTS are short peptide sequences located at one end of the precursor protein, able to drive the protein into the mitochondrion. After entering the organelle, MTS are cleaved, thus releasing the fused protein and allowing its localization and functionality. In addition to the aforementioned proteins, this method can be used to deliver restriction enzymes able to digest mutant DNA deriving from mitochondrial disease producing specific elimination of the altered DNA [ 193 ]. PTD are specific domains of about 10–16 residues, that are able to cross and deliver cargos through biological membranes. In particular, arginine-rich peptides have the ability to penetrate cell membranes and to bring exogenous proteins into the cells [ 194 ]. A well-known example is the PTD from HIV-1 TAT protein, which consists of 11 amino acids including six arginine and two lysine residues [ 195 ]. Such domains can be fused with peptides, proteins or other species such as liposomes or low-molecular weight compounds, and penetrate barriers (even the blood-brain barrier) to deliver cargo molecules to the desired tissue both in vivo and in vitro [ 196 ]. PTD can reach the cytosol, but is not able to specifically target organelles. PTD and MTS can then be combined to achieve efficient cytoplasmic and mitochondrial protein delivery as shown by Shokolenko et al. , who prepared a fusion protein consisting in exonuclease III protein conjugated to TAT and MTS. The protein was successfully delivered into breast cancer cell mitochondria making the mitochondrial DNA prone to oxidative stress [ 197 ]. Several protein toxins in bacteria are composed of different domains that control different functions. One of the domains is responsible for the selective binding to target receptors; a second domain assures the permeation of the complex into the cytosol via receptor-mediated endocytosis; a third domain is the real toxin and can affect intracellular enzymatic processes leading to cell death [ 198 ]. Once the toxin has been removed or inactivated, the other domains can be exploited to deliver molecules inside cells. Clostridium botulinum neurotoxins are the most dangerous for humans. They target neurons and cause flaccid paralysis, muscle coordination disorders and breathing muscles paralysis, thus leading to death. However, once the active domain of the toxin has been removed, the rest can be used to selectively deliver proteins to neurons [ 199 ]. The specificity of botulinum binding domain for cholinergic nerve terminals can be exploited to improve the gene delivery by viral vectors to motor neurons in gene therapy of amyotrophic lateral sclerosis (ALS). The botulinum-binding domain can be fused with streptavidin thus making it possible to bind any biotinylated viral vector carrying the gene of interest and to deliver it to the motor nerve terminal [ 200 ]. Recently, was also demonstrated that genetic modification can be carried out on Clostridium botulinum toxin delivery domain to make it target non-neuronal cells in order to broaden its therapeutic field of application. The modified toxin was used to target and cleave SNAP23, a non-neuronal SNARE (Soluble N -ethylmaleimide-sensitive factor Attachment protein REceptors) protein that mediates vesicle-plasma membrane fusion processes involved in human hypersecretion diseases, thus proving the feasibility of its application in the treatment of simi

Blood vessels increase their permeability when affected by solid tumors or by inflammatory or infectious processes [ 213 ]. The vessels become leaky, thus allowing particles to cross the wall and permeate the interstitial space. The size of the particles can vary from 10−500 nm, thus including liposomes or micelles. The nature of the disease affects the porosity of the vasculature, allowing for control over diffusion of the drug; the choice of a properly sized carrier would allow the drug to extravasate from the blood vessel ( Figure 8 ). Moreover, tumor cells lack an effective lymphatic drainage system. Both aspects facilitate structures with a size up to approximately 200 nm to accumulate in tumor tissue. Maeda and co-workers named this phenomenon “Enhanced Permeability and Retention (EPR) effect” and widely investigated this method as a targeting solution for cancer, especially when using macromolecular drugs [ 214 , 215 , 216 , 217 , 218 , 219 , 220 ]. Figure 8 Illustration of the EPR effect. Only particles smaller than the cut-off size can permeate the vessel walls and extravasate to the tissues, eventually resulting in enhanced retention and accumulation of particles due to slow lymphatic clearance. To exploit this targeting method, drug carriers should circulate in blood long enough to provide acceptable accumulation of the active molecule in the area of interest. PEG-coated liposomes loaded with highly toxic anticancer drugs such as anthracyclines circulate for a long time in the blood flow, thus exerting a positive effect in cancer therapy with reduced side effects [ 221 ]. The PEG coating inhibits liposome uptake by the reticuloendothelial system and significantly extends liposome residence time in the bloodstream. Polyphosphazenes-platinum (II) conjugates afford high tumor selectivity by EPR effect,allowing for controlled release of the active platinum moiety, avoiding problems due to unfavorable pharmacokinetics and short circulation time of platinum-based anticancer drugs [ 222 ]. Mouse model studies showed how the EPR effect is responsible for the accumulation of PEGylated anti-inflammatory antibody in inflamed joints after intravenous administration [ 223 ].

The use of targeting moieties is an active strategy that relies on specific interactions at target sites; such interactions include antibody-antigen and ligand-receptor interactions. The “magic bullet” initially theorized by Paul Ehrlich has evolved in a three-parts system composed of a therapeutic agent, a carrier for the drug and a targeting moiety combined together. The targeting moiety has to be specific for the area of interest, thus making the distribution of the drug independent from the EPR effect. This targeting strategy provides evident advantages such as high specificity for the injured area and reduced side effects, but it can show its best performances in cancer therapy, especially for the cure of delocalized tumors or tumors in their early stages of development, when the vasculature is still immature. Potential targeting moieties include antibodies and their fragments [ 243 ], aptamers (protein binding DNA) [ 244 , 245 , 246 ], peptides [ 247 , 248 ], proteins such as transferrin [ 249 , 250 ] or lectins [ 251 ], saccharides [ 252 ], hormones [ 253 ], glycoproteins [ 254 ] and vitamins, especially folate [ 255 ]. In spite of the potential benefits of targeted nanocarriers, these systems have some drawbacks such as the cost and stability of the targeting moiety. To justify the increased cost, the moiety must significantly increase the efficacy of the nanovector. Peptides and aptamers are often more stable than carbohydrates or antibodies [ 51 ], they still can undergo proteolysis in vivo ; they may lose their targeting ability or elicit an immune response. Similarly, nuclease activity can degrade aptamers [ 244 ]. Another concern is that the targeting ligand itself could elicit an immunogenic response in a patient, although this issue is more prominent for antibodies [ 256 ]. One further drawback of targeted delivery is the effect the uptake pathway may have on these systems. Targeted liposomes are often taken up and transported to the harsh environment of lysosomes [ 256 ]. Another problem of targeted drug delivery is the depth of penetration into the target tissue. It has been reported that targeted liposomes bind to the first few cell layers after extravasation from the vasculature and retard the entry of following liposomes [ 257 ]. This phenomenon is correlated with the size of the nanovector and binding affinity of the targeting ligand; bigger nanovectors and stronger binding ligands penetrate shorter distances [ 258 ]. Despite many of these still standing challenges, targeted nanocarriers represent a promising strategy for future development of targeted delivery therapeutics. Antibodies are highly selective for the relevant antigen, and this feature can be exploited for precise delivery of drugs to desired tissues. Moreover, it is now possible to create monoclonal antibodies, i.e., antibodies designed to be specific for almost any substance, obtained from a single clone of an immune cell. They can be engineered in several ways in order to meet specific requirements from different biological environments [ 259 ]. Antibodies are proteins composed of IgG, which contains an antigen-binding fragment (Fab, responsible for specific antigen binding) and a complement-fixing fragment (Fc, responsible for fixing complement for in vivo biological response). Recombinant antibody technology allows for the preparation of a library of antibodies from which the ones with the required properties can be selected. Monoclonal antibodies can be used to deliver peptide radiopharmaceuticals through the BBB for imaging brain tumors, especially during their early stages when the BBB is still intact. Imaging is also useful to detect extracellular amyloid in order to monitor other neurological disorders like Alzheimer. The small size of these short radiolabeled peptides provide fast blood clearance and suitable pharmacokinetics, and different peptide sequences, both natural or synthetic, can be labeled with iodine, technetium, indium, gallium, carbon or fluorine [ 260 ]. The 83–14 monoclonal antibody to the human insulin receptor, tagged with streptavidin, was used to deliver the biotinylated 125 I-Aβ 1-40 ( 125 I-labeled 40-residue β-amyloid peptide) to the brain of rhesus monkeys, showing good uptake of the radiolabel from the brain and a 90% clearance after 48 hours [ 261 ]. Cystatin, a protein inhibitor of cysteine proteases with potential antitumoral activity, was incorporated in PLGA nanoparticles [ 262 ], which were further surface-modified with a monoclonal antibody recognizing a specific antigen overexpressed in invasive breast tumor cells (MCF-10A neoT). In vitro tests showed that the immunonanoparticles were able to recognize and target the antigen on MCF-10A neoT cells in a co-culture with other cells. Following endocytosis, cystatin delivered by immunonanoparticles effectively inhibited intracellular cathepsin B only in the target cells. Polymeric nanoparticles displaying tumor necrosis factor on their