Journey to the Center of the Cell: Current Nanocarrier Design Strategies Targeting Biopharmaceuticals to the Cytoplasm and Nucleus

Abstract

New biopharmaceutical molecules, potentially able to provide more personalized and effective treatments, are being identified through the advent of advanced synthetic biology strategies, sophisticated chemical synthesis approaches, and new analytical methods to assess biological potency. However, translation of many of these structures has been significantly limited due to the need for more efficient strategies to deliver macromolecular therapeutics to desirable intracellular sites of action. Engineered nanocarriers that encapsulate peptides, proteins, or nucleic acids are generally internalized into target cells via one of several endocytic pathways. These nanostructures, entrapped within endosomes, must navigate the intracellular milieu to orchestrate delivery to the intended destination, typically the cytoplasm or nucleus. For therapeutics active in the cytoplasm, endosomal escape continues to represent a limiting step to effective treatment, since a majority of nanocarriers trapped within endosomes are ultimately marked for enzymatic degradation in lysosomes. Therapeutics active in the nucleus have the added challenges of reaching and penetrating the nuclear envelope, and nuclear delivery remains a preeminent challenge preventing clinical translation of gene therapy applications. Herein, we review cutting-edge peptide- and polymer-based design strategies with the potential to enable significant improvements in biopharmaceutical efficacy through improved intracellular targeting. These strategies often mimic the activities of pathogens, which have developed innate and highly effective mechanisms to penetrate plasma membranes and enter the nucleus of host cells. Understanding these mechanisms has enabled advances in synthetic peptide and polymer design that may ultimately improve intracellular trafficking and bioavailability, leading to increased access to new classes of biotherapeutics.

Introduction

Recent advances in drug design coupled with the evolution of the human genome project have enabled identification and production of a wide variety of biologics, including nucleic acids, peptides, and proteins. These biological macromolecules hold significant potential for treating various genetic and acquired diseases. However, their large size, biochemical properties, and surface charge pose numerous delivery challenges [1, 2]. Engineered nanocarriers represent a leading class of treatment strategies used to successfully deliver biopharmaceuticals to target cells both safely and effectively [3]. In many cases, these nanocarriers must deliver cargoes to specific target sites within the cell, including the cytosol and the nucleus, to fully realize therapeutic activity (Figure 1). Unfortunately, directing engineered nanocarriers within cells continues to be a major barrier to translation of many biopharmaceutical treatment strategies. This fact is supported by numerous preclinical studies demonstrating that following delivery to target cells, only a small fraction of nanocarriers actually transport the therapeutic payloads to the intended intracellular destination [4, 5]. Furthermore, amongst the several dozen nanotechnology-based therapies on the market (reviewed in [6]), a majority are liposomal drugs and polymer-drug conjugates whose primary benefits are improved bioavailability and/or reduced dosing. The lack of robust strategies to target the intracellular space has precluded widespread clinical application of nucleic acids and many other macromolecules.

Figure 1.

Intracellular trafficking of engineered nanocarriers and their therapeutic cargos following endocytosis.

Endocytic uptake has been established as one of the main mechanisms for nanocarrier internalization by target cells. Carriers are internalized into vesicles known as endosomes via invagination of the plasma membrane. Endosomes transport cargoes to various destinations within the cell, and as part of this process, ultimately fuse with other compartments such as lysosomes, where their contents are subsequently degraded by digestive enzymes [7]. Recent studies have observed that the majority of internalized nanocarriers often become trapped in endocytic vesicles [4, 8, 9] and many are also believed to eventually traffic to lysosomes. The result of these processes is that most nanocarriers are degraded, disassembled, or recycled before reaching the desired intracellular target, which significantly impacts therapeutic efficacy [10]. Therefore, it is essential that engineered nanocarriers be designed to avoid lysosomal degradation while effectively targeting the desired cellular compartments.

Numerous research groups have investigated different methods to overcome lysosomal degradation. Some approaches have attempted to deliver macromolecular therapeutics directly to the cytosol (see reviews [11-14]), avoiding the endocytic pathway altogether. For example, many groups have used physical methods such as microinjection to directly deliver exogenous proteins, cDNA constructs, peptides, drugs and particles into transfection-challenged cells. With precisely controlled dosage and timing, microinjection has been applied in both basic research and clinical practice.[15] Sonoporation and electroporation have also been explored extensively in gene delivery, allowing the transport of nucleic acids directly into the cytoplasm by creating pores in the membrane.[16] Synthetic and naturally-derived membranolytic materials including cell penetrating and amphipathic peptides, membrane-translocating peptides, or polyanions such as poly(β-l-malic acid) modified with hydrophobic groups have also been explored for direct delivery of DNA, siRNA, and oligonucleotides to the cytoplasm.[17-20] Exosomes are a class of cell-derived nanocarriers that have recently received particular interest in nucleic acid delivery applications. Exosomes are immunologically inert if purified from a compatible cell source and possess an intrinsic ability to cross biological barriers. Exosomes are increasingly recognized as major players in cell-to-cell communication, transmitting important information by activation of cell surface receptors and subsequent fusion with the target cell to transfer signaling molecules. Exosomes can also transfer information via endocytic internalization.[21]

The majority of studies have focused on the design of nanocarriers with the capacity to induce endosomal release [22]. Design parameters often take inspiration from nature, such as viruses, bacteria, and other microorganisms, which have evolved efficient systems to escape the endosome. For macromolecular therapeutics active in the nucleus (e.g. plasmid DNA), additional carrier design strategies must be considered in order to effectively penetrate the nuclear envelope. Retrograde trafficking as well as nuclear translocation and retention mechanisms used by viruses, toxins, and native receptors provide useful models [23-26], and approaches to harness these mechanisms in nanocarriers may provide necessary advances in nuclear targeting design.

Engineered nanocarriers can be broadly classified into four categories: peptides, polymers, lipids, and inorganic nanoparticles. Although lipid- and nanoparticle-based systems have shown efficacy and enormous promise for delivery of some classes of drugs ([27, 28] and [29-31], respectively), they have raised significant toxicity concerns in multiple applications [32-34]. Additionally, lipid-based systems, in particular, have been hindered by insufficient control over size, composition, and colloidal stability [3], limiting potential in commercial manufacturing. In contrast, peptide and polymer-based nanocarriers possess scalable syntheses as well as excellent structural tailorability and the ability to directly mimic cell-regulatory structures found in nature. Already, these features provide improved biocompatibility and bioavailability in recent clinical studies with RNA nanocarriers [35, 36], and the tunability in size and composition of such nanoassemblies lends enormous potential for advancement in a variety of other intracellular delivery applications.

Herein, we focus on recent advances in peptide- and polymer-based design used to enhance intracellular delivery of biological macromolecules. In particular, we briefly review the mechanisms governing intracellular trafficking, and we provide a targeted overview of key materials approaches used to achieve successful cytosolic delivery of biopharmaceuticals. In addition, we summarize the primary cellular barriers hindering nuclear delivery, and we critically examine recent design strategies and mechanistic studies providing new insight into the nuclear delivery process.

Endocytic Uptake

Endocytosis is a class of energy-dependent processes used by cells both to absorb nutrients and defend themselves against invading foreign material. Upon arrival at the plasma membrane, engineered nanocarriers can initiate cellular entry via similar mechanisms by interacting with various receptors on the cell surface, or alternatively, by interacting directly with the membrane. Two main classes of endocytosis exist: phagocytosis and pinocytosis. Phagocytosis primarily occurs in specialized cells (i.e. phagocytes), and refers to the process by which cells engulf large particles [10]. Pinocytosis, in contrast, occurs in all cells, and regulates uptake of various fluids and solutes. Most nanomaterials have been shown to initiate cellular entry through pinocytosis. Multiple types of pinocytosis have been established, and pinocytic mechanisms often vary depending on cellular origin and function. Pinocytosis encompasses both non-specific internalization (i.e. macropinocytosis, see reviews [37-39]) as well as receptor-mediated uptake. For nanocarriers in particular, the two most well documented classes of receptor-mediated uptake are clathrin-dependent endocytosis and caveolae-mediated endocytosis. The mechanisms of these two cellular internalization pathways are briefly described below. The authors refer the reader to several recent reviews for more detailed discussions of both cellular uptake mechanisms [39-41], as well as analyses of how these mechanisms are employed by engineered nanocarriers to enter cells [10, 42-44].

Clathrin-Dependent Endocytosis

Clathrin-dependent endocytosis is constitutively active in all types of mammalian cells, and is the best characterized method of endocytic uptake [10, 42]. To describe this mechanism in brief, ligands (e.g. nutrients, growth factors, and sometimes pathogens) bind to cell surface receptors, leading to the clustering of these receptors in regions on the plasma membrane known as coated pits. These coated pits are formed by the assembly of various proteins, including clathrin-1, on the cytosolic face of the plasma membrane. The pits containing the ligand-receptor clusters are engulfed and pinched off of the plasma membrane, forming clathrin-coated vesicles that are typically between 10 and 200 nm depending on cell and cargo type [45]. Once inside the cell, the clathrin coat is shed and recycled back to the plasma membrane, while the resulting vesicles fuse to form early endosomes. At this stage, the endosome lumen becomes slightly acidic (pH 6.0), and the enclosed ligands are shuttled to their appropriate intracellular destinations via transport along the actin and tubulin cytoskeletal networks. Other ligands are recycled back to the plasma membrane and out of the cell via exocytosis. The pH of the lumen continues to acidify (pH ~5.5) as early endosomes fuse with late endosomes, which ultimately transport enclosed cargoes to lysosomes [10, 46]. Once fused with lysosomes (pH ~5.0), any remaining endocytosed material is degraded by the acidic pH and digestive enzymes [47, 48]. Therefore, engineered drug carriers internalized via clathrin-mediated uptake encounter two critical obstacles that hinder their therapeutic efficacy: entrapment within endosomal vesicles and/or degradation in the lysosomal compartment.

Caveolae-Mediated Endocytosis

Caveolae are a subset of cholesterol-rich lipid rafts, active in a wide variety of cells types, and particularly prevalent in muscle cells, endothelial cells, and fibroblasts [40]. Mechanisms of caveolar uptake were first elucidated by following the trafficking patterns of the SV40 virus [49]. Since then, other pathogens have also been shown to enter cells via caveolar internalization, including cholera toxin B and the shiga toxin [39]. Internalization is believed to be dynamin- and actin-dependent [41, 50-52], and occurs via the budding of caveolae from the plasma membrane into vesicles of around 50 nm in diameter. Vesicular budding is augmented by the caveolin family of proteins, particularly caveolin-1 [53]. Caveolar vesicles can fuse with early endosomes and traffic through acidifying pathways leading to lysosomes. Alternatively, caveolae can fuse with neutral vesicular structures known as caveosomes [49, 53]. Caveosomes do not undergo acidification, and have been shown (in some cases) to avoid fusion with lysosomes [54, 55]. Transport through caveosomes allows pathogens to traffic in a retrograde fashion, through the Golgi and/or endoplasmic reticulum (ER), avoiding digestive degradation [56-58]. Such innate ability to bypass lysosomes makes the caveolar uptake pathway an appealing target when delivering biologics, and several types of nanocarriers have been shown to enter cells primarily via caveolae [42, 59, 60]. However, carriers and their associated cargos must still escape these endocytic vesicles in order to effectively reach their therapeutic site of action within the cell.

Targeting the Intracellular Space: Delivering Biopharmaceuticals to the Cytoplasm

Following endocytic uptake, escape from the internalized vesicle is required for nanocarriers transporting biopharmaceuticals that are active in the cytosol. Failing to escape these membrane-bound structures will result in loss of therapeutic efficacy, and in most cases, lead to degradation of the nanocarrier and cargo in the lysosomal compartment. Therefore, it is essential that engineered nanocarriers targeting the cytosol be designed with the capacity to induce endosomal escape once internalized into the cell.

Inspiration for the design of endosomolytic or membranolytic nanocarriers comes from natural pathogens, which have evolved their own strategies to efficiently infect cells. The majority of these microorganisms enter cells via membrane fusion and/or pore formation, gaining direct access to the intracellular space [61, 62]. Enveloped viruses, such as the influenza virus [63] and the herpes simplex virus [64], are able to fuse their viral envelopes with the lipid bilayer in either the plasma membrane or in endomembrane vesicles. Following fusion, these viruses release their protein capsid and viral genome into the cytosol. Non-enveloped viruses, such as the poliovirus [65] and adenovirus [66], are thought to invade the cytosol either by lysing vesicular membranes or by generating escape pores in these vesicles. In addition, bacterial exotoxins, including diphtheria [67] and listeriolysin O [68, 69], rely on endosome acidification to escape the endosome and infect cells. As the pH decreases, protonation of the surface bacterial peptides occurs, causing them to shift in conformation to form amphipathic α-helices. This allows the peptides to interact with the phospholipid membrane, facilitating escape via pore formation, membrane fusion, and/or lysis. The mechanisms of these endosomal escape strategies are detailed below, along with more recently established artificial methods for escaping the endosome following cellular uptake.

Membrane Fusion

This type of endosomal escape involves destabilization of the phospholipid membrane by fusogenic, or fusion, peptides. These peptide sequences are typically, but not exclusively, 20-30 amino acids long, moderately hydrophobic, and often found at the N-terminal domain of proteins near the surface of enveloped viruses [70]. Following endocytosis, these proteins undergo a conformational change during endosome acidification, exposing hydrophobic fusion peptides. At acidic pH, these peptides retain α-helical structure, which allow them to interact with and insert themselves into the endosomal membrane. Consequently, the viral and endosomal membranes fuse, allowing the viral capsid to be translocated into the cytosol.

Pore Formation

Pore formation is used by both non-enveloped viruses and the majority of bacterial exotoxins as an effective strategy for endosomal escape. Pore formation in the endosomal membrane relies on the relationship between membrane tension, which controls pore opening, and line tension, which controls pore closing. This is best illustrated by the barrel-stave model [71], where amphipathic α-helical peptides interact with and form pores in the interior of the membrane. These peptides bind the phospholipid bilayer, increasing membrane tension and creating the initial pore [72]. This action progressively recruits additional α-helical peptides, which have high affinity for the rim of the pore. Rim binding reduces both line tension and the number of peptides causing internal membrane tension, resulting in a stable pore [72, 73]. Within the pores, the hydrophobic groups of the peptides interact with the lipid membrane core, while the hydrophilic groups face each other in the pore lumen [74]. As more peptides are recruited, pore size increases, resulting in the escape of the associated virus/toxin.

Proton Sponge Effect

The proton sponge effect is a hypothesized endosomal escape technique mediated by synthetic cationic nanocarriers that possess a high buffering capacity. It is a widely reported mechanism for the endosomal escape of engineered delivery structures. Cationic nanocarriers that contain a larger number of secondary and tertiary amines with a pK_a between 5 and 7 are able to buffer endosomal acidification by taking up protons [75]. Further pumping of protons into the endosome is accompanied by an influx of chloride ions and water. This causes osmotic swelling and eventual bursting of the endosome, releasing the nanocarrier into the cytosol [75]. Although numerous publications support this buffering technique for inducing endosomal escape [22, 76-80], its validity has been questioned. Several studies demonstrate that amine containing polymers do not induce changes in pH as previously suggested, and argue that the osmotic pressure built during endosome acidification is theoretically insufficient to cause membrane bursting on its own [81].

Design Strategies to Achieve Endosomal Escape

Researchers have employed numerous design strategies to achieve effective endosomal escape and cytoplasmic delivery of engineered nanocarriers. These approaches can be broadly classified into two categories: (1) systems that utilize natural techniques employed by various viruses and bacteria; and (2) artificial systems designed to induce endosome escape in response to physical or biological signals. Table 1 lists some of the most effective design strategies employed to date for achieving successful transport of biologics into the cytoplasm. In the following paragraphs, both the structure and escape mechanism of these nanocarriers are described in detail.

Table 1.

Endosomal escape agents and mechanisms.

Category	Escape Agent	Proposed Mechanism	Reference
Peptides and proteins	Viral
	Hemagglutinin (HA2) domain, influenza virus	Fusion	[76-80]
	diINF-7 peptide analog, influenza virus	Fusion	[81-86]
	glycoprotein 41 (gp41), human immunodeficiency virus (HIV)	Fusion	[93-95]
	trans-activating transcriptional (TAT) domain, HIV	Unknown	[100-109]
	penton base, adenovirus	Unknown	[110]
	vp1 capsid protein, rhinovirus	Pore	[111,112]
	Bacterial Toxins
	Listeriolysin O (LLO), Listeria monocytogenes	Pore	[113-121]
	Diphtheria (DT), Corynebacterium diphtheria	Fusion	[124,125]
	Animal defense toxins
	Melittin, bee venom	Pore	[127-133]
	Penetratin (Antp), Drosophila homeoprotein	Pore	[136,137]
	Transportan (galanin & mastoparan), neuropeptide & wasp venom	Pore	[136,137]
	Synthetic
	GALA	Fusion	[141-144]
	KALA	Fusion	[145-148]
	Oligoarginine	Fusion	[149-151]
Polymer- based	pH-sensitive
	polyethylenimine (PEI)	Proton-sponge	[156,158-164]
	polyamidoamine (PAMAM) dendrimers	Proton-sponge	[157,165-167]
	imidazole, poly(L-histidine)	Proton-sponge	[15,170]
	CALAA-01, cyclodextran system	Proton-sponge	[28,173-177]
	alkyl acrylic acids (PEAA, PPAA, PBAA)	Destabilizing	[178-181]
	methacrylates (PDMAEMA)	Destabilizing	[182-184]
	Dynamic PolyConjugate™ (DPC)	Destabilizing	[185-189]
	Thermo-responsive
	Pluronic F-127	Bursting	[190-193]

Nanocarriers Inspired by Nature

A wide variety of peptides and proteins, derived from different viruses and bacteria, form the primary group of materials used to escape the endosome. Improved understanding of the biological mechanisms that drive endosomal escape has enabled the production of synthetic peptides and polymers capable of recapitulating natural membrane fusion or pore formation techniques to achieve cytosolic delivery.

Virus-Inspired Agents

The fusogenic or pore forming domains of various viruses have been identified and extensively studied for their ability to facilitate the delivery of engineered nanocarriers to the cytoplasm. The best characterized fusogenic peptide is found within HA2 subunit the hemagglutinin (HA) glycoprotein domain on the surface of the influenza virus [63, 82]. At low pH, the N-terminal domain of this subunit is activated, promoting fusion with the lipid bilayer and eliciting membrane destabilization in the endosome. Consequently, several studies have explored the effects of attaching this fusogenic peptide domain to engineered nanocarriers to improve endosomal escape [83-85]. It has been shown that the HA2 peptide can significantly enhance gene transfer when conjugated to lipoplexes [86] and poly-L-lysine (PLL) polyplexes by augmenting endosome escape [87]. Analogs of the natural HA2 fusogenic peptide have also been used to facilitate endosome escape. The glutamic acid enriched diINF-7 peptide analog was shown to improve the cytosolic delivery of both siRNA [88, 89] and proteins [90]. It has also been used to enhance the gene delivery capability of various cationic polymers [91-93].

Glycoprotein 41 (gp41) is a transmembrane protein, and part of the human immunodeficiency virus (HIV) envelope glycoproteins responsible for infecting host cells [94, 95]. The C-terminal peptide domain of gp41 has been shown to have high membrane association [96] and lytic properties [97]. Although the fusion step is not well understood, the C-terminal peptides of gp41 have been shown to adopt amphipathic α-helical conformations at low pH, with the potential to form membrane pores [98, 99]. These peptides have been shown to efficiently deliver oligonucleotides to cells via direct cytosolic delivery [100]. In addition, Pun and co-workers have demonstrated that covalent conjugation of the gp41 membrane lytic peptide domain to polyethylenimine PEI [101] and more recently, cationic block copolymers [102], enhances nucleic acid delivery by facilitating endosome escape.

Cell-penetrating peptides (CPPs), or protein transduction domains, have garnered significant attention over the past two decades as a means to effectively deliver nanocarriers and their associated biomacromolecules, given their ability to penetrate cellular membranes. A wide variety of both synthetic and naturally-derived CPPs have been described [103, 104]. These peptides are typically short (~30 amino acids) and positively charged. One of the first, and best characterized CPPs comes from the HIV-1 trans-activating transcriptional (TAT) protein [105]. The transducing domain is confined to a 48-60 region comprised of basic amino acids (mostly arginine and lysine). The exact mechanism by which this protein transports across lipid membranes is unknown [106], and its ability to escape the endosome when carrying cargo is still debated [103]. However, the TAT CPP has been widely studied for its capacity to mediate the cytoplasmic delivery of both DNA [107-109] and protein [110]. Enhanced macromolecular delivery combined with endosomal escape have been achieved by coupling the TAT CPP to membrane disruptive peptides and polymers, including the influenza HA2 and INF-7 analog peptides [111, 112], histidine and cysteine residues [113], and PEI (both in vitro [114] and in vivo [115, 116]).

The membrane destabilizing peptide domains of various non-enveloped viruses have also been explored for their ability to enhance endosomal escape of engineered nanocarriers. The penton base, a highly hydrophobic protein region of the adenovirus, is known to cause lipid membrane disruption at acidic pH, although the exact mechanism is not fully understood. However, the adenovirus serotype 5 (Ad5) penton protein has been shown to mediate gene delivery when conjugated to poly-lysine [117], and this effect was attributed to its endosome escape functionality. Similarly, the N-terminal peptide domain of the vp1 capsid protein, derived from the non-enveloped human rhinovirus (HRV), was shown to enhance gene transfer via pore formation when added to DNA polyplexes [118, 119].

Bacteria and Animal Defense Toxins

In addition to viral proteins and peptides, the membrane lytic portions of various bacterial exotoxins, as well as animal-derived defense toxins, have been exploited to enhance the translocation of engineered nanocarriers into the cytoplasm. For example, the cholesterol-dependent toxin produced by the Listeria monocytogenes bacterium, listeriolysin O (LLO), is a potent hemolysin able to destabilize lipid membranes by inducing pore formation at acidic pH, without damaging the host cell [68]. Kyung-Dall Lee and coworkers were the first to report on the ability of this protein to enhance cytosolic delivery of engineered nanocarriers [120]. Since then, LLO has been used in a number of investigations to enhance the delivery of biological macromolecules, including lipid-encapsulated proteins [121, 122], as well as lipoplexes [123, 124] and polyplexes for gene delivery [125-127]. More recently, mutants of the LLO protein with enhanced pore-forming activity have been identified, allowing lower concentrations of LLO to be used in macromolecular delivery formulations, thereby enhancing endosomal escape while reducing cytotoxicity [128].

One of the few types of the bacterial exotoxins that utilizes a membrane fusion process instead of pore formation to escape the endosome is the diphtheria toxin (DT) produced by Corynebacterium diphtheria. As the endosome acidifies, the transmembrane (T) domain undergoes a conformational change, rendering it able to insert itself into the endosomal membrane and transfer the catalytic (C) domain into the cytosol [129, 130]. Coupling the T domain to cationic polymers has been shown to significantly enhance endosomal escape and increase gene transfection efficiency [131, 132].

Several animal defense toxins have also been shown to be effective at stimulating endosome escape. However, their use in biologic drug delivery has been limited due to their high cytotoxic effects. One of the most commonly used defense toxins is melittin, a 26 cationic amino acid CPP derived from the venom of honey bees. This peptide is amphipathic and able to destabilize lipid membranes by adopting an α-helical conformation [133]. Early studies using this peptide demonstrated that enhanced gene delivery could be achieved following its conjugation to PEI, due to improved endosomal escape [134, 135]. However, the toxic effect of the peptide promoted the design of melittin-PEI conjugates that only exhibited membrane lytic activity at endosomal pH, thereby increasing intracellular translocation while reducing cytotoxicity [136, 137]. Recent studies have explored other methods of reducing the cytotoxic effects of this peptide, such as reducing melittin concentration [138], masking its activity with an anhydride that is cleaved only at endosomal pH [139], or transforming melittin into a sulfhydryl-polymerized peptide able to condense DNA and stabilize its lytic activity until it is exposed to the reducing endosome environment [140].

Other animal-derived membrane lytic CPPs also have been incorporated into synthetic nanocarriers. Penetratin (also termed Antp) is a basic 16 amino acid peptide derived from the third α-helix of the Antennapedia homeodomain transcription factor of Drosophila. Antp has been shown to efficiently transport across lipid membranes [141]. Another CPP, transportan, is a peptide comprised of 27 basic amino acid residues that is generated by conjugating the N-terminal sequence of the neuropeptide galanin with the pore-forming peptide mastoparan, found in wasp venom [142]. Both penetratin and transportan CPPs have been used extensively to enhance the cytosolic delivery of a wide variety of macromolecules, including peptides, proteins, peptide nucleic acids, oligonucleotides, plasmid DNA, and RNA, both in vitro and in vivo (see reviews [143, 144]).

Synthetic Peptides

With an improved understanding of the chemical and biological mechanisms behind both viral and toxin entry into host cells, researchers have developed a wide variety of synthetic peptides capable of enhancing the endosomal escape of engineered nanocarriers. These synthetic CPPs are typically short sequences (about 30 amino acids or less) that are cationic and/or amphipathic in nature.

Two novel peptides termed GALA and KALA, inspired by the membrane destabilization properties of the influenza HA2 protein domain, have been synthesized and extensively studied for their endosomal escape properties. GALA possesses a single amino acid substitution of glycine for glutamic acid and is amphipathic, undergoing a conformation change to an α-helix at endosomal pH to allow it to destabilize lipid membranes [145, 146]. KALA is similar to GALA, with reduced glutamic acid content and with some of the alanine residues replaced with lysine. KALA is cationic and able to bind and deliver nucleic acids, yet still possesses the ability to retain α-helical conformation at acidic pH and thereby maintain lytic activity [147]. Both peptides have been applied to a wide variety of macromolecular delivery applications. The fusogenic properties of the GALA peptide have been shown to enhance drug and gene delivery by facilitating endosomal escape when co-delivered with liposomal formulations [148, 149] or when anchored to the liposome surface [150, 151]. Conjugation of the cationic KALA peptide to various polymers including PEI [152, 153], polylysine [154], and other cationic peptides [155] also enhances gene delivery via membrane destabilization.

Synthetic oligoarginine CPPs are another promising class of nanocarriers for delivering various types of biological macromolecules due to their ability to efficiently interact with cellular membranes and stimulate robust cellular uptake. Studies have focused on using these peptides alone [156] or coupled to various membrane disruptive reagents, such as the influenza HA2 subunit [157] and its INF7 analog [158], for enhanced siRNA and protein delivery.

The design of various synthetic and naturally derived peptides has significantly improved the cytosolic delivery of biological macromolecules. Their small size, simplicity, and ease of preparation impart substantial versatility, allowing them to be easily conjugated to different peptides, polymers, or lipids. Conjugation is achieved via a variety of covalent (e.g. disulfide or amide bonds) or non-covalent (e.g. electrostatic) interactions with the biologic drug [159] and/or nanocarrier being delivered. Such flexibility, coupled with their ability to penetrate and cross lipid-bilayers, indicate that these peptides possess great therapeutic potential. Improved cellular targeting and extracellular stability is crucial to their in vivo success. This has been explored by conjugating these peptides to multifunctional nanocarrier delivery systems displaying multiple targeting and stealth (e.g. polyethylene glycol) ligands. However, care must be taken when designing these elegant, yet sometimes complicated, delivery systems. In particular, it should be noted that there are size- and structure-based limitations governing the capacity to exploit some naturally-occurring endosomal escape mechanisms; hence, there is a need for control over the physiochemical properties of the nanocarrier to ensure efficient use of any given escape mechanism. Additionally, nanocarrier design cannot be so complex that it inhibits manufacturing scalability and/or translatability. Finally, it is important to consider cell-specificity when designing a carrier system, since available uptake and trafficking pathways will vary depending on cell phenotype. This may affect the types of endosome escape mechanisms available to engineered nanocarriers. Taking these factors into consideration will aid in the design of novel peptide-guided nanocarrier systems that can deliver biomacromolecules with a potency equivalent to their viral counterparts.

Polymer-Based Nanocarriers

Recent advances in polymer system design have led to the development of new types of engineered nanocarriers capable of both encapsulating biologic drugs and responding to various physical and biological signals to escape the endosome. Research in this area has been motivated by the innate immunogenicity of many naturally-derived peptides. Stimuli-responsive biomaterials have been extensively applied to targeted chemotherapeutics and other types of small molecule delivery (see reviews [160-162]). Herein, we discuss their application to the cytosolic delivery of biological macromolecules via endosome destabilization.

pH-Sensitive Polymers

Multiple studies have focused on the design of polymer and lipid formulations that exhibit pH-induced chemical shifts or structural changes able to trigger endosomal escape. Some of the first polymers used for this purpose were polyethylenimine (PEI) [163] and polyamidoamine (PAMAM) dendrimers [164], which have both been extensively explored for DNA delivery. The protonable amines present in these polymers give them an innate buffering capacity, allowing them to escape acidifying endosomes via the hypothesized proton-sponge effect described above. PEI polymers have been examined in numerous applications in vitro and in vivo, in a wide variety of tissues and cell-types [22, 165-167]. PEI is also being studied in clinical trials for its use in localized cancer gene therapy [2]. Due to its inherent cytotoxicity, various modifications to the PEI polymer have been investigated. Notable efforts include polyethylene glycol (PEG)-PEI block co-polymers [168], diacrylate and disulfide cross-linked PEIs [169, 170], and alkylated PEIs [171], with each modification being designed to enhance delivery while reducing cytotoxicity. PAMAM dendrimers have also been shown to be effective at stimulating the delivery of a variety of nucleic acids following endosome escape [172-174].

Despite the widespread use of these commercially available polymers, their endosome buffering capacities and delivery properties are still sub-optimal when compared to naturally-derived agents. This has prompted the design of many types of polymers specifically intended to overcome certain intracellular barriers to effective biopharmaceutical delivery, with a primary focus on enhancing endosome escape. Initial efforts focused on designing polymers that could more effectively take advantage of the proton-sponge mechanism. Imidazole, which possesses a pK_a ~6, is found in natural biomolecules such as the amino acid histidine [175]. Imidazole-containing polymers have been have been extensively studied for their ability to enhance the delivery of nucleic acids [22]. Most notably, the conjugation of imidazole groups to cyclodextrin-containing polymers (CDPs) became the first targeted siRNA delivery system to enter clinical trials [35].

The CDP delivery system was introduced by Davis and co-workers in 1999 to deliver plasmid DNA [176]. The CDP is a short polycation, containing CD groups separated by amidine charge centers capable of binding and condensing nucleic acids into small polyplex structures ~100 nm in diameter. The addition of imidazole groups to the terminal ends of the polymers were shown to significantly enhance nucleic acid delivery in part by enhancing endosomal buffering and escape [177]. Pun and Davis further modified the surface of the polyplexes with adamantane-PEG (AD-PEG) conjugates, which are able to form inclusion complexes with the CD [178]. The AD-PEG conjugates could also be modified with targeting ligands, such as galactose [178] and transferrin (AD-PEG-Tf) [179], to help facilitate cellular uptake. Altogether, these components were shown to efficiently self-assemble with nucleic acids into nanoparticles containing a significant nucleic acid payload [180].

The transferrin-targeted CDP delivery system has been extensively investigated to deliver siRNA for cancer treatment. In initial studies using murine models, CDP nanocarriers exhibited potent anti-tumor efficacy, silencing cancer associated genes such as the EWS-FLI1 fusion gene in Ewing’s sarcoma [181] and the M2 subunit of ribonucleotide reductase (RRM2) [182]. The safety and clinical translatability of the CDP-siRNA nanocarrier was then evaluated in nonhuman primates. Nanoparticles administered at doses of 3 and 9 mg siRNA/kg (body weight) were shown to be effective and well tolerated [183]. These results led to the initiation of drug candidate CALAA-01 in Phase I clinical trials in 2008. CALAA-01 is a transferrin-targeted CDP-based delivery system targeting siRNA to RRM2 mRNA in patients with solid cancers (Figure 2 a-c). Results from these trials indicated a reduction in RRM2 mRNA levels, and CALAA-01 represents the first example of a systemically administered siRNA formulated nanoparticle eliciting the RNA interference mechanism in humans (Figure 2 d-f) [184].

Figure 2.

Structure and activity of CALAA-01 siRNA nanocarriers, targeting RRM2 in solid tumors. (a) Nanocarriers are comprised of four self-assembling components. (b) CALAA-01 nanocarriers are injected into patients. (c) CALAA-01 circulates systemically, accumulating in the “leaky” tumor vasculature. (d) Nanocarriers interact with tranferrin receptors on the tumor cell surface, stimulating receptor-mediated endocytosis. (e) Following a one month dosing cycle to patients with metastatic melanoma, tumor biopsies show a reduction in positive RRM2 staining. (f) The same patients also show a reduction in RRM2 mRNA and protein expression, as assessed by quantitative real-time reverse-transcriptase polymerase chain reaction and western blot analysis, respectively. Panels (a) – (d) reprinted with permission from [28] (Davis ME. Mol Pharm 2009; 6: 659-68). Copyright 2009 American Chemical Society. Panels (e) and (f) reprinted by permission from [177] (Macmillan Publishers Ltd: Nature 2010; 464: 1067-1070, copyright 2010).

Other types of pH-sensitive, endosome-disruptive polymers have also been developed. For example, work by Stayton, Hoffman, and co-workers generated a family of α-alkyl acrylic acid polymers, including polyethylacrylic acid (PEAA), polypropylacrylic acid (PPAA), and polybutylacrylic acid (PBAA) [185]. These ‘smart’ polymers are defined by their pH-sensitive carboxylic acid and anhydride groups as well as their membrane interacting hydrophobic groups, which impart membrane disruptive properties [186]. Additionally, incorporation of reactive groups (e.g. free thiols and biotin) within the polymer can be used for direct conjugation to both biologics as well as cell targeting ligands [185]. Several studies have demonstrated the ability of this polymer family to improve the cytosolic bioavailability of conjugated proteins and peptides [185] via enhanced endosome destabilization. These polymers have also been shown to enhance nucleic acid delivery in vitro when conjugated to polymeric gene delivery systems [187, 188].

Related to the α-alkyl acrylic acid family of polymers, poly[2-(dimethylamino)ethyl methacrylate] (PDMAEMA) also has been extensively studied as a gene transfer reagent, given its inherent cationic charge and innate buffering capacity. Its utility in gene delivery was first shown by Hennink and co-workers, who attributed the observed enhancements in transfection efficiency with PDMAEMA to its capacity to cause endosome disruption [189, 190]. Since then various groups have modified the standard PDMAEMA structure in an attempt to further improve gene transfer capability [191].

Another interesting class of pH-sensitive polymers that has seen substantial clinical advancement is the Dynamic PolyConjugates^™ (DPC) platform developed by the Arrowhead Research Corporation for siRNA delivery. The DPC systems are unique in that they are simple and effective nucleic acid delivery vehicles, in which siRNA is conjugated to a membrane disruptive amphiphilic polymer, polyvinylether (PBAVE) that contains both positively charged amines and hydrophobic groups [192]. The amines are masked with a proprietary maleamate derivative, termed “CDM”, to which both PEG shielding agents and targeting ligands are attached. The masking agents are incorporated by pH-sensitive linkers, allowing the acidifying environment of the endosome to activate the polymer’s membrane lytic capability through linker cleavage. The siRNA therapeutics are conjugated to the polymer via reducible disulfide bonds, which are cleaved in the cytosol. Hence, exposure to the endosomal environment triggers both endosomal escape and RNA interference (RNAi) activation through two separate responsive release mechanisms.

First generation DPCs were designed to target hepatocytes for the treatment of liver diseases, using N-acetylgalactosamine targeting ligands. These nanocarriers were successful at silencing specific liver genes following intravenous injection in mice [193]. An interesting class of next generation DPCs focuses on co-injection of cholesterol-conjugated siRNA and masked PBAVE polymer, instead of direct siRNA-polymer conjugation (Figure 3) [194]. Following either co-injection or separate injection, the two components did not interact with one another during circulation in the bloodstream, yet were both able to target hepatocytes and colocalize in endosomes. The substantial success of this co-delivery technique in mice and non-human primates led to Arrowhead Research Corporation’s first drug candidate, ARC-520, designed to treat chronic hepatitis B virus (HBV) infection. The ARC-520 drug targets two conserved HBV transcripts, using two cholesterol-conjugated siRNAs co-injected with a masked melittin-like peptide [195]. ARC-520 has just been approved for Phase II clinical trials, following successful administration in healthy volunteers as a part of Phase I clinical studies. ARC-520 doses as high as 4.0 mg/kg (body weight) were shown to be safe and well-tolerated [196]. Preliminary Phase II results indicate an observed reduction in serum markers of disease progression in patients with chronic HBV infection following treatment with ARC-520 at 2.0 mg/kg [196].

Figure 3.

Structure and activity of ARC-520 DPC for the treatment of chronic HBV infection. (a) ARC-520 is formulated as a two component system and coinjected intravenously. Following receptor-mediated endocytosis and endosome escape in infected hepatocytes the siRNA (b) reduces viral antigen production, promoting effective restoration of host immunity. (c) ARC-520 treatment on days 1 and 15 in chronic HBV-infected chimpanzees led to a 95% reduction in circulating HBV DNA and ~90% reduction in hepatitis e- and s-antigens (HBeAg and HBsAg respectively). Figure reproduced with permission from Arrowhead Research Corporation.

Thermo-Responsive Materials

Recently, a new generation of temperature-sensitive polymers have been developed to enhance the cytosolic delivery of macromolecules via membrane disruption. In response to heat or cold shock treatments, these polymers undergo rapid volume expansion, resulting in a physical bursting of the endosome. Several types of temperature sensitive polymers have been developed [197, 198] and subsequently explored in various drug delivery applications. For example, hydrogel nanoparticles synthesized by cross-linking Pluronic F-127 with PEI exhibited significant swelling upon decreasing the temperature from 37 to 20 °C [199]. This observed swelling was thermally reversible. The enhanced gene silencing efficiency achieved when utilizing these particles to encapsulate and deliver siRNA-PEG conjugates in vitro following cold shock treatment was attributed to the physical bursting of the endosomal membrane [200]. In vitro cytotoxicity studies indicated that both nanocarrier administration and cold shock treatment were well tolerated. Although cold shock treatment is not directly applicable to in vivo applications, macromolecular drug delivery using these types of nanocarriers may be useful in various ex vivo cell-based therapies.

DNA Scaffolds

DNA molecules have also been used to build a variety of nanoscale structures and devices, with a number of promising potential applications. These approaches take advantage of the sequence specificity and resulting spatial addressability of DNA to produce DNA nanoarchitectures that can be used for the organization of proteins, peptides, and viral capsids into nanoparticles.[201, 202] Several of these DNA-directed assemblies have exhibited unique and improved functional properties, such as increased enzyme-cascade activities due to spatially positioned enzyme pairs.[203] DNA nanostructures with the capacity to bypass cellular membrane barriers have also been created for applications in drug delivery.[204] For example, in one such study, dendrimers built of DNA were coupled to antibodies directed against selected cell-surface receptors, and these hybrid structures were found to escape the endosome.[204]

Polymer-based nanocarriers represent a promising class of macromolecular delivery vehicles due to their inherent tailorability and ease of preparation. The clinical advancement of the CALAA-01 and Dynamic PolyConjugate delivery systems supports such potential. These types of nanocarriers also demonstrate how scalable and translatable carrier design plays an important role in improving effectiveness. With advances in synthetic technology, targeting ligands, biologic encapsulating agents, stealth coatings, and membrane-lytic groups can all potentially be incorporated into a single polymer design strategy. This is an essential step forward in developing new nanomaterials that can mimic the effectiveness of natural pathogens. However, as these new technologies progress, understanding how engineered nanocarriers interact with and engage cellular machinery will be crucial. Investigations aimed toward gaining fundamental insights into native biological interactions will help further enhance polymer-based nanocarrier efficacy.

Targeting the Intracellular Space: Delivering Biopharmaceuticals to the Nucleus

The delivery of biological macromolecules to the nucleus also is central to various biopharmaceutical delivery strategies as well as genetic engineering applications. Nanocarriers destined for the nucleus must take advantage of the cellular trafficking machinery to reach the perinuclear cytoplasm, and ultimately, to enter the nucleus via nuclear pores or post-mitotic redistribution. Various aspects of these processes present significant challenges, and in particular, the densely packed cytoplasm necessitates the use of active transport strategies to reach the nuclear membranes [205], and the nuclear membrane itself presents one of the most significant cellular transport barriers precluding efficacy of gene medicines and other biopharmaceuticals [206, 207].

Design strategies for nanocarriers targeted to the nucleus also take inspiration from mechanisms employed by pathogens as well as native proteins. Viruses have evolved diverse strategies to hijack the natural intracellular transport machinery to ensure successful infection. The majority of these viruses, as well as other pathogens and toxins, exploit the microtubule network or actin cytoskeleton for efficient transfer to the nuclear membranes. Once in the nuclear periphery, most viruses use the nuclear pore complex (NPC) in order to transfer their genomes into the nucleus; however, increasing evidence suggests that some microorganisms may bypass the nuclear pore by fusing directly with the nuclear membrane [208]. The nuclear import of proteins is regulated by NPCs. Small proteins are able to freely diffuse through NPCs, whereas larger macromolecules require association with nuclear transporter proteins, through a nuclear localization sequence (NLS), prior to NPC transport. The mechanisms of these nuclear trafficking strategies are detailed below.

Cytoskeleton-Mediated Transport to the Perinuclear Space

Cells have extensive microtubule (MT) networks that can act as tracks to efficiently move cellular components along polarized filaments toward destinations including the nucleus and plasma membrane [209]. Microtubules are hollow cylinders which support the intracellular transport of vesicles, organelles, and chromosomes. The orientation of the MT network facilitates transport to the nucleus along MTs whose “minus” ends are located near the nucleus and “plus” ends are located in the cell periphery. MTs were originally thought to exclusively transport organelles and vesicles, whereas smaller molecules such as proteins were believed to simply diffuse through the cytoplasm; however, MTs were also shown to regulate the localization and nuclear import of proteins [210]. Another cytoskeletal component that can mediate transport of organelles, as well as endocytic and secretory vesicles, is actin. Actin is involved in many cellular functions such as maintenance of cell structure, cell motility, cytokinesis, and movement of cargo [211]. These highly dynamic cytoskeletal components, which constantly undergo cycles of polymerization/depolymerization, have different sizes and shapes and contribute to various aspects of cellular functions [211].

Recent studies have indicated that cellular cytoskeletons and their associated signaling pathways also regulate different phases of the viral life cycle [212]. MTs are essential for many viruses to localize to the nuclear periphery [213], and viruses rely on MTs in general for their directional transport in the cytoplasm [214-216]. Viruses also have evolved mechanisms to disrupt and hijack actin filaments for cellular trafficking and nuclear entry [217].

Prior to nuclear entry, viruses, cellular proteins, and pathogens typically use one of several strategies to access the nuclear import machinery. Many viruses use the endomembrane network to reach lysosomes, where the low pH triggers a conformational change of viral fusion proteins that results in membrane fusion between the viral envelope and vesicle membrane [208]. Cellular receptors, such as EGFR, as well as several types of viruses and toxins are known to use the Sec61 translocon to retrotranslocate from the luminal side of endomembrane vesicles to the cytoplasmic face, prior to entry into the nucleus [218-220]. Sec61 also retrotranslocates other types of toxins that are trafficked from the cell surface to the ER as an essential part of the intoxication process [221]. The Sec61 translocon performs a similar function with misfolded proteins, which are retrotranslocated from the ER to the cytosol for degradation as part of the ER-associated degradation pathway.

Nuclear Structure and Entry

The nucleus in eukaryotic cells is separated from the cytoplasm by the nuclear envelope, which consists of two chemically distinct membranes, the inner and outer membranes, which are separated by the perinuclear cisterna. Transport of molecules between the cytoplasm and nucleus occurs through NPCs, which are multi-protein membrane transport structures that are widely distributed throughout the nuclear envelope [222]. Unless mitosis is occurring, molecules must pass through these channels in order to access the nucleus. Small molecules up to 9 nm are able to freely diffuse through the pore channels, whereas molecules up to 39 nm can only enter these channels through active transport [223]. These molecules are selectively transported into and out of the nucleus by a signal-mediated process [224] involving an NLS [225]. NLS-mediated nuclear protein import initially involves energy-independent recognition of the NLS-containing protein by the NLS chaperone, a heterodimeric complex consisting of the NLS-binding protein importin-α and the NPC-docking protein importin-β [226], and the subsequent docking of the bound complex at the NPC. This is followed by the translocation of the complex through the NPC and release within the nucleus.

The prevailing strategies for nuclear entry of viruses are similar to those for cellular proteins, although the nuclear import pathways of viruses differ depending upon their size and structure. Intact or partially uncoated viruses that are sufficiently small (e.g. parvoviruses) can pass through the NPC by associating with the nuclear import receptors [227]. In contrast, large viruses (e.g. herpes virus) release their genomes from the capsid prior to nuclear import. Specifically, upon reaching the nucleus, the capsid docks to the NPC [228]. Once docked, the capsid undergoes a conformational change known as uncoating and releases the DNA into the nucleus through the NPC [227]. Complex viruses, such as lentiviruses, uncoat their RNA genomes in the cytoplasm and reverse-transcribe their RNA genomes into DNA. The DNA is then recoated into a complex with viral proteins known as the preintegration complex (PIC) [229]. The PIC drives the genome into the nucleus through the NPC. Most capsid proteins also carry NLSs that allow nuclear import of newly synthesized viral proteins for assembly of progeny viral particles inside the nucleus [230].

Nuclear Entry During Mitosis

The nucleus disassembles during mitosis, and the NPCs disassociate. Nuclear envelope membrane proteins then diffuse throughout the ER membrane, as they are no longer tethered to the pore complexes, nuclear membrane, or chromatin [23]. In combination, these events result in the breakdown of the membrane barrier that separates the nucleus and cytoplasm, and nuclear proteins that were not bound to membranes are free to mix with the cytosol of the dividing cell [209]. At the end of mitosis, the nuclear envelope reassembles on the surface of chromatin, while NPCs begin to reassemble and actively reimport proteins that contain NLSs. Some studies have also shown that the ER membrane forms the source of the newly forming nuclear membrane [231], which wraps around chromosomes until the nuclear envelope is reformed.

While most viruses deliver their genomes to the nucleus through NPC-mediated entry as described, some viruses are only able to access the nucleus of a host cell during mitosis when the nuclear envelope is disassembled. For example, the retrovirus murine leukemia virus (MLV) can only access the nucleus during mitosis because the viral complex is too large for the NPC [232]. These viruses presumably wait for the dispersion of the nuclear membrane that occurs at mitosis, and become included within the nucleus during the reformation of the nuclear membrane in the daughter cells. The papillomavirus also requires cell division to achieve nuclear entry, based on evidence demonstrating that host cells need to pass through early prophase for successful onset of transcription of the viral genome [233].

Nuclear Access by Nanocarriers

The general mechanisms for nuclear uptake by nanocarriers are not fully understood, but given the size of polyplexes, lipoplexes, and other drug delivery nanostructures, it is unlikely that the NPC is used in the same way that it is used by native proteins and pathogens. For example, polyplexes made with branched or linear polyethylenimine (PEI) [234], and also compact DNA particles assembled with lysine and PEG copolymers [235], have been shown to enter the nucleus in non-dividing cells, suggesting active transport through the NPC, yet these polyplexes and particles are typically larger than the size limitations for NPC-based entry. Conjugation of NLSs to various types of nanocarriers enhances nuclear uptake in some cases [236], although the effects are not generalizable and appear to depend on specific aspects of the NLS, nanocarrier structure, and cell type [237]. Multiple reports show that mitosis significantly enhances the delivery of plasmid DNA as well as delivery of DNA within lipoplexes and polyplexes, with 30- to more than 500- fold higher transfection efficiencies reported when cells are exposed to delivery structures during S or G2 phase as compared with G1 phase [238-240]. This suggests that transfection is likely facilitated by the temporary breakdown of the nuclear membrane, yet efficient design strategies to exploit this effect remain significantly limited. Figure 4 summarizes the primary strategies that have been documented or proposed for the nuclear delivery of nanocarriers. In the following paragraphs, both the structure and nuclear entry mechanisms of nanocarriers are described in detail.

Figure 4.

Nuclear entry of engineered nanocarriers and their therapeutic cargos following intracellular trafficking.

Design Strategies to Target the Nucleus

Targeting engineered nanocarriers to the nucleus continues to be a significant challenge. Similar to the efforts focused on enhancing endosomal escape, researchers have investigated numerous design strategies to achieve effective nuclear delivery. The inclusion of an NLS continues to be the most commonly studied approach. Other reports have explored different ways of nuclear targeting without utilizing a NLS, such as the use of peptide sequences that interact with proteins that are known to translocate between the cytoplasm and nucleus. Table 2 lists some of the most effective design strategies employed to date for achieving successful biologic drug transport into the nucleus.

Table 2.

Nuclear localization agents and mechanisms.

Category	Nuclear Import Agent	Proposed Mechanism	Reference
Peptides and proteins	Viral agents
	SV40	NPC	[241]
	Native cellular proteins
	Histone H1	NPC	[242, 243]
	Histone H3	NPC/unknown	[244]
	p50	Enhance MT transport; NPC	[245]
	NfKB	NPC	[246]
	Modified peptides
	S413-PV	Membrane Permeabilization	[247]
	Lactosylated poly-l-lysine	NPC - lectins	[248]
Polymer- based	Polyethylenimine (PEI)	Membrane Permeabilization	[239]
	poly-(glycoamidoamine)s (PGAAs)	Membrane Permeabilization	[249]
	Glycofect	Retrograde transport/Sugar signaling	[250]
	A-C3	NPC	[251]

NLS-Mediated Design Strategies

Multiple reports show that adding an NLS to nanocarriers can increase their nuclear localization in non-dividing cells [252]. The best characterized transport signals are the classical NLS sequences for nuclear protein import, which consist of either a monopartite or bipartite stretch of basic amino acids [253]. One of the most widely explored classical NLS peptides is the sequence derived from the SV40 large T antigen, which was shown to increase plasmid DNA accumulation by a factor of two-fold within the nucleus of non-dividing TC7 cells by enhancing transport through the NPC [254]. This peptide also has been coupled to cationic DNA nanocarriers [255] or polycations [241] and shown to increase transfection efficiencies by three-fold or two-fold, respectively. The SV40 NLS was also studied as part of a fusion peptide, S4₁₃-PV, containing the SV40 NLS and a cell-penetrating sequence from the Dermaseptin S4 peptide. S4₁₃-PV enhanced DNA transfection by up to 50%, and this effect was suggested to result from transient membrane destabilization induced by the peptide [247].

Other strategies have explored the use of non-classical NLS peptides or NLS-containing proteins to increase nuclear entry. Chen and coworkers have used protamine, a naturally-occurring, cationic protein with an arginine-rich NLS, in combination with DNA nanocarriers [256]. This polypeptide enhanced nuclear localization of plasmid DNA, which increased transfection efficiency. Complexes of plasmids with high mobility group-1 (HMG-1) proteins or the nuclear protein nucleoplasmin have been shown to increase nuclear localization of the plasmids by three-fold and gene expression by a five-fold [257]. Nucleoproteins themselves were tested for their potential as carriers of nucleic acid drugs. Histones [258] and p50 [245] contain both NLSs and intrinsic positive charges to condense DNA; therefore, these proteins have been explored as a substitute for synthetic polymeric carriers. These proteins have been found to condense plasmid DNA and increase trafficking to the nucleus. For example, Böttger and colleagues used histone H1, complexed with plasmid DNA (pDNA), and demonstrated a two-fold improvement in transfection. The authors proposed that H1 mediates DNA transfection via DNA binding and delivery into the nucleus [259]. Another study found that association of NLS-carrying p50 with pDNA facilitated not only nuclear entry of the DNA, but also its migration through the cytoplasm along MTs towards the nucleus, mediated by the nuclear transport receptor that recognizes the p50 NLS [245].

Non-NLS Design Strategies

Various studies also have explored non-NLS sequences for their capacity to induce transport into the nucleus. For example, sugar residues can transport cargoes into the nucleus, and the pathway used by neoglycoproteins to enter the nucleus appears to be different from that used by NLS-bearing proteins. Unlike NLS-mediated nuclear import, the transport of neoglycoproteins from the cytosol to the nucleus does not use the pathway of NLS-bearing proteins and their associated cytosolic factors [260]. Rather, this process works by utilizing sugar-binding proteins (lectins) as a shuttle between the cytoplasm and nucleus [260].

Targeting with lactose [248] and mannose [261] showed significant improvement in nuclear delivery of DNA. For example, a study involving lactosylated poly-L-lysine/cDNA polyplexes showed that incorporation of lactose residues induced nuclear localization by allowing nanocarriers to bind to a potential lectin-like shuttling protein with galactose/lactose specificity, and this binding interaction was suggested to trigger the nuclear internalization of the complex. Another study also demonstrated the cell cycle-independent nuclear translocation of lactosylated polylysine complexes, which was attributed to the presence of the lactose residues on the polylysine, which were also thought to work through a lectin-like protein-mediated nuclear internalization [262]. Studies have also developed conjugates containing carbohydrate-binding proteins linked to pDNA, and these conjugates were shown to enter the nucleus, possibly through a sugar-dependent mechanism [248, 250]. Reineke and coworkers designed a glycopolymer, poly-(galactaramidopentaethylenetetramine), or Glycofect, that increased gene transfection when complexed with pDNA by inducing unique trafficking behaviors, possibly from the presence of saccharide units [249].

Other studies have indicated that direct nuclear permeabilization is a possible route of nuclear import, and that polymers with the highest nuclear envelope permeability displayed the highest expression efficiency [263]. In one such example, the Reineke group hypothesized that cationic polymers may be capable of disrupting the nuclear envelope, since they are also capable of disrupting the plasma membrane. Their studies have shown that the polycations with the highest amount of protein expression, PEI and T4₄₃, a poly-(glycoamidoamine), are capable of inducing nuclear membrane permeability; however, they also have the highest level of cytotoxicity [249].

Improving Nuclear Delivery with Engineered Nanocarriers

Despite recent advances, the ability to efficiently deliver nanocarriers to the nucleus remains limited, and improved efficacy requires increased understanding of how nanostructures interact with the cellular machinery. To this end, one key challenge in designing improved nanocarrier structures has been the limited understanding of the cellular trafficking steps leading to the nucleus, and whether endosomal escape is required to enable nuclear entry by nanostructures. Recent studies in our lab have used histone-targeted polyplexes to analyze endomembrane transport leading to the nucleus, and these studies identified novel, membrane-assisted mechanisms to enhance nuclear partitioning in dividing populations of cells. Our lab linked histone H3 tail sequences to PEI polyplexes and showed that the H3 tails shuttled these polyplexes via caveolae to the perinuclear Golgi and ER, avoiding recycling vesicles and lysosomes in the process (Figure 5). The polyplexes remained associated with ER vesicles/membranes until mitosis, when they were redistributed into the nucleus, potentially facilitated by post-mitotic redistribution of ER membranes along with the membrane-bound polyplexes [264].

Figure 5.

H3-targeted polyplex trafficking following caveolar-mediated endocytic uptake in vitro. H3-targeted polyplexes harness endomembrane trafficking pathways similar to native pathogens and partially regulated by histone methyltransferases (H3K4MTs). Polyplexes accumulate in Golgi/ER vesicles and are shuttled into the nucleus following mitosis.

Another important challenge in nuclear delivery is the limited understanding of the nanocarrier-cargo interface during nuclear import, and to what extent the cargo itself may play a role in driving delivery. A series of studies have shown the key role played by specific DNA sequences in determining nuclear import of plasmids, and these analyses suggest the importance of more integrated approaches in cargo and nanocarrier design. In particular, pDNA promoter/enhancer sequences are recognized by transcription factor proteins in the cytoplasm, and various reports demonstrate that incorporating such sequences into plasmids can be used to enhance nuclear delivery [254, 265, 266].

In one such example, a DNA sequence coding for a binding site for NFkB, a transcription factor with the capacity to shuttle between the cytoplasm and nucleus, increased the nuclear transport of pDNA complexed with PEI or histidylated polylysine, resulting in an increase in transfection [246]. David Dean and coworkers conducted a series of seminal studies to explore in depth how specific promoter and enhancer sequences affected the extent of nuclear localization by ‘naked’ plasmids [267]. In particular, they demonstrated that when specific eukaryotic sequence elements, including a short region of the SV40 enhancer termed a ‘DNA nuclear targeting sequence,’ or DTS, were added to pDNA, it greatly enhanced transfection [254]. This is due in part to the binding sites the DTSs contain for a number of ubiquitously expressed transcription factors. Transcription factors contain NLSs for their nuclear import, and association of these endogenously expressed proteins with plasmids produces a ‘coating’ of the NLSs required for import. The Dean lab recently demonstrated that promoter sequences and cytoplasmic factors derived from specific tissues can be used for cell-specific nuclear import [268], wherein use of these tissue/cell-specific promoters restricted transfection to only those cell populations containing the unique set of complementary transcription factors. By screening promoters that are transcriptionally active only in a desired cell type, this lab has also shown that it is possible to uncover new sequences that function for cell-specific nuclear import [269]. Additionally, these authors have shown that the MT network is required for directed plasmid trafficking to the nucleus, and transcription factor binding sites incorporated onto pDNA allow for enhanced rates of intracellular trafficking and nuclear accumulation [270].

A recent study, utilizing a novel diblock copolymer (A-C3) comprised of poly(2-dimethylaminoethyl acrylate) (PDMAEA) and an anionic polymer poly(acrylic acid), with a second block of P(N-3-(1H-imidazol-1-yl)propyl) acrylamide (PImPAA) and poly(butyl acrylate) (PBA) to package plasmid DNAs, shed direct light on the nanocarrier-cargo interface. The authors demonstrated dissociation of the large plasmid DNA/polymer complexes prior to nuclear entry, most likely on the nuclear membrane or in the nuclear pore [251]. Transfection was inhibited when a nuclear pore-blocking agent was added, supporting the hypothesis that entry involved plasmid uptake through the NPC and not nuclear membrane breakdown during cellular division. These analyses provided evidence that the nuclear entry pathway for many nucleic acids may, in fact, involve the active import of the cargo separate from the carrier. Moreover, these studies demonstrate the potential benefits that may arise from synergistic design strategies seeking to harness both the nanocarrier and the cargo in the delivery process.

With a better understanding of how the cellular machinery involved in both endomembrane transport and nuclear import, significant improvements can be achieved in the design of gene delivery vehicles and other nanocarriers destined for the nucleus. Overcoming the nuclear barrier will be a key objective of future research, especially for in vivo and clinical applications of nucleic acid drugs.

Conclusions

Biological macromolecules possess immense potential in their ability to treat human disease. However, their successful application is fully dependent upon the design of delivery vehicles that confer high intracellular bioavailability. As described above, an extensive variety of natural and synthetic peptides, as well as engineered polymers, have been developed to achieve this requirement. Unfortunately, nanocarriers still fall short of their viral counterparts in their ability to deliver macromolecules in sufficient quantities to the intracellular sites of action. A better understanding of the natural cellular mechanisms involved in cellular trafficking, endosomal escape, and nuclear import is crucial to achieving effective macromolecular delivery. Additionally, more quantitative assessments of endosomal escape and nuclear delivery need to be established in order to develop improved delivery techniques. The recent clinical advances achieved with multifunctional polymer formulations demonstrate that possessing both tailorability and the capacity to interact favorably with the numerous biological signals encountered during the delivery process are essential to non-viral vehicle design. Combining multifunctional biomimetic carrier design with natural mechanistic insight will help elucidate the complex journey taken by engineered nanocarriers en route to their active site within the cell, greatly enhancing the therapeutic efficacy of biological macromolecules.