Accepted Manuscript

Title: Genome editing in the mammalian brain using the

CRISPR–Cas system

Author: Jun Nishiyama

PII: S0168-0102(18)30152-4
DOI: https://doi.org/10.1016/j.neures.2018.07.003
Reference: NSR 4183

To appear in: Neuroscience Research

Received date: 29-3-2018

Revised date: 3-7-2018
Accepted date: 9-7-2018

Please cite this article as: Nishiyama J, Genome editing in the mammalian
brain using the CRISPR–Cas system, Neuroscience Research (2018),
https://doi.org/10.1016/j.neures.2018.07.003

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Review article

Genome editing in the mammalian brain using the CRISPR–Cas system

PT
Jun Nishiyama a,*

RI
SC
a
Max Planck Florida Institute for Neuroscience, One Max Planck Way, Jupiter, FL 33458, USA.

U
N
A
* Corresponding author: at Max Planck Florida Institute for Neuroscience, One Max Planck
M

Way, Jupiter, FL 33458, USA.

D

Email: Jun.Nishiyama@mpfi.org
TE

The total number of pages: 31

EP

The total number of figures: 2

CC

The total number of tables: 1

A

Highlights:
 Genome editing technologies have recently been applied to neuroscience.
 CRISPR–Cas systems have enabled rapid and efficient gene modifications in the brain.
 Endogenous genes can be precisely edited in mitotic cells and postmitotic neurons.
  CRISPR–Cas systems will facilitate our understanding of gene functions in the brain.

ABSTRACT

PT
Recent advances in genome editing technologies such as the clustered regularly interspaced short

RI
palindromic repeats (CRISPR)-associated endonuclease Cas9 have enabled the rapid and

SC
efficient modification of endogenous genomes in a variety of cell types, accelerating biomedical

research. In particular, precise genome editing in somatic cells in vivo allows for the rapid

U
generation of genetically modified cells in living animals and holds great promise for the

N
possibility of directly correcting genetic defects associated with human diseases. However,
A
because of the limited suitability and efficiency of these technologies in the brain, especially in
M

postmitotic neurons, the practical application of genome editing technologies has been largely

limited in the field of neuroscience. Recent technological advances overcome significant

D

challenges facing genome editing in the brain and have enabled us to precisely edit the genome
TE

in both mitotic cells and mature postmitotic neurons in vitro and in vivo, providing powerful
EP

means for studying gene function and dysfunction in the brain. This review highlights the

development of genome editing technologies for the brain and discusses their applications,
CC

limitations, and future challenges.

A

Abbreviations:
AAV, Adeno-associated virus; BAC, Bacterial artificial chromosome; CRISPR, Clustered

regularly interspaced short palindromic repeats; DSB, Double-strand breaks; ESC, Embryonic

stem cell; GFP, Green fluorescent protein; HDR, Homology-directed repair; HR, Homologous

recombination; IUE, In utero electroporation; NHEJ, Non-homologous end joining; NLS,

PT
Nuclear localization sequence; NMDAR, N-methyl-D-aspartate-type glutamate receptor; PAM,

Protospacer adjacent motif; ssODNs, Single-stranded DNA oligonucleotides; TALENs,

RI
Transcription activator-like effector nucleases; ZFN, zinc-finger nucleases;.

SC
U
Keywords: Genome editing; CRISPR; Cas9; HDR; NHEJ; Postmitotic neurons; SLENDR;

vSLENDR. N
A
M

1. Introduction
D
TE

The development and subsequent rapid advance of DNA sequencing technology have greatly

expanded our knowledge of the genetic information and its relationship with diseases.
EP

Conversely, the advent of genetic engineering has permitted the manipulation of DNA sequences,

thereby decoding genetic information written in the genome. In particular, gene-targeting

CC

technology, such as homologous recombination (HR) between exogenous DNA and the

endogenous genomic sequence in mouse embryonic stem cells (ESCs), has been widely used to
A

generate mutant mice, providing a powerful means to explore genotype–phenotype relationships

in vivo. Furthermore, the biological function of a gene, especially in a complex system such as

the brain, often depends on the specific regions and timing of its expression. Therefore,
conditional gene knockout mice are often generated using the Cre/loxP system for applications in

neuroscience. However, the generation of mutant animals using conventional gene-targeting

technology, especially when combined with the Cre/loxP system, is laborious, time-consuming,

and costly, and more importantly, it requires ESCs, which are not available for most mammalian

PT
species. In addition, because of the inherent low efficiency of HR, the practical application of

gene-targeting techniques has been mainly limited to cultured cells such as ESCs in vitro.

RI
The recent development of genome editing tools based on engineered nucleases is

SC
revolutionizing genetic engineering (Cong et al., 2013; Jinek et al., 2012; Mali et al., 2013b).

These engineered nucleases, namely, zinc-finger nucleases (ZFNs), transcription activator-like

U
effector nucleases (TALENs), and the clustered regularly interspaced short palindromic repeats
N
(CRISPR)-associated endonuclease Cas– enable targeted and efficient genome modification in
A
many organisms and cell types in vitro and in vivo (Cox et al., 2015; Hsu et al., 2014; Shmakov
M

et al., 2017; Sternberg and Doudna, 2015). In particular, the CRISPR nucleases such as Cas9
D

from bacterial adaptive immune systems can be easily used to target virtually any genomic
TE

region, providing one of the most popular approaches for genome editing. Here, I overview

CRISPR–Cas-mediated genome editing systems in mammalian cells, highlight the recent

EP

development and applications of CRISPR–Cas-mediated genome editing in the brain, and

CC

discuss its future challenges to better understand brain functions and disorders.
A

2. Overview of CRISPR–Cas-mediated genome editing

Engineered nucleases, including ZFNs, TALENs, and Cas9, can induce DNA double-strand

breaks (DSBs) at specific genomic loci, which are subsequently repaired by one of at least two

endogenous cellular DNA repair pathways: non-homologous end joining (NHEJ) and homology-
directed repair (HDR) (Fig. 1A) (Cox et al., 2015; Doudna and Charpentier, 2014; Hsu et al.,

2014; Shmakov et al., 2017; Sternberg and Doudna, 2015). In NHEJ, DSB ends are directly

rejoined without the requirement for sequence homology. Although NHEJ-mediated DSB repair

can be accurate, repeated repair of the same DSB by NHEJ eventually results in the formation of

PT
unpredictable patterns of insertions and deletion mutations (indels) at the break site (Cox et al.,

2015). Indels in the coding sequence of the target gene can shift the translational reading frame,

RI
thus introducing a premature stop codon. Therefore, NHEJ can be used to knockout endogenous

SC
genes. By contrast, the HDR pathway directs a precise recombination event between a

homologous DNA donor template and the damaged DNA site. Thus, HDR can be used to

U
precisely introduce sequence insertions, deletions, or mutations by encoding desired changes in

N
the donor template. The donor template can either be in the form of double-stranded DNA or
A
single-stranded DNA oligonucleotides (ssODNs) with homology arms flanking the insertion
M

sequence. ssODNs have been used for the efficient targeted insertion of small DNA fragments or

introduction of single nucleotide mutations.

D
TE

ZFNs and TALENs use protein motifs for DNA sequence recognition, and these motifs are often

extremely difficult to engineer. By contrast, Cas9 uses a short guide RNA sequence to recognize
EP

the target double-stranded DNA (dsDNA) via Watson–Crick base pairing, making it much easier

and simpler to design highly specific constructs (Fig.1B) (Cong et al., 2013; Jinek et al., 2012;
CC

Mali et al., 2013b). Together with two separate RNAs called CRISPR-RNA (crRNA) and trans-

activating crRNA (tracrRNA) or a chimeric single-guide RNA (sgRNA) consisting of crRNA

A

and tracrRNA, Cas9 binds to dsDNA in a site-specific manner. The 20-nt guide sequence in the

crRNA defines DNA target specificity while the tracrRNA facilitates Cas9 recruitment to the

target DNA. The only requirement for Cas9 target recognition is the presence of a short sequence
motif proximal to the DNA target known as a protospacer adjacent motif (PAM) (Cox et al.,

2015; Hsu et al., 2014; Shmakov et al., 2017; Sternberg and Doudna, 2015). The most commonly

used Cas9 from Streptococcus pyogenes (SpCas9) recognizes a 5′-NGG-3′ PAM, whereas other

Cas9 orthologs have different PAM requirements for sufficient DNA cleavage (e.g., 5′-

PT
NNGRRT-3′ PAM for Staphylococcus aureus Cas9 [SaCas9]). Therefore, Cas9 can be directed

to any genomic locus containing a PAM sequence by simply changing the guide RNA-targeting

RI
sequence. Cas9 subsequently cleaves both strands of the DNA target sequence using its HNH

SC
and RuvC nuclease domains. Because of its simplicity, high efficiency, and wide applicability

over other nucleases, the CRISPR–Cas9 system has been the most widely used genome editing

U
technology.

N
The success of DNA editing using the CRISPR–Cas9 system has prompted the extensive
A
exploration and characterization of novel CRISPR–Cas effectors with features distinct from Cas9,
M

leading to the expansion of CRISPR–Cas toolkit. For example, Cas12a (previously Cpf1) is a
D

single-RNA-guided endonuclease of the CRISPR–Cas system (Zetsche et al., 2015) (Fig. 1B).
TE

This system requires no tracrRNA, reducing experimental complexity. In addition, unlike the G-

rich PAM sequence of the Cas9 systems (e.g., NGG for SpCas9) that lies 3′ of its target
EP

sequence, Cas12a recognizes a T-rich PAM, namely, TTTV, on the 5′ side of the target (Kim et
CC

al., 2017b). Thus, Cas12a expands the targeting range of the CRISPR–Cas system and facilitates

genome editing, especially in organisms with AT-rich genomes. Side-by-side comparisons of

A

genome editing efficiency and specificity revealed that Cas12a orthologs exhibit similar or

slightly lower efficiency but fewer off-target effects when compared with SpCas9 (Kim et al.,

2016).
In addition to DNA-editing tools based on Cas9 and Cas12, RNA-targeting CRISPR systems

were recently discovered. These RNA-guided RNases, including Cas13a (previously C2c2) and

Cas13b, lack the RuvC domain common to both Cas9 and Cas12, but they contain two higher

eukaryotes and prokaryotes nucleotide-binding (HEPN) RNase domains (Abudayyeh et al.,

PT
2016; East-Seletsky et al., 2016) (Fig. 1B). Cas13b, similar to Cas12a, is guided by a single

crRNA, but it cleaves single-stranded RNA with a preference for targets with protospacer

RI
flanking sequences or PFS (A, U, or C for Leptotrichia shahii Cas13a or LshCas13a). Cas13

SC
exhibits efficient RNA interference (RNAi) with minimum off-target effects when

heterologously expressed in mammalian cells (Fig. 1B, see also 3.2) (Cox et al., 2017).

U
Furthermore, the introduction of point mutations into nuclease domains results in the nuclease
N
deficient mutants of Cas proteins, which still retain DNA/RNA binding ability. By fusing with
A
various protein effector domains, the catalytically dead Cas mutants have been widely used as
M

molecular scaffolds for many applications, including transcriptional control (Gilbert et al., 2014;
D

Gilbert et al., 2013; Konermann et al., 2015; Maeder et al., 2013; Mali et al., 2013a; Perez-Pinera
TE

et al., 2013; Tanenbaum et al., 2014; Zheng et al., 2018; Zhou et al., 2018), epigenetic regulation

(Liu et al., 2016; Liu et al., 2018), and ‘base editing’ in which a specific DNA/RNA base is
EP

converted into another at a targeted locus (Cox et al., 2017; Gaudelli et al., 2017; Komor et al.,
CC

2016) (Fig.1C). Overall, the systems based on catalytically dead Cas proteins do not cleave DNA,

thereby avoiding deleterious effects of indel mutations.

A

The development of versatile genome editing tools based on various CRISPR–Cas effectors has

rapidly extended the repertoire of targeted genome modifications. The continued exploration and

characterization of the enormous diversity of the natural CRISPR–Cas system will undoubtedly

lead to further expansion of CRISPR–Cas-mediated genome editing tools (Shmakov et al., 2017).
3. In vivo genome editing in the brain

CRISPR–Cas9-mediated genome editing has dramatically accelerated the generation of

genetically engineered animals because of its high efficiency, simplicity, and flexibility.

CRISPR–Cas9 reagents can be delivered into ESCs or directly into fertilized eggs (zygotes),

PT
leading to the generation of mutant animals carrying targeted genetic modifications.

Alternatively, genome editing can be performed in somatic cells in vivo, in which the editing

RI
occurs in an organ in live animals. In vivo genome editing provides an ideal platform for rapidly

SC
generating genetically modified cells in living animals and holds great promise for correcting

disease-causing genetic defects in humans. Indeed, in vivo genome editing bypasses laborious

U
time- and cost-consuming processes to generate and cross genetically engineered animals, and it
N
can be directly applied to thousands of existing transgenic lines, enabling rapid and flexible
A
analysis of edited cells. In addition, the local delivery of genome editing machinery into cells in
M

vivo permits time-and tissue-specific genome editing in living organism. This is particularly
D

important for the translational application of genome editing because many human disorders
TE

occur in specific organs and tissues over the course of diseases.

EP

3.1 Delivery of the CRISPR–Cas system in the brain in vivo

One major obstacle to implementing in vivo genome editing in the brain is the efficient delivery
CC

of genome editing machinery to target cells in vivo.

A

Various viral vectors, including adeno-associated virus (AAV), lentivirus, adenovirus, and

retrovirus, have been used to deliver CRISPR–Cas9-mediated genome editing machinery into

cells. In particular, AAV permits the long-term persistent expression of transgenes with low

immunogenicity and toxicity in both dividing and non-dividing cells in the brain in vivo.
However, AAV has a limited transgene capacity (up to 4.7–5 kb) (Wu et al., 2010) . Because the

SpCas9 coding sequence comprises more than 4.2 kb, it has been challenging to package both

SpCas9 and the other components required for gene editing (e.g., SpCas9 and sgRNA expression

cassettes for NHEJ-mediated gene knockout or SpCas9, sgRNA expression cassettes, and the

PT
donor template for HDR-mediated gene knock-in) into a single AAV. Recently, smaller Cas9

orthologs, including SaCas9 (approximately 3.2 kb) and Campylobacter jejuni Cas9 (CjCas9,

RI
approximately 3.0 kb), have been successfully harnessed to package Cas9 and the sgRNA

SC
expression cassettes into a single AAV (Kim et al., 2017a; Ran et al., 2015). Although these

Cas9 orthologs exhibit comparable genome editing efficiency, the PAM sequences recognized by

U
SaCas9 (5′-NNGRRT-3′) and CjCas9 (5′-NNNNACAC-3′ or 5′-NNNNRYAC-3′) occur less

N
frequently than those recognized by SpCas9 (5ʹ-NGG-3ʹ), limiting the number of targetable sites
A
in the genome. Alternatively, SpCas9 and sgRNA expression cassettes can be packaged into two
M

separate AAVs. In this dual-AAV strategy, SpCas9 can be packaged with extremely short

promoters, including truncated MeCP2 and the short form of elongation factor 1α promoters
D

(235 and 255 bp, respectively), as well as minimal polyadenylation signals (48 bp). By co-
TE

infecting two AAVs expressing SpCas9 and sgRNAs, the dual-AAV system permits efficient
EP

CRISPR-Cas9-mediated gene knockout (Swiech et al., 2015) (see 3.2) and gene knock-in

(Nishiyama et al., 2017; Suzuki et al., 2016) (see 3.3) in various brain regions.
CC

In utero electroporation (IUE) is a rapid and efficient method for delivering transgenes into
A

neuronal progenitors in the brain in vivo. IUE enables the efficient introduction of multiple

constructs into the same cells without strict size limitations. Indeed, multiple plasmids including

a plasmid encoding a transgene longer than 12 kb or a 125 kb bacterial artificial chromosome

(BAC) clone have been successfully co-introduced into neuronal progenitors in vivo via
electroporation (Barnabe-Heider et al., 2008; Nishiyama et al., 2012). In addition, three plasmids,

each encoding a different fluorescent protein, were demonstrated to be co-introduced in 99% of

transfected neurons by IUE (Nishiyama et al., 2012). By taking advantages of these features,

multiple constructs required for CRISPR–Cas9-mediated gene knockout (Chen et al., 2015;

PT
Kalebic et al., 2016; Shinmyo and Kawasaki, 2017; Shinmyo et al., 2017; Straub et al., 2014)

(see 3.2) and gene knock-in (Mikuni et al., 2016; Tsunekawa et al., 2016; Uemura et al., 2016)

RI
(see 3.3) can be readily and efficiently introduced into neuronal progenitors in the embryonic

SC
brain through IUE .

The in vivo delivery of genetically encoded Cas9 with viral vectors and plasmids might cause

U
undesirable side-effects such as the genomic integration of constructs, immune responses elicited
N
by persistent expression of the bacterial Cas9 protein, and off-target editing (Staahl et al., 2017).
A
These are particularly important limitations to the translational use of the CRISPR–Cas9 system.
M

To overcome these concerns, preassembled Cas9 ribonucleoprotein complexes (RNPs)

D

consisting of purified Cas9 protein in complex with a sgRNA can be harnessed for genome
TE

editing. In contrast to plasmid- or viral-based vectors, Cas9 RNPs do not require cellular

transcription/translation machinery for their functional expression and thereby act immediately
EP

after entering cells. In addition, Cas9 RNPs can greatly reduce the possibility of side effects

caused by the long-term expression of Cas9, as they can be quickly cleared from the cells via the
CC

degradation system. Indeed, RNPs have been found to provide rapid nuclease action with high
A

efficiency and low off-target effects for genome editing (Kim et al., 2014; Lin et al., 2014). Cas9

RNPs can be directly delivered into a variety of cells and tissues including neuronal progenitors

and cortical organotypic slice cultures via electroporation and microinjection (Kalebic et al.,

2016). Although Cas9 RNPs have no innate cell-penetrating activity, a recent study illustrated
that adding multiple SV40 nuclear localization sequences (NLSs) resulted in the generation of

cell-permeable Cas9 RNPs (Staahl et al., 2017). By locally injecting engineered SV40 NLS-

tagged Cas9 RNPs into the adult mouse brain in vivo, efficient genome editing was successfully

achieved in multiple brain regions, including the hippocampus, striatum, and cortex.

PT
Interestingly, engineered Cas9 RNPs exhibit preferential neurotropism by unknown mechanisms,

leading to neuron-specific genome editing in the brain in vivo.

RI
3.2 CRISPR–Cas-mediated gene perturbations

SC
The most direct method for studying gene function is reducing or completely disrupting gene

U
expression and exploring the resulting phenotypes in vitro or in vivo. Recently, several

N
techniques have been developed to rapidly perturb gene expression in vitro and in vivo based on
A
CRISPR–Cas systems.
M

CRISPR–Cas9 enables rapid and stable gene knockout through error-prone NHEJ in virtually
D

any types of cells. In 2014, two different groups first reported that CRISPR–Cas9-mediated
TE

NHEJ could be used to efficiently generate gene knockout in postmitotic neurons (Incontro et al.,

2014; Straub et al., 2014). Both groups targeted the Grin1, the gene encoding the GluN1 subunit
EP

of the N-methyl-D-aspartate-type glutamate receptor (NMDAR), using different guide RNAs

through IUE or biolistic transfections. Intrudingly, the NMDAR-mediated current completely

CC

abolished in most tested neurons, suggesting the efficient disruption of both genomic alleles

encoding GluN1. The extremely high gene knockout efficiency may be attributed to the long-
A

term persistent expression of Cas9 and sgRNAs in non-dividing neurons. Importantly, by co-

expressing cDNA encoding GluN1, the phenotype was completely rescued, confirming the

specificity of CRISPR–Cas9-mediated gene knockout. In addition, CRISPR–Cas9-mediated gene

knockout can be applied to investigate phenotypes at both the cellular and behavioral levels in

the brain in vivo. For example, Rett syndrome is a progressive neurodevelopmental disorder

caused by a loss-of-function mutation in the X-linked gene MeCP2. To study the MeCP2

function in the brain in vivo, two AAVs expressing SpCas9 and sgRNA to disrupt MeCP2 were

PT
stereotaxically injected into the dorsal dentate gyrus (DG) of the mouse hippocampus. Two to

three weeks after AAV delivery, the number of MeCP2-positive cells was decreased by

RI
approximately 70% in the injected dentate gyrus, leading to the specific impairment of

SC
contextual memory without affecting other cognitive functions (Swiech et al., 2015). Overall,

CRISPR–Cas9 system permits a rapid and efficient gene perturbation in specific brain regions in

U
vivo, facilitating the interrogation of gene functions in the brain (Chen et al., 2015; Kalebic et al.,

2016; Shinmyo et al., 2016; Shinmyo et al., 2017). N

A
These loss-of-function experiments may also be performed by using RNAi. RNAi reduces the
M

translation of mRNAs by inducing the degradation of target gene transcripts, enabling reversible
D

perturbation of gene expression without affecting genomic sequences. Very recently, techniques
TE

based on CRISPR–Cas system have been developed as alternative and better means of RNAi.

For example, Cas13 is a CRISPR-associated RNA-guided RNase that can be engineered for
EP

efficient RNA depletion in mammalian cells (Fig.1B) (Abudayyeh et al., 2016; Cox et al., 2017;
CC

East-Seletsky et al., 2016). Upon screening a number of Cas13 family enzymes, the Cas13b

ortholog from Prevotella sp. P5-125 (PspCas13b) was found to exhibit the most efficient RNA
A

depletion activity. PspCas13b displayed significantly higher efficiency than RNAi with position-

matched small hairpin RNAs (shRNAs). Intrudingly, PspCas13b exhibited no detectable off-

target activity, whereas shRNAs had hundreds of off-targets (Cox et al., 2017). Although the

Cas13 system has only been tested in bacteria, mammalian cells, and plant cells, the Cas13-based
RNA editing system might enable specific, efficient, reversible gene perturbation in the brain in

vivo and provide a tool to avoid permanent off-target indels through CRISPR–Cas-mediated

DNA editing.

Instead of directly manipulating DNA/RNA, catalytically dead Cas9 (or dCas9) has been used to

PT
control endogenous gene expression with high specificity and efficiency in a different way. For

example, a robust transcriptional repression has been achieved by fusing dCas9 with

RI
transcriptional repressors such as KRAB (Gilbert et al., 2014; Gilbert et al., 2013). Recently, this

SC
system was successfully applied to neurons in vitro and in vivo. By expressing single lentivirus

vector encoding dCas9 fused with KRAB (dCas9-KRAB) together with sgRNA expression

U
cassettes into neurons, endogenous gene expression was efficiently suppressed by more than
N
90% in dissociated cultures (Zheng et al., 2018). Importantly, the dCas9-KRAB system showed
A
significantly higher knock-down efficiency and specificity than that obtained by shRNA. To
M

manipulate synaptic transmission in the brain in vivo, this system was used to suppress the
D

expression of synaptotagmin I (Syt1), a Ca2+ sensor required for neurotransmitter release. By

TE

injecting lentivirus encoding Syt1-targeting dCas9-KRAB constructs under the control of

CaMKIIα or VGAT promoters, synaptic transmission in the excitatory or inhibitory neurons was
EP

effectively suppressed in the DG in the hippocampus. Interestingly, suppression of synaptic

CC

transmission in excitatory and inhibitory neurons in the mouse DG resulted in memory

impairment and enhancement, respectively. These results suggest that the dCas9-KRAB system
A

efficiently suppresses endogenous gene expression in the brain in vivo. By contrast, dCas system

has also been used for transcriptional activation by fusing with transcriptional activators such as

VP64 in various cell-types including neurons and astrocytes in the brain (Gilbert et al., 2014;

Gilbert et al., 2013; Konermann et al., 2015; Maeder et al., 2013; Mali et al., 2013a; Perez-Pinera
et al., 2013; Zhou et al., 2018). Overall, the dCas9 mediated system enables us to conditionally

decrease or increase the gene expression with minimum off-target activity in vitro and in vivo,

providing a powerful tool to study gene functions in the brain.

PT
3.3 CRISPR–Cas-mediated gene knock-in

RI
In contrast to NHEJ, HDR in principle can be used for genome editing in an error-free manner.

SC
Furthermore, HDR enables the introduction of sequence insertions, deletions, or mutations by

encoding any desired changes in the donor template. However, HDR has substantially lower

U
efficiency than NHEJ, limiting the utility of HDR-mediated precise genome editing in many
N
systems. More importantly, HDR is believed to occur primarily in the S and G2 phases of the
A
cell cycle in mitotically dividing cells and considered rare in postmitotic cells such as matured
M

neurons. Thus, HDR-mediated precise genome editing has been a challenge in the brain, in
D

which most neurons are postmitotic.

TE

Recent studies illustrated that dividing neuronal progenitors retain intrinsic activity to induce

HDR in vivo. By efficiently delivering the genome editing machinery into dividing neuronal
EP

progenitors via IUE, HDR-mediated genome editing was successfully achieved in the embryonic
CC

mouse brain (Mikuni et al., 2016; Tsunekawa et al., 2016; Uemura et al., 2016). By inserting a

sequence encoding an epitope tag, such as a small human influenza hemagglutinin (HA) tag, into
A

a gene of interest, Mikuni et al. (2016) developed a simple and generalizable technique to image

endogenous proteins with high specificity, resolution, and contrast in single cells in mammalian

brain tissue (Fig. 2A). Using this strategy, termed SLENDR (single-cell labeling of endogenous

proteins by CRISPR–Cas9-mediated HDR), they succeeded in visualizing the subcellular

localization of a variety of nuclear, cytoskeletal, vesicular, cytosolic, and membrane proteins.

Importantly, genome editing occurs extremely rapidly (within a few days) after IUE only in a

sparse subset of cells, allowing rapid and high-contrast visualization of endogenous proteins in

single cells in dense brain tissue. In addition, SLENDR is highly precise and specific to sgRNA,

PT
as no gene insertion was observed when using incorrect sgRNA targeting a different gene. The

overall knock-in efficiency in SLENDR was as high as 5% of transfected cells for HA tag knock-

RI
in. Thus, SLENDR provides a rapid, precise, and scalable technique to image endogenous

SC
proteins in the brain tissue. The drawback of SLENDR is that the technique might lead to

unintended indels through NHEJ in transfected cells if DSBs were not repaired through HDR.

U
This problem may cause a large impact, particularly in the case of N-terminal tagging, as indels

N
near the N-terminal region may lead to frameshift mutations that cause a premature stop codon.
A
To minimize the effect of on-target NHEJ, CRISPR–Cas9 cleavage sites can be located in the
M

non-coding region upstream of the start codon or downstream from the stop codon. Indeed, using

this strategy, the target protein was not deleted in most (~99%) of the transfected cells, and the
D

expression of gene products was not significantly altered.

TE

SLENDR provides a simple and generalizable platform to probe endogenous proteins in brain
EP

tissue, enabling broad applications to explore molecular function in the brain (Fig. 1B). For
CC

example, immunoelectron microscopy allows nanoscale visualization of endogenous proteins

with defined ultrastructures in cells. However, the lack of reliable antibodies against each target
A

protein compatible with immunoelectron microscopy severely limits its application to a variety

of proteins. By contrast, in SLENDR, high-quality antibodies (e.g., anti-HA antibody) can be

used to detect tags fused to various endogenous proteins. Indeed, by combining SLENDR with

immunoelectron microscopy, the localization of endogenous HA-tagged CaMKIIβ was clearly

visualized in synapses at nanometer scale. Furthermore, SLENDR enables efficient multiplex

genome editing for simultaneously modifying multiple targets. Because IUE permits efficient co-

transfection of various constructs into single cells, genome editing constructs targeting two

endogenous genes can be efficiently delivered into neuronal progenitors. Thus, using SLENDR,

PT
it is possible to create knock-in and/or knockout alleles at two distinct endogenous genes in

single cells. For example, by introducing SLENDR constructs to insert two different tags into

RI
two distinct genes, simultaneous labeling of two endogenous proteins can be achieved in single

SC
cells. This technique would be particularly useful for co-localization assays of a pair of

endogenous proteins. Furthermore, by combining single-cell knockout and knock-in, it is

U
possible to investigate the cell-autonomous effects of a gene on endogenous proteins. Finally,

N
SLENDR enables monitoring of the dynamics of an endogenous protein in living tissue by
A
inserting a fluorescent protein into a gene of interest. Consistent with previous reports that a long
M

sequence insertion through HDR is less efficient, the efficiency of green fluorescent protein

(GFP) knock-in into CaMKIIα and CaMKIIβ was significantly lower than that of HA knock-in
D

(less than 1%). Nevertheless, the dynamics of endogenous CaMKIIα and CaMKIIβ was clearly
TE

visualized with GFP at a single-spine resolution in organotypic slice cultures using two-photon
EP

microscopy.
CC

4. Genome editing in postmitotic neurons

A

One of the major goals of genome editing technology is precisely editing the genome in targeted

tissues and cells in living animals. For example, precise correction of genetic mutations via HDR

in the brain in vivo would permit the therapeutic use of genome editing in patients with various

genetic brain disorders. However, many adult tissues such as the brain and heart consist of
postmitotic cells, which are considered to have little to no HDR activity. Thus, there is an urgent

need for precise and efficient genome editing strategies in postmitotic cells.

To circumvent the poor HDR activity of postmitotic cells, NHEJ-mediated gene ligation has

been used for sequence insertion. Because external DNA fragments can be ligated at DSB sites,

PT
by adding donor DNAs, targeted sequence insertion has been successfully achieved at DSB sites

generated by engineered nucleases (Auer et al., 2014; Maresca et al., 2013; Suzuki et al., 2016).

RI
This strategy is mediated by NHEJ, as it does not require homology between the donor and target,

SC
enabling sequence insertion in both dividing and non-dividing cells. These techniques, termed

Obligate Ligation-Gated Recombination (ObLiGaRe) or homology-independent targeted

U
integration (HITI), employ in situ linearization of the circular donor template, which contains the
N
same nuclease recognition site as the targeted genomic locus (Maresca et al., 2013; Suzuki et al.,
A
2016). Because engineered nucleases generate DNA fragments from the circular donor template,
M

the donor template can be bidirectionally inserted into DSB sites. To increase the possibility of
D

obtaining correctly ligated products, they used the inverted orientation of the nuclease
TE

recognition sequences in the donor template. This leads to eliminate the sensitivity of correctly

ligated products to the nuclease, whereas products with the inverted orientation containing the
EP

intact nuclease recognition sequences are repeatedly digested by the nuclease. Using NHEJ-
CC

mediated gene ligation, Suzuki et al., succeeded in introducing a transgene into mitotic and

postmitotic neurons in the mouse brain. The GFP sequence was effectively introduced into β3-
A

tubulin in 10–15% of transfected neurons by delivering genome editing machinery via IUE into

neuronal progenitors in the embryonic brain or AAV into postmitotic neurons in the adult brain.

When they used plasmid- and AAV-based vectors in a dissociated culture system, indel

mutations at on-target sites, likely caused by NHEJ, were observed in approximately10 and 30%
of GFP knock-in neurons, respectively. Intriguingly, using AAV-based vectors, they succeeded

in restoring a missing exon of Mertk in the Royal College of Surgeons rat, a rodent model of

inherited forms of human retinal degeneration (retinitis pigmentosa). This resulted in increased

expression of Mertk and increased outer nuclear layer thickness and electroretinography

PT
responses, indicating a partial rescue of the phenotype (Suzuki et al., 2016). Overall, NHEJ-

mediated gene ligation techniques enable the efficient insertion of exogenous DNA fragments

RI
into DSB sites in various cell types including dividing and non-dividing cells including

SC
postmitotic neurons. However, it should be noted that various types of indels can be introduced

at on-target sites through NHEJ, and DNA fragments can be non-specifically inserted into DSBs

U
at on-target and off-target sites.

N
Recently, HDR-mediated genome editing became possible in postmitotic cells using AAV. AAV
A
has been widely used for gene delivery in the brain, as AAV can infect both dividing and non-
M

dividing cells with high efficiency and low toxicity. More importantly, AAV has been reported
D

to serve as a donor template for HR at much higher efficiencies than conventional transfection or
TE

electroporation approaches. This is presumably due to the efficient nuclear delivery and single-

strand nature of the AAV genome and the effect of inverted terminal repeats (Barzel et al., 2015;
EP

Gaj et al., 2016; Russell and Hirata, 1998). Although the efficiency of AAV-mediated gene
CC

targeting by itself is insufficient for somatic tissues in vivo, especially postmitotic cells, the

induction of DSBs by endonuclease can increase the frequency of AAV-mediated gene targeting
A

by several orders of magnitude (Dever et al., 2016; Gaj et al., 2016; Porteus et al., 2003).

Therefore, by combining CRISPR–Cas9-mediated DNA cleavage and the efficient delivery of

donor template with AAV, HDR has been successfully induced in mature postmitotic neurons as

well as mitotic cells in the mouse brain (Nishiyama et al., 2017). AAV encoding sgRNA and
HDR template was combined with either SpCas9 knock-in mice (Chiou et al., 2015) or SpCas9-

expressing AAV. Using this strategy, a sequence encoding an epitope tag or fluorescent protein

was successfully inserted into an endogenous gene in dissociated and organotypic slice cultures

in vitro and various brain areas and cell types in developing, adult, and aged brains in vivo.

PT
Intrudingly, the overall knock-in efficiency reached up to approximately 30% of infected cells in

organotypic slice cultures and up to approximately 15% in the cerebral cortex in vivo. Thus, this

RI
strategy, termed viral-mediated SLENDR (vSLENDR), enables us to perform efficient HDR-

SC
mediated genome editing in virtually any cell types and areas in the brain irrespective of age. In

addition, vSLENDR might be directly adapted to other mammalian tissue and species, possibly

U
permitting precise genome engineering in a variety of organs and organisms. Indeed, a recent

N
study illustrated that, by combining AAV and CRISPR–Cas9, HDR-mediated genome editing
A
was successfully achieved in non-dividing cardiomyocytes (Ishizu et al., 2017). HDR-mediated
M

genome editing is advantageous over NHEJ-based techniques, as it enables precise introduction

of any desired changes, such as point mutations and deletions, incorporated in the template DNA
D

(Table 1). Furthermore, although external DNA fragments can be non-specifically inserted into
TE

DSB sites using NHEJ-based techniques, HDR enables highly specific genome editing at the
EP

specific genomic locus homologous to the template. This allows for simultaneous genome

editing at two or more different target genomic loci. Thus, vSLENDR provides a precise,
CC

flexible, and versatile tool for genome engineering in the brain.

A
5. Future perspectives

The recent progress in genome editing technologies based on the CRISPR–Cas system has

permitted efficient modification of endogenous genomes in mitotic cells and postmitotic neurons

in the brain.

PT
In particular, (v)SLENDR enables HDR-mediated genome editing in virtually any brain area,

RI
cell type, and age of interest. In combination with advanced molecular and optical techniques,

SC
(v)SLENDR can reveal the localization and dynamics of many endogenous molecules in the

brain with unprecedented accuracy and detail. The expression levels of endogenous proteins are

U
generally much lower than those of overexpressed proteins, potentially limiting the utility of

N
SLENDR to image some low-abundance endogenous proteins. To overcome this limitation, the
A
detection sensitivity could be increased by inserting multiple copies of epitope tags or more
M

antigenic probes such as Spaghetti-monster, which is compatible with fixed tissue (Viswanathan

et al., 2015). To image low-abundance endogenous proteins in living tissue, other signal
D

amplification methods could be used. These include SunTag, which can recruit up to 24 copies
TE

of GFP (Tanenbaum et al., 2014) and tandem arrays of self-complementing split fluorescent
EP

proteins (Kamiyama et al., 2016). By combining SLENDR, signal amplification methods, and

super-resolution microscopy, it may be possible to image endogenous proteins with high

CC

sensitivity at sub-diffraction resolution in fixed and living neurons. Furthermore, two-photon

fluorescent lifetime imaging microscopy (2pFLIM), which combines two-photon microscopy

A

with fluorescence lifetime measurement, has proven useful for imaging molecular interactions in

brain tissue with high spatial and temporal resolution (Nishiyama and Yasuda, 2015). By

labeling two endogenous proteins with different fluorescent proteins using (v)SLENDR, it would

be possible to image the dynamics of endogenous protein–protein interactions in brain tissue

using 2pFLIM. Although the insertion of a long sequence such as one encoding a fluorescent

protein is less efficient, the relatively high knock-in efficiency of vSLENDR may enable

simultaneous labeling of two endogenous proteins with different fluorescent proteins.

In addition to basic neuroscience, in vivo genome editing has great potential for translational

PT
research aiming to treat various genetic disorders in the brain. Because a number of different

types of deleterious mutations were identified in various diseases, correction of mutations can be

RI
achieved through a number of approaches based on in vivo genome editing (Cox et al., 2015).

SC
For example, a pathogenic gene product produced by a gain-of-function mutation might be

treated by disrupting the target gene through NHEJ. Alternatively, for some deleterious loss-of-

U
function or gain-of-function mutations, therapeutic efficacy might be achieved by precisely
N
replacing the pathogenic allele with the wild-type sequence through HDR.
A
M

However, to achieve these goals, several challenges must be addressed. The major challenge is

increasing the specificity and efficiency of genome editing. CRISPR–Cas9 may cause unwanted
D

mutations at off-target sites when genomic loci partially match the sgRNA targeting sequence.
TE

These off-target mutations may complicate the interpretation of experimental results and limit
EP

the application of genome editing technologies for translational purposes. The recently

developed SpCas9 mutants such as eSpCas9, SpCas9-HF1, and HypaCas9 have minimal off-
CC

target effects while retaining comparable on-target cleavage activity, resulting in greatly

increased specificity (Chen et al., 2017; Kleinstiver et al., 2016; Slaymaker et al., 2016). In
A

addition, a very recent study succeeded in developing an SpCas9 mutant named xCas9 that can

recognize a broader range of PAM sequences including 5′-GAA-3′, 5′-GAT-3′, and 5′-NG-3′

with substantially reduced off-target activity and increased on-target activity (Hu et al., 2018).

Because these results suggest that there is no trade-off among the editing efficiency, PAM
compatibility, and target specificity of the CRISPR–Cas9 system, it might be possible to further

evolve various CRISPR–Cas-mediated nucleases to enhance their specificity and efficiency. In

addition to off-target effects, the effects of indels at the on-target site should be minimized for

gene knock-in regardless of whether it is mediated by NHEJ or HDR. Because virtually all

PT
genome editing technologies rely on the intrinsic DNA repair system, NHEJ-mediated indels

would be inevitably generated at on-target sites to varying degrees when gene knock-in does not

RI
occur as intended. However, for HDR-mediated knock-in, NHEJ-mediated indel formation at the

SC
on-target site may be suppressed by manipulating the intrinsic activities of DNA repair pathways.

Recently, various strategies for stimulating HDR and suppressing NHEJ have been developed

U
using pharmacological or genetic methods (Canny et al., 2018; Chu et al., 2015; Maruyama et al.,

N
2015; Song et al., 2016), raising the possibility to reduce on-target indel formations. In addition,
A
it is highly desired to improve knock-in efficiency for large DNA fragments. A recent study has
M

shown that 200kb BAC DNA could be incorporated into the genome in rat zygotes by combining

NHEJ-mediated gene ligation and two ssODNs bridging the junctions between the template and
D

targeted genome (Yoshimi et al., 2016). Therefore, whether any of these strategies is compatible
TE

with gene knock-in in the brain should be carefully evaluated. The exact mechanism by which
EP

AAV combined with CRISPR–Cas9-mediated DSBs enables efficient HDR in postmitotic

neurons remains elusive. The knock-in efficiency might be further increased by clarifying the
CC

detailed mechanisms of AAV-mediated genome editing. Another important challenge is

developing an optimized delivery method for genome editing in specific cell types in the brain.
A

The brain consists of densely packed populations of numerous types of cells with distinct

functions. In addition, the dysfunction of specific cell types in specific brain areas is associated

with many genetic disorders. Therefore, cell type-specific modification of endogenous genes
would provide an ideal tool for exploring the complexity of the brain and potentially treating

brain disorders. Precise genome editing in a specific brain region is possible by injecting AAV

encoding HDR machinery into specific brain regions using vSLENDR. Therefore, the production

of Cas9-expressing AAV driven by small cell type-specific promoters or Cre-dependent AAV

PT
expressing Cas9 combined with hundreds of different Cre driver lines (Gerfen et al., 2013;

Taniguchi et al., 2011) would enable cell type-specific genome editing in defined brain areas.

RI
The CRISPR–Cas9 system has made genome editing a routine application for many

SC
neuroscientists. The continued development of versatile genome editing tools based on the

natural diversity of Cas nucleases, together with the improvement of the genome editing

U
specificity, efficiency, and delivery methods, will improve our understanding of complex brain

functions and disorders.

N
A
M
D

Acknowledgments
TE

I thank Drs. Toshihiro Nomura and Yoshihisa Nakahata for helpful comments on the manuscript.

This work was supported by Max Planck Florida Institute For Neuroscience.
EP
CC
A
References

Abudayyeh, O.O., Gootenberg, J.S., Konermann, S., Joung, J., Slaymaker, I.M., Cox, D.B.,
Shmakov, S., Makarova, K.S., Semenova, E., Minakhin, L., Severinov, K., Regev, A., Lander,
E.S., Koonin, E.V., Zhang, F., 2016. C2c2 is a single-component programmable RNA-guided
RNA-targeting CRISPR effector. Science 353, aaf5573.

Auer, T.O., Duroure, K., De Cian, A., Concordet, J.P., Del Bene, F., 2014. Highly efficient

PT
CRISPR/Cas9-mediated knock-in in zebrafish by homology-independent DNA repair. Genome
Res. 24, 142-153.

RI
Barnabe-Heider, F., Meletis, K., Eriksson, M., Bergmann, O., Sabelstrom, H., Harvey, M.A.,
Mikkers, H., Frisen, J., 2008. Genetic manipulation of adult mouse neurogenic niches by in vivo
electroporation. Nat Methods 5, 189-196.

SC
Barzel, A., Paulk, N.K., Shi, Y., Huang, Y., Chu, K., Zhang, F., Valdmanis, P.N., Spector, L.P.,
Porteus, M.H., Gaensler, K.M., Kay, M.A., 2015. Promoterless gene targeting without nucleases

U
ameliorates haemophilia B in mice. Nature 517, 360-364.

N
Canny, M.D., Moatti, N., Wan, L.C.K., Fradet-Turcotte, A., Krasner, D., Mateos-Gomez, P.A.,
Zimmermann, M., Orthwein, A., Juang, Y.C., Zhang, W., Noordermeer, S.M., Seclen, E.,
A
Wilson, M.D., Vorobyov, A., Munro, M., Ernst, A., Ng, T.F., Cho, T., Cannon, P.M., Sidhu,
S.S., Sicheri, F., Durocher, D., 2018. Inhibition of 53BP1 favors homology-dependent DNA
repair and increases CRISPR-Cas9 genome-editing efficiency. Nat. Biotechnol. 36, 95-102.
M

Chen, F., Rosiene, J., Che, A., Becker, A., LoTurco, J., 2015. Tracking and transforming
neocortical progenitors by CRISPR/Cas9 gene targeting and piggyBac transposase lineage
D

labeling. Development 142, 3601-3611.

TE

Chen, J.S., Dagdas, Y.S., Kleinstiver, B.P., Welch, M.M., Sousa, A.A., Harrington, L.B.,
Sternberg, S.H., Joung, J.K., Yildiz, A., Doudna, J.A., 2017. Enhanced proofreading governs
CRISPR-Cas9 targeting accuracy. Nature 550, 407-410.
EP

Chiou, S.H., Winters, I.P., Wang, J., Naranjo, S., Dudgeon, C., Tamburini, F.B., Brady, J.J.,
Yang, D., Gruner, B.M., Chuang, C.H., Caswell, D.R., Zeng, H., Chu, P., Kim, G.E., Carpizo,
D.R., Kim, S.K., Winslow, M.M., 2015. Pancreatic cancer modeling using retrograde viral vector
CC

delivery and in vivo CRISPR/Cas9-mediated somatic genome editing. Genes Dev. 29, 1576-
1585.
A

Chu, V.T., Weber, T., Wefers, B., Wurst, W., Sander, S., Rajewsky, K., Kuhn, R., 2015.
Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene
editing in mammalian cells. Nat. Biotechnol. 33, 543-548.

Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D., Wu, X., Jiang, W.,
Marraffini, L.A., Zhang, F., 2013. Multiplex genome engineering using CRISPR/Cas systems.
Science 339, 819-823.
Cox, D.B., Platt, R.J., Zhang, F., 2015. Therapeutic genome editing: prospects and challenges.
Nat. Med. 21, 121-131.

Cox, D.B.T., Gootenberg, J.S., Abudayyeh, O.O., Franklin, B., Kellner, M.J., Joung, J., Zhang,
F., 2017. RNA editing with CRISPR-Cas13. Science 358, 1019-1027.

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., Uchida, N., Hendel, A., Narla, A., Majeti, R.,

PT
Weinberg, K.I., Porteus, M.H., 2016. CRISPR/Cas9 beta-globin gene targeting in human
haematopoietic stem cells. Nature 539, 384-389.

Doudna, J.A., Charpentier, E., 2014. Genome editing. The new frontier of genome engineering

RI
with CRISPR-Cas9. Science 346, 1258096.

East-Seletsky, A., O'Connell, M.R., Knight, S.C., Burstein, D., Cate, J.H., Tjian, R., Doudna,

SC
J.A., 2016. Two distinct RNase activities of CRISPR-C2c2 enable guide-RNA processing and
RNA detection. Nature 538, 270-273.

U
Gaj, T., Epstein, B.E., Schaffer, D.V., 2016. Genome Engineering Using Adeno-associated
Virus: Basic and Clinical Research Applications. Mol. Ther. 24, 458-464.
N
Gaudelli, N.M., Komor, A.C., Rees, H.A., Packer, M.S., Badran, A.H., Bryson, D.I., Liu, D.R.,
A
2017. Programmable base editing of A*T to G*C in genomic DNA without DNA cleavage.
Nature 551, 464-471.
M

Gilbert, L.A., Horlbeck, M.A., Adamson, B., Villalta, J.E., Chen, Y., Whitehead, E.H.,
Guimaraes, C., Panning, B., Ploegh, H.L., Bassik, M.C., Qi, L.S., Kampmann, M., Weissman,
J.S., 2014. Genome-Scale CRISPR-Mediated Control of Gene Repression and Activation. Cell
D

159, 647-661.
TE

Gilbert, L.A., Larson, M.H., Morsut, L., Liu, Z., Brar, G.A., Torres, S.E., Stern-Ginossar, N.,
Brandman, O., Whitehead, E.H., Doudna, J.A., Lim, W.A., Weissman, J.S., Qi, L.S., 2013.
CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442-
EP

451.

Hsu, P.D., Lander, E.S., Zhang, F., 2014. Development and applications of CRISPR-Cas9 for
CC

genome engineering. Cell 157, 1262-1278.

Hu, J.H., Miller, S.M., Geurts, M.H., Tang, W., Chen, L., Sun, N., Zeina, C.M., Gao, X., Rees,
H.A., Lin, Z., Liu, D.R., 2018. Evolved Cas9 variants with broad PAM compatibility and high
A

DNA specificity. Nature.

Incontro, S., Asensio, C.S., Edwards, R.H., Nicoll, R.A., 2014. Efficient, complete deletion of
synaptic proteins using CRISPR. Neuron 83, 1051-1057.

Ishizu, T., Higo, S., Masumura, Y., Kohama, Y., Shiba, M., Higo, T., Shibamoto, M., Nakagawa,
A., Morimoto, S., Takashima, S., Hikoso, S., Sakata, Y., 2017. Targeted Genome Replacement
via Homology-directed Repair in Non-dividing Cardiomyocytes. Scientific reports 7, 9363.
Jinek, M., Chylinski, K., Fonfara, I., Hauer, M., Doudna, J.A., Charpentier, E., 2012. A
programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science
337, 816-821.

Kalebic, N., Taverna, E., Tavano, S., Wong, F.K., Suchold, D., Winkler, S., Huttner, W.B.,
Sarov, M., 2016. CRISPR/Cas9-induced disruption of gene expression in mouse embryonic brain
and single neural stem cells in vivo. EMBO reports 17, 338-348.

PT
Kamiyama, D., Sekine, S., Barsi-Rhyne, B., Hu, J., Chen, B., Gilbert, L.A., Ishikawa, H.,
Leonetti, M.D., Marshall, W.F., Weissman, J.S., Huang, B., 2016. Versatile protein tagging in
cells with split fluorescent protein. Nature communications 7, 11046.

RI
Kim, D., Kim, J., Hur, J.K., Been, K.W., Yoon, S.H., Kim, J.S., 2016. Genome-wide analysis
reveals specificities of Cpf1 endonucleases in human cells. Nat. Biotechnol. 34, 863-868.

SC
Kim, E., Koo, T., Park, S.W., Kim, D., Kim, K., Cho, H.Y., Song, D.W., Lee, K.J., Jung, M.H.,
Kim, S., Kim, J.H., Kim, J.S., 2017a. In vivo genome editing with a small Cas9 orthologue
derived from Campylobacter jejuni. Nature communications 8, 14500.

U
Kim, H.K., Song, M., Lee, J., Menon, A.V., Jung, S., Kang, Y.M., Choi, J.W., Woo, E., Koh,
N
H.C., Nam, J.W., Kim, H., 2017b. In vivo high-throughput profiling of CRISPR-Cpf1 activity.
Nat Methods 14, 153-159.
A
Kim, S., Kim, D., Cho, S.W., Kim, J., Kim, J.S., 2014. Highly efficient RNA-guided genome
M

editing in human cells via delivery of purified Cas9 ribonucleoproteins. Genome Res. 24, 1012-
1019.

Kleinstiver, B.P., Pattanayak, V., Prew, M.S., Tsai, S.Q., Nguyen, N.T., Zheng, Z., Joung, J.K.,
D

2016. High-fidelity CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects.

Nature 529, 490-495.
TE

Komor, A.C., Kim, Y.B., Packer, M.S., Zuris, J.A., Liu, D.R., 2016. Programmable editing of a
target base in genomic DNA without double-stranded DNA cleavage. Nature 533, 420-424.
EP

Konermann, S., Brigham, M.D., Trevino, A.E., Joung, J., Abudayyeh, O.O., Barcena, C., Hsu,
P.D., Habib, N., Gootenberg, J.S., Nishimasu, H., Nureki, O., Zhang, F., 2015. Genome-scale
CC

transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 517, 583-588.

Lin, S., Staahl, B.T., Alla, R.K., Doudna, J.A., 2014. Enhanced homology-directed human
genome engineering by controlled timing of CRISPR/Cas9 delivery. eLife 3, e04766.
A

Liu, X.S., Wu, H., Ji, X., Stelzer, Y., Wu, X., Czauderna, S., Shu, J., Dadon, D., Young, R.A.,
Jaenisch, R., 2016. Editing DNA Methylation in the Mammalian Genome. Cell 167, 233-247
e217.

Liu, X.S., Wu, H., Krzisch, M., Wu, X., Graef, J., Muffat, J., Hnisz, D., Li, C.H., Yuan, B., Xu,
C., Li, Y., Vershkov, D., Cacace, A., Young, R.A., Jaenisch, R., 2018. Rescue of Fragile X
Syndrome Neurons by DNA Methylation Editing of the FMR1 Gene. Cell 172, 979-992 e976.
Maeder, M.L., Linder, S.J., Cascio, V.M., Fu, Y., Ho, Q.H., Joung, J.K., 2013. CRISPR RNA-
guided activation of endogenous human genes. Nat Methods 10, 977-979.

Mali, P., Aach, J., Stranges, P.B., Esvelt, K.M., Moosburner, M., Kosuri, S., Yang, L., Church,
G.M., 2013a. CAS9 transcriptional activators for target specificity screening and paired nickases
for cooperative genome engineering. Nat. Biotechnol. 31, 833-838.

Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E., Church, G.M.,

PT
2013b. RNA-guided human genome engineering via Cas9. Science 339, 823-826.

Maresca, M., Lin, V.G., Guo, N., Yang, Y., 2013. Obligate ligation-gated recombination
(ObLiGaRe): custom-designed nuclease-mediated targeted integration through nonhomologous

RI
end joining. Genome Res. 23, 539-546.

Maruyama, T., Dougan, S.K., Truttmann, M.C., Bilate, A.M., Ingram, J.R., Ploegh, H.L., 2015.

SC
Increasing the efficiency of precise genome editing with CRISPR-Cas9 by inhibition of
nonhomologous end joining. Nat. Biotechnol. 33, 538-542.

U
Mikuni, T., Nishiyama, J., Sun, Y., Kamasawa, N., Yasuda, R., 2016. High-Throughput, High-
Resolution Mapping of Protein Localization in Mammalian Brain by In Vivo Genome Editing.
Cell 165, 1803-1817. N
A
Nishiyama, J., Hayashi, Y., Nomura, T., Miura, E., Kakegawa, W., Yuzaki, M., 2012. Selective
and regulated gene expression in murine Purkinje cells by in utero electroporation. Eur. J.
M

Neurosci. 36, 2867-2876.

Nishiyama, J., Mikuni, T., Yasuda, R., 2017. Virus-Mediated Genome Editing via Homology-
Directed Repair in Mitotic and Postmitotic Cells in Mammalian Brain. Neuron 96, 755-768 e755.
D

Nishiyama, J., Yasuda, R., 2015. Biochemical Computation for Spine Structural Plasticity.
TE

Neuron 87, 63-75.

Perez-Pinera, P., Kocak, D.D., Vockley, C.M., Adler, A.F., Kabadi, A.M., Polstein, L.R.,
EP

Thakore, P.I., Glass, K.A., Ousterout, D.G., Leong, K.W., Guilak, F., Crawford, G.E., Reddy,
T.E., Gersbach, C.A., 2013. RNA-guided gene activation by CRISPR-Cas9-based transcription
factors. Nat Methods 10, 973-976.
CC

Porteus, M.H., Cathomen, T., Weitzman, M.D., Baltimore, D., 2003. Efficient gene targeting
mediated by adeno-associated virus and DNA double-strand breaks. Mol. Cell. Biol. 23, 3558-
3565.
A

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., Koonin, E.V., Sharp, P.A., Zhang, F., 2015. In vivo genome editing
using Staphylococcus aureus Cas9. Nature 520, 186-191.

Russell, D.W., Hirata, R.K., 1998. Human gene targeting by viral vectors. Nat. Genet. 18, 325-
330.
Shinmyo, Y., Kawasaki, H., 2017. CRISPR/Cas9-Mediated Gene Knockout in the Mouse Brain
Using In Utero Electroporation. Current protocols in neuroscience 79, 3 32 31-33 32 11.

Shinmyo, Y., Tanaka, S., Tsunoda, S., Hosomichi, K., Tajima, A., Kawasaki, H., 2016.
CRISPR/Cas9-mediated gene knockout in the mouse brain using in utero electroporation.
Scientific reports 6, 20611.

Shinmyo, Y., Terashita, Y., Dinh Duong, T.A., Horiike, T., Kawasumi, M., Hosomichi, K.,

PT
Tajima, A., Kawasaki, H., 2017. Folding of the Cerebral Cortex Requires Cdk5 in Upper-Layer
Neurons in Gyrencephalic Mammals. Cell reports 20, 2131-2143.

Shmakov, S., Smargon, A., Scott, D., Cox, D., Pyzocha, N., Yan, W., Abudayyeh, O.O.,

RI
Gootenberg, J.S., Makarova, K.S., Wolf, Y.I., Severinov, K., Zhang, F., Koonin, E.V., 2017.
Diversity and evolution of class 2 CRISPR-Cas systems. Nature reviews. Microbiology 15, 169-
182.

SC
Slaymaker, I.M., Gao, L., Zetsche, B., Scott, D.A., Yan, W.X., Zhang, F., 2016. Rationally
engineered Cas9 nucleases with improved specificity. Science 351, 84-88.

U
Song, J., Yang, D., Xu, J., Zhu, T., Chen, Y.E., Zhang, J., 2016. RS-1 enhances CRISPR/Cas9-
N
and TALEN-mediated knock-in efficiency. Nature communications 7, 10548.
A
Staahl, B.T., Benekareddy, M., Coulon-Bainier, C., Banfal, A.A., Floor, S.N., Sabo, J.K., Urnes,
C., Munares, G.A., Ghosh, A., Doudna, J.A., 2017. Efficient genome editing in the mouse brain
M

by local delivery of engineered Cas9 ribonucleoprotein complexes. Nat. Biotechnol. 35, 431-434.

Sternberg, S.H., Doudna, J.A., 2015. Expanding the Biologist's Toolkit with CRISPR-Cas9.
Molecular cell 58, 568-574.
D

Straub, C., Granger, A.J., Saulnier, J.L., Sabatini, B.L., 2014. CRISPR/Cas9-mediated gene
TE

knock-down in post-mitotic neurons. PloS one 9, e105584.

Suzuki, K., Tsunekawa, Y., Hernandez-Benitez, R., Wu, J., Zhu, J., Kim, E.J., Hatanaka, F.,
EP

Yamamoto, M., Araoka, T., Li, Z., Kurita, M., Hishida, T., Li, M., Aizawa, E., Guo, S., Chen, S.,
Goebl, A., Soligalla, R.D., Qu, J., Jiang, T., Fu, X., Jafari, M., Esteban, C.R., Berggren, W.T.,
Lajara, J., Nunez-Delicado, E., Guillen, P., Campistol, J.M., Matsuzaki, F., Liu, G.H.,
CC

Magistretti, P., Zhang, K., Callaway, E.M., Belmonte, J.C., 2016. In vivo genome editing via
CRISPR/Cas9 mediated homology-independent targeted integration. Nature 540, 144-149.

Swiech, L., Heidenreich, M., Banerjee, A., Habib, N., Li, Y., Trombetta, J., Sur, M., Zhang, F.,
A

2015. In vivo interrogation of gene function in the mammalian brain using CRISPR-Cas9. Nat.
Biotechnol. 33, 102-106.

Tanenbaum, M.E., Gilbert, L.A., Qi, L.S., Weissman, J.S., Vale, R.D., 2014. A protein-tagging
system for signal amplification in gene expression and fluorescence imaging. Cell 159, 635-646.
Tsunekawa, Y., Terhune, R.K., Fujita, I., Shitamukai, A., Suetsugu, T., Matsuzaki, F., 2016.
Developing a de novo targeted knock-in method based on in utero electroporation into the
mammalian brain. Development 143, 3216-3222.

Uemura, T., Mori, T., Kurihara, T., Kawase, S., Koike, R., Satoga, M., Cao, X., Li, X.,
Yanagawa, T., Sakurai, T., Shindo, T., Tabuchi, K., 2016. Fluorescent protein tagging of
endogenous protein in brain neurons using CRISPR/Cas9-mediated knock-in and in utero
electroporation techniques. Scientific reports 6, 35861.

PT
Viswanathan, S., Williams, M.E., Bloss, E.B., Stasevich, T.J., Speer, C.M., Nern, A., Pfeiffer,
B.D., Hooks, B.M., Li, W.P., English, B.P., Tian, T., Henry, G.L., Macklin, J.J., Patel, R.,
Gerfen, C.R., Zhuang, X., Wang, Y., Rubin, G.M., Looger, L.L., 2015. High-performance probes

RI
for light and electron microscopy. Nat Methods 12, 568-576.

Wu, Z., Yang, H., Colosi, P., 2010. Effect of genome size on AAV vector packaging. Mol. Ther.

SC
18, 80-86.

Yoshimi, K., Kunihiro, Y., Kaneko, T., Nagahora, H., Voigt, B., Mashimo, T., 2016. ssODN-

U
mediated knock-in with CRISPR-Cas for large genomic regions in zygotes. Nature
communications 7, 10431.
N
Zetsche, B., Gootenberg, J.S., Abudayyeh, O.O., Slaymaker, I.M., Makarova, K.S.,
A
Essletzbichler, P., Volz, S.E., Joung, J., van der Oost, J., Regev, A., Koonin, E.V., Zhang, F.,
2015. Cpf1 is a single RNA-guided endonuclease of a class 2 CRISPR-Cas system. Cell 163,
M

759-771.

Zheng, Y., Shen, W., Zhang, J., Yang, B., Liu, Y.N., Qi, H., Yu, X., Lu, S.Y., Chen, Y., Xu,
D

Y.Z., Li, Y., Gage, F.H., Mi, S., Yao, J., 2018. CRISPR interference-based specific and efficient
gene inactivation in the brain. Nat. Neurosci. 21, 447-454.
TE

Zhou, H., Liu, J., Zhou, C., Gao, N., Rao, Z., Li, H., Hu, X., Li, C., Yao, X., Shen, X., Sun, Y.,
Wei, Y., Liu, F., Ying, W., Zhang, J., Tang, C., Zhang, X., Xu, H., Shi, L., Cheng, L., Huang, P.,
Yang, H., 2018. In vivo simultaneous transcriptional activation of multiple genes in the brain
EP

using CRISPR-dCas9-activator transgenic mice. Nat. Neurosci. 21, 440-446.

CC
A
Fig. 1. Genome editing mediated by engineered nucleases. (A) Schematic illustration of DNA

editing systems. Engineered nucleases, such as TALENs, ZFNs, CRISPR–Cas9, can induce

targeted DSBs in the genome that are subsequently repaired through either NHEJ or HDR. (B)

Comparison of common CRISPR–Cas systems. Schematic illustrations of CRISPR–Cas9,

PT
Cas12a, Cas13a, and Cas13b systems are shown. The crRNA and tracrRNA are shown in purple.

dsDNA and ssRNA are shown in gray. The functional and mechanistic characteristics of

RI
CRISPR–Cas systems are shown. AsCas12a, Cas12a from Acidaminococcus sp. BV3L6.

SC
LbCas12a, Cas12a from Lachnospiraceae bacterium ND2006. LwCas13a, Cas13a from

Leptotrichia wadei F0279. (C) Schematic illustrations of genome editing applications based on

U
catalytically dead Cas (dCas) mutants. By fusing dCas9 with transcriptional repressor and

N
transcriptional activator, a robust transcriptional repression and activation has been achieved
A
(left). By fusing dCas9 with the DNA methylation modification enzymes, targeted modification
M

of DNA methylation has been successfully achieved, leading to the modification of gene

expression (middle). By fusing dCas9 with cytidine or adenosine deaminases, it has been
D

realized that specific DNA base is converted into another at a targeted genomic locus (right).
TE
EP
CC
A
A
CC
EP
TE
D
M
A
N
U
SC
RI
PT
Fig. 2. Genome editing in the brain using SLENDR. (A) Schematic illustration of SLENDR.

Genome editing machinery can be delivered into neuronal progenitors through IUE or mitotic

and postmitotic cells via viral infection. By inserting a sequence encoding an epitope tag into a

gene of interest in a sparse subset of cells in the brain, SLENDR enables the imaging of

PT
endogenous proteins with high specificity and contrast in brain tissue. (B) SLENDR can be

exploited in diverse applications, some of which is shown. Pictures adapted from (Mikuni et al.,

RI
2016).

SC
U
N
A
M
D
TE
EP
CC
A
Table 1. Comparison of strategies for in vivo genome editing in the brain.

HITI SLENDR vSLENDR

Mechanism NHEJ HDR HDR
Mitotic and postmitotic Mitotic neuronal Mitotic and postmitotic
Target cells
cells progenitors cells

PT
10-15 % < 5% (HA knock-in) 10-15%
Efficiency*
(GFP knock-in) < 1% (GFP knock-in) (HA/GFP knock-in)
Speciticity △ ○ ○

RI
Knock-in allele may
on-target indels Knock-in allele is intact Knock-in allele is intact
contain indels
off-target indels Possible Possible Possible

SC
off-target insertion Possible Minimum Minimum
△ ○ ○
Multiplex labeling DNA fragments can be

U
non-specifically Possible Possible
inserted into DSB sites
○ △ △
Ease of vector
construction Homology arm is not
required
N
Homology arm is
required
Homology arm is
required
A
Costs Low-High Low High
* Efficiency in the brain in vivo
M
D
TE
EP
CC
A