2.1. What Is Prime Editing?

Prime editing is a versatile genome editing method that “writes” new genetic information into precise DNA locations. This method differs from CRISPR/Cas9 systems in that it employs a catalytically impaired Cas9 endonuclease fused to a reverse transcriptase that is programmed via prime editing guide RNA (pegRNA) [Reference Anzalone, Randolph and Davis1]. pegRNA encodes both the desired edit and the target DNA site. The prime editing method enables DNA insertions, deletions, and substitutions without requiring double-strand DNA breaks or exogenous donor DNA templates. By avoiding the sporadic DNA repair associated with double-strand breaks, prime editing can improve the accuracy of gene editing in vitro and theoretically also in vivo.

Unfortunately, the in vivo efficacy of prime editing systems is yet to be adequately demonstrated. To achieve widespread in vivo use, effective delivery of the prime editing machinery is required. Lentiviral systems have been proposed as transporters for the base editor 3 (BE3) system as its large size would not fit within generic adenoviral systems [Reference Rittiner, Moncalvo, Chiba-Falek and Kantor4]; however, viral delivery systems often lead to mutagenic and carcinogenic side effects [Reference Nayerossadat, Ali and Maedeh5]. Therefore, a nonviral delivery system is preferred for patient safety reasons. Unfortunately, transfection efficiency is often lower for such systems. Ultimately, the development of effective nonviral delivery systems for prime editors will likely require innovation and a thorough evaluation of delivery system requirements.

4.1. Liposomes

Liposomes are vesicles composed of at least one lipid bilayer [Reference Sercombe, Veerati, Moheimani, Wu, Sood and Hua11]. They are generally spherical, yet can assume other shapes with proper engineering. The bilayer structure of liposomes effectively serves as a barrier between the internal components and external surrounding fluid, allowing therapeutic agents to be protected during transport. The phospholipid assembly also enables hydrophilicity on both sides of the membrane, allowing the loading of water-soluble drugs within the liposome interior and the loading of lipophilic compounds by housing them within the bilayer. Drugs exhibiting an intermediate partition coefficient (logP) can segregate between the two phases. The liposomal compartmental space can house smaller liposomes, enabling unique architectures and the development of multilamellar liposome types. Specific lipids and other components can be tailored to increase the rigidity and stability of the liposomes or to ensure a slow-release vessel. Furthermore, mechanisms for targeting and tracking the vesicles can be incorporated within the liposomes throughout the majority of the synthesis process. Accordingly, liposomes should undoubtedly be considered when contemplating a straightforward nano-enabled delivery approach for new commercially relevant therapeutics.

With regard to the delivery of prime editing machinery, liposomes have several distinct advantages. First, the size of the liposomes can be optimized for cargo delivery, and the interior compartment space can house both prime editing protein and pegRNA. In fact, delivery of large protein-RNA complexes has been previously demonstrated using liposomes [Reference Cho, Ryu and Lee12]. Second, several liposomal products have passed the US Food and Drug Administration (FDA) regulations and are commercially available. Adapting these currently approved liposomes for prime editing delivery would allow for a more direct path through the regulatory process. Finally, more advanced liposomal gene delivery systems have been engineered to enhance cellular [Reference Alshehri, Grabowska and Stolnik13], nuclear [Reference Iwasa, Akita, Khalil, Kogure, Futaki and Harashima14], and mitochondrial uptake [Reference Kang and Ko15] and could be examined for design innovation purposes.

However, liposomal systems often have several drawbacks. While liposomes mimic natural membranes, they are still foreign materials in the body and are known to be cleared by the mononuclear phagocytic system [Reference van Etten, ten Kate, Snijders and Bakker-Woudenberg16]. Efforts to use synthetic phospholipids and incorporate polyethylene glycol (PEG) coatings have somewhat lengthened the time to full blood clearance [Reference Immordino, Dosio and LJIjon17]; however, there are concerns accompanying these techniques regarding their ability to inhibit bloodstream extravasation. In addition, liposomal stability is a concern. Phospholipids sporadically jump from one membrane to another, leading to the occasional merging and coalescence of vesicles. Liposomes are not different and may often fuse membranes with unintended cells or bursts when in close proximity to other membranes. One estimate is that up to 30% of liposomal contents can be leaked in this manner, leading to large amounts of nontargeted API exposure [Reference Russell, Hultz and Searson18]. In vivo fate of liposomes can be significantly affected by the interaction between liposomes and cells where they can be absorbed or undergo endocytosis. Stability during storage also remains an issue with most liposomal formulations requiring frozen storage conditions. If a specific prime editing therapy could initiate severe side effects without directed localization, liposomes may not provide the stability needed to adequately reduce off-target effects. Nonetheless, liposomes are likely to be strong candidate carriers for low-toxicity prime editing therapies.

4.2. Micelles

Micelles are the fundamental building blocks of emulsion-based formulations. A micelle is a three-dimensional assembly involving multiple amphiphilic surfactants. The hydrophobic ends (tails) of the multiple surfactant molecules arrange themselves near one another in order to minimize contact with water molecules, leading to a structure in which the hydrophilic ends encounter the water molecules at the periphery. Like liposomes, micelles are generally spherical, yet rod and planar structures can be obtained using surfactants with proper head-to-tail volume ratios and distinct solution conditions. By exchanging the solvent, inverse micelles can also be formed where the hydrophilic ends cluster and the hydrophobic ends interact with the solvent. Pharmaceutical formulations take advantage of both micelle types in the form of water-in-oil (W/O) emulsions and oil-in-water (O/W) emulsions. Advanced water-in-oil-in-water (W/O/W) and oil-in-water-in-oil (O/W/O) emulsions have also been developed to inhibit globular coalescence. Micelles carry molecular cargo by thermodynamically stabilizing the molecules in the core. Generally, fat-soluble agents are poorly soluble in surrounding solvents, leading to micelles being major applications in the transport of fat-soluble nutrients and drugs. Because of the simplicity of the system and the long history of using detergents, micelle systems are not always recognized as nanoparticle delivery systems, yet they have been shown to be effective in the delivery of genes and specific biologics [Reference Nishiyama, Bae, Miyata, Fukushima and Kataoka19]. For example, a recent study has demonstrated the potential of polyplex micelles in delivering Cas9 mRNA and guide RNA for in vivo genome editing in the mouse brain [Reference Abbasi, Uchida and Toh20].

However, with regard to their potential for prime editing delivery, micelles face many challenges. First, micelles generally exhibit a size limitation, usually ranging from 2 to 20 nm, which can hinder the delivery applicability of larger macromolecules. In fact, amphiphilic block copolymers often self-assemble into micellar structures themselves [Reference Gaucher, Dufresne, Sant, Kang, Maysinger and Leroux21], but rarely surpass the size limit as a stable structure. This inherent size limitation strongly negates the potential use of micelles to deliver protein-RNA complexes. Second, the stability of a three-dimensional supramolecular structure is difficult to maintain without covalent crosslinking. As such, internal contents are often “spilled” from micelles during delivery [Reference Trivedi and Kompella22]. Third, micelles can activate the immune system and trigger rapid clearance. While PEGylation can decrease clearance, the presence of a hydrophilic chain can disrupt micellar stability, even in the case of amphiphilic polymeric micelles. Finally, as certain adverse reactions have been associated with strong surfactants, toxicity must always be considered when designing micelle-based systems. For these reasons, micelles are not recommended for the formulation of genome prime editing technologies.

4.3. Exosomes

Exosomes are similar in structure to liposomes, being membrane-bound vesicles; however, one intriguing difference sets them apart. Exosomes are derived from the endosomal compartment of cells and carry unique cell biomarkers that are characteristic of the cell of origin. Therefore, exosomes often display inherent targeting molecules on the outer surface of the membrane, enabling improvement in target cell uptake [Reference Sancho-Albero, Navascués, Mendoza, Sebastián, Arruebo and Martín-Duque23]. Exosomes are known to undergo endocytosis or fusion with the plasma membrane of target cells [Reference Escrevente, Keller, Altevogt and Costa24, Reference Horodecka and Düchler25], and internalized exosomes are degraded after delivering the cargo into the cytosol [Reference Horodecka and Düchler25].

Exosomes can be engineered in a variety of sizes, and various exosome-loading procedures have been developed [Reference Liu and Su26]. Recently, exosomes showed the potential in delivering prime editing proteins and gRNA. A recent study demonstrated that CIRSPR/Cas9 protein and sgRNA can be packaged into exosomes that in turn successfully transduced cells in vitro [Reference Ye, Zhang and Xie27]. Nanomedic, an exosome-based CRISPR/Cas9 delivery system, showed promising efficiency of gene editing in various hard-to-transfer cell types, including human iPS cells, neurons, and myoblasts [Reference Gee, Lung and Okuzaki28].

While it may initially seem that exosomes are an obvious choice for prime editing delivery over liposomes, certain drawbacks limit their utility, although not entirely. First, exosome engineering remains in its infancy, and the ability to design an exosome with firm-targeting capabilities is still somewhat only theoretical. Furthermore, alterations to exosomes after synthesis, for purposes such as bloodstream cloaking, tend to result in loss of desired exosomal properties [Reference Luan, Sansanaphongpricha, Myers, Chen, Yuan and Sun29]. Finally, the key biomarkers present in exosomes often lead to robust cytotoxicity, immune responses, and direct uptake by the reticuloendothelial system.

4.4. Polymer-Based Systems

Polymers are long-chain molecules composed of many repeating subunits (monomers). Theoretically, polymers assume a three-dimensional architecture in solution owing to interactions with the solvent environment. Hydrophilic polymers are commonly employed to improve the solubility of drugs with low solubility, while hydrophobic polymer components are often used for drug stabilization. Many different polymer types exist and can enable a wide range of properties relevant to drug delivery, namely, polymers can be covalently attached to drugs or utilize noncovalent interactions for drug transport. Furthermore, polymer nanosystems can be designed with various architectures, sizes, and compositions. As such, many different systems can be engineered to fully encapsulate both prime editing protein and pegRNA. Additionally, polymers enable facile attachment of in vivo tracking moieties, targeting molecules, and environmentally responsive entities and are thus often considered the most versatile platform for drug delivery. Upon arrival to the target cells, the most of polymer-based delivery vehicle undergo biodegradation or breakdown to initiate the delivery cargo into the cells.

While a polymer-based system may be a strong candidate for prime editing delivery, several factors should be considered. First, covalent attachment of a polymer to a protein may lead to incapacitation of the protein [Reference Wright, Page and Konkolewicz30], and nearly all covalent attachments of polymers to RNA lead to loss of function. This being noted, polymer encapsulation is likely to be the best approach for prime editing delivery. It is not directly relevant to discuss all polymer types in this article; however, it should be mentioned that certain polymer types may reduce bloodstream interactions and/or immune responses to the nanosystem. For regulatory purposes, the determination of in vivo polymer degradation and excretion profiles is essential. Most regulatory agencies tend to look more favorably on liposomal systems than polymer-based systems, yet a few polymer-based systems have achieved US FDA approval [Reference Ventola31]. However, any additional modifications to the polymer delivery system (e.g., to enhance tracking, transport, and targeting) will steepen the climb for regulatory approval. For research and grant-obtaining purposes, polymer-based systems can likely produce the best results, but from a commercialization standpoint, polymer nanosystems might introduce too many hurdles to merit their investigation.

4.5. Dendrimers

One of the most unique nanoparticle systems is the use of dendrimers, which are macromolecules composed of repeatedly branched chains. Their divergent synthesis begins with a single molecule at the core that then branches in multiple directions. Four branches become eight in the next stage of synthesis (called a generation), then sixteen, thirty-two, and so on. The dendrimer structure is unique in that the entire structure exhibits polymeric flexibility, yet the dendrimer surface exhibits an extremely high charge density. Dendrimers are known to chelate small molecules and ions within their branched structures, and their high charge density permits penetration across difficult in vivo barriers, such as the blood-brain barrier [Reference Santos, Xavier and Leite32].

Dendrimers face similar issues as those of micelles when being evaluated for prime editing delivery. In theory, no dendrimer size limit should exist; however, in practice, it is very difficult to produce dendrimers beyond six generations. Quite simply, the charge density becomes too large, and steric repulsion forces restrict further branching. Therefore, the probability of developing a dendrimer system with sufficient cargo space to deliver both prime editing protein and pegRNA molecule is quite small.

4.6. Rigid Nanoparticle Systems

In this article, rigid nanoparticles are defined as nanoparticles composed of any material (organic, inorganic, metallic, etc.) that elicits a rigid morphology, including silica, metal oxide, and palladium nanoparticles. Essentially, rigid nanoparticle systems utilize conventional nanoparticles as delivery vessels. These systems can be designed using either top-down or bottom-up approaches. Top-down techniques tend to result in samples with high polydispersity in size and shape, whereas bottom-up methods can produce monodisperse samples with unique architectures. While surfaces can be chemically modified for the attachment of both proteins and nucleic acids, the attached molecules are on the surface of the particles and therefore presented to bloodstream components and rapidly disabled or cleared. Many of these systems rely solely on the adsorption of the APIs without chemical modification. However, these systems are likely to fail as prime editing delivery systems for the same reason.

On the contrary, certain rigid nanoparticle systems, such as silica, can be produced with a hollow core using bottom-up approaches, allowing for the encapsulation of various molecules [Reference Hadipour Moghaddam, Yazdimamaghani and Ghandehari33]. As a result of the required etching processes, these hollow nanoparticles have a porous shell. Loading of proteins and nucleic acids into these particles could prove difficult depending on the pore size. Furthermore, if the pore size is too large, the molecular cargo will not be retained within the hollow core during transport. Methods have been developed to place an external shell around the hollow rigid nanoparticles to block pores and fully encapsulate drugs [Reference El-Toni, Habila and Labis34]; however, a nanoparticle system of such complexity has never attained federal regulatory approval. In addition, rigid nanoparticle systems often undergo surface dissolution to some extent in aqueous environments, leading to the release of various ions. In many cases, particularly with metallic nanoparticles, released ions pose considerable cytotoxic threats [Reference Hahn, Fuhlrott, Loos and Barcikowski35]. Overall, rigid nanoparticle systems should be avoided for prime editing delivery unless a strong rationale justifies their use.