biomolecules

Review
Recent Advances in CRISPR/Cas9 Delivery Strategies
Bon Ham Yip
Vector Development and Production Laboratory, St. Jude Children’s Research Hospital,
Memphis, TN 38105, USA; bonham.yip@stjude.org; Tel.: +1-9018965417




Received: 3 May 2020; Accepted: 28 May 2020; Published: 30 May 2020


Abstract: The clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system has
revolutionized the field of gene editing. Continuous efforts in developing this technology have
enabled efficient in vitro, ex vivo, and in vivo gene editing through a variety of delivery strategies.
Viral vectors are commonly used in in vitro, ex vivo, and in vivo delivery systems, but they can
cause insertional mutagenesis, have limited cloning capacity, and/or elicit immunologic responses.
Physical delivery methods are largely restricted to in vitro and ex vivo systems, whereas chemical
delivery methods require extensive optimization to improve their efficiency for in vivo gene editing.
Achieving a safe and efficient in vivo delivery system for CRISPR/Cas9 remains the most challenging
aspect of gene editing. Recently, extracellular vesicle-based systems were reported in various studies
to deliver Cas9 in vitro and in vivo. In comparison with other methods, extracellular vesicles offer
a safe, transient, and cost-effective yet efficient platform for delivery, indicating their potential for
Cas9 delivery in clinical trials. In this review, we first discuss the pros and cons of different Cas9
delivery strategies. We then specifically review the development of extracellular vesicle-mediated
gene editing and highlight the strengths and weaknesses of this technology.

Keywords: CRISPR/Cas9; gene editing; delivery; extracellular vesicles; virus-like particles

1. Introduction
Clustered regularly interspaced short palindromic repeats (CRISPR) were first discovered as
an adaptive immune system effector in prokaryotes [1]. CRISPR comprises a group of small DNA
sequences found in the genomes of prokaryotes that were acquired from previous infections by
bacteriophages [1]. It offers a defense mechanism for prokaryotes to fight against reinfection by similar
bacteriophages. Subsequent development of this technology into a gene-editing tool in eukaryotic
cells [2,3] enabled the application of gene editing for human diseases. The CRISPR/Cas9 system is
composed of a target-specific single guide RNA (sgRNA) and a Cas9 endonuclease. A target-specific
sgRNA, formed by the fusion of a CRISPR RNA (crRNA) and a transactivating CRISPR RNA, directs
the Cas9 protein to a target site for cleavage, creating a double-strand break (DSB). Because target
recognition is based on RNA–DNA interactions, CRISPR/Cas9 has the advantages of the easy design
of genomic targets and multiplexing over that of zinc finger nucleases (ZFNs) and transcription
activator-like effector nucleases (TALENs). In contrast with ZFNs and TALENs, which require
laborious protein engineering steps for each new editing target, the Cas9 nuclease simply requires
a target-specific sgRNA for each editing target.
The target specificity of Cas9 is determined by the spacer sequence of crRNAs (~20 nucleotides)
and adjacent protospacer adjacent motifs (PAMs) [4]. More precisely, the seed sequence located in the 30
end of the spacer sequence (10–12 base pairs adjacent to the PAM) is critical for correct targeting [4].
Cas9 will cleave only when sufficient homology is present between the seed region and the target
DNA. However, off-target cleavage occurs when DNA sequences contain a few mismatches but share
some homology with the seed region of the sgRNA [4]. Research efforts are focused on minimizing

Biomolecules 2020, 10, 839; doi:10.3390/biom10060839 www.mdpi.com/journal/biomolecules

Biomolecules 2020, 10, 839 2 of 16

the off-target effects associated with CRISPR/Cas9. For example, truncated gRNAs (< 20 nucleotides)
reportedly reduce off-target effects without affecting on-target genome editing [5]. Moreover, gene
editing with a Cas9 nickase (a mutant that creates a single-strand break in DNA) and two sgRNAs, each
cleaving at different sites of the target, dramatically reduces off-target effects [6]. The delivery of a Cas9
ribonucleoprotein (RNP) complex into cells produces fewer off-target effects than does the delivery of
DNA plasmids expressing Cas9 and sgRNA [7,8]. Because the Cas9 protein is short-lived and is able
to immediately cleave the target DNA, Cas9 RNP delivery mitigates the propensity of Cas9-induced
off-target effects.
The delivery of Cas9 into cells is an important consideration in gene editing. Cas9 can be delivered
in the forms of DNA, mRNA, or protein (Figure 1A). Each format has pros and cons. The delivery of
Cas9 in the form of plasmid DNA offers a cost-effective option. Only a standard laboratory set-up
is required for plasmid preparation. Plasmid DNA-driven Cas9 expression also yields a longer
expression time in cells, which may be advantageous if sustained expression is required for editing.
However, because transcription and translation are required for the synthesis of the Cas9 protein
(Figure 1A), the plasmid DNA format has the slowest onset of editing when compared with that
of mRNA and protein. Sustained expression of Cas9 in cells also increases the chance of off-target
effects [9]. Furthermore, plasmid DNA poses a risk of insertional mutagenesis [10].
The delivery of Cas9 by mRNA enables the faster onset of gene editing than that by plasmid DNA
because transcription is not required anymore (Figure 1A). Because mRNA is highly unstable and prone
to degradation by RNases, this format only permits transient Cas9 expression. Chemical modifications
of mRNA are available to enhance its stability after delivery [11]. Although transient Cas9 expression
may compromise gene editing efficiency, it also reduces the chance of off-target effects [9].
The delivery of Cas9 via protein enables immediate gene editing in the nucleus (Figure 1A),
resulting in higher gene editing efficiency than that of DNA and mRNA [12]. However, the protein
delivery of Cas9 in cells is the most transient of the formats, but the chance of off-target effects is
also minimal [12]. The cost of protein delivery is also higher than that of DNA and mRNA delivery.
Importantly, delivering the Cas9 protein, which is of bacterial origin, into cells may induce the carryover
of bacterial endotoxin and trigger serious immunologic responses. This aspect is a key safety concern
of using Cas9 in clinical trials [13].
The first clinical trial using CRISPR/Cas9 technology was approved in 2016 [14]. In that trial,
the CRISPR/Cas9-mediated knockout of PD1 was performed in patient blood cells to reactivate T cells
for the treatment of lung cancer [14]. Since then, researchers began investigating the potential of using
CRISPR/Cas9 for the treatment of other diseases in clinical trials. At the time of writing this review,
ClinicalTrials.gov (https://clinicaltrials.gov/) has listed more than 20 clinical trials using CRISPR/Cas9
to treat solid tumors, hematologic malignancies, and genetic disorders.
Biomolecules 2020,10,
Biomolecules2020, 10,839
x 3 3ofof1615

Figure1.1.Extracellular
Figure Extracellularvesicle-mediated
vesicle-mediateddelivery
deliveryofofCas9
Cas9ribonucleoproteins
ribonucleoproteins(RNPs).
(RNPs).(A)
(A)Cas9
Cas9can
canbebe
delivered in the forms of DNA, mRNA, or protein. The protein format enables the immediate
delivered in the forms of DNA, mRNA, or protein. The protein format enables the immediate action action
ofofCas9
Cas9when
whenititisisininthe
thenucleus.
nucleus.The
Thetransduction
transductionofofextracellular
extracellularvesicles
vesiclesreleases
releasespre-loaded
pre-loadedCas9
Cas9
RNPs
RNPs into cells for efficient gene editing. (B) Structural differences between a lentivirus,aavirus-like
into cells for efficient gene editing. (B) Structural differences between a lentivirus, virus-like
particle,
particle,and
anda avesicle.
vesicle.

To date, many strategies are available for Cas9 delivery. These can be classified into viral and
nonviral vector-based approaches. In general, nonviral vector-based approaches include physical
and chemical methods. Because extracellular vesicles (EVs) resemble viruses lacking genomes
(Figure 1B), the EV-based delivery method represents a compromise between viral and nonviral
delivery approaches and possesses the strengths of both approaches (Table 1). Understanding the pros
and cons of each delivery strategy is paramount to choosing the most appropriate delivery method
for specific applications. In clinical trial settings, stringent safety requirements should be considered,
in addition to delivery and editing efficiencies. In this review, we first discuss the advances in
CRISPR/Cas9 gene editing and the various Cas9 delivery strategies available today. Researchers are
developing and optimizing novel strategies to improve the safety and efficiency profiles of Cas9
delivery. We also discuss the development of EVs for Cas9 delivery and the potential of this strategy
for achieving safe and efficient gene editing.
Biomolecules 2020, 10, 839 4 of 16

Table 1. Comparison of common clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 delivery strategies.

Viral Delivery Non-Viral Delivery

Strategy Cell
Lipid-Based Gold
LV AAV AV EV Microinjection Electroporation Penetrating
Nanoparticle Nanoparticle
Peptide
Cas9 Delivery DNA, mRNA or DNA, mRNA DNA, mRNA
DNA DNA DNA Protein Protein Protein
Format protein or protein or protein
Delivery
+++ ++ ++ ++ + +++ + + ++
Efficiency
Safety
+++ + ++ + + + + + +
Concern
Cost + ++ ++ + +++ +++ + + ++
Technical
+ ++ +++ + +++ + ++ + ++
Requirement
Efficient Non-integrating;
Efficient
delivery; transient Direct delivery; FDA-approved;
Major Non- Non- delivery; No risk of No risk of
Large exposure; Dosage more Low stress to
Advantages integrating integrating Easy to virus virus
cloning multiplexible; controllable the cells
operate
capacity all-in one format
Random Technical Cell viability
Limited Limited
Major integration; immune challenging; issue; Variable efficiency depends on cell types;
cloning quantification
Limitations Insertional response in vivo work in vivo work requires extensive optimization
capacity method
mutagenesis not feasible difficult
Major in vitro and in vitro, ex vivo in vitro and ex in vitro and ex in vitro and in vitro and in vitro and
in vivo in vivo
Applications ex vivo and in vivo vivo vivo in vivo in vivo in vivo
AV, adenovirus; AAV, adeno-associated virus; EV, extracellular vesicle; LV, lentivirus; + denotes low; ++ denotes medium; +++ denotes high.
Biomolecules 2020, 10, 839 5 of 16

2. Advances in CRISPR/Cas9 Gene Editing

Researchers have been diversifying and optimizing CRISPR/Cas9 applications to achieve various
types of editing in the genome. By exploiting the right DNA repair pathway, CRISPR/Cas9 can
accomplish gene disruption, deletions, knockins, or targeted editing. The non-homologous end
joining (NHEJ) DNA repair pathway is active in any phase of the cell cycle, with increased activity
as cells progress from the G1 to G2/M phase [15]. Although the efficiency of NHEJ is high, it is
highly error-prone [16]. It frequently introduces insertions or deletions at DSBs, resulting in frameshift
mutations that disrupt gene products [16]. Therefore, the NHEJ pathway is commonly used for
generating gene knockouts [17]. A complete gene deletion can also be achieved if two sgRNAs are
used, with one targeting the start and the other targeting the end of a gene, to generate two DSBs.
In contrast, the homology-directed repair (HDR) pathway is primarily used to generate gene
knockins and for targeted gene editing [17]. The HDR pathway is a more precise pathway than
NHEJ [17]. To repair DSBs, HDR requires a donor template, comprising the knockin sequence flanked
by 50 and 30 homology arms containing homologous sequences to both sides of the DSBs. Although
the HDR pathway is less error-prone than NHEJ, it has a much lower efficiency than does NHEJ [17].
Unlike NHEJ, which is active in any phases of the cell cycle, the HDR pathway is active only in the S
and G2/M phases [18]. This limits HDR-mediated gene editing to actively dividing cells, restricting its
application in the clinical setting, as stem cells are maintained in a quiescent state [19].
The recent discovery of homology-independent targeted integration (HITI) enabled gene knockin
generation in both dividing and nondividing cells via NHEJ [20]. When a desired transgene is flanked
by the Cas9/gRNA target sequence in a donor vector, the Cas9 cleavage of the donor vector and
the genome occurs simultaneously. The DSBs generated in both the vector and the genome trigger
the NHEJ repair pathway, which integrates the desired transgene into the genome. Because NHEJ is
active in all cell cycle phases [15], HITI is a more efficient approach to generate gene knockins than is
HDR. However, several challenges must be addressed before HITI can be used in the clinical setting.
The knockin efficiency of HITI is currently less than 5% in nondividing cells in most cases [21], and
the potential off-target effects of Cas9 may lead to the integration of transgenes at off-target locations.
Nevertheless, these off-target effects can be minimized by the stringent selection of Cas9/gRNA target
sequences and the use of a high-fidelity version of Cas9 nuclease [22,23].
Microhomology-mediated end joining (MMEJ) is an alternative end-joining pathway exploited
in CRISPR/Cas9 gene editing [24]. MMEJ is active when microhomology (5–25 bp) is present
upstream and downstream of DSBs. This allows the annealing of two microhomology sequences,
resulting in the deletion of the intervening sequence [25]. Recently, Nakade et al. developed
an MMEJ-based method, termed precise integration into target chromosome (PITCh), to achieve
the targeted knockin of transgenes [24]. Cleavage by Cas9 at the PITCh donor vector and the genome
exposes their microhomology sequences, which trigger the MMEJ-mediated integration of transgenes
into the genome at the DSBs [24]. MMEJ is active during the M and early S phase, when HDR
is inactive [26]. Importantly, MMEJ is two to three times more efficient than is HDR at achieving
the targeted knockin of transgenes [24].
Irrespective of which DNA repair pathway is employed, the generation of DSBs after the Cas9
cleavage of the genome poses a safety concern about genotoxicity [27]. A base-editing strategy that
bypasses DNA cleavage has therefore revolutionized the gene editing of point mutations [28,29].
The fusion of a catalytically dead mutant Cas9 (dCas9) protein to cytidine deaminase mediates
the conversion of C > T or G > A [28], providing a promising way to correct C or G bases in
the genome [27]. However, a lack of deaminases for A or T limited the application of this technology.
Approximately 1 year later, however, adenosine deaminase was synthesized via the protein evolution
and engineering of a tRNA adenosine deaminase [29]. Adenosine deaminase was subsequently
reported to mediate the conversion of A > G or T > C in the genome [29]. Therefore, base editing by
the fusion of dCas9 to cytidine deaminase or adenosine deaminase may lead to a safe and efficient
approach to editing point mutations.
Biomolecules 2020, 10, 839 6 of 16

Although base editing can mediate the conversion of the four transition mutations, it cannot
convert transversion mutations. Recently, Anzalone et al. reported a versatile prime-editing strategy
that can achieve targeted insertions, deletions, and conversions of all 12 combinations of point mutations
without the need for a donor template [30]. Prime editing requires two components: Cas9 nickase
and a prime editing guide RNA (pegRNA) [30]. The pegRNA is an extended version of sgRNA,
containing a primer binding site to permit the hybridization of the 30 end of the nicked genomic
DNA and a reverse transcriptase (RT) template containing the desired edit to provide a template for
the synthesis of the edited information [30]. The catalytically impaired Cas9 nickase is coupled to an
RT and introduces a single-strand nick to genomic DNA to facilitate the binding of the 30 end nick
to the primer binding site of the pegRNA [30]. The RT therefore reverse transcribes the sequence
information, including the edit from the RT template to DNA [30]. Currently, prime editing appears to
be superior to other editing strategies in terms of its efficiency, genotoxicity, and versatility in gene
editing [30]. Nevertheless, further investigation of this strategy in more cell types and the optimization
of the delivery strategy is warranted.

3. Common CRISPR/Cas9 Delivery Strategies

The delivery of the CRISPR/Cas9 system for efficient gene editing is challenging. The Cas9
protein has a molecular weight of approximately 160 kDa [31], and after forming an RNP complex,
the long phosphate backbone of the sgRNA imparts a net negative charge to the complex [32]. Both
of these properties make it difficult for the Cas9 RNP to cross the cell membrane. Moreover, once
inside cells, both the Cas9 protein and sgRNA must survive the degradation processes in the cell
and translocate into the nucleus to enable gene editing. Therefore, choosing an appropriate delivery
strategy for the CRISPR/Cas9 system is critical to achieving efficient and precise gene editing. If it is to
be used in clinical settings, the safety profile must also be considered to avoid or minimize insertional
mutagenesis. To date, the delivery strategies for CRISPR/Cas9 can be broadly classified into viral
or nonviral approaches, depending on whether viral transduction is used. The nonviral approach
includes various physical and chemical delivery strategies. Each of these methods has its own pros
and cons that should be considered for each gene editing application (Table 1).

3.1. Viral Vectors

Adeno-associated viruses (AAVs) are a common viral vector used for gene delivery. The unique
properties of AAVs (i.e., replication-defective, nonintegrating into the genome, and low immunogenicity
in humans) have triggered huge interest in their potential as delivery vehicles, especially for in vivo
applications [33,34]. After transduction, the AAV genomes remain episomal in the nucleus, which are
gradually diluted by cell division. Therefore, the episomal delivery of transgenes by AAV provides
a safe option to transiently express genes [35].
CRISPR/Cas9 can be delivered by AAVs in two ways: First, AAVs can serve as a vehicle to deliver
Cas9, sgRNAs, and/or donor templates into cells by transduction [36]. Indeed, AAVs not only enable
in vivo CRISPR/Cas9 genome editing but are also useful in in vitro applications, especially when
genome integration is a safety concern and electroporation is not an option for the cell type of interest.
However, AAV vectors are limited by their low cloning capacity (< 4.7 kb). A study of metabolic liver
disease in mice used two separate AAV vectors, one expressing Cas9 and the other expressing an sgRNA
and a donor DNA sequence, to achieve a gene editing event [36]. The delivery of the commonly used
Streptococcus pyogenes Cas9 by AAVs is challenging because of its large size (~4.2 kb). A smaller strain
of Cas9 from Staphylococcus aureus (SaCas9; ~3.15 kb) is a more feasible option [37]. However, SaCas9 is
restricted by the availability of suitable PAM sequences for targeting [38]. Second, AAV vectors can be
used as a donor template for gene knockin through the HDR pathway [39,40]. The knockin efficiency
of AAV donor templates is higher than that for nonviral targeting methods [39]. Similarly, the limited
cloning capacity of recombinant AAVs can be circumvented by splitting large transgenes into two
separate AAV vectors to enable sequential homologous recombination [40]. Another disadvantage
Biomolecules 2020, 10, 839 7 of 16

of AAVs is their low efficiency in gene targeting. Specific homologous recombination only occurs in
~0.1% to 1% of the total cell population under optimal conditions [41]. Currently, the AAV-based gene
editing trials registered on ClinicalTrials.gov only use ZFNs to insert a corrective copy of the gene
into the genomes of patients with hemophilia B [42] or mucopolysaccharidosis types I and II [43].
Because AAV-based delivery is expected to become increasingly popular, clinical trials with AAV-based
CRISPR/Cas9 gene editing may be forthcoming.
Lentiviruses (LVs) are another viral vector used for CRISPR/Cas9 delivery. LV vectors have a more
generous cloning capacity (< 8 kb) than do AAV vectors, which enables the cloning of both Cas9 and
sgRNA into a single LV vector. The production of LVs is also less laborious than that of AAVs. The LV
transduction process is highly efficient in a wide variety of cell types in both dividing and nondividing
cells [44]. These advantages indicate that LV vectors are an optimal vehicle for delivery in vitro and
ex vivo [44]. However, random integration into host cell genomes is the biggest challenge associated
with LV systems. The integration of LVs in the vicinity of oncogenes may lead to their activation,
resulting in tumorigenesis [45]. This precludes the LV-mediated delivery of CRISPR/Cas9 for in vivo
gene editing in clinical trials [46]. Indeed, several tragedies in clinical trials were reported due to
insertional mutations introduced by retroviruses [47–50], indicating the potential danger of using LVs
in patients. The development of integration-defective lentiviruses with plasmids expressing mutant
integrase may increase the safety of LV transduction [51]. Nevertheless, a variable level of background
integration occurs and appears to be unavoidable [52,53].
Adenoviruses (AVs) are widely used in clinical trials for gene delivery [54]. AVs transduce both
dividing and nondividing cells and, most importantly, do not integrate into host cell genomes [54].
The major challenge of using AVs for delivery is that they trigger a high level of innate immune
responses in host cells, resulting in the inflammation of tissues and subsequent removal of AV
vectors [55]. The production of AVs is also laborious [56], which limits the application and efficiency of
this strategy.
The efficient delivery of CRISPR/Cas9 by viral vectors generally results in a higher percentage
of editing than by other methods. Although this is advantageous in most cases, in certain disease
conditions, such as retinal diseases and spinal cord injuries, a modest level of editing or reprogramming
in a fraction of the cells can achieve therapeutic effects [57,58]. Over-editing, therefore, may create
safety issues in these scenarios. The efficiency of gene editing required should be considered based on
each disease condition.

3.2. Nonviral Physical Methods

Microinjection is the physical method of injecting Cas9 and sgRNAs directly into cells with
a microscope and needle. Because the needle pierces through the cell membrane to directly deliver
the cargoes into the nucleus, the molecular weight of Cas9, which is usually an issue in viral
vector-mediated delivery due to its limited cloning capacity, is not an obstacle in microinjection.
Moreover, manual injection enables the controlled dosing of cargoes into cells. However, microinjection
is laborious and technically challenging, rendering this technique low throughput. Furthermore,
the requirement of a microscope for injection excludes this technique from in vivo patient work. Indeed,
most microinjection applications are used in animal zygotes for the generation of transgenic animal
models [59–61].
Electroporation is a popular physical delivery method. It applies pulses of electrical currents to
stimulate the transient opening of pores in cell membranes, permitting the delivery of cargoes into
cells. Electroporation is commonly used in in vitro and ex vivo gene editing because it efficiently
delivers cargoes into a wide variety of cell types. This is advantageous over standard transfection
methods, which are usually hampered in difficult-to-transfect cell types, such as primary cells.
Indeed, electroporation-mediated ex vivo gene editing fostered the development of stem cell therapies,
especially for the treatment of hematologic malignancies [62,63]. Patient hematopoietic stem/progenitor
cells after ex vivo modification are transplanted back into patients for treatment [63]. Although in vivo
Biomolecules 2020, 10, 839 8 of 16

electroporators are currently available and are reported to successfully accomplish gene editing in
animals [64–66], the application of electroporation in patients for in vivo gene editing is still not
generally feasible. Moreover, the cost associated with electroporation-mediated gene editing is
usually high because the extensive optimization of Cas9-to-sgRNA ratios and specific electroporation
conditions for each cell type are required. Importantly, the strong electrical current generated by
electroporation results in a high percentage of cell death, indicating that this method may not be
suitable for stress-sensitive cell types.

3.3. Nonviral Chemical Methods

Lipid-based nanoparticles (LNPs) are commonly used for nucleic acid delivery [67]. Liposomes
are spherical structures composed of lipid bilayers formed in aqueous solutions. Because both nucleic
acids and cell membranes are negatively charged, the repulsion between them prevents the entry
of nucleic acids into cells. The encapsulation of negatively charged nucleic acids into positively
charged liposomes thereby facilitates the fusion of the complexes across cell membranes into cells [67].
The CRISPR/Cas9 system can be delivered in the format of DNA (“all-in-one” plasmid), mRNA (Cas9
and sgRNA), or protein (RNP). Currently, the Lipofectamine reagent is the most popular choice for
LNP formation. The successful delivery of the CRISPR/Cas9 system with Lipofectamine transfection
in vitro and in vivo for gene editing has been demonstrated [12,68–71]. Non-lipid polymeric reagents,
such as polyethylenimine and poly-L-lysine, are also commonly used to generate nanoparticles for
the delivery of CRISPR/Cas9 cargoes [72,73]. Similarly, polymeric reagents mediate the encapsulation of
CRISPR/Cas9 cargoes into positively charged complexes to enable endocytosis into cells [74]. Because
viruses are not involved in nanoparticle-mediated delivery, this method is a safer alternative. Moreover,
LNPs do not exert stress on cells to the same extent as electroporation. Consequently, this delivery
system is approved by the US Food and Drug Administration for drug delivery [75]. Indeed, both lipid-
and polymer-based reagents were used to deliver CRISPR/Cas9 in clinical trials for the treatment of
various diseases [10]. Nevertheless, the efficiency of this strategy, which relies solely on the endosomal
pathway, is low when compared to that of viral transduction and electroporation. For example,
chemical transfection methods result in less than 10% of eGFP expression in human embryonic stem
cells. Transgene expression also decreases over time after each cell division [76,77]. This limitation
restricts the application of this strategy to certain cell types.
Cell-penetrating peptides (CPPs) are short peptides with an intrinsic ability to translocate across cell
membranes. They have been exploited to facilitate the delivery of a variety of cargoes into cells [78–80].
CPPs can be conjugated to Cas9 and sgRNAs separately [8,81] or, in most cases, conjugated only
to Cas9 followed by complexing with sgRNAs to form RNPs before delivery [81,82]. CPPs offer an
ostensibly safer option for Cas9 RNP delivery because random integration and insertional mutagenesis
are not factors. However, CPP-based delivery is inefficient when compared to that of viral and physical
methods, resulting in a low percentage of gene editing [83]. Although CPP-based deliveries work well
in vitro and ex vivo [8,78–80,82], the involvement of multiple parameters at different stages (CPPs,
Cas9, sgRNA, and each cell type) introduces variability and requires extensive optimization. Hence,
CPP-based delivery is not ideal for achieving efficient in vivo gene editing.
Gold nanoparticles (AuNPs) can efficiently deliver Cas9 RNPs for gene editing [84,85]. Being
chemically inert [86], AuNPs do not trigger an immune response after delivery [85], which increases
their safety profile. In the CRISPR-Gold system developed by Lee et al., AuNPs (15 nm) are first
conjugated to 50 thiol modified single-stranded DNA sequences hybridized to single-stranded donor
DNA. This is followed by the loading of Cas9 RNPs to donor DNA and then coating with whole
particles of silicate and the polymer PAsp(DET) [85]. CRISPR-Gold was demonstrated to induce
HDR in cell lines and primary cells with approximately 4% efficiency [85]. These results indicate that
CRISPR-Gold is more efficient than Lipofectamine transfection or nucleofection for inducing HDR
in vitro [85]. Moreover, intramuscular injection of CRISPR-Gold in mice induced HDR to correct a point
mutation in the dystrophin gene in vivo at approximately 5% efficiency [85]. Although additional
Biomolecules 2020, 10, 839 9 of 16

research is required to improve the editing efficiency of this technique, it provides a safer alternative to
viral approaches for HDR-mediated gene editing.

4. Delivery of CRISPR/Cas9 Gene Editing by Extracellular Vesicles

Viral vector-based delivery is highly efficient, but LVs carry the risk of insertional mutagenesis and
AAVs have a limited cloning capacity for cargoes (Table 1). Physical and chemical methods, however,
are generally not effective in vivo, which limits their usage in clinical trials (Table 1). Recently, several
studies reported the efficient delivery of Cas9 RNPs in vitro and in vivo by EVs [87–91], indicating
the potential of using this platform in clinical settings.
The formation of EVs is based on the expression and self-assembly of the viral envelope and/or
viral structural proteins (Figure 1B) [90,92,93]. The term virus-like particles (VLPs) is used if both
the viral envelope and viral structural proteins are assembled into EVs. The protein of interest is
fused to the Gag polyprotein so that they can be incorporated simultaneously into a particle. During
the viral maturation process, the expression of the protease from Pol mediates cleavage along the Gag
polyprotein, releasing the protein of interest for delivery [94]. In contrast, the term vesicles should be
used when only the viral envelope proteins, such as the envelope glycoprotein of the vesicular stomatitis
virus (VSV-G), are assembled into a particle. Vesicles contain no or minimal viral structural proteins.
The formation of vesicles is not dependent on the Gag polyprotein and protease cleavage. Vesicles are
formed from the budding of the cell membrane when viral envelope proteins are overexpressed [90,92].
Nevertheless, in contrast with LVs, EVs do not contain any viral genome (Figure 1B). Therefore, EVs are
not integrated into host genomes and do not replicate [95]. This superior safety feature of EVs renders
them a safe version of viral delivery. Moreover, the transient exposure of Cas9 by EVs in cells greatly
reduces the chance of off-target effects due to long-term Cas9 expression [9]. The production of EVs is
also simple and cost-effective, because only the standard transfection of plasmids into packaging cells
is required. Indeed, VLPs have been exploited extensively in vaccine development [95]. VLP-based
vaccines deliver antigenic proteins into hosts to provoke an immune response, which is much safer than
the traditional approach of using attenuated viruses [95,96]. The delivery of other proteins, including
fluorophores, Cre recombinase, and human caspase 8 by VLPs has also been demonstrated [97]. Thus,
current research efforts are shifting to Cas9 protein delivery with EVs [87–91].
To reduce the off-target effects introduced by CRISPR/Cas9, Choi et al. demonstrated the packaging
of the Cas9 protein into VLPs for transient Cas9 exposure in cells [87]. The Cas9 gene was fused
to the Gag gene in a nonviral Gag/Pol expression vector. VLPs were produced by co-transfection
with wild-type Gag/Pol and VSV-G plasmids. The protease cleavage site between Cas9 and Gag in
the fusion plasmid allowed the release of the Cas9 protein during viral maturation [87]. Importantly,
VLPs preloaded with Cas9 reduced off-target effects, as compared with LVs expressing Cas9 after
integration into genomes [87], suggesting that transient Cas9 exposure with VLPs is advantageous.
This is the first study to demonstrate the packaging of Cas9 into VLPs for gene editing, although it did
not show the preloading of sgRNAs to Cas9 to form RNPs before delivery [87].
Mangeot et al. later developed this approach for gene editing with nanoblades, which are all-in-one
VLPs preloaded with Cas9 and sgRNAs [88]. With this strategy, donor templates can also be loaded
in VLPs for HDR [88]. Nanoblades were produced by the co-transfection of the retroviral murine
leukemia virus Gag-Cas9 expression plasmid with the wild-type Gag-Pol, sgRNA, VSV-G, and baboon
envelope plasmids [88]. Nanoblade-mediated gene editing was demonstrated in vitro in primary cells
and induced pluripotent stem cells (iPSCs), as well as in vivo in mice [88]. The possibility of loading
different sgRNAs into each nanoblade particle also suggested that gene editing is multiplexible [88].
The results of this study have opened the door for VLP-mediated gene editing in vitro and in vivo.
Compared to vesicle production, the wild-type Gag-Pol plasmid is required in VLP production
in order to provide protease to release Cas9. However, the competition between wild-type Gag
and Gag-Cas9 proteins during VLP packaging restricts the dosage of the Cas9 protein that can be
Biomolecules 2020, 10, 839 10 of 16

delivered [93]. Moreover, protease cleavage at the cryptic sites present in Cas9 may occur, resulting in
its degradation [98].
Several studies demonstrated the production of vesicles pre-loaded with Cas9 RNPs for gene
editing [89,90]. Campbell et al. used the Takara Guide-it CRISPR/Cas9 Gesicle Production System to
produce gene editing vesicles, which are called gesicles, through the overexpression of VSV-G and
the interaction between Cherry Picker red proteins and Cas9 RNPs [89]. In contrast, Montagna et
al. produced gene editing vesicles, which they termed VEsiCas9, by the co-transfection of the HIV-1
Gag-Cas9 expression plasmid with sgRNA and VSV-G plasmids [90]. The production in both studies
used the passive incorporation of Cas9 into the vesicles [89,90]. Recently, Gee et al. developed a robust
production system of active Cas9 incorporation into vesicles, which they termed NanoMEDIC [91].
Cas9 incorporation into a vesicle was induced by the addition of a ligand (AP21967) to trigger
the specific interaction between the FRB and FKBP12 domains [91]. In addition to gene editing,
Gee et al. demonstrated that NanoMEDIC successfully induced exon skipping in vitro and in vivo
in different models of Duchenne muscular dystrophy [91]. These results further demonstrated
the applicability of vesicles for in vivo therapy.
Taken together, these studies demonstrate the superior safety features of vesicle-mediated Cas9
RNP delivery, which do not integrate into the genome and provide transient Cas9 exposure [89–91].
Montagna et al. and Gee et al. also demonstrated the possibility of multiplexed gene editing with
vesicles [89,90]. Because protease cleavage is not required in vesicle production, the vesicle system
is not hampered by protease-mediated protein degradation or the competition between wild-type
Gag and Gag-Cas9 proteins during packaging, which are the limitations in the VLP-based system.
However, no data are currently available for the successful packaging of donor templates into vesicles
for HDR-mediated gene editing. Further investigation is warranted to develop complete all-in-one
preloaded vesicles for gene editing.

5. Conclusions and Future Perspectives

CRISPR/Cas9 gene editing technology is well-developed for in vitro applications. Without
safety concerns in patients, most of the delivery strategies are suitable options for efficient editing.
The off-target effects due to long-term Cas9 exposure are the biggest challenges for in vitro experiments.
LV transduction results in permanent Cas9 expression, which increases the likelihood of off-target
effects over that of other methods that impart transient Cas9 expression [9,87]. The delivery of plasmid
DNA expressing Cas9 is more prone to off-target effects than is the direct delivery of the Cas9 protein
by physical or chemical methods [7,8]. Longer Cas9 exposures to genomes are more susceptible to
off-target cleavage [9,87]. When a transient system is used to reduce the duration of Cas9 exposure,
the editing specificity of Cas9 is increased [9,87]. Because the EV-based delivery of Cas9 RNPs is
transient, it outperforms many delivery methods by reducing off-target effects. Moreover, cell viability
is not affected after transduction by EVs, but high levels of cell death are observed after electroporation.
EVs also have the efficiency of viral systems. Therefore, EV-based systems are generally more efficient,
and require less optimization, than do chemical-based systems.
The delivery of CRISPR/Cas9 is also rapidly progressing in ex vivo applications. The treatment of
hematopoietic diseases is made possible by harvesting patient hematopoietic stem cells for ex vivo
modification before the autologous transplantation of the edited/modified cells back into patients [99].
However, the cost and effort to harvest the stem cells from each patient for autologous transplants
negate the benefits of ex vivo cell therapy. Therefore, iPSCs have quickly become a popular cell
platform for gene editing research [100], and which provide an unlimited supply of materials for
gene editing and can be generated from easily accessible cell types, such as fibroblasts and peripheral
blood cells [100]. Recently, the idea of universal donor cells has directed even more attention to
the therapeutic values of iPSCs. By knocking out human leukocyte antigen class I and II in iPSCs,
the cell products generated after differentiation are considered “off the shelf” and compatible with all
patients [101,102]. This will streamline the process of ex vivo cell therapies. LVs and electroporation are
Biomolecules 2020, 10, 839 11 of 16

commonly used to deliver CRISPR/Cas9 ex vivo [103], but LVs have safety issues and electroporation
causes high levels of cell death. In the future, further development of EV-based delivery systems will
offer a safe and efficient gene editing in ex vivo cell therapy.
To date, the CRISPR/Cas9 system is not widely used in clinical trials. The trials that have
investigated CRISPR/Cas9 gene editing to date only do so at the ex vivo level, by either modifying
stem cells or T cells before transplanting them back into patients (ClinicalTrials.gov). Most current
Cas9 delivery methods either have safety issues or low efficiency that preclude them from in vivo
applications in patients. Based on the results from recent studies, EVs offer a transient, multiplexible,
and all-in-one delivery platform for gene editing. Non-integrating EVs have no risk of insertional
mutagenesis. The transient exposure of Cas9 to the cells dramatically reduces the chance of off-target
effects. However, all of these studies used ultracentrifugation to concentrate EVs [87–90], which is not
a scalable method or compatible with good manufacturing practice settings. Further investigation
using scalable methods of purification and concentration, such as fast protein liquid chromatography,
is needed. Because the genome quantitation of the titers of EVs is not possible with PCR-based assays,
these studies used Western blots or dot blots to quantify the amount of Cas9 protein in EVs [87–90].
However, these methods may not be sensitive enough to detect low concentrations of Cas9. Moreover,
the amount of Cas9 protein detected, which varies by the number of Cas9 protein copies packaged into
each EV particle, does not reflect the actual titers of EVs produced. Without knowing the actual yield of
EVs, the optimization of their production is difficult. Therefore, a precise method for EV quantitation
is needed. Furthermore, EV-based systems have not yet been demonstrated to deliver base editors for
gene editing. Because base editing and prime editing are safer approaches than conventional editing,
in terms of genotoxicity, investigation into the possibility of their delivery by EVs is crucial. Taken
together, CRISPR/Cas9-mediated gene editing is evolving rapidly, including its delivery methods.
The EV-based system offers a safe and transient method to deliver Cas9 in vitro and in vivo. Further
development and the optimization of this delivery platform will open the door for CRISPR/Cas9 gene
editing in future clinical trials.

Funding: This review received no external funding.

Conflicts of Interest: The author declares no conflict of interest.

References
1. Lander, E.S. The Heroes of CRISPR. Cell 2016, 164, 18–28. [CrossRef] [PubMed]
2. Mali, P.; Yang, L.; Esvelt, K.M.; Aach, J.; Guell, M.; DiCarlo, J.E.; Norville, J.E.; Church, G.M. RNA-guided
human genome engineering via Cas9. Science 2013, 339, 823–826. [CrossRef] [PubMed]
3. Cong, L.; Ran, F.A.; Cox, D.; Lin, S.; Barretto, R.; Habib, N.; Hsu, P.D.; Wu, X.; Jiang, W.; Marraffini, L.A.; et al.
Multiplex genome engineering using CRISPR/Cas systems. Science 2013, 339, 819–823. [CrossRef] [PubMed]
4. Doudna, J.A.; Charpentier, E. Genome editing. The new frontier of genome engineering with CRISPR-Cas9.
Science 2014, 346, 1258096. [CrossRef] [PubMed]
5. Fu, Y.; Sander, J.D.; Reyon, D.; Cascio, V.M.; Joung, J.K. Improving CRISPR-Cas nuclease specificity using
truncated guide RNAs. Nat. Biotechnol. 2014, 32, 279–284. [CrossRef]
6. Ran, F.A.; Hsu, P.D.; Lin, C.Y.; Gootenberg, J.S.; Konermann, S.; Trevino, A.E.; Scott, D.A.; Inoue, A.;
Matoba, S.; Zhang, Y.; et al. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing
specificity. Cell 2013, 154, 1380–1389. [CrossRef]
7. Kim, S.; Kim, D.; Cho, S.W.; Kim, J.; Kim, J.S. Highly efficient RNA-guided genome editing in human cells
via delivery of purified Cas9 ribonucleoproteins. Genome Res. 2014, 24, 1012–1019. [CrossRef]
8. Ramakrishna, S.; Kwaku Dad, A.B.; Beloor, J.; Gopalappa, R.; Lee, S.K.; Kim, H. Gene disruption by
cell-penetrating peptide-mediated delivery of Cas9 protein and guide RNA. Genome Res. 2014, 24, 1020–1027.
[CrossRef]
9. Wu, X.; Kriz, A.J.; Sharp, P.A. Target specificity of the CRISPR-Cas9 system. Quant. Biol. 2014, 2, 59–70.
[CrossRef]
Biomolecules 2020, 10, 839 12 of 16

10. Chen, F.; Alphonse, M.; Liu, Q. Strategies for nonviral nanoparticle-based delivery of CRISPR/Cas9
therapeutics. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2020, 12, e1609. [CrossRef]
11. Yin, H.; Song, C.Q.; Suresh, S.; Wu, Q.; Walsh, S.; Rhym, L.H.; Mintzer, E.; Bolukbasi, M.F.; Zhu, L.J.;
Kauffman, K.; et al. Structure-guided chemical modification of guide RNA enables potent non-viral in vivo
genome editing. Nat. Biotechnol. 2017, 35, 1179–1187. [CrossRef] [PubMed]
12. Liang, X.; Potter, J.; Kumar, S.; Zou, Y.; Quintanilla, R.; Sridharan, M.; Carte, J.; Chen, W.; Roark, N.;
Ranganathan, S.; et al. Rapid and highly efficient mammalian cell engineering via Cas9 protein transfection.
J. Biotechnol. 2015, 208, 44–53. [CrossRef]
13. You, L.; Tong, R.; Li, M.; Liu, Y.; Xue, J.; Lu, Y. Advancements and Obstacles of CRISPR-Cas9 Technology in
Translational Research. Mol. Methods Clin. Dev. 2019, 13, 359–370. [CrossRef] [PubMed]
14. Cyranoski, D. Chinese scientists to pioneer first human CRISPR trial. Nature 2016, 535, 476–477. [CrossRef]
15. Mao, Z.; Bozzella, M.; Seluanov, A.; Gorbunova, V. DNA repair by nonhomologous end joining and
homologous recombination during cell cycle in human cells. Cell Cycle 2008, 7, 2902–2906. [CrossRef]
[PubMed]
16. Ceccaldi, R.; Rondinelli, B.; D’Andrea, A.D. Repair Pathway Choices and Consequences at the Double-Strand
Break. Trends Cell Biol. 2016, 26, 52–64. [CrossRef]
17. Ran, F.A.; Hsu, P.D.; Wright, J.; Agarwala, V.; Scott, D.A.; Zhang, F. Genome engineering using
the CRISPR-Cas9 system. Nat. Protoc. 2013, 8, 2281–2308. [CrossRef]
18. Ranjha, L.; Howard, S.M.; Cejka, P. Main steps in DNA double-strand break repair: An introduction to
homologous recombination and related processes. Chromosoma 2018, 127, 187–214. [CrossRef]
19. Li, L.; Bhatia, R. Stem cell quiescence. Clin. Cancer Res. 2011, 17, 4936–4941. [CrossRef]
20. Suzuki, K.; Tsunekawa, Y.; Hernandez-Benitez, R.; Wu, J.; Zhu, J.; Kim, E.J.; Hatanaka, F.; Yamamoto, M.;
Araoka, T.; Li, Z.; et al. In vivo genome editing via CRISPR/Cas9 mediated homology-independent targeted
integration. Nature 2016, 540, 144–149. [CrossRef]
21. Suzuki, K.; Izpisua Belmonte, J.C. In vivo genome editing via the HITI method as a tool for gene therapy. J.
Hum. Genet. 2018, 63, 157–164. [CrossRef] [PubMed]
22. Vakulskas, C.A.; Dever, D.P.; Rettig, G.R.; Turk, R.; Jacobi, A.M.; Collingwood, M.A.; Bode, N.M.; McNeill, M.S.;
Yan, S.; Camarena, J.; et al. A high-fidelity Cas9 mutant delivered as a ribonucleoprotein complex enables
efficient gene editing in human hematopoietic stem and progenitor cells. Nat. Med. 2018, 24, 1216–1224.
[CrossRef] [PubMed]
23. Kleinstiver, B.P.; Pattanayak, V.; Prew, M.S.; Tsai, S.Q.; Nguyen, N.T.; Zheng, Z.; Joung, J.K. High-fidelity
CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects. Nature 2016, 529, 490–495.
[CrossRef] [PubMed]
24. Nakade, S.; Tsubota, T.; Sakane, Y.; Kume, S.; Sakamoto, N.; Obara, M.; Daimon, T.; Sezutsu, H.; Yamamoto, T.;
Sakuma, T.; et al. Microhomology-mediated end-joining-dependent integration of donor DNA in cells and
animals using TALENs and CRISPR/Cas9. Nat. Commun. 2014, 5, 5560. [CrossRef] [PubMed]
25. Kim, S.I.; Matsumoto, T.; Kagawa, H.; Nakamura, M.; Hirohata, R.; Ueno, A.; Ohishi, M.; Sakuma, T.; Soga, T.;
Yamamoto, T.; et al. Microhomology-assisted scarless genome editing in human iPSCs. Nat. Commun. 2018,
9, 939. [CrossRef] [PubMed]
26. Sakuma, T.; Nakade, S.; Sakane, Y.; Suzuki, K.T.; Yamamoto, T. MMEJ-assisted gene knock-in using TALENs
and CRISPR-Cas9 with the PITCh systems. Nat. Protoc. 2016, 11, 118–133. [CrossRef]
27. Bukowska, B.; Karwowski, B.T. The Clustered DNA Lesions—Types, Pathways of Repair and Relevance to
Human Health. Curr. Med. Chem. 2018, 25, 2722–2735. [CrossRef]
28. Komor, A.C.; Kim, Y.B.; Packer, M.S.; Zuris, J.A.; Liu, D.R. Programmable editing of a target base in genomic
DNA without double-stranded DNA cleavage. Nature 2016, 533, 420–424. [CrossRef]
29. Gaudelli, N.M.; Komor, A.C.; Rees, H.A.; Packer, M.S.; Badran, A.H.; Bryson, D.I.; Liu, D.R. Programmable
base editing of A*T to G*C in genomic DNA without DNA cleavage. Nature 2017, 551, 464–471. [CrossRef]
30. Anzalone, A.V.; Randolph, P.B.; Davis, J.R.; Sousa, A.A.; Koblan, L.W.; Levy, J.M.; Chen, P.J.; Wilson, C.;
Newby, G.A.; Raguram, A.; et al. Search-and-replace genome editing without double-strand breaks or donor
DNA. Nature 2019, 576, 149–157. [CrossRef]
31. Jinek, M.; Jiang, F.; Taylor, D.W.; Sternberg, S.H.; Kaya, E.; Ma, E.; Anders, C.; Hauer, M.; Zhou, K.; Lin, S.;
et al. Structures of Cas9 endonucleases reveal RNA-mediated conformational activation. Science 2014, 343,
1247997. [CrossRef] [PubMed]
Biomolecules 2020, 10, 839 13 of 16

32. Sun, W.; Ji, W.; Hall, J.M.; Hu, Q.; Wang, C.; Beisel, C.L.; Gu, Z. Self-assembled DNA nanoclews for the efficient
delivery of CRISPR-Cas9 for genome editing. Angew. Chem. Int. Ed. Engl. 2015, 54, 12029–12033. [CrossRef]
[PubMed]
33. Carter, B.J. Adeno-associated virus and the development of adeno-associated virus vectors: A historical
perspective. Mol. Ther. 2004, 10, 981–989. [CrossRef] [PubMed]
34. Hastie, E.; Samulski, R.J. Adeno-associated virus at 50: A golden anniversary of discovery, research, and
gene therapy success—A personal perspective. Hum. Gene Ther. 2015, 26, 257–265. [CrossRef] [PubMed]
35. Daya, S.; Berns, K.I. Gene therapy using adeno-associated virus vectors. Clin. Microbiol. Rev. 2008, 21,
583–593. [CrossRef] [PubMed]
36. Yang, Y.; Wang, L.; Bell, P.; McMenamin, D.; He, Z.; White, J.; Yu, H.; Xu, C.; Morizono, H.; Musunuru, K.;
et al. A dual AAV system enables the Cas9-mediated correction of a metabolic liver disease in newborn mice.
Nat. Biotechnol. 2016, 34, 334–338. [CrossRef]
37. Ran, F.A.; Cong, L.; Yan, W.X.; Scott, D.A.; Gootenberg, J.S.; Kriz, A.J.; Zetsche, B.; Shalem, O.; Wu, X.;
Makarova, K.S.; et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 2015, 520, 186–191.
[CrossRef]
38. Ma, D.; Xu, Z.; Zhang, Z.; Chen, X.; Zeng, X.; Zhang, Y.; Deng, T.; Ren, M.; Sun, Z.; Jiang, R.; et al. Engineer
chimeric Cas9 to expand PAM recognition based on evolutionary information. Nat. Commun. 2019, 10, 560.
[CrossRef]
39. Eyquem, J.; Mansilla-Soto, J.; Giavridis, T.; van der Stegen, S.J.; Hamieh, M.; Cunanan, K.M.; Odak, A.;
Gonen, M.; Sadelain, M. Targeting a CAR to the TRAC locus with CRISPR/Cas9 enhances tumour rejection.
Nature 2017, 543, 113–117. [CrossRef]
40. Bak, R.O.; Porteus, M.H. CRISPR-Mediated Integration of Large Gene Cassettes Using AAV Donor Vectors.
Cell Rep. 2017, 20, 750–756. [CrossRef]
41. Deyle, D.R.; Russell, D.W. Adeno-associated virus vector integration. Curr. Opin. Mol. 2009, 11, 442–447.
42. George, L.A. Hemophilia gene therapy comes of age. Blood Adv. 2017, 1, 2591–2599. [CrossRef] [PubMed]
43. Sawamoto, K.; Chen, H.H.; Almeciga-Diaz, C.J.; Mason, R.W.; Tomatsu, S. Gene therapy for
Mucopolysaccharidoses. Mol. Genet. Metab. 2018, 123, 59–68. [CrossRef] [PubMed]
44. Kotterman, M.A.; Chalberg, T.W.; Schaffer, D.V. Viral Vectors for Gene Therapy: Translational and Clinical
Outlook. Annu. Rev. Biomed. Eng. 2015, 17, 63–89. [CrossRef] [PubMed]
45. Popescu, N.C.; Zimonjic, D.; DiPaolo, J.A. Viral integration, fragile sites, and proto-oncogenes in human
neoplasia. Hum. Genet. 1990, 84, 383–386. [CrossRef]
46. Rothe, M.; Modlich, U.; Schambach, A. Biosafety challenges for use of lentiviral vectors in gene therapy. Curr.
Gene Ther. 2013, 13, 453–468. [CrossRef]
47. Check, E. Gene therapy put on hold as third child develops cancer. Nature 2005, 433, 561. [CrossRef]
48. Hacein-Bey-Abina, S.; Garrigue, A.; Wang, G.P.; Soulier, J.; Lim, A.; Morillon, E.; Clappier, E.; Caccavelli, L.;
Delabesse, E.; Beldjord, K.; et al. Insertional oncogenesis in 4 patients after retrovirus-mediated gene therapy
of SCID-X1. J. Clin. Investig. 2008, 118, 3132–3142. [CrossRef]
49. Hacein-Bey-Abina, S.; Von Kalle, C.; Schmidt, M.; McCormack, M.P.; Wulffraat, N.; Leboulch, P.; Lim, A.;
Osborne, C.S.; Pawliuk, R.; Morillon, E.; et al. LMO2-associated clonal T cell proliferation in two patients
after gene therapy for SCID-X1. Science 2003, 302, 415–419. [CrossRef]
50. Check, E. Regulators split on gene therapy as patient shows signs of cancer. Nature 2002, 419, 545–546.
[CrossRef]
51. Liu, K.C.; Lin, B.S.; Gao, A.D.; Ma, H.Y.; Zhao, M.; Zhang, R.; Yan, H.H.; Yi, X.F.; Lin, S.J.; Que, J.W.; et al.
Integrase-deficient lentivirus: Opportunities and challenges for human gene therapy. Curr. Gene Ther. 2014,
14, 352–364. [CrossRef] [PubMed]
52. Banasik, M.B.; McCray, P.B., Jr. Integrase-defective lentiviral vectors: Progress and applications. Gene Ther.
2010, 17, 150–157. [CrossRef] [PubMed]
53. Qamar Saeed, M.; Dufour, N.; Bartholmae, C.; Sieranska, U.; Knopf, M.; Thierry, E.; Thierry, S.; Delelis, O.;
Grandchamp, N.; Pilet, H.; et al. Comparison Between Several Integrase-defective Lentiviral Vectors Reveals
Increased Integration of an HIV Vector Bearing a D167H Mutant. Mol. Nucleic Acids 2014, 3, e213. [CrossRef]
[PubMed]
Biomolecules 2020, 10, 839 14 of 16

54. Lee, C.S.; Bishop, E.S.; Zhang, R.; Yu, X.; Farina, E.M.; Yan, S.; Zhao, C.; Zheng, Z.; Shu, Y.; Wu, X.; et al.
Adenovirus-Mediated Gene Delivery: Potential Applications for Gene and Cell-Based Therapies in the New
Era of Personalized Medicine. Genes Dis. 2017, 4, 43–63. [CrossRef]
55. Muruve, D.A. The innate immune response to adenovirus vectors. Hum. Gene 2004, 15, 1157–1166. [CrossRef]
56. Imperiale, M.J.; Kochanek, S. Adenovirus vectors: Biology, design, and production. Curr. Top. Microbiol.
Immunol. 2004, 273, 335–357.
57. Cai, B.; Sun, S.; Li, Z.; Zhang, X.; Ke, Y.; Yang, J.; Li, X. Application of CRISPR/Cas9 technologies combined
with iPSCs in the study and treatment of retinal degenerative diseases. Hum. Genet. 2018, 137, 679–688.
[CrossRef]
58. Su, Z.; Niu, W.; Liu, M.L.; Zou, Y.; Zhang, C.L. In vivo conversion of astrocytes to neurons in the injured
adult spinal cord. Nat. Commun. 2014, 5, 3338. [CrossRef]
59. Horii, T.; Arai, Y.; Yamazaki, M.; Morita, S.; Kimura, M.; Itoh, M.; Abe, Y.; Hatada, I. Validation of
microinjection methods for generating knockout mice by CRISPR/Cas-mediated genome engineering. Sci.
Rep. 2014, 4, 4513. [CrossRef]
60. Ma, Y.; Shen, B.; Zhang, X.; Lu, Y.; Chen, W.; Ma, J.; Huang, X.; Zhang, L. Heritable multiplex genetic
engineering in rats using CRISPR/Cas9. PLoS ONE 2014, 9, e89413. [CrossRef]
61. Long, C.; McAnally, J.R.; Shelton, J.M.; Mireault, A.A.; Bassel-Duby, R.; Olson, E.N. Prevention of muscular
dystrophy in mice by CRISPR/Cas9-mediated editing of germline DNA. Science 2014, 345, 1184–1188.
[CrossRef] [PubMed]
62. Dever, D.P.; Bak, R.O.; Reinisch, A.; Camarena, J.; Washington, G.; Nicolas, C.E.; Pavel-Dinu, M.; Saxena, N.;
Wilkens, A.B.; Mantri, S.; et al. CRISPR/Cas9 beta-globin gene targeting in human haematopoietic stem cells.
Nature 2016, 539, 384–389. [CrossRef] [PubMed]
63. Romero, Z.; DeWitt, M.; Walters, M.C. Promise of gene therapy to treat sickle cell disease. Expert Opin. Biol.
2018, 18, 1123–1136. [CrossRef] [PubMed]
64. de Melo, J.; Blackshaw, S. In Vivo Electroporation of Developing Mouse Retina. Methods Mol. Biol. 2018,
1715, 101–111.
65. Saito, T. In vivo electroporation in the embryonic mouse central nervous system. Nat. Protoc. 2006, 1,
1552–1558. [CrossRef]
66. Li, Q.; Qian, C.; Zhou, F.Q. Investigating Mammalian Axon Regeneration: In Vivo Electroporation of Adult
Mouse Dorsal Root Ganglion. J. Vis. Exp. 2018. [CrossRef]
67. Pensado, A.; Seijo, B.; Sanchez, A. Current strategies for DNA therapy based on lipid nanocarriers. Expert
Opin. Drug Deliv. 2014, 11, 1721–1731. [CrossRef]
68. Horii, T.; Tamura, D.; Morita, S.; Kimura, M.; Hatada, I. Generation of an ICF syndrome model by efficient
genome editing of human induced pluripotent stem cells using the CRISPR system. Int. J. Mol. Sci. 2013, 14,
19774–19781. [CrossRef]
69. Sakuma, T.; Nishikawa, A.; Kume, S.; Chayama, K.; Yamamoto, T. Multiplex genome engineering in human
cells using all-in-one CRISPR/Cas9 vector system. Sci. Rep. 2014, 4, 5400. [CrossRef]
70. Schwank, G.; Koo, B.K.; Sasselli, V.; Dekkers, J.F.; Heo, I.; Demircan, T.; Sasaki, N.; Boymans, S.; Cuppen, E.;
van der Ent, C.K.; et al. Functional repair of CFTR by CRISPR/Cas9 in intestinal stem cell organoids of cystic
fibrosis patients. Cell Stem Cell 2013, 13, 653–658. [CrossRef]
71. Zuris, J.A.; Thompson, D.B.; Shu, Y.; Guilinger, J.P.; Bessen, J.L.; Hu, J.H.; Maeder, M.L.; Joung, J.K.; Chen, Z.Y.;
Liu, D.R. Cationic lipid-mediated delivery of proteins enables efficient protein-based genome editing in vitro
and in vivo. Nat. Biotechnol. 2015, 33, 73–80. [CrossRef] [PubMed]
72. Zuckermann, M.; Hovestadt, V.; Knobbe-Thomsen, C.B.; Zapatka, M.; Northcott, P.A.; Schramm, K.; Belic, J.;
Jones, D.T.; Tschida, B.; Moriarity, B.; et al. Somatic CRISPR/Cas9-mediated tumour suppressor disruption
enables versatile brain tumour modelling. Nat. Commun. 2015, 6, 7391. [CrossRef] [PubMed]
73. Zhen, S.; Hua, L.; Liu, Y.H.; Gao, L.C.; Fu, J.; Wan, D.Y.; Dong, L.H.; Song, H.F.; Gao, X. Harnessing
the clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated Cas9 system to
disrupt the hepatitis B virus. Gene Ther. 2015, 22, 404–412. [CrossRef] [PubMed]
74. Longo, P.A.; Kavran, J.M.; Kim, M.S.; Leahy, D.J. Transient mammalian cell transfection with polyethylenimine
(PEI). Methods Enzym. 2013, 529, 227–240.
75. Allen, T.M.; Cullis, P.R. Liposomal drug delivery systems: From concept to clinical applications. Adv. Drug
Deliv. Rev. 2013, 65, 36–48. [CrossRef]
Biomolecules 2020, 10, 839 15 of 16

76. Eiges, R.; Schuldiner, M.; Drukker, M.; Yanuka, O.; Itskovitz-Eldor, J.; Benvenisty, N. Establishment of human
embryonic stem cell-transfected clones carrying a marker for undifferentiated cells. Curr. Biol. 2001, 11,
514–518. [CrossRef]
77. Moore, J.C.; van Laake, L.W.; Braam, S.R.; Xue, T.; Tsang, S.Y.; Ward, D.; Passier, R.; Tertoolen, L.L.; Li, R.A.;
Mummery, C.L. Human embryonic stem cells: Genetic manipulation on the way to cardiac cell therapies.
Reprod. Toxicol. 2005, 20, 377–391. [CrossRef]
78. Henriques, S.T.; Costa, J.; Castanho, M.A. Translocation of beta-galactosidase mediated by the cell-penetrating
peptide pep-1 into lipid vesicles and human HeLa cells is driven by membrane electrostatic potential.
Biochemistry 2005, 44, 10189–10198. [CrossRef]
79. Presente, A.; Dowdy, S.F. PTD/CPP peptide-mediated delivery of siRNAs. Curr. Pharm. Des. 2013, 19,
2943–2947. [CrossRef]
80. Lim, S.; Koo, J.H.; Choi, J.M. Use of Cell-Penetrating Peptides in Dendritic Cell-Based Vaccination. Immune
Netw. 2016, 16, 33–43. [CrossRef]
81. Suresh, B.; Ramakrishna, S.; Kim, H. Cell-Penetrating Peptide-Mediated Delivery of Cas9 Protein and Guide
RNA for Genome Editing. Methods Mol. Biol. 2017, 1507, 81–94. [PubMed]
82. Ngwa, V.M.; Axford, D.S.; Healey, A.N.; Nowak, S.J.; Chrestensen, C.A.; McMurry, J.L. A versatile
cell-penetrating peptide-adaptor system for efficient delivery of molecular cargos to subcellular destinations.
PLoS ONE 2017, 12, e0178648. [CrossRef] [PubMed]
83. Lino, C.A.; Harper, J.C.; Carney, J.P.; Timlin, J.A. Delivering CRISPR: A review of the challenges and
approaches. Drug Deliv. 2018, 25, 1234–1257. [CrossRef] [PubMed]
84. Mout, R.; Ray, M.; Yesilbag Tonga, G.; Lee, Y.W.; Tay, T.; Sasaki, K.; Rotello, V.M. Direct Cytosolic Delivery
of CRISPR/Cas9-Ribonucleoprotein for Efficient Gene Editing. ACS Nano 2017, 11, 2452–2458. [CrossRef]
[PubMed]
85. Lee, K.; Conboy, M.; Park, H.M.; Jiang, F.; Kim, H.J.; Dewitt, M.A.; Mackley, V.A.; Chang, K.; Rao, A.;
Skinner, C.; et al. Nanoparticle delivery of Cas9 ribonucleoprotein and donor DNA in vivo induces
homology-directed DNA repair. Nat. Biomed. Eng. 2017, 1, 889–901. [CrossRef]
86. Zhang, X. Gold Nanoparticles: Recent Advances in the Biomedical Applications. Cell Biochem. Biophys. 2015,
72, 771–775. [CrossRef]
87. Choi, J.G.; Dang, Y.; Abraham, S.; Ma, H.; Zhang, J.; Guo, H.; Cai, Y.; Mikkelsen, J.G.; Wu, H.; Shankar, P.;
et al. Lentivirus pre-packed with Cas9 protein for safer gene editing. Gene Ther. 2016, 23, 627–633. [CrossRef]
88. Mangeot, P.E.; Risson, V.; Fusil, F.; Marnef, A.; Laurent, E.; Blin, J.; Mournetas, V.; Massourides, E.;
Sohier, T.J.M.; Corbin, A.; et al. Genome editing in primary cells and in vivo using viral-derived Nanoblades
loaded with Cas9-sgRNA ribonucleoproteins. Nat. Commun. 2019, 10, 45. [CrossRef]
89. Campbell, L.A.; Coke, L.M.; Richie, C.T.; Fortuno, L.V.; Park, A.Y.; Harvey, B.K. Gesicle-Mediated Delivery of
CRISPR/Cas9 Ribonucleoprotein Complex for Inactivating the HIV Provirus. Mol. Ther. 2019, 27, 151–163.
[CrossRef]
90. Montagna, C.; Petris, G.; Casini, A.; Maule, G.; Franceschini, G.M.; Zanella, I.; Conti, L.; Arnoldi, F.;
Burrone, O.R.; Zentilin, L.; et al. VSV-G-Enveloped Vesicles for Traceless Delivery of CRISPR-Cas9. Mol.
Nucleic Acids 2018, 12, 453–462. [CrossRef]
91. Gee, P.; Lung, M.S.Y.; Okuzaki, Y.; Sasakawa, N.; Iguchi, T.; Makita, Y.; Hozumi, H.; Miura, Y.; Yang, L.F.;
Iwasaki, M.; et al. Extracellular nanovesicles for packaging of CRISPR-Cas9 protein and sgRNA to induce
therapeutic exon skipping. Nat. Commun. 2020, 11, 1334. [CrossRef] [PubMed]
92. Mangeot, P.E.; Dollet, S.; Girard, M.; Ciancia, C.; Joly, S.; Peschanski, M.; Lotteau, V. Protein transfer into
human cells by VSV-G-induced nanovesicles. Mol. Ther. 2011, 19, 1656–1666. [CrossRef] [PubMed]
93. Briggs, J.A.; Simon, M.N.; Gross, I.; Krausslich, H.G.; Fuller, S.D.; Vogt, V.M.; Johnson, M.C. The stoichiometry
of Gag protein in HIV-1. Nat. Struct. Mol. Biol 2004, 11, 672–675. [CrossRef] [PubMed]
94. Bell, N.M.; Lever, A.M. HIV Gag polyprotein: Processing and early viral particle assembly. Trends Microbiol.
2013, 21, 136–144. [CrossRef] [PubMed]
95. Fuenmayor, J.; Godia, F.; Cervera, L. Production of virus-like particles for vaccines. New Biotechnol. 2017, 39,
174–180. [CrossRef] [PubMed]
96. Mohsen, M.O.; Zha, L.; Cabral-Miranda, G.; Bachmann, M.F. Major findings and recent advances in virus-like
particle (VLP)-based vaccines. Semin. Immunol. 2017, 34, 123–132. [CrossRef] [PubMed]
Biomolecules 2020, 10, 839 16 of 16

97. Kaczmarczyk, S.J.; Sitaraman, K.; Young, H.A.; Hughes, S.H.; Chatterjee, D.K. Protein delivery using
engineered virus-like particles. Proc. Natl. Acad. Sci. USA 2011, 108, 16998–17003. [CrossRef]
98. Aoki, T.; Miyauchi, K.; Urano, E.; Ichikawa, R.; Komano, J. Protein transduction by pseudotyped lentivirus-like
nanoparticles. Gene Ther. 2011, 18, 936–941. [CrossRef]
99. Mamcarz, E.; Zhou, S.; Lockey, T.; Abdelsamed, H.; Cross, S.J.; Kang, G.; Ma, Z.; Condori, J.; Dowdy, J.;
Triplett, B.; et al. Lentiviral Gene Therapy Combined with Low-Dose Busulfan in Infants with SCID-X1. N.
Engl. J. Med. 2019, 380, 1525–1534. [CrossRef]
100. Savic, N.; Schwank, G. Advances in therapeutic CRISPR/Cas9 genome editing. Transl. Res. 2016, 168, 15–21.
[CrossRef]
101. Mattapally, S.; Pawlik, K.M.; Fast, V.G.; Zumaquero, E.; Lund, F.E.; Randall, T.D.; Townes, T.M.; Zhang, J.
Human Leukocyte Antigen Class I and II Knockout Human Induced Pluripotent Stem Cell-Derived Cells:
Universal Donor for Cell Therapy. J. Am. Heart Assoc. 2018, 7, e010239. [CrossRef] [PubMed]
102. Zeng, J.; Tang, S.Y.; Toh, L.L.; Wang, S. Generation of “Off-the-Shelf” Natural Killer Cells from Peripheral
Blood Cell-Derived Induced Pluripotent Stem Cells. Stem Cell Rep. 2017, 9, 1796–1812. [CrossRef] [PubMed]
103. Pinto, C.; Estrada, M.F.; Brito, C. In Vitro and Ex Vivo Models—The Tumor Microenvironment in a Flask.
Adv. Exp. Med. Biol. 2020, 1219, 431–443. [PubMed]

© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).