CRISPR 101 Ebook Final

CRISPR 101:
A Desktop Resource
Created and Compiled by Addgene

January 2016 (1st Edition)
www.addgene.org
CRISPR 101: A Desktop Resource (1st Edition)
CRISPR 101: INTRODUCTION TO ADDGENES DESKTOP RESOURCE

If you work in molecular biology chances are youve heard of CRISPR. This amazing technology, best known
for its use in fast and easy genome editing, has accelerated research in many basic and applied disciplines. As
the tools developed around CRISPR have grown, so too have the resources available at Addgene thanks to the
generous contributions of depositors working in the field. To better help you understand how to use these resources, weve been steadfast in our efforts to keep the CRISPR educational content on both our website and
our blog up to date.
Now, to help you best utilize CRISPR and the plasmids kindly deposited by your colleagues weve compiled
this comprehensive eBook. Whether youre looking to use CRISPR for the first time, need some advice on a
particular CRISPR technique, or would simply like to learn more about how CRISPR came about, we hope that
youll find something interesting in this handy new resource.
If you have any questions about the eBook, have suggestions for new content, or would just like to send us a
note, please contact us at blog@addgene.org. Happy reading and good luck with your experiments!
~ The Addgene Team
www.addgene.org
blog.addgene.org
www.twitter.com/addgene
www.linkedin.com/company/addgene
www.facebook.com/addgene
www.youtube.com/user/addgenemedia
2 | Page
Table of Contents
TABLE OF CONTENTS
Page
Section
CRISPR 101: Introduction to Addgenes Desktop Resource
Chapter 1: Genome Engineering Overview?
10
Chapter 2: What is CRISPR?
18
Components of CRISPR/Cas9
21
CRISPR Glossary
23
The PAM Requirement and Expanding Beyond SpCas9
26
Cpf1: A Cas9 Homolog
28
Enhancing CRISPR Targeting Specificity with eSpCas9 and SpCas9-HF1
30
Chapter 3: Using CRISPR in Your Experiments
38
Which Cas9 Do I Choose for My CRISPR Experiment?
42
Using Cas9 to Generate Knock-Outs via NHEJ
46
Editing with Homology Directed Repair
50
Using Cas9 Nickases for Enhanced Specificity
52
How to Design Your gRNA for CRISPR Genome Editing
57
The CRISPR Software Matchmaker
63
Validating Your Genome Edit
67
Genome-wide Screening Using CRISPR/Cas9
70
Regulating Gene Expression with CRISPR
73
Tagging Genes with CRISPR
76
Using CRISPR to Visualize the Genome
79
CRISPR Purification
82
Chapter 4: CRISPR Expression Systems and Delivery Methods
89
Mammalian Expression Systems and Delivery Methods
11
31
83
Introduction to Genome Engineering

History of CRISPR
Planning Your CRISPR Experiment
Multiplexible CRISPR Expression Systems
3 | Page
Table of Contents
TABLE OF CONTENTS (CONT)

Page
Section
92
Using SaCa9 for Easier AAV-Based CRISPR Delivery
95
Adenoviral Delivery of CRISPR/Cas9 Aims to Expand Genome Engineering to Primary Cells
97
An Elegans Approach to Better CRISPR/Cas9 Editing Efficiency
100
Even More Elegant: Single Injection CRISPR/Cas9 in C. elegans
103
Another Pathway Into Cells: ITOP Protein Only Transduction
105
Chapter 5: Therapeutic Applications of CRISPR
108
Treating DMD Mice with CRISPR Gene Editing
112
Chapter 6: CRISPR Biosafety
120
CRISPR Concerns Beyond the Lab Bench
122
Acknowledgements and Final Words
106
117
Using CRISPR/Cas9 to Edit Disease Out of the Genome
Transgenic Organisms, Cas9 Gene Drives, and Appropriate Safeguards
4 | Page
Chapter 1 - Genome Engineering Overview
GENOME ENGINEERING OVERVIEW
5 | Page
INTRODUCTION TO GENOME ENGINEERING

By Mary Gearing | August 26, 2015
Genome engineering has made it possible to not only

dissect complex gene interactions, but also to build
new pathways through synthetic biology. The past few
decades have seen tremendous advances in both the
number and feasibility of genome engineering techniques,
many of which are available from Addgene.
Genome engineering was born in the late 1970s,
when multiple groups showed that exogenous DNA
could be taken up by yeast or bacteria and randomly
integrated into the genome. Subsequent work showed
that this process could also occur in a targeted fashion.
Addgene depositor Mario Capecchi realized that DNA
microinjection into a cells nucleus would stimulate cellular
homologous recombination, permitting targeted genome
modification. In 1989, he, Martin Evans and Oliver
Smithies created the first knockout mouse, a watershed
moment for genome engineering.
Cre-lox
Cre Recombinase Complexed to DNA (1).
The late 1980s also marked the introduction of Cre-lox recombination, a system derived from P1 bacteriophage
now widely used to control gene expression. Today, Cre recombinase under the control of various promoters,
or in its inducible form, provides sophisticated spatiotemporal control of gene expression, especially in mouse
transgenics.
The Cre-lox system is a technology that can be used to induce site-specific recombination events. The system
consists of two components derived from the P1 bacteriophage: the Cre recombinase and a loxP recognition
site. The P1 bacteriophage uses these components as part of its natural viral lifecycle, and researchers have
adapted the components for use in genome manipulation.
Cre recombinase, originally named because it causes recombination (although later referred to as the cyclization recombinase), is a 38 kDa protein responsible for intra- and inter-molecular recombination at loxP
recognition sites. A key advantage of the system is that Cre acts independently of any other accessory proteins
or co-factors, thus allowing for broad applications in a variety of experiments.
LoxP (locus of X(cross)-over in P1) sites are 34-base-pair long recognition sequences consisting of two 13-bp
long palindromic repeats separated by an 8-bp long asymmetric core spacer sequence. The asymmetry in the
core sequence gives the loxP site directionality, and the canonical loxP sequence is ATAACTTCGTATA-GCATACATTATACGAAGTTAT. The loxP sequence does not occur naturally in any known genome other than P1 phage,
and is long enough that there is virtually no chance of it occurring randomly. Therefore, inserting loxP sites at
deliberate locations in a DNA sequence allows for very specific manipulations as shown in the figure below.
6 | Page
INTRODUCTION TO GENOME ENGINEERING (CONT)

DNA Manipulations Possible with Cre-lox
Inversion: If the loxP sites are on the same DNA
strand and are in opposite orientations, recombination results in an inversion and the region of DNA
between the loxP sites is reversed. Deletion: If
the sites face in the same direction, the sequence
between the loxP sites is excised as a circular piece
of DNA (and is not maintained). Translocation: If the
sites are on separate DNA molecules, a translocation event is generated at the loxP sites.
Homologous Recombination and the Journey Towards CRISPR/Cas9

Homologous recombination is a cornerstone of genome engineering, but with the caveat that it occurs at
low frequencies, limiting editing efficiency. To improve editing rates, researchers hijacked the function of
endonucleases, which create difficult-to-repair DNA double-stranded breaks. Targeting these enzymes to a
given locus results in DNA cleavage and forces the cell to undergo either non-homologous end joining (NHEJ)
or homology-directed repair (HDR). NHEJ occurs if no DNA repair template is provided, and its error-prone
nature often leads to inactivating mutations. In the case of HDR, a repair template specifies desired genomic
modifications, enabling precise editing. In addition to introducing point mutations or recombination sites, repair
templates can also be used to introduce a gene of interest into a given locus.
Zinc Finger Nucleases

Zinc finger nucleases (ZFNs) represent the first
step towards efficient, targeted nucleases.
To create a ZFN, a series of zinc fingers is
designed to bind to a specific genomic locus,
and subsequently fused to FokI nuclease. Paired
ZFNs recognizing two adjacent sites cleave DNA,
initiating HDR. The utility of ZFNs is limited by their long synthesis time and nonmodular assembly process.
Although computational tools helped improve targeting, its not possible to design suitable ZFN pairs for every
genomic locus.
TALENs
First reported in 2011, TALENs represented a

huge step forward for genome engineering. This
modular system is based on TAL effector DNA
binding proteins, isolated from Xanthomonas spp,
fused to FokI endouclease. Whereas it might take
an experienced scientist six weeks to make a
ZFN, a newbie could make a TALEN in just a few
days! TALEN technolgy was rapidly adopted by the
research community, with the Golden Gate TALEN
kit becoming Addgenes most popular kit. The customizable DNA-binding properties of TALENs also enabled
the design of custom transcription factors to modulate gene expression.
7 | Page

CRISPR
Just when we thought genome
engineering couldnt get any better,
along came the CRISPR (Clustered
Regularly Interspaced Short
Palindromic Repeats)/Cas9 system.
CRISPR is an important component
of the bacterial immune system that
allows bacteria to remember and
destroy phages. In genome engineering
applications, Cas9 endonuclease
is targeted by guide RNA (gRNA)
sequence homology to a given locus,
where it induces a double stranded break. Like ZFNs and TALENs, CRISPR/Cas9 employs HDR, but the use of
RNA to specify editing makes the system much less expensive and time-consuming, as well as more precise
and scalable. For this reason, CRISPR/Cas9 has proved to be incredibly valuable for high-throughput genome
engineering. CRISPR/Cas9 can also target multiple loci in one organism, and like TALENs, the system has also
been adapted for other functions. CRISPRs are even more accessible to the research community than TALENs,
with new papers using this technology published every week.
Other Important Genome Engineering Tools

With the success of CRISPR, its easy to forget about other genome engineering methods...but you shouldnt!
Another HDR-based method, recombineering, is commonly used in E. coli to make edits to the genome or
a bacterial artificial chromosome (BAC). 50 bp homology arms flanking the repair template specify the site
of recombination, catalyzed by phage recombinases. Since repair templates can be quickly generated using
PCR, recombineering is easily scalable just like CRISPR. In addition to its applications in bacterial genome
engineering, recombineering is also useful for creating BAC-based repair templates for other HDR genome
engineering methods.
Outside of HDR, an exciting new gene transfer method uses the Sleeping Beauty transposon, reconstructed
from fossils of ancient fish. Transposons are mobile DNA elements, and are thus ideal for gene transfer. This
system is divided across two plasmids, one containing the gene of interest (GOI) flanked by inverted repeats,
the other carrying the transposase. Following cotransfection, the transposase cleaves the GOI from the plasmid
and facilitates a double strand break to allow the GOI to integrate into the genome. Sleeping Beauty was
named 2009s Molecule of the Year, and it represents a robust alternative to viral vector-mediated gene transfer.
Conclusion
With multiple robust and efficient genome engineering methods at our fingertips, we have entered a Golden
Age of genome engineering. Current work focuses on refining these techniques to ensure high specificity and
activity, whatever the desired target locus (or loci) may be, with the hope that these methods will be useful
clinically. What we at Addgene find most exciting is the democratization of genome engineering, which has and
will continue to allow researchers all over the world to use these tools in their research. Continue reading to
catch up on all things CRISPR.
8 | Page

Further Reading
1. Guo, Feng, Deshmukh N. Gopaul, and Gregory D. Van Duyne. Structure of Cre recombinase complexed with DNA in a site-specific recombination synapse. Nature 389.6646 (1997): 40-46. PubMed
PMID: 9288963.
2. Addgenes Genome Engineering Guide
9 | Page
Chapter 2 - What is CRISPR?
WHAT IS CRISPR?
10 | Page
HISTORY OF CRISPR
By Mary Gearing | January, 2015
The CRISPR revolution has began, and it shows no signs of slowing down. This system, which is key to
prokaryotic adaptive immunity, has proven to be especially amenable to genome engineering. CRISPR/Cas
offers flexibility, as well as easy multiplexing and scaling, far beyond the capacities of previous systems.
Prokaryotes have long utilized CRISPR/Cas as a powerful defensive strategy against viral invaders, and this
system is proving to be just as useful (if not more so) for research applications.
CRISPR/Cas: An Adaptive Immune System
Figure 1. An overview of CRISPR/Cas as a bacterial adaptive immune system. When foreign viral or plasmid DNA enters the cell, a Cas complex recognizes it and cleaves it into small fragments (1), adding a new spacer (2) to the end of the CRISPR array. This array contains small pieces of DNA from past
invaders (3), but does not contain PAM sites, so it is recognized as self. The CRISPR array is transcribed into a long RNA (4) that is subsequently cleaved
into mature crRNAs (5). These crRNAs direct the Cas complex to the foreign DNA based on sequence specificity (6), allowing the DNA to be cleaved and
destroyed (7). Wikipedia, accessed 25 November 2013. Author: James Atmos (3).
11 | Page
HISTORY OF CRISPR (CONT)
CRISPR (Clustered Regularly Interspaced Short Palindromic Repeat) sequences were initially discovered in the
E. coli genome in 1987 (1), but their function as a safeguard against bacteriophages was not elucidated until
2007 (2). Initial experiments exposed S. thermophilus, a bacterium important for yogurt and cheese production,
to predatory phages to test if exogenous phage DNA could be incorporated into the bacterial genome as
part of the CRISPR repeats. Cas (CRISPR associated) genes, which code for polymerases, nucleases, and
helicases, were also disrupted to determine their various roles in this process. Scientists hypothesized that
prokaryotes had developed an adaptive immune system - utilizing various Cas genes to not only store a
record of invading phages but also to destroy the phage upon re-exposure (2,3) (Figure 1). More specifically,
specialized Cas proteins snip foreign DNA into small fragments approximately 30 bp in length and paste them
into the CRISPR sequence. Separate Cas proteins then express and process the CRISPR loci to generate
CRISPR RNAs (crRNAs). Through sequence homology, these crRNAs guide a Cas nuclease to the specified
exogenous genetic material, located next to the species-specific protospacer adjacent motif (PAM). The
CRISPR/Cas complex binds to the foreign DNA and cleaves it to destroy the invader.
CRISPR systems, found in 95% of archaeal and 48% of bacterial genomes, are highly diverse, with variation in
PAM sequences and the number and type of Cas proteins (4). Makarova et al.s classification defines 5 types
and 16 subtypes based on shared characteristics and evolutionary similarity. These are grouped into two large
classes based on the structure of the effector complex that cleaves genomic DNA (5). The Type II CRISPR/Cas
system was the first harnessed for genome engineering, with Type V following in 2015.
Fighting Back: Evolution of Anti-CRISPR Genes in Phage

The CRISPR/Cas adaptive immune system seems like a winning strategy to ensure prokaryotic survival, but it is
not invincible. In the original S. thermophilus experiments, virulent phage 858 was able to generate single point
mutations in the S1 spacer region, preventing the Cas nucleases from re-identifying it, and thus circumventing
the CRISPR/Cas defense mechanism (3). Surprisingly, only a few genes that neutralize CRISPR/Cas have been
identified. A 2013 study found five distinct anti-CRISPR genes in bacteriophages infecting Pseudomonas
aeruginosa. Addition of one of the five anti-CRISPR genes allowed a phage to evade P. aeruginosas Type
I-F CRISPR/Cas system, while mutating a phages anti-CRISPR gene had the opposite effect (6). These
genes were only able to deactivate the Type I-F system and are not translatable to other CRISPR subtypes.
Subsequent work found anti-CRISPRs inhibiting Type I-E CRISPR/Cas in P. aeruginosa, but these genes do not
12 | Page

inhibit type I-E activity in other bacteria such as E. coli (7).
Later work has shown that these anti-CRISPR genes employ varied mechanisms to block CRISPR/Cas. Of the
three proteins studied, two prevented the CRISPR-Cas complex from binding to DNA, but did so by binding
different complex subunits. Another anti-CRISPR binds to the helicase-nuclease Cas3, thus preventing its
recruitment to the CRISPR-Cas complex (8).
The high diversity and near ubiquity of CRISPR systems suggests that there may be a variety of undiscovered
anti-CRISPR genes, but these genes will likely provide limited protection, as seen with the previously
characterized anti-CRISPRs. It will be important to monitor this molecular arms race to see what new strategies
both bacteria and phage develop to oppose one another. Like the current CRISPR systems, these battles may
provide previously unimagined tools for genome engineering.
Genome Engineering Applications

In 2012, Jinek et al. first demonstrated that CRISPR/Cas could be programmed for targeted DNA cleavage
in vitro (9). In 2013, Cong et al. and Mali et al. described CRISPR/Cas-based genome editing in cell culture
(10, 11). Less than four years later, PubMed lists more than 2600 CRISPR-related publications, many of which
detail work to improve the tools specificity, orthogonality, and multiplexibility in various species, as well as the
development of new applications. The first CRISPR papers described two main categories of genome edits.
First, the double-stranded break induced by Cas9 could be repaired through non-homologous endjoining
(NHEJ), an error-prone process that produces insertions/deletions disrupting a target gene. On the other hand,
if a repair template with homology arms was supplied, the break could be repaired according to this template,
allowing for precise gene editing. These edits can include mutations seen in human patients, protein tags, or
loxP/FRT sites. Homology-directed repair (HDR) allows users to create a variety of modifications in endogenous
loci with unprecedented speed and specificity.
CRISPR/Cas technology development has occurred at a truly astonishing pace, with much work directed
towards increasing specificity. A nickase mutant, which cuts only one strand of DNA, is now commonly used
with paired gRNAs to lower off-target cleavage frequency (10). gRNA length has also been optimized; truncated
gRNAs with <20 base homology display less off-target activity (12). Two exciting new Cas9 variants display
low off-target activity due to rationally designed mutations. Mutations in enhanced SpCas9 (eSpCas9) reduce
the enzymes helicase activity, lowering cleavage at off-target sites without compromising on-target editing
efficiency (13). Similarly, SpCas9-HF contains engineered mutations in the SpCas9 residues that normally form
hydrogen bonds with DNA. Like eSpCas9, SpCas9-HF lowers off-target effects to nearly undetectable levels
(14).
Multiple Cas9 orthologs from other bacterial species, such as Staphylococcus aureus, have also been
characterized (15). These orthologs have distinct PAM sites that do not cross react, allowing multiple Cas9s to
be used together. SpCas9 mutants with alternative PAM sites have also been identified. One mutant, D1135E,
retains the same 3-NGG PAM, but is more selective for this PAM sequence than wt SpCas9, which may reduce
the frequency of off-target effects (16). Cpf1, another CRISPR nuclease, recognizes a 5-TTN PAM, improving
targeting in AT-rich genomes (17). This expansion of possible PAM sequences moves us closer to being able
to target every locus in every genome. PAM diversity will also improve HDR applications, as editing efficiency
increases when the PAM is in close proximity to the desired edit site.
CRISPR/Cas has truly revolutionized genetics in a variety of model organisms. C. elegans was one of the
first organisms to which CRISPR was applied - today, you can generate a protein fusion or loss of function
13 | Page

mutations in C. elegans with just a few hours of hands-on time (18)! Zebrafish and Drosophila are also easily
modified using CRISPR (19,20). For those interested in mammalian systems, CRISPR has made it easier than
ever to generate knockout or mutant animals; injecting RNA encoding Cas9 and and the appropriate gRNA
into mouse embryos results in efficient germline editing (21). Combining SaCas9 with adeno-associated viral
vectors has also made postnatal genome editing a reality in mice (15).
Genome editing with CRISPR/Cas is not only simple, but also scalable. Multiple gRNAs targeting multiple loci
can be easily expressed in the same cell or organism (10, 11). The use of RNA to target a locus makes CRISPR
especially amenable to genetic screens. Pooled gRNA libraries can be used to identify genes that are important
to a given phenotype (22, 23). Current libraries are available for gene knockouts, as well as transcriptional
activation or repression (24). Combined with the power of next-generation sequencing, CRISPR/Cas9 is the
most robust system yet for genome-wide screening.
Using CRISPR for Non-editing Applications

Scientists have also used the targeting capability of CRISPR Cas9 to perturb and study biology without altering
DNA sequence. For instance, a catalytically dead Cas9 mutant (dCas9) has been especially useful, with fusion
proteins available to help researchers activate or deactivate transcription of given genes (25). Fluorescently
labeled dCas9 helps visualize genomic loci of interest prompting new work into nuclear organization (26).
Epitope-tagged dCas9 can also be used to purify a genomic locus and its associated proteins or RNAs, a
process called enCHIP (27). Multiple methods have been developed to tag endogenous proteins, enabling
specific protein isolation without the need for a custom antibody (28, 29).
Future Directions
Its amazing to see how much progress has been made in just a few years, with the skyrocketing popularity of
CRISPR echoing that of RNAi and iPS cells and quickly surpassing that of its predecessors ZFNs and TALEs.
Genome engineers continue to work to develop a highly specific, programmable platform well-suited for various
biological and translational technologies. Biotechnology companies are exploring therapeutic applications
of CRISPR to treat genetic disease, with the caveat that off-target editing risk must be made very low or
nonexistent. Three studies published concurrently in Science demonstrate in vivo treatment of mice suffering
from Duchenne muscular dystrophy; these studies represent the first evidence of successful postnatal genome
editing in a disease model (30, 31, 32). Scientists continue to wrestle with the possibility of postnatal and
germline editing, with the International Summit of Human Gene Editing convened in 2015 to discuss the ethics
associated with CRISPR editing in agricultural and therapeutic applications.
Despite the ethical controversies surrounding non-research applications, its clear that CRISPR is a truly
disruptive technology for basic research. The beauty of CRISPR/Cas is that its adaptable to almost any
model system or biological topic, and you dont need to be an expert to see results! Addgene has empowered
researchers to harness previous experimental successes and further develop the CRISPR/Cas toolkit by
posting lab protocols, providing tips from experts in the field, and enabling access to multiple plasmids used for
various applications. We hope this eBook will further facilitate the rapid use and development of CRISPR tools!
14 | Page

Further Reading
1. Ishino Y, Shinagawa H, Makino K, Amemura M, Nakata A. Nucleotide sequence of the iap gene,
responsible for alkaline phosphatase isozyme conversion in Escherichia coli, and identification of the gene
product. J Bacteriol. 1987 Dec;169(12):5429-33. PubMed PMID: 3316184.
2. Barrangou R, Fremaux C, Deveau H, Richards M, Boyaval P, Moineau S, Romero DA, Horvath P.
CRISPR provides acquired resistance against viruses in prokaryotes. Science. 2007 Mar 23;315(5819):1709-12.
PubMed PMID: 17379808.
3. Atmos J. Diagram of the possible mechanism for CRISPR. Photo: Wikipedia. 2009 Sept 15. http://
en.wikipedia.org/wiki/File:Crispr.png
4. IGM, Universite Paris-Sud. CRISPRdb. Last updated 2014 Aug 05. http://crispr.u-psud.fr/crispr/
CRISPRdatabase.php
5. Makarova KS, Wolf YI, Alkhnbashi OS, Costa F, Shah SA, Saunders SJ, Barrangou R, Brouns SJ,
Charpentier E, Haft DH, Horvath P, Moineau S, Mojica FJ, Terns RM, Terns MP, White MF, Yakunin AF, Garrett
RA, van der Oost J, Backofen R, Koonin EV. An updated evolutionary classification of CRISPR-Cas systems.
Nat Rev Microbiol. 2015 Nov;13(11):722-36. PubMed PMID: 26411297.
6. Bondy-Denomy J, Pawluk A, Maxwell KL, Davidson AR. Bacteriophage genes that inactivate the
CRISPR/Cas bacterial immune system. Nature. 2013 Jan 17;493(7432):429-32. PubMed PMID: 23242138.
7. Pawluk A, Bondy-Denomy J, Cheung VH, Maxwell KL, Davidson AR. A new group of phage antiCRISPR genes inhibits the type I-E CRISPR-Cas system of Pseudomonas aeruginosa. MBio. 2014 Apr
15;5(2):e00896. PubMed PMID: 24736222.
8. Bondy-Denomy J, Garcia B, Strum S, Du M, Rollins MF, Hidalgo-Reyes Y, Wiedenheft B, Maxwell KL,
Davidson AR. Multiple mechanisms for CRISPR-Cas inhibition by anti-CRISPR proteins. Nature. 2015 Oct
1;526(7571):136-9. PubMed PMID: 26416740.
9. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E. A programmable dual-RNA-guided
DNA endonuclease in adaptive bacterial immunity. Science. 2012 Aug 17;337(6096):816-21. PubMed PMID:
22745249.
10. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA, Zhang F.
Multiplex genome engineering using CRISPR/Cas systems. Science. 2013 Jan; 339 (6121): 81923. PubMed
PMID: 23287718. PubMed Central PMCID: PMC3795411.
11. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM. RNA-guided human
genome engineering via Cas9. Science. 2013 Jan;339 (6121): 8236. PubMed PMID: 23287722. PubMed
Central PMCID: PMC3712628.
12. Fu Y, Sander JD, Reyon D, Cascio VM, Joung JK. Improving CRISPR-Cas nuclease specificity using
truncated guide RNAs. Nat Biotechnol. 2014 Mar;32(3):279-84. PubMed PMID: 24463574. PubMed Central
PMCID: PMC3988262.
13. Slaymaker IM, Gao L, Zetsche B, Scott DA, Yan WX, Zhang F. Rationally engineered Cas9 nucleases
with improved specificity Science. 2016 Jan 1;351(6268):84-8. PubMed PMID: 26628643.
14. Kleinstiver BP, Pattanayak V, Prew MS, Tsai SQ, Nguyen NT, Zheng Z, Joung JK. High-fidelity CRISPRCas9 nucleases with no detectable genome-wide off-target effects. Nature. 2016 Jan 6. doi: 10.1038/
nature16526. PubMed PMID: 26735016.
15. Ran FA, Cong L, Yan WX, Scott DA, Gootenberg JS, Kriz AJ, Zetsche B, Shalem O, Wu X, Makarova KS,
Koonin EV, Sharp PA, Zhang F. In vivo genome editing using Staphylococcus aureus Cas9. Nature. 2015 Apr
9;520(7546):186-91. PubMed PMID: 25830891. PubMed Central PMCID: PMC4393360
15 | Page

16.
Kleinstiver BP, Prew MS, Tsai SQ, Topkar VV, Nguyen NT, Zheng Z, Gonzales AP, Li Z, Peterson RT, Yeh
JR, Aryee MJ, Joung JK. Engineered CRISPR-Cas9 nucleases with altered PAM specificities. Nature. 2015 Jul
23;523(7561):481-5. PubMed PMID: 26098369.
17. Zetsche B, Gootenberg JS, Abudayyeh OO, Slaymaker IM, Makarova KS, Essletzbichler P, Volz SE,
Joung J, van der Oost J, Regev A, Koonin EV, Zhang F. Cpf1 Is a Single RNA-Guided Endonuclease of a Class
2 CRISPR-Cas System. Cell. 2015 Oct 22;163(3):759-71. PubMed PMID: 26422227.
18. Dickinson DJ, Pani AM, Heppert JK, Higgins CD, Goldstein B. Streamlined Genome Engineering with a
Self-Excising Drug Selection Cassette. Genetics. 2015 Aug;200(4):1035-49. PubMed PMID: 26044593.
19. Jao LE, Wente SR, Chen W. Efficient multiplex biallelic zebrafish genome editing using a CRISPR
nuclease system. Proc Natl Acad Sci U S A. 2013 Aug 20;110(34):13904-9. PubMed PMID: 23918387.
20. Port F, Chen HM, Lee T, Bullock SL. Optimized CRISPR/Cas tools for efficient germline and somatic
genome engineering in Drosophila. Proc Natl Acad Sci U S A. 2014 Jul 22;111(29):E2967-76. PubMed PMID:
25002478.
21. Wang H, Yang H, Shivalila CS, Dawlaty MM, Cheng AW, Zhang F, Jaenisch R. One-step generation
of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell. 2013 May
9;153(4):910-8. PubMed PMID: 23643243.
22. Wang T, Wei JJ, Sabatini DM, Lander ES. Genetic Screens in Human Cells Using the CRISPR/
Cas9 System. Science. 2014 Jan 3;343(6166):80-4. PubMed PMID: 24336569. PubMed Central PMCID:
PMC3972032.
23. Shalem O, Sanjana NE, Hartenian E, Shi X, Scott DA, Mikkelson T, Heckl D, Ebert BL, Root DE,
Doench JG, Zhang F. Genome-Scale CRISPR-Cas9 Knockout Screening in Human Cells. Science. 2014 Jan
3;343(6166):84-7. PubMed PMID: 24336571. PubMed Central PMCID: PMC4089965.
24. Gilbert LA, Horlbeck MA, Adamson B, Villalta JE, Chen Y, Whitehead EH, Guimaraes C, Panning B,
Ploegh HL, Bassik MC, Qi LS, Kampmann M, Weissman JS. Genome-Scale CRISPR-Mediated Control of Gene
Repression and Activation. Cell. 2014 Oct 8. PubMed PMID: 25307932. PubMed Central PMCID: PMC4253859.
25. Gilbert LA, Larson MH, Morsut L, Liu Z, Brar GA, Torres SE, Stern-Ginossar N, Brandman O,
Whitehead EH, Doudna JA, Lim WA, Weissman JS, Qi LS. CRISPR-mediated modular RNA-guided regulation
of transcription in eukaryotes. Cell. 2013 Jul 18;154(2):442-51. PubMed PMID: 23849981. PubMed Central
PMCID: PMC3770145.
26. Chen B, Gilbert LA, Cimini BA, Schnitzbauer J, Zhang W, Li GW, Park J, Blackburn EH, Weissman JS, Qi
LS, Huang B. Dynamic imaging of genomic loci in living human cells by an optimized CRISPR/Cas system. Cell.
2013 Dec 19;155(7):1479-91. PubMed PMID: 24360272. PubMed Central PMCID: PMC3918502.
27. Fujita T, Fujii H. Efficient isolation of specific genomic regions and identification of associated proteins
by engineered DNA-binding molecule-mediated chromatin immunoprecipitation (enChIP) using CRISPR.
Biochem Biophys Res Commun. 2013 Sep 13;439(1):132-6. PubMed PMID: 23942116.
28. Savic D, Partridge EC, Newberry KM, Smith SB, Meadows SK, Roberts BS, Mackiewicz M, Mendenhall
EM, Myers RM. CETCh-seq: CRISPR epitope tagging ChIP-seq of DNA-binding proteins. Genome Research
2015 Oct;25(10):1581-9. PubMed PMID: 26355004.
29. Dalvai M, Loehr J, Jacquet K, Huard CC, Roques C, Herst P, Ct J, Doyon Y. A Scalable GenomeEditing-Based Approach for Mapping Multiprotein Complexes in Human Cells. Cell Rep. 2015 Oct 20;13(3):62133. PubMed PMID: 26456817.
30. Nelson CE, Hakim CH, Ousterout DG, Thakore PI, Moreb EA, Rivera RM, Madhavan S, Pan X, Ran
FA, Yan WX, Asokan A, Zhang F, Duan D, Gersbach CA. In vivo genome editing improves muscle function in a
mouse model of Duchenne muscular dystrophy. Science. 2015 Dec 31. pii: aad5143. PubMed PMID: 26721684.
31. Tabebordbar M, Zhu K, Cheng JK, Chew WL, Widrick JJ, Yan WX, Maesner C, Wu EY, Xiao R, Ran FA,
Cong L, Zhang F, Vandenberghe LH, Church GM, Wagers AJ. In vivo gene editing in dystrophic mouse muscle
and muscle stem cells. Science. 2015 Dec 31. pii: aad5177. PubMed PMID: 26721686.
32. Long C, Amoasii L, Mireault AA, McAnally JR, Li H, Sanchez-Ortiz E, Bhattacharyya S, Shelton JM,
16 | Page

Bassel-Duby R, Olson EN. Postnatal genome editing partially restores dystrophin expression in a mouse model
of muscular dystrophy. Science. 2015 Dec 31. pii: aad5725. PubMed PMID: 26721683.
17 | Page
COMPONENTS OF CRISPR/CAS9
By Joel McDade, Tyler J. Ford | October, 2015
At their most basic level, CRISPR/Cas9 genome editing systems use a non-specific endonuclease (Cas9 or
closely related Cpf1) to cut the genome and a small RNA (gRNA) to guide this nuclease to a user-defined cut
site. In this section we will go into a bit more detail about the components of these major players. After reading
this section, we hope you will be caught up on much of the major CRISPR lingo and will be able to describe the
functions of the various CRISPR/Cas9 components. If you ever need to quickly look up any CRISPR associated
terminology, be sure to hop down to the CRISPR glossary. Please note that while this section is intended to
provide a general overview of CRISPR components, new Cas9 variants are being discovered all time and the
requirements of these different systems can vary (for example, read our Cpf1 section for some of the interesting
properties of this exciting new nuclease tool). You can keep up on all the exciting new developments in CRISPR
research on our blog.
The Endonuclease
While native CRISPR/Cas systems have a variety of
enzymes responsible for processing foreign DNA as
well as the RNA guides required for endonuclease
function, when used for genome editing, the only
CRISPR protein required is the Cas9 endonuclease
or a variant thereof. This individual protein has all the
components necessary to:

1. Bind to a Guide RNA
The guide RNA, which will be described in more detail

below, enables Cas9 to cut a specific genomic locus of
many possible loci. Without binding to the guide RNA,
Cas9 cannot cut.

2. Bind to Target DNA in the Presence of a Guide RNA Provided that Target is Upstream (5) of a
Protospacer Adjacent Motif (PAM)
Cas9 endonuclease binding to the target genomic locus is mediated both by the target sequence contained
within the guide RNA and a 3-base pair sequence known as the Protospacer Adjacent Motif or PAM. In order
for dsDNA to be cut by Cas9, it must contain a PAM sequence immediately downstream (3) of the site targeted
by the guide RNA. In the absence of either the guide RNA or a PAM sequence, Cas9 will neither bind nor cut
the target. Cas9 homologs from different organisms or Cas9 mutants developed in a variety of labs (see table
below) have different PAM requirements. These different PAM requirements allow researchers to target many
different genomic loci.

3. Cleave Target DNA Resulting in a Double-Strand Break (DSB)
Cas9 and its variants have two endonuclease domains: the n-terminal RuvC-like nuclease domain and
the HNH-like nuclease domain near the center of the protein. Upon target binding, Cas9 undergoes a
conformational change that positions the nuclease domains to cleave opposite strands of the target DNA. Thus,
the end result of Cas9-mediated DNA damage is a DSB within the target DNA ~3-4 nucleotides upstream of the
PAM sequence.
18 | Page
COMPONENTS OF CRISPR/CAS9 (CONT)

Cas9 Species/Variants and PAM Sequences
Species/Variant of Cas9
PAM Sequence
Streptococcus pyogenes (SP); SpCas9
NGG
SpCas9 D1135E variant
NGG (reduced NAG binding)
SpCas9 VRER variant
NGCG
SpCas9 EQR variant
NGAG
SpCas9 VQR variant
NGAN or NGNG
Staphylococcus aureus (SA); SaCas9
NNGRRT or NNGRR(N)
Neisseria meningitidis (NM)
NNNNGATT
Streptococcus thermophilus (ST)
NNAGAAW
Treponema denticola (TD)
NAAAAC
Cpf1 (from various species)
TTN
Additional Cas9s from various species
PAM sequence may not be characterized
The Synthetic Guide RNA or gRNA (Sometimes sgRNA)

In the native Type II CRISPR/Cas system, Cas9 is guided to its target sites with the aid of two RNAs: the crRNA
which defines the genomic target for Cas9, and the tracrRNA which acts as a scaffold linking the crRNA to
Cas9 and facilitates processing of mature crRNAs from pre-crRNAs derived from CRISPR arrays. In most
systems used for CRISPR-mediated genome editing, these two small RNAs have been condensed into one
RNA sequence known as the guide RNA (gRNA) or single guide RNA (sgRNA). Throughout the remainder of
this eBook, well refer to this RNA complex as the gRNA. The gRNA contains both the 20 nucleotide target
sequence to direct Cas9 to a specific genomic locus and the scaffolding sequence necessary for Cas9 binding.
When using CRISPR/Cas9 for genome editing, researchers simply need to express a gRNA designed to direct
Cas9 to their target sequence of choice (see tips for designing a gRNA in a later section) and their prefered
Cas9 variant (with the appropriate PAM sequence) to modify the desired genomic locus.
Some Historical Notes

CRISPR arrays in bacterial genomes consist of repeated elements separated by unique sequences. When
researchers first discovered these arrays, they did not know their functions and simply called the repeated
elements direct repeats and the unique stretches of DNA between them spacers (see figure on the next
page).
After years of research, we now know that each direct repeat, combined with its adjacent spacer, ultimately
encodes a single crRNA. The direct repeat regions contain sequences required for processing pre-crRNA into
mature crRNA and tracrRNA binding. The spacer regions, on the other hand, are the unique, foreign DNA target
sequences specific to each individual crRNA.
19 | Page
COMPONENTS OF CRISPR/CAS9 (CONT)

Bottom line, the direct repeat region combined with the tracrRNA forms the scaffold portion of a gRNA and
the spacer region forms the target sequence.
Overview of native CRISPR arrays and their processing for cleaving foreign DNA. A) CRISPR arrays as found in a bacterial genomes are transcribed into
pre-crRNAs containing both the spacer region and the direct repeat region. B) RNaseIII, the tracrRNA, and Cas9, bind to these transcripts and C) cleave
them leaving mature crRNAs bound to the Cas9/tracrRNA complex. D) The mature crRNA is used to guide the Cas9 complex to the target DNA which
is E) cleaved leaving a F) double-strand break. A gRNA is a researcher-designed hybrid of the tracrRNA and the crRNA. The direct repeat region
combined with the tracrRNA forms the scaffold portion of a gRNA and the spacer region forms the target sequence.
20 | Page
CRISPR GLOSSARY
By Joel McDade | October, 2015
Below you can find a table defining many of the common terms used in articles and discussions about CRISPR
research and CRISPR genome engineering. We hope that youll find this glossary useful and maybe even print
out a copy to keep at your desk for quick reference.
Term
Defenition
Cas
CRISPR Associated Protein, the Cas9 nuclease is the active enzyme for the Type II
CRISPR system
CRISPR
Clustered Regularly Interspaced Short Palendromic Repeat, a region in bacterial

genomes used in pathogen defense
CRISPRa
CRISPR Activation, using a dCas9 or dCas9-activator with a gRNA to increase

transcription of a target gene
CRISPRi
CRISPR Interference, using a dCas9 or dCas9-repressor with a gRNA to repress/

decrease transcription of a target gene
Cut
A DNA double strand break, the wild type function of Cas9
dCas9
Nuclease dead Cas9, an enzymatically inactive form of Cas9; Can bind, but cannot
cleave DNA
DSB
Double Strand Break, a break in both strands of DNA, through the use of Cas9 or
two Cas9-nickases targeting opposite strands
Dual Nickase/
Double Nick
A method to decrease off-target effects by using a single Cas9 nickase and 2

different gRNAs, which bind in close proximity on opposite strands of the DNA, to
create a DSB
enChIP
Engineered DNA-binding molecule-mediated ChIP, using a tagged-dCas9+gRNA to

purify specific genomic regions to identify molecules associated with the genomic
regions
Genetic
modification or
manipulation
Any genetic perturbation, including genetic knock-out, gene activation, or gene

repression
gRNA
Guide RNA, a synthetic fusion of the endogenous bacterial crRNA and tracrRNA;
Provides both targeting specificity and scaffolding/binding ability for Cas9
nuclease; Does not exist in nature; Also referred to as single guide RNA or
sgRNA
gRNA scaffold
sequence
The sequence within the gRNA that is responsible for Cas9 binding; Does not
include the 20bp spacer/targeting sequence that is used to guide Cas9 to target
DNA
21 | Page
CRISPR GLOSSARY (CONTD)

Term
Defenition
gRNA targeting
sequence
The 20 nucleotides that precede the PAM sequence in the genomic DNA; What
gets put into a gRNA expression plasmid; Does NOT include the PAM sequence or
the gRNA scaffold sequence
HDR
Homology Directed Repair, a DNA repair mechanism that uses a template to repair
nicks or DSBs
InDel
Insertion/Deletion, a type of mutation that can result in the disruption of a gene by

shifting the ORF and/or creating premature stop codons
NHEJ
Non-Homologous End-Joining; A DNA repair mechanism that often introduces

InDels
Nick
A break in only one strand of a double stranded DNA; Normally repaired by HDR
Nickase
Cas9 that has one of the two nuclease domains inactivated; Can be either the
RuvC or HNH domain; Capable of cleaving only one strand of target DNA
Off-target
effects or offtarget activity
Cas9 cleavage at undesired locations due to gRNA targeting sequence with

sufficient homology to recruit Cas9 to unintended genomic locations; Can be
minimized by double-nick
On-target
activity
Cas9 cleavage at a desired location due to gRNA targeting sequence with

sufficient homology to recruit Cas9 to desired genomic locations
ORF
Open Reading Frame; The codons that make up a gene
PAM
Protospacer Adjacent Motif; Necessary for Cas9 to bind target DNA; Must
immediately follow the target sequence
PCR
Polymerase Chain Reaction; Used to amplify a specific sequence of DNA
sgRNA
Single guide RNA; A synthetic RNA composed of a targeting sequence and

scaffold sequence derived from endogenous bacterial crRNA and tracrRNA; Used
to target Cas9 to a specific genomic locus in genome engineering experiments;
Also referred to as a gRNA
Target
sequence
Genomic target of the gRNA targeting sequence; The 20 nucleotides that are
incorporated into the gRNA plus the PAM sequence; Not to be confused with the
gRNA targeting sequence
22 | Page
THE PAM REQUIREMENT AND EXPANDING CRISPR BEYOND SpCAS9

By Joel McDade | Nov 12, 2015
Diverse genomes and genomic targets require a

variety of tools to engineer them effectively. Read
on to learn how a variety of natural and engineered
forms of Cas9 can be used to expand CRISPRs
reach to new genomic loci.
The Need for More PAM Sequences

CRISPR/Cas9 can be used to modify any desired
genomic target provided that sequence is unique
compared to the rest of the genome and is located
just upstream of a Protospacer Adjacent Motif (PAM
sequence). A PAM sequence is a short stretch
of DNA (typically 3-5 nucleotides) that serves as
a binding signal for Cas9 and the presence of this sequence is a strict requirement for Cas9-mediated DNA
cleavage. While PAM sequences for the commonly used S. pyogenes Cas9 (5NGG3) are abundant throughout
the human genome, they are not always positioned correctly to target particular genes for Cas9-mediated
knockout or modification. Furthermore, a target sequence may have high homology elsewhere in the genome.
These additional sequences (so-called off-targets) may be unintentionally mutated during attempts to use
CRISPR mediated genome engineering on the gene of interest. The lack of suitable PAM sequences is of
particular concern when trying to edit a gene using Homology Directed Repair (HDR), since HDR-mediated
gene editing is most efficient when target sites are located in close proximity to the region to be edited.
Therefore, the requirement for a single acceptable PAM sequence remains a technical hurdle in using CRISPR
to modify genomic loci across the entire human genome. In this section, we will cover two ways to circumvent
this limitation: 1) the use of novel S. pyogenes Cas9 variants with varying PAM sequences and 2) the use of
Cas9 homologs derived from species other than S. pyogenes.
Synthetic S. pyogenes Cas9s

Kleinstiver et al. (1) performed a series of positive selection screens in bacteria to identify mutants of S.
pyogenes Cas9 that were able to cleave target DNA sequences located upstream of either NGA or NGC PAM
sequences. From these screens, they identified four novel SpCas9 variants with altered PAM binding specificity:
SpCas9 Variant
Mutations (relative to SpCas9)
PAM sequence (5 to 3)
D1135E variant
D1135E
NGG
VQR variant
D1335V, R1335Q and T1337R
NGAN or NGNG
EQR variant
D1135E, R1335Q and T1337R
NGAG
VRER variant
D1135V, G1218R, R1335E and T1337R
NGCG
Notably, the D1135E variant, which still recognizes the canonical S. pyogenes PAM sequence (5NGG3), is
far more selective for the NGG PAM over the NGA PAM compared to wild-type SpCas9. This variant may
therefore increase the specificity of genome modifications at DNA targets adjacent to NGG PAM sequences
23 | Page
THE PAM REQUIREMENT AND EXPANDING CRISPR BEYOND SpCAS9 (CONT)

when used in place of wild-type SpCas9. The remaining variants (VQR, EQR and VRER) recognize novel
PAM sequences (shown above). Importantly, the VQR, EQR and VRER Cas9 variants are capable of cleaving
genomic DNA when expressed in mammalian cells and zebrafish embryos and can be used to modify genomic
loci that cannot be modified using wild-type SpCas9. The number of off-target cleavage events for the VQR
and VRER variants is similar to wild-type SpCas9 for several genomic targets in human cells, indicating that the
variants are likely just as selective as wild-type SpCas9. Inclusion of the VQR, EQR and VRER SpCas9 variants
effectively doubles the targeting range of CRISPR/Cas9 within the human genome.
Characterization of Cas9 from Additional Bacterial Species

Over 20 additional Cas9 homologs derived from a variety of bacterial species have been isolated (many by
Feng Zhangs lab at the Broad Institute) and the PAM binding specificity of at least 4 have been characterized
(see table):
Cas9 Species
PAM sequence (5 to 3)
Streptococcus pyogenes (Sp)
NGG
Staphylococcus aureus (Sa)
NGRRT or NGRRN
Neisseria meningitidis (NM)
NNNNGATT
Streptococcus thermophilus (St)
NNAGAAW
Treponema denticola (Td)
NAAAAC
~20 additional Cas9 species
PAM sequence may not be characterized
Non-SpCas9s bind a variety of PAM sequences, which may make them useful when no suitable SpCas9 PAM
sequence is present within your gene of interest. Furthermore, non-SpCas9s may have other characteristics
that make them more useful than SpCas9 for a specific application. For example, the coding sequence
for Cas9 from Staphylococcus aureus (SaCas9) is about 1 kilobase smaller than SpCas9, which allows for
packaging into adeno-associated virus (AAV). AAV-mediated delivery of SaCas9 has already been used to
target liver cells of mice in-vivo, and the list of applications is certain to expand in the near-future. It is important
to remember that non-SpCas9s are only compatible with the tracrRNA and crRNA (or synthetic gRNA) derived
from the same species.
Characterization of Cas9 from Additional Bacterial Species

Isolation of novel CRISPR proteins or modification of existing CRISPR reagents has and will continue to
dramatically increase the number of CRISPR applications. While S. pyogenes Cas9 has already been modified
to enable gene knockout, repression and activation in a variety of cell types, it is certainly possible that novel
Cas9s could be modified for similar approaches. Isolation of novel CRISPR proteins will also increase gene
editing efficiencies and possibilities. An example is the recent identification of Cpf1, a non-Cas9 CRISPR
nuclease that generates double strand breaks in target genes resulting in the formation of sticky ends rather
than blunt ends (which is the case for all Cas9 proteins). As the number of CRISPR reagents continues to grow,
so too will the number of reagents available through Addgene!
24 | Page
THE PAM REQUIREMENT AND EXPANDING CRISPR BEYOND SpCAS9 (CONT)
Further Reading
1. Kleinstiver, Benjamin P., et al. Engineered CRISPR-Cas9 nucleases with altered PAM specificities. Nature (2015). Pubmed PMID: 26098369.
25 | Page
CPF1: A CAS9 HOMOLOG

By Mary Gearing | October 14, 2015
CRISPR genome editing is no longer limited to Cas9

- theres a new nuclease in town! A recent paper from
Feng Zhangs lab describes two genes from the Cpf1
family that display cleavage activity in mammalian cells.
Like Cas9 nucleases, Cpf1 family members contain a
RuvC-like endonuclease domain, but they lack Cas9s
second HNH endonuclease domain. Cpf1 cleaves
DNA in a staggered pattern and requires only one RNA
rather than the two (tracrRNA and crRNA) needed by
Cas9 for cleavage. For multiple reasons, Cpf1 may be
even better suited for genome editing than Cas9 - read
on to learn more about Cpf1.
How Was Cpf1 Found and Tested?

Class 2 CRISPR systems, including the type II Cas9based system, require a single-component nuclease
to mediate cleavage rather than the multi-subunit
complex employed by class 1 systems. A putative new
Image Attribution: Zetsche, Bernd, et al. Cpf1 is a single RNA-guided enclass 2 nuclease, Cpf1 (CRISPR from Prevotella and
donuclease of a class 2 CRISPR-Cas system. Cell (2015). PubMed PMID:
Francisella), was recently annotated in several genomes 26422227.
and is classified as a novel, type V CRISPR system.
Like Cas9, Cpf1 contains a RuvC-like endonuclease domain, but it lacks Cas9s other HNH endonuclease
domain, indicating that Cpf1 functions differently. Since Cpf1 loci are widely distributed across bacterial
species, Zetsche et al. hypothesized that Cpf1 might represent a functional CRISPR nuclease that could
be adapted for genome editing. The use of a different nuclease could potentially overcome some of Cas9s
shortcomings - namely its blunt double stranded cleavage and G-rich PAM requirement.
Zetsche et al. started from square one to characterize the Cpf1 nucleases. Using Francisella Cpf1 (FnCpf1),
they employed an E. coli plasmid depletion assay to discover FnCpf1s PAM sequence requirements. Cpf1s
preferred PAM is 5-TTN, differing from that of Cas9 (3-NGG) in both genomic location and GC-content. After
sequencing and searching for cellular RNAs important for Cpf1 function, they found that mature crRNAs for
Cpf1-mediated cleavage are 42-44 nucleotides in length, about the same size as Cas9s, but with the direct
repeat preceding the spacer rather than following it. The Cpf1 crRNA is also much simpler in structure than
Cas9s; only a short stem-loop structure in the direct repeat region is necessary for cleavage of a target. Cpf1
also does not require an additional tracrRNA.
Once they had determined the minimal elements of CRISPR-Cpf1, Zetsche et al. turned to characterizing its
cleavage pattern. Again, they were in for a surprise! Whereas Cas9 generates blunt ends 3 nt upstream of the
PAM site, Cpf1 cleaves in a staggered fashion, creating a 5 nucleotide 5 overhang 18-23 bases away from the
PAM. With this information, they turned to a cell culture system to see if any Cpf1 nucleases would exhibit in
vivo activity in mammalian cells. From 16 diverse Cpf1 candidates, Zetsche et al. found two that display robust
cleavage activity similar to that of Cas9. These two nucleases, AsCpf1 and LbCpf1 (1307 and 1228 amino acids
long respectively), both cleave in a staggered pattern similar to FnCpf1.
26 | Page
CPF1: A CAS9 HOMOLOG (CONT)

Potential Advantages of Cpf1 Over Cas9
Type II CRISPR systems based on Cas9 were thought to be the simplest CRISPR systems and the easiest
to adapt to genome editing, but the introduction of type V Cpf1-driven systems has added another option to
the CRISPR toolbox. Cpf1s staggered cleavage pattern opens up the possibility of directional gene transfer,
analogous to traditional restriction enzyme cloning. Sticky-end mediated gene transfer would be particularly
helpful for targeting non-dividing cells, which are difficult to modify through homology-directed repair (HDR).
Cpf1 also expands the number of sites that can be targeted by CRISPR to AT-rich regions or AT-rich genomes
that lack the 3-NGG PAM sites favored by SpCas9.
Since Cpf1 doesnt require a tracrRNA, guide RNAs are only ~42 nt long. Direct synthesis of these gRNAs
should be significantly cheaper than that of the ~100 nt crRNA/tracrRNA hybrid guides needed for Cas9
function. Since both Cpf1 and its guide RNAs are smaller than their SpCas9 counterparts, they will also be
easier to deliver in low-capacity vectors, such as adeno-associated viral (AAV) vectors.
Zetsche et al. also suggest that Cpf1 may improve the frequency of HDR over non-homologous endjoining
(NHEJ). Cas9-mediated NHEJ usually destroys the PAM site due to its proximity to the cleavage site, preventing
future edits. In contrast, since Cpf1 cleaves relatively far away from the PAM, NHEJ might retain the PAM site.
Therefore, if HDR does not initially occur after Cpf1-mediated cleavage, the continued presence of the PAM
may give Cpf1 the ability to cleave again and possibly mediate HDR. This second chance mechanism might
improve the frequency of desired HDR edits, but the possibility has not yet been experimentally confirmed. To
prevent new editing post-HDR, repair templates should remove the PAM sequence.
The application of Cpf1 to genome editing is exciting both in terms of basic science and translational
applications. This putative type V CRISPR system proves we have a lot more to learn about CRISPR biology,
and that future work may harness other nucleases superior for genome editing applications. Wherever the
CRISPR journey leads us, Addgene will strive to provide all of the plasmids and resources youll need to apply
CRISPR technology to your own research!
Further Reading
1.
Zetsche, Bernd, et al. Cpf1 is a single RNA-guided endonuclease of a class 2 CRISPR-Cas system.
Cell (2015). PubMed PMID: 26422227.
2. Makarova, Kira S., et al. An updated evolutionary classification of CRISPR-Cas systems. Nature
Reviews Microbiology (2015). PubMed PMID: 26411297.
27 | Page
ENHANCING CRISPR TARGETING SPECIFICITY WITH eSpCAS9 AND SpCAS9-HF1

By Tyler J. Ford | Dec 16, 2015
As evidenced by all the CRISPR publications, press, and plasmids out there, its obvious that CRISPR is a
ground-breaking technology thats already had a huge impact on research and will be affecting our everyday
lives very soon. Not only is CRISPR having effects on various biological disciplines, the base technology itself
is constantly improving. Cas9 variants have been modified for genome editing, activating gene expression,
visualizing genomic loci, and much more. Now, researchers from the Zhang and Joung labs have improved the
on-target specificity of the Cas9 nuclease with two independently discovered CRISPR variants: eSpCas9 and
SpCas9-HF1.
The Off-target Problem

It is well known that CRISPR/Cas9 genome editing can result in unwanted changes at non-target sites. Means
to decrease these so-called off-target effects have included using a Cas9 nickase variant, lowering Cas9
expression, and truncating gRNA sequences used for targeting; however, these options can be cumbersome,
can lower on-target efficiency, and sometimes even increase off-target effects, respectively. Recognizing these
issues, researchers from the Zhang lab at the Broad Institute and the Joung lab at Harvard Medical School set
out to decrease the off target effects of CRISPR/Cas9 by altering the nuclease activity of the Cas9 nuclease
itself (1).
Design and Testing of eSpCas9

Looking at the structure of the Cas9 nuclease (PDB ID: 4OO8 and 4UN3), Slaymaker et al. hypothesized that
Cas9 cutting efficiency increases when target DNA strand separation is stabilized (2,3). Stable strand separation
is maintained by 2 sets of interactions :
1. Interactions between the non-target strand and a positively charged groove formed by the Cas9 HNH and
RuvC nuclease domains
2. Interactions between the target strand and the gRNA (see figure below)
Because off-target sequences have less complementarity to the gRNA, they have a higher propensity to reform
double helices and decrease Cas9 cutting efficiency. However, this is not always enough to keep off-target
sites from being cut. Slaymaker et al reasoned that, if they decreased the positive charge in the HNH/RuvC
groove, then they could weaken interactions between the groove and the negatively charged DNA and therefore
destabilize basal strand separation. This decreased strand separation combined with the weak separation
afforded by gRNA binding at non-target sites would theoretically decrease off-target cutting.
Strand separation and target DNA binding to the Cas9 nuclease. Stable strand separation is maintained through interactions
between the non-target strand and the HNH/RuvC groove and interactions between the target strand and the gRNA.
28 | Page
ENHANCING CRISPR TARGETING SPECIFICITY WITH eSpCAS9 AND SpCAS9-HF1 (CONT)

To decrease the electropositivity of the HNH/RuvC groove, Slaymaker et al. made a variety of alanine
substitutions throughout the groove in 32 separate Cas9 mutants. When tested for their ability to cut the EMX1
locus in human embryonic kidney (HEK) cells, 11 of these initial mutants retained the on-target activity of wt
Cas9 while decreasing off-target activity at known off-target loci. The authors went on to combine the best
of these mutations and tested the combination mutants for their on-target and off-target activity at multiple
genomic locations in HEK cells. Further testing of two of the mutants, SpCas9(K855A) and eSpCas9(1.1),
followed by a technique for detecting genome-wide double strand breaks (BLESS), revealed that these mutants
do not cause off-target effects at unanticipated sites and, as predicted, decrease off-target effects genomewide.
Further work by Slaymaker et al. showed that these mutants are not toxic to HEK cells and that similar
mutations can improve the specificity of Cas9 derived from S. aureus. This means you can likely apply similar
rational mutations to your Cas9 homolog of choice and get improved on-target specificity.
Design and Testing of SpCas9-HF1

Kleinstiver et al. and the Joung Lab similarly thought that if they could weaken sequence independent
interactions between Cas9 and DNA, then they could diminish off-target cutting (4). However, instead of
weakening interaction sbetween the non-target strand and Cas9, Kleinstiver et al disrupted interactions
between Cas9 and the phophate backbone of the target DNA strand via mutations N497A, R661A, Q695A, and
Q926A (SpCas9-HF1 contains all four mutations). Although the precise mutations were different than those
found in eSpCas9, the effect on specificity was similar; SpCas9-HF1 generated fewer off-target cuts when
compared to WT SpCas9 across a variety of genomic sites.
Future Possibilities
With thier enhanced specificity, eSpCas9 and SpCas9-HF1 should enable researchers to make precise edits in
mammalian cells and may decrease worries about off target effects in applied and/or therapeutic settings. Will
combining the different mutations found in each of these great tools further enhance genome editing specificty?
We hope one of you will be able to answer this question very soon! We look forward to seeing how the research
community makes use of these tools and are excited to see continued improvements in genome engineering!
Further Reading
1. Slaymaker, Ian M., et al. Rationally engineered Cas9 nucleases with improved specificity. Science
(2015): aad5227. PubMed PMID: 26628643.
2. Nishimasu, Hiroshi, et al. Crystal structure of Cas9 in complex with guide RNA and target DNA. Cell
156.5 (2014): 935-949. PubMed PMID: 24529477. PubMed Central PMCID: PMC4139937.
3. Anders, Carolin, et al. Structural basis of PAM-dependent target DNA recognition by the Cas9
endonuclease. Nature 513.7519 (2014): 569-573. PubMed PMID: 25079318. PubMed Central PMCID:
PMC4176945.
4. Kleinstiver, Benjamin P., et al. High-fidelity CRISPRCas9 nucleases with no detectable genome-wide
off-target effects. Nature (2016). PubMed PMID: 26735016.
29 | Page
Chapter 3 - Using CRISPR in Your Experiments
USING CRISPR IN YOUR EXPERIMENTS
30 | Page
PLANNING YOUR CRISPR EXPERIMENT

By Joel McDade | Decmber, 2015
Develop a Biological Question
Select Your Desired Manipulation
Knockout
Edit
Activation Repression
Genome-Wide or single edit?
Select Expression System and Delivery
Transient
Viral
RNA Riboprotein
Design gRNAs and Perform Cloning
Deliver CRISPR Components

and Validate Your Genome Edit
31 | Page
PLANNING YOUR CRISPR EXPERIMENT (CONT)

Getting Started
CRISPR/Cas9 is a powerful system that enables researchers to manipulate the genome of target cells like never
before. This section will provide a general framework to get you started using CRISPR/Cas9 in your research.
Although we will focus on using CRISPR in mammalian cells, many of these principles apply to using CRISPR
in other organisms. We will go more in-depth into many of the items discussed here later in the eBook and provide links that allow you to jump do the appropriate chapters throughout this section. First, consider the genetic
manipulation that is necessary to model your specific disease or process of interest. For instance, do you want
to:
Generate complete and permanent loss of gene expression or function (i.e. knock-out)?
Express a mutated version of the gene (e.g. point mutant)?
Increase or decrease expression of a target gene?
Once you have a clear understanding of what you are trying to do, you are ready to start navigating the different
reagents that are available for your particular experiment.
Select Your Desired Genetic Manipulation

Different genetic manipulations require different CRISPR components. Selecting a specific genetic manipulation
can be a good way to narrow down which reagents are appropriate for a given experiment. Make sure to check
whether reagents are available to carry out your specific genetic manipulation in your specific model organism.
There may not be a perfect plasmid for your specific application, and in such a case, it may be necessary to
customize an existing reagent to suit your needs. You can learn more about selecting a Cas9 for your specific
experiment here. Many CRISPR plasmids can be found organized by function on our website.
Genetic
Manipulation
Application
Knock-out
Edit
Cas9
gRNA
Additional Considerations
Permanently disrupt Cas9 (or Cas9

nickase)
gene function in a
particular cell type
or organism without
regard for specific
mutation
Single (or dual)

gRNA targeting
5 exon or
essential protein
domains
Dual-nickase approach
increases specificity but is
less efficient. eSpCas9 can
also be used to increase
specificity.
Generate a specific Cas9 (or Cas9

user-defined
nickase)
sequence change
in a particular gene,
such as generating
a point mutation or
inserting a tag
Single (or dual) Requires DNA template for

gRNA targeting HDR; Efficiency is reduced
the region
compared to knock-out
where the edit
should be made
32 | Page

Genetic
Manipulation
Application
Cas9
gRNA
Additional Considerations
Repress or
Interfere
(CRISPRi)
Reduce expression
of a particular
gene(s) without
permanently
modifying the
genome
dCas9 or
dCas9repressor
(such as
dCas9-KRAB)
gRNA(s)
targeting
promoter
elements of
target gene
dCas9-KRAB is more
effective than dCas9 alone
for mammalian cell lines
Activate
(CRISPRa)
Increase expression
of an endogenous
gene(s) without
permanently
modifying the
genome
dCas9activator (such
as dCas9VP64)
gRNA(s)
targeting
promoter
elements of
target gene
Many different activators

exist, all derived from S.
pyogenes dCas9
Select Expression System

To use the CRISPR system, you will need to express both a gRNA and Cas9 in your target cells. The expression system you need will depend upon your specific application. For example, certain cell types are easier to
transfect than others. For easy-to-transfect cell types (e.g. HEK293 cells), transfection with standard transfection reagents may be sufficient to express both your gRNA and Cas9. For more difficult cells (e.g. primary cells),
viral delivery of CRISPR reagents may be more appropriate. See Chapter 4 on CRISPR expression systems and
delivery methods for more information
The table below summarizes the major expression systems and variables for using CRISPR in mammalian cells.
Some of the variables include:
Species of Cas9 and gRNA

Species of promoter and expression pattern of promoter for Cas9 and gRNA
Presence of a selectable marker (drug or fluorophore)
Delivery method
Expression
System
Components of System
Application
Mammalian
expression
vector

Cas9 promoter can be constitutive (CMV, EF1alpha,
CBh) or inducible (Tet-ON); U6 promoter is typically used
for gRNA

May contain reporter gene (e.g. GFP) to identify
and enrich positive cells, or selection marker to generate
stable cell lines (for details, see Addgenes Plasmids 101:
Mammalian vectors)
Transient or stable
expression of Cas9
and/or gRNA in a
mammalian cell line that
can be transfected at
high efficiency
33 | Page

Expression
System
Application
Lentiviral
transduction

Cas9 and gRNA can be present in a single lentiviral
transfer vector or separate transfer vectors

May contain reporter gene (e.g. GFP) or selection
marker to identify and enrich positive cells

Packaging and envelope plasmids provide the
necessary components to make lentiviral particles (for
details about lentivirus, see Addgenes Lentivirus guide)

Stable, tunable
expression of Cas9
and/or gRNA in a wide
variety of mammalian
cell lines

Useful for
difficult to transfect cell
types and can be used
in vivo

A common
choice for conducting
genome-wide screens
using CRISPR/Cas9
AAV
transduction

Only compatible with SaCas9 (packaging limit
~4.5kb)

CRISPR elements are inserted into an AAV transfer
vector and used to generate AAV particles (for details, see
Addgenes AAV guide)

Transient or
stable expression of
SaCas9 and/or gRNA

Infects dividing
and non-dividing cells

AAV is least toxic
method for in vivo viral
delivery
Cas9 mRNA
and gRNA
Plasmids containing gRNA and Cas9 are used in in

vitro transcription reactions to generate mature Cas9
mRNA and gRNA, then delivered to target cells (i.e.
microinjection or electroporation)

Transient
expression of CRISPR
components

Expression
decreases as RNA is
degraded within the cell

Can be used for
generating transgenic
embryos
Cas9-gRNA
riboprotein
complexes

Transient
Purified Cas9 protein and in vitro transcribed gRNA are
combined to form a Cas9-gRNA complex and delivered to expression of CRISPR
components
cells using cationic lipids

Expression
decreases as gRNA
and Cas9 protein are
degraded within the cell
34 | Page

Additional Addgene Resources:
Depositor protocols for plasmids at Addgene

JOVE Video Reprint: Protocol for genomic deletions in mammalian cell lines
Select Your Target Sequence and Design Your gRNA

Once you have selected your CRISPR components and method of delivery, you are ready to select a target
sequence and design your gRNA. A general overview of how to design a gRNA is presented below. More
details on gRNA design can be found in our How to Design Your gRNA section. An overview of CRISPR
software tools for gRNA design can be found in the CRISPR Software Matchmaker section.
A. Know your cell line and genome sequence
The cell line you choose determines a variety of factors related to your experiment. The genomic sequence
used to design gRNAs will depend upon the gene in question and the species from which your cells were
derived. When possible, you should sequence the region you are planning to modify prior to designing your
gRNA, as sequence variation between your gRNA targeting sequence and target DNA may result in reduced
cleavage. The number of alleles for each gene may vary depending on the specific cell line or organism, which
may affect the observed efficiency of knock-out or knock-in using CRISPR/Cas9.
B. Select gene and genetic element to be manipulated
In order to manipulate a given gene using CRISPR/Cas9, you will have to identify the genomic sequence for
the gene you are trying to target. However, the exact region of the gene you target will depend on your specific
application. For example:
To activate or repress a target gene using dCas9-activators or dCas9-repressors, gRNAs should be

targeted to the promoter driving expression of your gene of interest.

For genetic knock-outs, gRNAs commonly target 5 constitutively expressed exons, which reduces
the chances that the targeted region will be removed from the mRNA due to alternative splicing. Exons near
the N-terminus are also often targeted since frameshift mutations here will increase the likelihood that a
nonfunctional protein product is produced.

Alternatively, gRNAs can be designed to target exons that code for known essential protein domains.
The benefit of this approach is that non-frameshift mutations such as insertions or deletions (which commonly
result from DSBs) are more likely to alter protein function when they occur in protein domains that are essential
for protein function.

For gene editing experiments using HDR, it is essential that the target sequence be close to the location
of the desired edit. In this case, it is necessary to identify the exact location where the edit should occur and
select a target sequence nearby.
C. Select gRNAs based on predicted on-target and off-target activity
A PAM sequence is absolutely necessary for Cas9 to bind target DNA. As such, one must identify all PAM
sequences within the genetic region to be targeted (PAM is 5 NGG 3 for SpCas9). If there are no PAM
sequences within your desired target, you may want to consider using a Cas9 from a different species or a S.
pyogenes Cas9 variant that binds a PAM sequence present in your genomic target (see additional Cas9 variants
35 | Page

and PAM sequences). Once all possible PAM sequences and putative target sites have been identified, it is time
to choose which site is likely to result in the most efficient on-target cleavage.
Clearly, the gRNA targeting sequence needs to match the target sequence, but it is also critical to ensure that
the gRNA targeting sequence does NOT match additional sites within the genome. In a perfect world, your
gRNA targeting sequence would have perfect homology to your target with no homology elsewhere in the
genome. Realistically, a given gRNA targeting sequence will have additional sites throughout the genome where
partial homology exists. These sites are called off-targets and should be considered when designing a gRNA
for your experiments. In general, off-target sites are not cleaved as efficiently when mismatches occur near the
PAM sequence, so gRNAs with no homology or those with mismatches close to the PAM sequence will have
the highest specificity. Cas9 variants eSpCas9 and SpCas9-HF1, was designed by the Zhang lab at the Broad
Institute to reduce off-target activity.
In addition to off-target activity, it is also important to consider factors that maximize cleavage of the desired
target sequence (on-target activity). It is now understood that two gRNA targeting sequences, each having
100% homology to the target DNA may not result in equivalent cleavage efficiency. In fact, cleavage efficiency
may increase or decrease depending upon the specific nucleotides within the selected target sequence. For
example, gRNA targeting sequences containing a G nucleotide at position 20 (1 bp upstream of the PAM) may
be more efficacious than gRNAs containing a C nucleotide at the same position in spite of being a perfect
match for the target sequence. Therefore, close examination of predicted on-target and off-target activity of
each potential gRNA targeting sequence is necessary to design the best gRNA for your experiment.
Several gRNA design programs have been developed that are capable of locating potential PAM and target
sequences and ranking the associated gRNAs based on their predicted on-target and off-target activity (see
The CRISPR Software Matchmaker Section). Additionally, many plasmids containing validated gRNAs are
now available through Addgene. These plasmids contain gRNAs that have been used successfully in genome
engineering experiments and have been published in peer-reviewed journals. Using validated gRNAs can save
your lab valuable time and resources when carrying out experiments using CRISPR/Cas9.
Browse Plasmids: Validated gRNAs
D. Synthesize and clone desired gRNAs
Once you have selected your target sequences it is time to design your gRNA oligos and clone these oligos
into your desired vector. In many cases, targeting oligos are synthesized, annealed, and inserted into plasmids
containing the gRNA scaffold using standard restriction-ligation cloning. However, the exact cloning strategy
will depend on the gRNA vector you have chosen, so it is best to review the protocol associated with the
specific plasmid in question (see CRISPR protocols from Addgene depositors).
E. Deliver Cas9 and gRNA
Your method of delivery will be determined by the specific expression system that you have chosen. It may be
necessary to optimize a protocol for delivery of gRNA and Cas9 to your target cells, as transfection efficiency
will vary based on the method of delivery and the cell type tested. Read the Chapter 4 on CRISPR expression
systems and delivery methods for more information.
F. Validate genetic modification
36 | Page
Chapter 3 - Using CRISPR in Your Experimentsb

Once you have successfully delivered the gRNA and Cas9 to your target cells, it is time to validate your
genome edit. Delivery of Cas9 and gRNA to wild-type cells will result in several possible genotypes within the
resulting mutant cell population. A percentage of cells may be wild-type due to either (1) a lack of gRNA and/
or Cas9 expression, or (2) a lack of efficient target cleavage in cells expressing both Cas9 and gRNA. Cells that
have undergone modification (mutant cells) can be homozygous (modification of both alleles) or heterozygous
(modification of a single allele). Furthermore, in mutant cells containing two mutated alleles, each mutated
allele may be different owing to the error-prone nature of NHEJ. Even for gene editing experiments using HDR,
not every mutated allele will contain the desired edit as a large percentage of DSBs are still repaired by NHEJ.
Therefore, the end result of most experiments is a heterogeneous population of cells containing a wide variety
of mutations or edits within target genes.
How do you determine that your desired edit has occurred? The exact method necessary to validate your edit
will depend upon your specific application, and in some cases, new approaches must be devised. However,
there are several common ways to verify that your cells contain your desired edit, including but not limited to:
1.
Mismatch-cleavage assay (for NHEJ repaired DSBs): Provides a semi-quantitative readout of the
percentage of alleles that have been mutated within a mixed cell population. The region of interest is PCR
amplified, PCR products are denatured-renatured, treated with a nuclease that cleaves DNA heteroduplexes,
and run on an agarose gel to identify DNA fragments.
2.
PCR and restriction digest (for HDR repaired DSBs): For small nucleotide edits that introduce a novel
restriction site. The region of interest is PCR amplified, digested with the appropriate restriction enzyme and run
on an agarose gel to identify DNA fragments.
3.
PCR amplification and gel electrophoresis (for HDR or NHEJ): For large deletions or insertions, the
region of interest can be PCR amplified using primers that (A) flank the region of interest (deletions or small
insertions) or (B) span the genome-insert boundary (insertions only). The PCR product is then run on an agarose
gel to determine whether the edit was successful.
4.
PCR amplification, subcloning and Sanger sequencing (for HDR or NHEJ): Provides semi-quantitative
assessment of targeting frequency and exact sequence of targeted alleles. Involves PCR amplification of
targeted region from DNA, subcloning into a plasmid, and screening individual clones.
5.
PCR amplification and next-generation sequencing (for HDR or NHEJ): Provides quantitative
assessment of the genome edits in your target sequence and can also be used to examine off-target effects.
6.
More information on each of these techniques can be found in the Validating your Genome Edit section.
37 | Page
WHICH CAS9 DO I CHOOSE FOR MY CRISPR EXPERIMENT?

By Joel McDade | Jan 19, 2016
The advent of CRISPR/Cas9 made it easier than

ever to efficiently make precise, targeted genome
modifications. Cas9 has been modified to enable
researchers to knock out, activate, repress or even
image your favorite gene. But, with such a wide variety
of Cas9-based reagents available, how do you choose
which Cas9 is right for your particular experiment?
This section will help familiarize you with the wide array
of Cas9s available through Addgenes repository and
make it easy to select the Cas9 reagent that is right for
you.
The first thing to do in any CRISPR experiment is
identify what, exactly, you are trying to do. Are you
trying to permanently knock-out a gene in a cell type
or organism? Are you trying to reduce expression
of a particular gene without permanently modifying
the genome? Does it make more sense to try and
activate at a particular locus? What about modifying
the epigenome at a particular location? As you might
expect, the answer to this question will substantially affect your decisions about which Cas9 you need for your
experiment. Below is a brief summary of a few of the common genetic manipulations one can carry out using
Cas9 and the specific Cas9s that can be used for each.
Repair of Cas 9-Induced Breaks by NHEJ

Carrying out a genetic knockout using standard SpCas9
Knock-out cells or animals are created by co-expressing a gRNA specific to the gene to be targeted and
the endonuclease Cas9. The genomic target can be any ~20 nucleotide DNA sequence, provided it meets
two conditions: 1) The sequence is unique compared to the rest of the genome. 2) The target is present
immediately upstream of a Protospacer Adjacent Motif (PAM). SpCas9 is a common choice when there is a
suitable target site and little concern for off-target effects (e.g. optimal gRNA design with minimal homologous
sites throughout the genome). Addgene carries a wide variety of SpCas9 containing plasmids that have
been optimized for carrying out genome engineering experiments in bacteria, yeast, worms, drosophila and
mammals.
Concerns about specificity? Consider using high-specificity Cas9 variants.
The specificity of Cas9-mediated cleavage can be enhanced by properly designing your gRNA sequence. That
is, choosing a target sequence that has minimal homology elsewhere in the genome. A variety of approaches
have been used to further enhance Cas9 specificity. For example, Cas9-nickase (Cas9n) takes advantage of
the fact that Cas9 makes double-strand breaks (DSBs) through the combined activity of two nuclease domains,
RuvC and HNH. Converting one of the two critical enzymatic residues to an alanine (D10A or H840A) generates
a nicking Cas9 that cannot generate a double-strand break. Thus, two properly targeted Cas9n molecules
are required to efficiently create DSBs at the target locus, which greatly enhances specificity compared to wild38 | Page
WHICH CAS9 DO I CHOOSE FOR MY CRISPR EXPERIMENT? (CONT)

type SpCas9.
While Cas9n is certainly more specific than wild-type SpCas9, DSBs are still detectable at target sites when
only one gRNA is expressed with Cas9n. This means its possible for Cas9n to bind and cause an indel
at one of the off target sites for either of its gRNAs. One can overcome this limitation through the use of
nuclease dead Cas9 (dCas9) fused to the non-specific endonuclease FokI. FokI only cleaves target DNA
when dimerized. Therefore, dCas9-FokI essentially requires proper targeting of two dCas9-FokI molecules at
the target site before any cleavage occurs; dCas9-FokI is much less likely to cut at an off target specified by a
single gRNA that Cas9n.
An obvious limitation of the Cas9n or dCas9-FokI approach is that they both necessitate two suitable target
sequences in close proximity in order to efficiently generate a DSB. Several labs, including Feng Zhangs lab
at the Broad Institute and Keith Joungs group at MGH, have used structural biology to identify key residues
that mediate Cas9s ability to cleave off-target sites. So-called enhanced Cas9 (Zhang) or high-fidelity
Cas9 (Joung) have comparable cleavage activity to wild-type SpCas9 at target loci but have greatly reduced
off-target activity. These Cas9 variants enhance specificity without requiring two or more adjacent target sites
within the target locus. More information on the enhanced specificity Cas9 variants can be found in our section
entitled Enhancing CRISPR Targeting Specificity with eSpCas9 and SpCas9-HF1.
What do I do when there is no 5NGG3 PAM sequence present?
Synthetic Cas9s with novel PAM recognition
Through a series of positive selection screens in bacteria, Keith Joungs group identified mutants of S.
Pyogenes Cas9 (VQR, EQR and VRER Cas9 variants) that recognize novel non-NGG PAM sequences.
Importantly, the VQR, EQR and VRER Cas9 variants are capable of modifying genomic loci that cannot be
modified using wild-type SpCas9, and their specificity for the PAM variants is similar to wild-type SpCas9 for
several genomic targets in human cells. Including the VQR, EQR and VRER SpCas9 variants effectively doubles
the targeting range of CRISPR/Cas9 within the human genome. More information on the various synthetic
Cas9s available through Addgene can be found in our section entitled The PAM Requirement and Expanding
CRISPR Beyond SpCas9.
Non S. pyogenes Cas9s
Additional Cas9 homologs have been isolated from a wide variety of bacterial species and many bind PAM
sequences other than the typical NGG PAM sequence. So called non-Sp Cas9s may be more suitable for
your experiment for reasons other than the PAM sequence. For example, the coding sequence for Cas9 from
Staphylococcus aureus (SaCas9) is about 1 kilobase smaller than SpCas9, which allows for packaging into
adeno-associated virus (AAV), the current gold standard for gene therapy. It is important to remember that nonSpCas9s are only compatible with the tracrRNA and crRNA (or synthetic gRNA) derived from the same species.
More information on the various non-SpCas9s that are available through Addgene can be found in our section
entitled The PAM Requirement and Expanding CRISPR Beyond SpCas9.
39 | Page

Non-Cas9 CRISPR endonuclease, Cpf1
A new CRISPR endonuclease was recently identified by the Zhang lab at the Broad institute. This non-Cas9
CRISPR endonuclease, termed Cpf1, is a class II CRISPR endonuclease that is capable of cleaving target DNA
in an RNA dependent manner. What makes Cpf1 so interesting is not the similarities to Cas9 (it has many)
but the differences. The PAM sequence for Cpf1 is 5 TTN 3 and is located 5 to the target site (in contrast
to the Cas9 PAM which is 3 to the target site). The endonuclease contains two enzymatic residues similar
to Cas9 (D917 and E1006) but both residues are located in the RuvC domain (Cpf1 lacks an HNH domain)
and mutation of either residue completely abolishes DNA cleavage. In contrast to SpCas9, which can be
converted into a nickase by mutating a single enzymatic residue. Intriguingly, Cpf1 cleavage results in a 5
nucleotide 5 overhang 18 base pairs from the PAM sequence. This is different from Cas9 cutting, which results
in blunt DNA ends 3 base pairs distal to the PAM sequence. Whether the staggered cutting pattern of Cpf1
will allow researchers to avoid using the inefficient HDR-dependent pathway for gene editing remains to be
demonstrated, regardless, it is likely that the most exciting applications of Cpf1 are yet to come. You can find
the Cpf1 plasmids and links to the publication on the Addgene website and read a little more about it in the
CPf1 section of this eBook.
Activation and Repression of Target Genes Using Cas9

One unique aspect of Cas9 is its ability to bind target DNA independently of its ability to cleave target DNA.
In other words, Cas9 containing mutations that disrupt DNA cleavage (D10A and H840A for SpCas9), is still
capable of binding target DNA. Further, so-called nuclease dead Cas9 or dCas9 can be used as a platform
to deliver various cargo to specific DNA loci by fusing them directly to dCas9. This property of dCas9 has
been exploited to localize a diverse array of proteins to target genes, including transcriptional activators and
transcriptional repressors (see our section on Cas9-based gene regulation).
Repressing target genes using dCas9-based repressors
Early experiments using dCas9 in bacteria demonstrated that targeting dCas9 to transcriptional start sites was
sufficient to repress or knock-down transcription by blocking transcription initiation. In mammalian cells,
robust repression requires targeting dCas9 tagged with transcriptional repressors to the promoter region of
the gene of interest. You may want to consider using dCas9-based repressors including dCas9-KRAB when
knocking out your target gene is toxic to cells. Plasmids for repressing target genes in a variety of species and
cell types can be found on Addgenes website.
Activating target genes using dCas9-based activators
The array of dCas9-based activators is quite diverse. The simplest activator consists of dCas9 fused to a
single transcriptional activation domain (typically VP64). A second generation of activators has recently been
developed and these alter gene expression using a few different approaches. For example, the SunTag
system facilitates recruitment of multiple activation domains to the same genetic locus through co-expression
of epitope tagged dCas9 and antibody-activator fusion proteins. The Synergistic Activation Mediator complex
consists of a dCas9-VP64 fusion and a modified gRNA that is capable of interacting with an additional RNAbinding transcriptional activator. Additional plasmids for activating target genes in a variety of species and cell
types can be found on Addgenes website.
40 | Page

OK, Ive Selected a Cas9! What Do I Do Next?
You have selected the Cas9 that is right for your experiment. What do you do next? Here are some
considerations that will help you move your CRISPR experiment forward.
Designing or selecting a gRNA - Addgene may already carry a validated gRNA for your favorite gene from
your favorite species. Validated gRNAs have been used successfully in experiments and published in peer
reviewed journals and will save your lab time and money when starting a CRISPR experiment. If you need to
generate a new gRNA for your experiment, you can select from a variety of empty gRNA vectors and design
your gRNA targeting sequence using one of the many freely available gRNA design programs.
What type of expression system or delivery method should I use? Selecting the appropriate Cas9 reagent for
your experiment is only half the battle - now you have to select the appropriate expression system and deliver
Cas9 to your target cells! Addgene carries a variety of Cas9 containing plasmids that have been optimized for
expression in different species and cell types including (but not limited to) bacteria, yeast, plants, drosophila,
worms and mammals (Browse plasmids by model organism, here). For difficult to transfect cell types, you may
want to consider using lentivirus to deliver Cas9 to your target cells. More information on the various delivery
systems for mammalian cells can be found in our section entitled, Mammalian Expression Systems and
Delivery Methods.
Validating your genome edit - Once you have delivered Cas9 and a gRNA to your target cells, it is time to
confirm that your target sequence has been modified! The exact protocol may vary depending on your specific
experiment, but a broad overview of the various ways in which you can verify your genome edit can be found in
section entitled Validating Your Genome Edit.
41 | Page
USING CAS9 TO GENERATE KNOCK-OUTS VIA NHEJ

By Joel McDade, David Wyatt, Dale Ramsden, Tyler J. Ford | Compiled on November 17th, 2015
CRISPR/Cas9 can be used to generate knock-out

cells or animals by co-expressing a gRNA specific to
the gene to be targeted and the endonuclease Cas9.
The genomic target can be any ~20 nucleotide DNA
sequence, provided it meets two conditions:
1.
The sequence is unique compared to the rest
of the genome.
2.
The target is present immediately upstream of a
Protospacer Adjacent Motif (PAM).
The PAM sequence is absolutely necessary for target
binding and the exact sequence is dependent upon
the species of Cas9 (5 NGG 3 for Streptococcus
pyogenes Cas9). We will focus on Cas9 from S.
pyogenes as it is currently the most widely used in
genome engineering. Once expressed, the Cas9
protein and the gRNA form a riboprotein complex
through interactions between the gRNA scaffold
domain and surface-exposed positively-charged
grooves on Cas9. Cas9 undergoes a conformational
change upon gRNA binding that shifts the molecule
from an inactive, non-DNA binding conformation, into
an active DNA-binding conformation. Importantly,
the spacer sequence of the gRNA remains free to
interact with target DNA. The Cas9-gRNA complex
will bind any genomic sequence with a PAM, but
the extent to which the gRNA spacer matches the
target DNA determines whether Cas9 will cut. Once
the Cas9-gRNA complex binds a putative DNA target, a seed sequence at the 3 end of the gRNA targeting
sequence begins to anneal to the target DNA. If the seed and target DNA sequences match, the gRNA will
continue to anneal to the target DNA in a 3 to 5 direction.
Cas9 will only cleave the target if sufficient homology exists between the gRNA spacer and target sequences.
The zipper-like annealing mechanics may explain why mismatches between the target sequence in the 3
seed sequence completely abolish target cleavage, whereas mismatches toward the 5 end are permissive
for target cleavage. The Cas9 nuclease has two functional endonuclease domains: RuvC and HNH. Cas9
undergoes a second conformational change upon target binding that positions the nuclease domains to cleave
opposite strands of the target DNA. The end result of Cas9-mediated DNA cleavage is a double strand break
(DSB) within the target DNA (~3-4 nucleotides upstream of the PAM sequence).
The resulting DSB is then repaired by one of two general repair pathways:
1.
2.
The efficient but error-prone Non-Homologous End Joining (NHEJ) pathway

The less efficient but high-fidelity Homology Directed Repair (HDR) pathway
42 | Page
USING CAS9 TO GENERATE KNOCK-OUTS VIA NHEJ (CONT)

The NHEJ repair pathway is the most active repair mechanism, capable of rapidly repairing DSBs, but
frequently results in small nucleotide insertions or deletions (Indels) at the DSB site. Here we will provide some
of the details behind NHEJ.
Non-Homologous End Joining

Unlike HDR, NHEJ is active throughout the cell cycle and has a higher capacity for repair, as there is no
requirement for a repair template (sister chromatid, homologue, or exogenously provided DNA) or extensive
DNA synthesis. NHEJ also finishes repair of most types of breaks in tens of minutes an order of magnitude
faster than HDR. NHEJ is consequently the principle means by which CRISPR/Cas9-introduced breaks are
repaired.
The following factors are required for NHEJ repair regardless of end structure, and dictate the major events of
the pathway:
1.
Broken ends are recognized by loading of the Ku70/Ku80 heterodimer
2.
Ku then acts as a scaffold for recruitment of a kinase (DNA-PKcs) and a two subunit DNA ligase
(XRCC4-ligase IV); together with some accessory factors (PAXX, XLF), this complex holds a pair of DNA ends
together, forming a paired end complex
3.
The paired end complex then ligates compatible DNA ends together, thus repairing the break
This is a simplified, streamlined version of this pathway and does not consider the missing or damaged
nucleotides that are common to biological sources of DSBs, and which need to be processed. Processing
occurs prior to ligation as incompatible DNA ends interfere with that step. Accordingly, NHEJ has a vast toolbox
of processing factors, including polymerases (Pol and Pol ), nucleases (Artemis), and structure-specific
end cleaning enzymes (Aprataxin, Tdp2) that function to make ends better substrates for ligation. Although
we do not describe these steps here, the processing of DNA ends tends to be the point where mutations are
introduced.
43 | Page

Repair of Cas 9-Induced Breaks by NHEJ
As illustrated above, NHEJ-mediated repair of Cas9-generated breaks is useful if the intent is to make a null
allele (knockout) in your gene of interest, as it is prone to generating indel errors. Indel errors generated in
the course of repair by NHEJ are typically small (1-10 bp) but extremely heterogeneous. There is consequently
about a two-thirds chance of causing a frameshift mutation. Of some importance, the deletion can be less
heterogeneous when constrained by sequence identities in flanking sequence (microhomologies).
It must be emphasized that NHEJ doesnt obligatorily introduce indels; given the end structure of the Cas9 DSB
(blunt or near-blunt ends without nucleotide damage) such products are rare, probably accounting for less than
5% of repair events. However, the products of accurate repair are easily re-cleaved while indel products arent,
so repeated cycles will favor accumulation of the latter products. As noted above, a single cycle of cleavage
and accurate repair should take less than an hour, thus a population of cells constitutively expressing a targeted
Cas9 should possess indels in the majority of their chromosomes within a day. Another factor expected to
impact repair is that the Cas9 protein doesnt immediately release from the broken end after cleavage, which
may interfere with loading of Ku and normal NHEJ activity.
NHEJ can also be engaged by variants of the canonical Cas9 approach. A pair of CRISPR guides that flank
regions of hundreds of base pairs or more can simultaneously introduce a pair of chromosome breaks, and
could result in deletion of the intervening DNA (pop-out deletions) if NHEJ joins the distal ends together.
Similarly, it may be possible to direct insertion of an exogenous DNA fragment at a Cas9 targeted break (or pair
of breaks) by NHEJ-dependent repair (pop-in insertion) provided a template containing compatible overhangs
is available. Cas9 can also be altered to generate a targeted single strand break; when two such breaks are
introduced near each other, in opposite strands, its presumed that this results in a DSB with long overhangs.
This double nickase strategy vastly reduces breaks and mutations at off-target sites, but it is not yet clear
how NHEJ engages this class of breaks (see the section on nickase for more information).
Many thanks to our guest contributors!
David Wyatt is a graduate student interested in determining how the structure of broken
ends impacts how they are repaired. He works in Dales lab.
Dale Ramsden is a member of the curriculum in Genetics and Molecular Biology, the
Dept. of Biochemistry and Biophysics and the Lineberger Comprehensive Cancer Center
at UNC Chapel Hill.
44 | Page

Further Reading
1. Bennardo, Nicole, et al. Limiting the persistence of a chromosome break diminishes its mutagenic potential. PLoS Genet 5.10 (2009): e1000683. PubMed PMID: 19834534. PubMed Central PMCID: PMC2752804.
2. Btermier, Mireille, et al. Is non-homologous end-joining really an inherently error-prone process. PLoS
Genet 10.1 (2014): e1004086. PubMed PMID: 24453986. PubMed Central PMCID: PMC3894167.
3. Waters, Crystal A., et al. The fidelity of the ligation step determines how ends are resolved during nonhomologous end joining. Nature Communications 5 (2014). PubMed PMID: 24989324. PubMed Central PMCID: PMC4107315.
4. Ghezraoui, Hind, et al. Chromosomal translocations in human cells are generated by canonical nonhomologous end-joining. Molecular Cell 55.6 (2014): 829-842. PubMed PMID: 25201414. PubMed Central PMCID: PMC4398060.
5. Ran, F. Ann, et al. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell 154.6 (2013): 1380-1389. PubMed PMID: 23992846. PubMed Central PMCID: PMC3856256.
45 | Page
EDITING WITH HOMOLOGY DIRECTED REPAIR

By Chari Cortez | March 12, 2015
DNA lesions are defined as sites of structural or basepairing damage in DNA. Perhaps the most nocuous
type of lesion results from breakage of both DNA
strands a double-strand break (DSB) as repair of
DSBs is paramount for genome stability. DSBs can be
caused by intracellular factors such as nucleases and
reactive oxygen species, or external forces such as
ionizing radiation and ultraviolet light; however, these
types of breaks occur randomly and unpredictably.
To provide some control over the location of the DNA
break, scientists have engineered plasmid-based
systems that can target and cut DNA at specified
sites. Regardless of what causes the DSB, the repair
mechanisms function in the same way.
Here, we will describe the general mechanism of homology directed repair with a focus on repairing breaks
engineered in the lab for genome modification purposes.
Mechanisms to Repair DNA Double-Strand Breaks

Genome stability necessitates the correct and efficient repair of DSBs. In eukaryotic cells, mechanistic repair of
DSBs occurs primarily by two pathways: Non-Homologous End-Joining (NHEJ) and Homology Directed Repair
(HDR). NHEJ (discussed in the previous section) is the canonical homology-independent pathway as it involves
the alignment of only one to a few complementary bases at most for the re-ligation of two ends, whereas HDR
uses longer stretches of sequence homology to repair DNA lesions. This section focuses on HDR, which is
considered to be the more accurate mechanism for DSB repair due to the requirement of higher sequence
homology between the damaged and intact donor strands of DNA. The process is error-free if the DNA
template used for repair is identical to the original DNA sequence at the DSB, or it can introduce very specific
mutations into the damaged DNA.
The three central steps of the HDR pathways are listed as follows:
1.
The 5-ended DNA strand is resected at the break to create a 3 overhang. This will serve as both a
substrate for proteins required for strand invasion and a primer for DNA repair synthesis.
2.
The invasive strand can then displace one strand of the homologous DNA duplex and pair with the
other; this results in the formation of the hybrid DNA referred to as the displacement loop (D loop).
3.
The recombination intermediates can then be resolved to complete the DNA repair process.
The invasion of the 3 single-stranded DNA (ssDNA) into the homologous DNA duplex (Step 2) is the defining
point of HDR. There are four different HDR pathways that can be employed to repair DSBs and the specific
mechanisms used in Steps 2 and 3 define the individual pathways as described below.
Mechanisms to Repair DNA Double-Strand Breaks
HDR can occur either non-conservatively or conservatively. The non-conservative method is composed
46 | Page
EDITING WITH HOMOLOGY DIRECTED REPAIR (CONT)
of the single-strand annealing (SSA) pathway and, in the interest of space, will not be discussed here. The
conservative methods, characterized by the accurate repair of the DSB by means of a homologous donor
(e.g., sister chromatid, plasmid, etc), is composed of three pathways: the classical double-strand break repair
(DSBR), synthesis-dependent strand-annealing (SDSA), and break-induced repair (BIR).
Classical Double-Strand Break Repair (DSBR)
In the classical DSBR pathway, the 3 ends invade an intact homologous template then serve as a primer for
DNA repair synthesis, ultimately leading to the formation of double Holliday junctions (dHJs). dHJs are fourstranded branched structures that form when elongation of the invasive strand captures and synthesizes DNA
from the second DSB end. The individual HJs are resolved via cleavage in one of two ways. Looking at the left
branch of the figure below, each junction resolution could happen on the crossing strand (horizontally at the
purple arrows) or on the non-crossing strand (vertically at the orange arrows). If resolved dissimilarly (e.g. one
junction is resolved on the crossing strand
and the other on the non-crossing strand),
a crossover event will occur; however, if
both HJs are resolved in the same manner,
this results in a non-crossover event. DSBR
is semi-conservative, as crossover events
are most common. This animation nicely
illustrates the DSBR pathway.
Synthesis-Dependent Strand-Annealing
Pathway (SDSA)
As illustrated on the right branch in the
figure to the left, SDSA is conservative, and
results exclusively in noncrossover events.
This means all newly synthesized sequences
are present on the same molecule. Unlike
DSBR, following strand invasion and D loop
formation in SDSA, the newly synthesized
portion of the invasive strand is displaced
from the template and returned to the
processed end of the non-invading strand at
the other DSB end. The 3 end of the noninvasive strand is elongated and ligated to
fill the gap, thus completing SDSA.
Break-Induced Repair (BIR) Pathway
Although BIR is not as well characterized as
either DSBR or SDSA, one central feature
of this pathway is the presence of only one
invasive end at a DSB that can be used for
repair. This single invasive strand invades
a homologous sequence and initiates both
47 | Page

leading and lagging strand synthesis, which results in the formation of one HJ.
This HJ is resolved by cleavage of the crossed strand. While this pathway may
not be immediately applicable in DSB-induced gene targeting or relevant to
plasmid-based genome engineering, it may have biological importance for the
repair of chromosome ends that have no second end that would enable DSBR
or SDSA.
Repair of DSBs and Genome Engineering
The advent of plasmid-based methods to induce DSBs engendered numerous

technologies that integrated homologous recombination into genome
engineering efforts. Early meganuclease-based technologies established
the foundation for plasmid-based genome engineering tools such as Zinc
Finger Nucleases (ZFNs) and TAL Effector Nucleases (TALENs). ZFNs and TALENs can both be used to
direct an endonuclease to a specific DNA locus targeted for modification. On the surface, the discovery and
development of RNA-guided CRISPR/Cas9 technology may just appear to be a new method for directing a
nuclease to cause a specific DSB; however, the ease of creating guides, the speed of the system, and the
overall versatility in application has not just reinvigorated genome engineering, but has really revolutionized the
field.General Considerations for Designing a Repair Template to Create Mutations
HDR templates used to create specific mutations or insert new elements into a gene require a certain amount
of homology surrounding the target sequence that will be modified. Scientists have been most successful using
homology arms that start at the CRISPR-induced DSB; however, there may be some wiggle room. In general,
the insertion sites of the modification should be no more than 100bp away from the DSB, ideally less than 10bp
away if possible, and the overall length of the homology arm is an important factor to consider when designing
these (more on this below). Longer distances of up to 200bp may work, but the efficiency will likely be lower
and you may need to introduce a selection marker to ensure the modification is present.
One important point to note is that the CRISPR/Cas9 system does not stop once a DSB is introduced and
repaired. As long as the gRNA target site/ PAM site remain intact, the Cas9 endonuclease will keep cutting and
the DSB will keep getting repaired through either NHEJ or HDR. This could be problematic if you are trying
to introduce a very specific mutation or sequence. To get around this, one may consider designing the HDR
template in such a way that will ultimately block further Cas9 targeting after the initial DSB is repaired. For
example, the PAM could be mutated such that it is no longer present, but the coding region of the gene is not
affected (i.e. a silent mutation).
The efficiency of HDR is generally low (<10% of modified alleles) even in cells that express Cas9, gRNA, and
an exogenous repair template. For this reason, many laboratories are attempting to artificially enhance HDR
by synchronizing the cells within the cell cycle stage (S-phase) when HDR is most active, or by chemically
or genetically inhibiting genes involved in NHEJ. The low efficiency of HDR has several important practical
implications. First, since the efficiency of Cas9 cleavage is relatively high and the efficiency of HDR is relatively
low, a portion of the Cas9-induced DSBs will be repaired via NHEJ. In other words, the resulting population of
cells will contain some combination of wild-type alleles, NHEJ-repaired alleles, and/or the desired HDR-edited
allele. Therefore, it is important to confirm the presence of the desired edit experimentally, and if necessary,
isolate clones containing the desired edit (see our validation section in Plan Your Experiment).
What Makes the Best Template: Plasmid DNA or Single-stranded Donor Oligonucleotide (ssODN)?
The size of your intended mutation is a big factor in deciding on a single- or double-stranded DNA repair
48 | Page

template. Historically, plasmids have been used as dsDNA templates when creating gene-targeting vectors;
however, ssDNA templates (ssODNs) have come into common use for smaller modifications as they tend to
have a higher efficiency. As a basic guideline, small mutations of up to ~50bp or single point mutations can
successfully be introduced using ssODN templates, while dsDNA plasmid-based templates should be used for
larger inserts such as fluorescent proteins or selection cassettes.
For ssODNs, the templates should be as long as possible with the Cas9-induced break point centered within
the template. Scientists have been successful with template lengths of ~100-200bp in total, with at least
40bp (but usually closer to 50-80bp) homology arms on either side of your intended mutation. Because target
sequence placement is PAM-dependent, it is not always possible to have the insertion site right next to the cut
site; however, they should be reasonably close (within ~20bp) to each other.
For larger inserts, dsDNA encompassing homology arms of 800bp each or larger should be used. Plasmids
are the most common source for providing dsDNA targets; however, they will need to be linearized before
transfection. This webpage is a great resource for designing a gene targeting vector.
Further Reading
1.
2.
3.
FAQ on Addgenes website: https://www.addgene.org/crispr/zhang/faq/#HR

CRISPR Forum Moderated by the Zhang Lab https://groups.google.com/forum/#!searchin/crispr/HDR
Maria Jasin and Rodney Rothstein. Repair of strand breaks by homologous recombination. Cold
Spring Harbor Perspectives in Biology 5.11 (2013): a012740. PubMed PMID: 24097900. PubMed Central PMCID: PMC3809576.
4. Pardo B, Gomez-Gonzalez B, and Aguilera A. DNA repair in mammalian cells: DNA double-strand
break repair: how to fix a broken relationship. Cell Mol Life Sci Mar;66(6) (2009): 1039-56. PubMed PMID:
19153654.
5. Shrivastav, Meena, Leyma P. De Haro, and Jac A. Nickoloff. Regulation of DNA double-strand break
repair pathway choice. Cell Research 18.1 (2008): 134-147. PubMed PMID: 18157161.
6. Pfeiffer, Petra, Wolfgang Goedecke, and Gnter Obe. Mechanisms of DNA double-strand break repair
and their potential to induce chromosomal aberrations. Mutagenesis 15.4 (2000): 289-302. PubMed PMID:
10887207.
7. Pques, Frdric, and James E. Haber. Multiple pathways of recombination induced by double-strand
breaks in Saccharomyces cerevisiae. Microbiology and Molecular Biology Reviews 63.2 (1999): 349-404.
PubMed PMID: 10357855. PubMed Central PMCID: PMC98970.
8. Ran, F. Ann, et al. Genome engineering using the CRISPR-Cas9 system. Nature Protocols 8.11 (2013):
2281-2308. PubMed PMID: 24157548. PubMed Central PMCID: PMC3969860.
9. Singh, Priti, John C. Schimenti, and Ewelina Bolcun-Filas. A mouse geneticists practical guide to
CRISPR applications. Genetics 199.1 (2015): 1-15. PubMed PMID: 25271304. PubMed Central PMCID:
PMC4286675.
10. Doudna JA and Sontheimer EJ. The use of CRISPR/Cas9, ZFNs, and TALENs, in generating site-specific genome alterations. Methods in Enzymology 546. (2014). PubMed PMID: 25398356.
11. Li, Kai, et al. Optimization of genome engineering approaches with the CRISPR/Cas9 system. PLoS
One (2014): e105779. PubMed PMID: 25166277. PubMed Central PMCID: PMC4148324.
12. Peng, Ying, et al. Making designer mutants in model organisms. Development 141.21 (2014): 40424054. PubMed PMID: 25336735. PubMed Central PMCID: PMC4302887.
49 | Page
USING CAS9 NICKASES FOR ENHANCED SPECIFICITY

By Joel McDade | October, 2015
CRISPR/Cas9 is highly specific when gRNAs are designed correctly, but specificity is still a major concern,
particularly as CRISPR is being developed for clinical use. The specificity of the CRISPR system is determined
in large part by how specific the gRNA targeting sequence is for the genomic target compared to the rest of the
genome. Ideally, a gRNA targeting sequence will have perfect homology to the target DNA with no homology
elsewhere in the genome. Realistically, a given gRNA targeting sequence will have additional sites throughout
the genome where partial homology exists. These sites are called off-targets and need to be considered when
designing a gRNA for your experiment. Learn how to design a gRNA in our How to Design Your gRNA for
CRISPR Genome Editing section.
In addition to optimizing gRNA design, specificity of
the CRISPR system can also be increased through
modifications to Cas9 itself. As discussed previously,
Cas9 generates double strand breaks (DSBs) through
the combined activity of two nuclease domains, RuvC
and HNH. The exact amino acid residues within each
nuclease domain that are critical for endonuclease
activity are known (D10A for HNH and H840A for
RuvC in S. pyogenes Cas9) and modified versions of
the Cas9 enzyme containing only one active catalytic
domain (called Cas9 nickase) have been generated.
Cas9 nickases still bind DNA based on gRNA
specificity, but nickases are only capable of cutting
one of the DNA strands, resulting in a nick, or single
strand break, instead of a DSB. DNA nicks are rapidly
repaired by HDR (homology directed repair) using the
intact complementary DNA strand as the template
(jump to our HDR section for more details). Thus, two
nickases targeting opposite strands are required to
generate a DSB within the target DNA (often referred
to as a double nick or dual nickase CRISPR
system). This requirement dramatically increases
target specificity, since it is unlikely that two off-target
nicks will be generated within close enough proximity
to cause a DSB. Therefore, if specificity and reduced
off-target effects are crucial, consider using the dual
nickase approach to create a double nick-induced
DSB. The nickase system can also be combined with
HDR-mediated gene editing for highly specific gene
edits.
Further Reading
1.
Cong, Le, et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339.6121 (2013):
2. Trevino, Alexandro E., and Feng Zhang. Genome Editing Using Cas9 Nickases. Methods Enzymol. 546
(2014): 161-74. PubMed PMID: 25398340.
3. Shen, Bin, et al. Efficient genome modification by CRISPR-Cas9 nickase with minimal off-target
50 | Page
USING CAS9 NICKASES FOR ENHANCED SPECIFICITY (CONT)

effects. Nature Methods 11.4 (2014): 399-402. PubMed PMID: 24584192.
51 | Page
HOW TO DESIGN YOUR gRNA FOR CRISPR GENOME EDITING

By John Doench | November 13, 2015
CRISPR technology has made it easier than ever both to make specific DNA sequences changes to the
genome and to perform genome-wide functional screens to identify genes involved in a phenotype of interest.
This section will discuss the differences between these approaches, as well as provide updates on how best to
design sgRNAs for your experiments.
Important Considerations Before You Start an Experiment with CRISPR

When designing experiments using CRISPR, the first choice centers on the type of edit you want to introduce: a
small insertion or deletion (indel) or a specific change (gene editing by homology directed repair [HDR]).
Using CRISPR to Create Indels
A small indel is most often used to disrupt the proteincoding capacity of a locus by introducing frameshift alleles
that result from the error-prone nature of non-homologous
end joining (NHEJ) following Cas9-mediated dsDNA breaks.
For this, all that is needed is Cas9 and the gRNA. It is clear
that this technique can be quite robust, with near-100%
efficiency possible, and many groups (eg. Shalem et al [1],
Wang et al [2], and Parnas et al [3]) have performed genomewide loss-of-function screens with this approach. When
using S. pyogenes Cas9, potential target sites are both
[5-20nt-NGG] and [5-CCN-20nt], as it is equally efficacious
to target the coding or non-coding strand of DNA. Thus,
for disruption of protein coding genes, potential target sites
occur ~1 every 8 nucleotides, and thus most genes have a
large number of gRNAs from which to choose.
Genome Editing via Homology Directed Repair
A specific change, such as the insertion of a fluorescent tag
or the introduction of a specific mutation, relies on HDR to
incorporate the new sequence into the DNA target site, and
thus also requires an exogenous template. HDR, however, is
still a very low-efficiency process, and usually involves the
need for single cell cloning and subsequent screening for
successful edits. This is a very time consuming process and should not be undertaken lightly! When targeting a
dsDNA break for HDR, the choice of target site is far more constrained by the desired location of the insertion;
efficiency decreases dramatically when the cut site is >30nt from the proximal ends of the repair template. This
means that, for gene editing, there are often far fewer potential gRNAs.
CRISPR Delivery Options

Once a target site has been identified, it is important to consider delivery options. For conducting genetic
screens using pooled libraries, the use of an integrating virus (e.g. lentivirus) is critical to the entire process;
at the end of the process you need to be able to determine what mutations caused the phenotypic effect
52 | Page
HOW TO DESIGN YOUR gRNA FOR CRISPR GENOME EDITING (CONT)

you screened for - this is easier to figure out if you can look for a genetic element that has stably integrated
into your cells. However, for generating a cellular model, long-term expression of CRISPR components is
not desirable, due to the potential for accumulation of off-target lesions. Transient expression options are the
appropriate choice for creating a stable cell line with your desired edit. Possible transient expression methods
include: transfection or electroporation of plasmid DNA, mRNA, or Cas9 protein pre-complexed to in vitro
transcribed gRNA, or the use of non-integrating viruses such as AAV or adenovirus.
If performing HDR, the repair template must be co-delivered with Cas9 and the gRNA as either a long, dsDNA
(e.g. a plasmid) or a single-stranded oligonucleotide. The choice between the two templates is largely dictated
by the size of the intended change, as introduction of small (< ~40) nucleotides changes can be accomplished
with a synthesized oligonucleotide of ~100 200 nts in length. Large inserts on the other hand, such as the
introduction of GFP, require a template with much larger homology arms.
Predicting gRNA On-Target Activity

Whether ones goal is gene disruption or gene editing, of one gene or genome-wide, being able to distinguish
effective from ineffective gRNAs can greatly streamline an experiment and simplify the interpretation of results.
Here at the Broad Institute, we examined sequence features that enhance on-target activity of gRNAs by
creating all possible gRNAs for a panel of genes and assessing, by flow cytometry, which sequences led to
complete protein knockout (4). By examining the nucleotide features of the most-active gRNAs from a set of
1,841 gRNAs, we derived scoring rules and built a website to help researchers design gRNAs against genes of
interest based on these rules.
More recently, we have both expanded our dataset and improved our computational modeling to derive a
more powerful set of rules to predict gRNA efficacy (Rule Set 2) (5). We measured the activity of more than
2,000 additional gRNAs to further strengthen the statistical power, and confirm the generalizability of activity
predictions. In collaboration with Nicolo Fusi and Jennifer Listgarten of Microsoft Research, we explored the
use of more-powerful computational modeling approaches. While our initial model (Rule Set 1) was based on a
fairly simple classification model, we found that the use of regression models in general, and gradient-boosted
regression trees in particular, greatly improved the power of our predictions. Web-based implementation of Rule
Set 2 is now available from both the Broad and Microsoft.
We tested the real-world impact of the use of these rules by designing a new genome-wide library, which
we named Avana (a grape used for making wine) for human and Asiago (a cheese) for mouse and compared
performance to the GeCKO library, which was developed before these rules were available. For both positive
and negative selection screens, we found that these new libraries were able to identify more hits with greater
statistical confidence, due to the increased consistency of different gRNAs targeting a gene, that is, more of
the gRNAs in the library were efficacious. This enables the use of smaller libraries, with fewer gRNAs per gene.
While it is of course true that using more gRNAs per gene in a screen provides more information, this comes at
the cost of screening and sequencing more cells, which puts some cellular models and experimental systems
out of reach. Thus, for many researchers, a primary screen that uses a smaller, high-activity genome-wide
library will be desirable. Towards this end, we have made new libraries, named Brunello for human (again, a
wine-making grape you can see where were going with this) and Brie for mouse, that take into account both
our newest on-target designs, as well as methods to avoid off-target sites, discussed below. Pooled libraries
can be requested from Addgene as they become avaliable.
53 | Page
Decreasing Off-Target Effects

Avoiding off-target effects of Cas9, that is, preventing cuts at other, unintended sites in the genome, is an
important step in designing gRNAs. Merely glancing through the literature shows that different groups have
come to wildly different conclusions as to the specificity of gRNAs. To take two examples, compare these titles:
Low Incidence of Off-Target Mutations in Individual CRISPR-Cas9 and TALEN Targeted Human Stem Cell
Clones Detected by Whole-Genome Sequencing(6)
High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells(7)
It is reasonable to ask, well, which is right? As usual, the truth lies in the details, which is another way of saying
that you cant judge a journal article by its title! Indeed, both titles are correct within the confines of each
study, but the generalizability is what matters most. For sure, some differences in these reports (and many
others) likely relate to differences in experimental systems, but probably most importantly, both of these papers
examined small numbers of gRNAs. Are there some really promiscuous gRNAs? For sure! Are there quite
specific ones? You bet! Of course, the same could be said for essentially any targeting technology there are
both really specific and really non-specific TALENs, siRNAs, antibodies, and small molecules.
Generalizability, then, needs to come from sampling from large numbers, and indeed, rules governing offtarget effects are beginning to be understood in more detail. First, direct physical detection of off-target sites
though techniques like GUIDE-Seq, developed by the Joung lab, have shown that some gRNAs have dozens
of detectable off-target sites, but that same study also found 1 gRNA, of 10 examined, that had zero offtarget sites by their technique (8). Further, they showed that existing heuristics to find and score off-targets in
fact miss many sites. They compared GUIDE-Seq results to two prediction sites, from Feng Zhangs lab and
Michael Boutross lab and discovered that neither program identified the vast majority of off-target sites found
by GUIDE-seq. Of course, at the time of launch, these servers were based on the best-available information at
the time, and the perfect should not be the enemy of the good.
More recently, we have examined off-target sites at much larger scale than previous studies and developed
the CFD (Cutting Frequency Determination - see paper for details) score to predict off-target sites with better
sensitivity and specificity than previous heuristics (5). In the course of this study, we also found that the search
algorithm itself plays a perhaps under-appreciated role in arriving at the right result. Because of its ease of
implementation and speed, many have used Bowtie2 to perform scans of the genome to find off-target sites
that contain small numbers of mismatches, but the Bowtie algorithm was not designed for quite this purpose,
and in fact misses many potential off-target sites, especially sites with more than 1 mismatch. Thus, both the
search metric and the scoring metric are critical for a comprehensive view of potential off-target sites.
Even as off-target activity becomes better understood, in the context of genetic screens, it is still critical
to require that multiple perturbations targeting a gene show consistency in order to conclude that the
observed phenotype is due to on-target activity. Indeed, all libraries build in this redundancy. For gene editing
approaches, however, where the goal is to introduce a specific change at a specific site, the choice of gRNAs
is often quite limiting. One method to decrease off-target effects with CRISPR technology is the use of two
gRNAs in combination with a mutated nickase version of Cas9. This approach has the benefit of increased
specificity and thus a reduced rate of off-target dsDNA breaks. One downside of this approach, though, is that
the requirement for two target sites will mean some specific locations are not suitable for creating a dsDNA
break. When possible, though, this is the preferred approach for gene editing. The characterization of novel
54 | Page

CRISPR systems, and the development of variants of Cas9 enzymes with alternative PAM requirements,
promises to make more and more of the genome readily accessible to on-demand dsDNA breaks.
Conclusions
In sum, selection of gRNAs for an experiment needs to balance maximizing on-target activity while minimizing
off-target activity, which sounds obvious but can often require difficult decisions. For example, would it be
better to use a less-active gRNA that targets a truly unique site in the genome, or a more-active gRNA with one
additional target site in a region of the genome with no known function? For the creation of stable cell models
that are to be used for long-term study, the former may be the better choice. For a genome-wide library to
conduct genetic screens, however, a library composed of the latter would likely be more effective, so long as
care is taken in the interpretation of results by requiring multiple sequences targeting a gene to score in order to
call that gene as a hit.
This is exciting time for functional genomics, with an ever-expanding list of tools to probe gene function. The
best tools are only as good as the person using them, and the proper use of CRISPR technology will always
depend on careful experimental design, execution, and analysis.
Many thanks to our guest contributor John Doench!
John Doench is Associate Director of the Genetic Perturbation Platform at the Broad
Institute and has worked with many Addgenies to help improve the understanding, curation, and explanation of our CRISPR resources. He really likes small RNAs.
Further Reading
1. Shalem, Ophir, et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science
2. Wang, Tim, et al. Genetic screens in human cells using the CRISPR-Cas9 system. Science 343.6166
(2014): 80-84. PubMed PMID: 24336569. PubMed Central PMCID: PMC3972032.
3. Parnas, Oren, et al. A genome-wide CRISPR screen in primary immune cells to dissect regulatory
networks. Cell 162.3 (2015): 675-686. PubMed PMID: 26189680. PubMed Central PMCID: PMC4522370.
4. Doench, John G., et al. Rational design of highly active sgRNAs for CRISPR-Cas9-mediated gene
inactivation. Nature Biotechnology (2014). PubMed PMID: 25184501. PubMed Central PMCID: PMC4262738.
5. Doench, John G., et al. Optimized sgRNA design to maximize activity and minimize off-target effects of
CRISPR-Cas9. Nature Biotechnology (2016). PubMed PMID: 26780180.
6. Veres, Adrian, et al. Low incidence of off-target mutations in individual CRISPR-Cas9 and TALEN
targeted human stem cell clones detected by whole-genome sequencing. Cell Stem Cell 15.1 (2014): 27-30.
7. Fu, Yanfang, et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in
human cells. Nature Biotechnology 31.9 (2013): 822-826. PubMed PMID: 23792628. PubMed Central PMCID:
55 | Page

PMC3773023.
8. Tsai, Shengdar Q., et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPRCas nucleases. Nature Biotechnology 33.2 (2015): 187-197. PubMed PMID: 25513782. PubMed Central
PMCID: PMC4320685.
56 | Page
THE CRISPR SOFTWARE MATCHMAKER

By Cameron MacPherson | November 3, 2015
This section was contributed by guest writer Cameron MacPherson at the Institut Pasteur.
CRISPR Software and the Piata Effect

Two years ago I was a part of a group (Biology of Host-parasite Interactions, Institut Pasteur, Paris) that
changed genome editing in the malaria community for the better (1). Given the timing, it shouldnt be a surprise
that the CRISPR system was involved. Today, that same laboratory enjoys a successful edit rate of over 90%
in their work editing the genome of Plasmodium falciparum (the parasite that causes malaria). I attribute their
success to technical expertise, thoughtful gRNA design, and the abnormally low GC content of the Plasmodium
falciparum genome. To put this last point into perspective, the Plasmodium falciparum genome contains only
0.66 million targetable NGG PAM sites whereas the human genome has about 300 million. With such a sparsely
targetable genome, off-targeting is less of a worry and on-targeting likely more efficient. These insights are hard
to appreciate without computational support. Indeed, rational gRNA design is not possible without relying on
some kind of pre-analysis. At the end of 2014, I began developing software to make gRNA design accessible
to all. At the time I thought there was room for improvement and a year later it became quite clear that others
thought the same. Since January 2013 there have been 33 CRISPR software tools published and documented
by OMICtools. It must be confusing for a newcomer to decide on the right software to use. With so much
choice the question I have is, how did we get here in the first place?
How CRISPR Innovation Affected

Software Development
CRISPR, like the miRNA bubble before it, offers an
interesting view into how rapid innovation affects the
research community. Ive likened it to the celebration
surrounding the piata. From the perspective of those
building the piata, their role is to fill the prop with
something of value. They are reviewers of content. On
the other hand, the goals of the piata beating party
goers are altogether different. They are filled with blind
anticipation. Their trust in the quality of the content is
implicitly defined by their trust in the piata builders the reviewers. After some physical exertion on the part
of the party goers, the piata bursts and its contents
spill to the ground. There is something akin to a frenzy as the crowd tries to weigh and determine the value
of every sweet or toy strewn across the floor. In the final wake it all settles down and the value of each item
changes from what the reviewers so painstakingly assessed to something a bit more representative of the
communities opinion. The trouble is, with so much choice, in those few moments after the piata burst, value
was subjectively ascribed relative to the basic needs of each individual; the community as a whole has yet to
settle on any true kind of value. Ive dubbed this the Piata Effect, it is the disconnect arising between those
reviewing the contents of the piata and the audience blindly receiving them; it results in highly similar content
and the absence of iterative design; it is caused by not having the time to assess public response; and, it is the
stage I think we are currently at in the CRISPR software space. The CRISPR research community hasnt had a
chance to develop a consensus on the best CRISPR software tools yet.
57 | Page
THE CRISPR SOFTWARE MATCHMAKER (CONT)

As a genome editing tool, the CRISPR/Cas9 technology has been surrounded by a whirlwind of research
activity and development since its inaugural year in 2012. Thats just 3 to 4 years ago. It is a bubble of
innovation (a piata), and software development has caught up with it. The peer review process has been
challenged with many closely spaced CRISPR software submissions. With few prior publications to go
on, a submitted manuscript could only have been viewed as a significant improvement. As a result, the
software piata has been filled with slight variations on a theme without many resources to review them.
Collectively, the pool of CRISPR software embodies solutions that facilitate most experimental applications in
CRISPR engineering. When compared to that collective utility, its easy to fault the lack of features in a single
application. The question is then, should we use all tools as some kind of a meta/Frankenstein app? Or perhaps
there is a clear winner, is there something that we can invest our time into learning and extract the most value?
This is only one of many decisions faced by those engaging in CRISPR design, but it is a significant hurdle
made worse by an ever increasing marketplace. Since 2013 about 11 tools have been added every year.
Breaking Down the Barriers: My View on Current gRNA Design Tools

The goal of this post is to provide some insight into available CRISPR software tools, what problems
various tools are trying to solve, and finally how we might proceed in the future. When we think of designing
experiments using CRISPR, there are two major areas where software can help. The foremost handles the
design of gRNA and represents the lions share of the currently available tools (we will focus on these tools in
this section). The second major area deals in post-experiment quality control, a good example is CRISPR-GA.
These tools assess repair events by type (NHEJ/HDR) and track indels at a single locus, usually the target site.
Multi-locus assessment would be required to fully evaluate off-targeting, but currently no software has been
designed for this purpose. This second area is extremely important and given the lack of attention to it, it is an
obvious place to focus further development. Quality control is, however, not a focus of this section.
I separate gRNA design tools into database and de novo solutions. Database tools allow us to view and get a
sense of which gRNA designs have previously worked and under what conditions. Such resources could be
highly valuable for automated rational design. The first of only three such databases, EENdb (published 1 Jan
2013), is a simple catalogue of reported gRNA designs. The second database, CrisprGE (published 27 Jun
2015), is noteworthy for its broad scope, curated content, and ease of access. The third, WGE (published 15
Sep 2015), is more similar to EENdb than CrisprGE, but is also a good example of how database tools can be
used to aid design. These databases are still young and will require a collaborative effort from the community in
order to succeed.
With respect to de novo gRNA design, I again separate it into two categories. On the one hand we have
software tools that apply some approximate or sensible rules to determine the value of one gRNA design over
another. On the other hand some software tools make use of empirically derived and/or prior knowledge to
inform on the qualities of what determines good gRNA design. Incidentally, the first software for gRNA design
(by Hsu et al., 2013, [2]) used a combination of both empirical and approximate methods. The details of their
scoring function can be found in their online tools documentation. Since Hsu et al., 2013 (2), other tools have
incorporated new parameters such as the Doench-Root score (by Doench et al., 2014, [3]) into their scoring
functions. Many tools have also opted to mix-and-match different scoring algorithms and parameters. This
tapestry of ideas is what makes choosing the right tool confusing and it means that there is no one-size-fits-all
approach. You will have to select a tool based on your project and not general opinion.
58 | Page
CLICK HERE TO USE THE CRISPR SOFTWARE MATCHMAKER

I could spend a lot of time running through each tool in detail, but that is already covered in the published
articles. Instead, I have broken down the tools into the individual features that make each tool unique. The
CRISPR Software Matchmaker is composed of these features and enables you to select the tool(s) based
on your project needs. There are 8 major categories found in the table, describing everything from basic to
advanced functionality as well as how the user is expected to interact with the tool (a screen shot and overview
of the table can be found above). These categories are further discussed below (definitions for all terms are also
available in the table):
Basic functions: These functions embody the main goals of the software and should be the first place
you look to determine a tools suitability. Functions of this category are the most common between
tools. Examples: single-target design, multi-target design, off-target aware, high mismatch limit,
approximate design, empirical design, single-PAM design, and multi-PAM design. Trends: By
the end of 2014, tools began moving away from designing exclusively for NGG PAM targets to allow for
arbitrary PAM definitions. These newer tools may still prove somewhat useful for one of the latest NTT
targeting Cas protein, Cpf1. Separately, on-target efficacy is still a concept that only a few tools are trying
59 | Page

to solve.
Advanced functions: These functions are not entirely necessary but should be considered extremely
useful, depending on design goals. Examples: feature aware, SNP aware, secondary structure
aware, and microhomology aware. Trends: Fewer tools are being released with advanced functionality,
it seems these kinds of features are delegated to other more suitable software tools such as ApE or
commercial workbenches such as MacVector or Benchling.
Utility functions: These functions help speed up the gRNA design process by removing repetitive tasks
and/or by providing features to help with the post-design process such as primer and plasmid design.
Examples: multiplex design, multi-method design, and single-method design. Trends: The most
common utility functions are batch design or multiplex features. However, tools aiding in primer design
are more and more common.
User interaction: Software design elements that fall into this category describe how the user is
expected to interact with the software. This category is important for users wishing to select tools
based on their comfort level with operating a computer. Examples: offline, online, CLI, and GUI.
Trends: Online tools dominate the software space but generally rely on less powerful algorithms to detect
off-targets. The few offline tools available mostly work on any computer but generally require working
knowledge of the command line or scripting.
Input flexibility: Software tools require some kind of data input in order to generate results. While
different types of input are more user-friendly than others, this is of less concern than data output as
different input types can be readily converted. Examples: organism, sequence, identifier, and
load. Trends: Sequence based input is by far the most common with most tools also requiring that the
user specify an organism.
Output diversity: Different tools provide results to the user in different formats. This can have a large
impact on downstream results. Examples: HTML, visual track, plot, tabular, interactive, and
save. Trends: Most tools provide tabular HTML formatted output. Surprisingly only 2 tools provide
FASTA output. A few notable tools allow the user to save the results and return later or share them.
Community exclusivity: Different communities develop software for their own needs. This often
results in software design that is applicable to only one, or a subset of organisms. In only a few cases
the software is organism agnostic. Most tools are exclusive to one or a subset of organisms. Examples:
organism agnostic, model organisms, small genome, large genome, and organism biased
scoring. Trends: Many tools can handle genomes of any size and are only limited by the time it takes
authors to add a genome to the tools repertoire. Organism agnostic tools are the better solution but are
often packaged as offline tools requiring expertise in one programming language or another.
Supported organisms: This section lists the genomes supported by each tool. The genomes are
denoted by organism name and genome assembly version. The names in the table appear as they do in
the software. Most gRNA design tools require the user to specify an organism. They do this because they
rely on pre built indices or databases in order to find and present the results to you as soon as possible.
Some tools dont require the organism to be specified and are truly organism agnostic. Others tools
require an organism to be specified but also allow you to build your own database; these tools will be
marked as organism agnostic and have currently available organisms listed in this section.
60 | Page

CRISPR Software Advice for Specific Users
You will find that some tools cater to specific use cases through the way their algorithms were developed
or by their focus on specific organisms. For this reason I generally recommend that laboratories that have
already attempted several CRISPR experiments evaluate the predictive power of each tool they are interested
in by comparing their past results to the designs suggested by each tool. For those who are entering gRNA
design for the first time, it is best to choose a tool based on immediate needs and re-evaluate after several
experiments. Alternatively, the databases mentioned above or, even better, first-hand knowledge from
laboratories employing CRISPR on the same organism could be used as a proxy for this last approach.
Top tool for first timers: You need to learn the language and what the parameters represent. The best
learn-by-doing tool is E-CRISP. The authors have made a huge effort so that their tool is both didactic
and functional. It is also the only tool I have come across to offer classes of parameters based on the
type of CRISPR experiment being pursued. You should also look at the glossary section of the CRISPR
Software Matchmaker, it provides a list of terms I think are important to define for the CRISPR software
space. It is unlikely that you will find these terms anywhere else, because I developed them for this post,
but they should give you an idea of what to look out for.
Top tools for bioinformaticians: For flexibility in your own analysis, you need access to raw data, the
target sites, off-target sites, scores and the statistics that go into calculating them. For this, some online
tools such as Cas-OFFinder will suffice. Cas-OFFinder provides a bare-bones data dump of ALL strings
in a genome using any IUPAC encoded pattern at any edit distance. It wont, however, compute any
score other than the edit distance. Added functionality is left to you to implement. You can also download
the software for use on any machine with OpenCL enabled hardware. This OpenCL dependency is a
severe limitation, but the tool is fast enough that, if youre engaged in heavy CRISPR design (Im talking
screens and/or very large genomes), then it is worthwhile buying a dedicated machine. On the offline
front, there is a Python script called SSFinder, but I find its implementation to be too slow for practical
use. The R based package CRISPRseek offers great utility and coupled with Bioconductor should be the
first choice for anyone already familiar with R. For the Java enthusiasts, take a look at sgRNAcas9.
Top tools for transferring CRISPR technology to a new organism: For obvious reasons you can rule
out any tool that limits you to a subset of specific genomes. These are most of the tools, but the good
news is that you only need one. ProtospacerWB was developed with the purpose of applying CRISPR
technologies to new organisms and will help you whether you have a full assembly or not. It is an offline
tool, but comes with a graphical user interface.
Best strategy for labs: Define what you need the software tool to do before going shopping for one.
Use the CRISPR Software Matchmaker to select the best tool based on your needs. Refine your criteria.
Repeat until you have found the best tool. Do several experiments and use the results to re-evaluate all
tools. Report your findings to help everyone else.
CRISPR technology has reached into so many different communities that it is all but impossible to fairly
judge individual tools. I believe at this time it is only possible to objectively break the tools into their individual
offerings and allow you to select them based on your needs. We will likely begin to see a contraction of
available tools with more and more features being integrated into so-called genome editing workbenches such
as Benchling (commercial, online), DESKGEN (commercial, online), or ProtospacerWB (academic, offline). While
it remains to be seen, I believe the future of CRISPR software is promising. Rapid innovation in this space has
61 | Page

put us in a good position to step back and cherry pick the features that will actually make a difference in our
experiments and lives. To do this effectively, there needs to be good communication between developers and
end users.
Many thanks to our guest contributor Cameron MacPherson!

Cameron R. MacPherson is Lead Data Scientist for the Milieu Intrieur project, Institut
Pasteur, Paris. You can follow him on Twitter @CMacPhD.
Further Reading
1. Ghorbal, Mehdi, et al. Genome editing in the human malaria parasite Plasmodium falciparum using the
CRISPR-Cas9 system. Nature Biotechnology 32.8 (2014): 819-821. PubMed PMID: 24880488.
2. Hsu, Patrick D., et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nature Biotechnology
3. Doench, John G., et al. Rational design of highly active sgRNAs for CRISPR-Cas9-mediated gene
inactivation. Nature Biotechnology 32.12 (2014): 1262-1267. PubMed PMID: 25184501. PubMed Central
PMCID: PMC4262738.
62 | Page
VALIDATING YOUR GENOME EDIT

By Melina Fan | July 30, 2015
Youve created your gRNA expression construct and used Cas9 to introduce it into your target cells. Hooray!
Youre ready to begin reading out data, right? Almost. In this section well explain how to verify that your cells
were appropriately edited. Well also cover the basic techniques for detecting insertion, deletion, and mutation
events.
Process Overview
The method for validating your genome edit will vary by species and the type of edit. In this post, we will focus
on diploid mammalian cells, but many of the principles will hold across different model organisms.
Introducing Cas9 and a gRNA into your cells (possibly along with a donor template) will result in a mixed
population of cells. Following the introduction of a Cas9-mediated double strand break (DSB) in mammalian
cells, cellular machinery repair the DSB by non-homologous end joining (NHEJ) or homology directed repair
(HDR). Repair via the NHEJ pathway predominates in mammalian cells resulting in the creation of indel errors,
short heterogeneous insertions, and deletions of nucleic acid sequences, at the site of the DSB. In addition
to the heterogeneity of indels introduced at Cas9-induced DSBs, allelic editing frequencies will vary as well.
The HDR pathway requires the presence of a repair template, which is used to fix the DSB in a more specific
manner. HDR faithfully copies the sequence of the repair template to the cut target sequence. Some cells will
not be edited, some will have one allele edited, and some will have both alleles edited.
63 | Page
VALIDATING YOUR GENOME EDIT (CONT)

The first step in the validation process is to quickly assess whether a significant number of the cells have been
edited (see Figure 1). For indels, this is visualized using a mismatch cleavage assay (see Figure 2). For HDR,
this is often visualized by a change in the restriction pattern at the site of interest or via a reporter readout. For
deletions (see Figure 3), this is visualized by a decrease in size of a PCR product produced by primers flanking
the region to be deleted.
Once you know that a portion of your cells have been edited, you can go on to create clonal cell lines. Serial
dilutions can be used to isolate single cells followed by an expansion period to generate these lines. If you have
a fluorescent protein marker on your plasmid, you can use FACS to enrich the cells that received Cas9 and your
gRNA. After expansion, assay each cell line and sequence the region of interest in order to validate the genome
edit as described in Figure 1. When possible, you should also assess protein expression via western blot as a
further form of validation.
Mismatch Cleavage Assay to Detect Indels

A mismatch cleavage assay is a quick and easy way to
detect indels. SurveyorTM nuclease is commonly used for
this purpose, as it cleaves both DNA strands 3 to any
mismatches. It can detect indels of up to 12 nucleotides
and is sensitive to mutations present at frequencies as
low as 1 in 32 copies.
Mismatch cleavage assays typically consist of four steps:
1) PCR amplify the region of interest
2) Denature the strands and rehybridize to allow for the
mutant and wild-type strands to anneal
3) Treat annealed DNA with SurveyorTM nuclease to cleave
heteroduplexes
4) Analyze DNA on an agarose gel or other instrument
that separates DNA based on size
Figure 2 illustrates how the assay works. In this example,
both +gRNA lanes contain cleaved fragments of the
expected sizes, indicating that the gRNAs successfully
produced indels in the target region. This assay is
often used semi-quantitatively, and in this case, gRNA1
appears to be more efficient at producing indels than
gRNA2.
Detect Homology Directer Repair

If you want to mutate your region of interest using HDR,
it is advisable to first determine whether your gRNA is
efficiently cutting your target sequence by creating indels
and conducting a mismatch cleavage assay. Once youve
selected your optimal gRNA, introduce it along with Cas9
and your repair template to drive HDR.
64 | Page

When designing your HDR donor template, plan ahead for detection of integration events. For instance,
you could purposefully introduce or remove restriction sites which would alter the digestion pattern of PCR
products. Alternatively, you could include a reporter element for detection of HDR at the DNA, RNA, or protein
level. A large insertion or deletion after integration could also be detected by a size change in the PCR product.
Single nucleotide changes can be quickly assayed using restriction digests if the polymorphism creates/
removes a restriction digest site. Otherwise, single nucleotide changes can be detected by TA cloning and
Sanger sequencing, next generation sequencing, or droplet digital PCR genotyping.
HDR events are generally less frequent than indels, so you will likely need to screen a larger number of colonies
to create a clonal line. The number of clones that need to be screened will depend on both your transfection/
transduction efficiency and HDR frequency. For example, if you have 40% transfection efficiency and 5% HDR
efficiency, approximately 0.4x0.05=0.02 or 2% of your cells will have the recombined region. Thus, you should
plan to screen at least 50 colonies. If you are able to select cells that have been successfully transfected/
transduced using a marker, then you may be able to test fewer colonies.
PCR to Detect Deletions

Most deletions are created by using two gRNAs that direct Cas9 to cleave out the intervening region of
DNA. The deletion can thus be detected by conducting a PCR using primers flanking the deleted region. The
workflow is similar to that described in Figure 1. Figure 3 provides an example of PCR results obtained by
screening a panel of clonal lines for deletions. In this example, clones 1, 5, and 7 are heterozygous for the
deletion and clone 4 is homozygous for the deletion.
65 | Page

Next Generation Sequencing to Validate Edits and Detect Off-Target Effects
If your lab has the resources, you can quantitatively assess genome edits in your target sequence and other
regions of the genome using next generation sequencing (NGS). NGS is a good option if you have a large
number of samples and/or want to simultaneously look at off-target changes. When using this method, it is
important to keep a set of control cells as you will need to compare the sequencing reads from your edited
sample to this untreated population. Software such as CRISPResso can help with the data analysis.
The techniques described in this post are not CRISPR-specific and can also be used for assessing genome
edits created by TALEN or Zinc Finger Nucleases. Regardless of what method you use, validating your edit is
time well spent as you prepare for your future experiments.
Thank you to David Scott (Dr. Feng Zhangs Lab), Joel McDade (Addgene), and Marcy Patrick (Addgene) for
helpful comments and edits.
Further Reading
1. Qiu, Peter, et al. Mutation detection using Surveyor nuclease. Biotechniques 36.4 (2004): 702-707.
2.
Ran, F. Ann, et al. Genome engineering using the CRISPR-Cas9 system. Nature Protocols 8.11 (2013):
66 | Page
GENOME-WIDE SCREENING USING CRISPR/CAS9

By Joel McDade | August 18, 2015
What genes are important for your phenotype? Many scientists (possibly you!) study diseases for which the
underlying genetic cause is not entirely known. Identifying which genes are important for a phenotype can lead
to a wealth of additional experiments investigating the role of individual genes or entire pathways in a particular
disease process and could aid in the development of novel therapies. While CRISPR/Cas9 is certainly not the
first means to carry out these so-called forward genetic screening experiments, it is certainly the most robust.
In this blog post, we will discuss how CRISPR libraries are being used to perform genome-wide screens and
highlight some of the reagents that have been made available through Addgene.
What Makes CRISPR So Special?

A major advantage of CRISPR/Cas9 over previous
genome editing techniques is its simplicity and
versatility. The CRISPR/Cas9 system consists of two
components: the non-specific endonuclease (Cas9)
and a single stranded guide RNA (gRNA). The ~20
nucleotide targeting sequence within the gRNA is
user defined, and can be easily modified to target Cas9
to virtually any genomic locus, provided the target
is unique compared to the rest of the genome and
located immediately 5 to a protospacer adjacent motif
(PAM) sequence. Co-delivery of wild-type Cas9 and
a gRNA generates a double-strand break in the target
DNA, which, when repaired through error-prone nonhomologous end joining (NHEJ), can result in a loss-offunction mutation within the target gene. CRISPR/Cas9
can also be used to activate or repress target genes
without permanently modifying the genome by using
nuclease dead Cas9 (dCas9) and dCas9-activators/
repressors.
What CRISPR-based Reagents

Are Available for Genome-wide
Screens?
The goal of a genome-wide screening experiment is

to generate and screen a population of mutant cells
to identify genes involved in a particular phenotype.
CRISPR/Cas9 can be readily scaled up for genomewide screening due to the broad range of potential
target sequences and ease of generating gRNAcontaining plasmids. CRISPR/Cas9 genome widescreening experiments commonly use lentivirus
to deliver a pooled population of gRNAs to Cas9
Figure 1: Like this jar of jellybeans, a pooled CRISPR library is a complex

mixture. Pooled libraries are composed of many different gRNA-containing plasmids, each with a different genomic target. The goal of a CRISPR
screen is to use Cas9 and a pool of gRNAs to identify genes that are
essential for a given phenotype. Image Credit: https://www.flickr.com/photos/72005145@N00/5600978712.
expressing cells. Pooled lentiviral CRISPR libraries (heretofore referred to simply as CRISPR libraries)
consist of a heterogeneous population of gRNA-containing lentiviral transfer vectors, each targeting a specific
67 | Page
GENOME-WIDE SCREENING USING CRISPR/CAS9 (CONT)

gene within the genome (see Figure 1). Individual gRNAs are designed in-silico using publicly available gRNA
design software and synthesized. Pooled gRNAs are then cloned into a lentiviral transfer vector, resulting in
the final CRISPR library. CRISPR libraries have been adapted to knock-out, activate or repress target genes
by combining a gRNA library with the aforementioned derivatives of Cas9. Several CRISPR libraries are now
available through Addgene.
Choosing the Library that Is Right for You

There are several factors to consider when selecting a CRISPR library for your experiments, such as: 1) what
species are your cells derived from? Currently, Addgene carries CRISPR libraries that target mouse or human
genes. 2) What is the genetic modification you are trying to make? Addgene carries CRISPR libraries for gene
knock-out (mouse and human), activation (mouse and human) and repression (mouse only). 3) Are you trying
to target every gene in the genome, or a specific class of genes? Addgene currently carries several genomewide CRISPR libraries and a selection of sub-libraries targeting specific classes of human genes. A welldeveloped biological question is absolutely necessary to ensure that you select the correct CRISPR library for
your experiment.
What Are the Steps Involved in a CRISPR Screen?

Performing a forward genetic screen using CRISPR
libraries is a multi-step process (see figure at left).
In most cases, CRISPR libraries are provided at a
concentration that is too low for experimental use.
Thus, the first step is to amplify your library to a
concentration that is sufficient to be used to generate
lentivirus. Be sure to check the representation of
your library (percentage of each gRNA pre- and postamplification) using next-generation sequencing.
Cas9-expressing cells are then transduced with
lentivirus containing the CRISPR library to generate
a heterogeneous population of mutant cells, with
each cell or sets of cells containing a mutation in a
different gene. Mutant cells are enriched using either
drug-selection or fluorescence-based cell sorting and
screened for a particular phenotype. For example,
mutant cells can be used in drug screens to identify
genes that confer drug-resistance. Here, mutant
Figure 2: The CRISPR library must be amplified (A) to be used to generate
lentivirus (B). Cas9-expressing cells or wild-type cells are treated with
cells are treated with a drug of interest and gRNA
lentivirus containing the gRNA library or gRNA library plus Cas9 to generate
distribution is analyzed in the drug-resistant population
mutant cells, respectively (C). Mutant cells are screened (D) and hits are
identified using next-generation sequencing (E).
compared to a non-treated control group. In this
scenario, gRNAs that are enriched correspond to
genes that confer drug-resistance when mutated. Findings from this type of experiment can shed light on the
mechanism by which cells gain resistance to drugs and can identify future therapeutic targets for diseases
causing uncontrolled cell-growth, such as cancer.
68 | Page
GENOME-WIDE SCREENING USING CRISPR/CAS9 (CONT)

Considerations and Tips for Successful Screens
Next-generation sequencing - CRISPR libraries contain thousands of gRNA plasmids, discerned only by a
unique barcode on each plasmid. As such, sequencing CRISPR libraries after amplification and after a screen
requires the use of next-generation sequencing.
Representation - Most libraries contain >3 gRNAs per target gene, and maintaining the distribution of each
gRNA within the population (representation) is absolutely essential. Loss of representation due to enrichment
or depletion of specific gRNAs can lead to skewed results.
Selecting a cell type - Theoretically, any cell type can be used in a CRISPR screen. However, maintaining
sufficient representation within your mutant population requires a massive amount of cells as starting material.
Therefore, cell types that are of low abundance are not particularly well suited for genome-wide screening.
Avoid false positives and false negatives As with any experiment, the use of appropriate controls, multiple
replicates and several cell types can strengthen your results. Furthermore, enrichment or depletion of multiple
gRNAs targeting the same gene can be strong evidence that a particular gene is actually important for a
given phenotype. Each hit from the screen should be independently validated to ensure that the desired
modification produces the phenotype you screened for in the first place.
With the proper experimental design and validation practices, CRISPR libraries can help you learn a lot about
your phenotype of interest.
69 | Page
REGULATING GENE EXPRESSION WITH CRISPR

By Tyler J. Ford and Joel McDade | January, 2016
In previous sections weve highlighted how CRISPR has made it easy for researchers to delete genes from or
edit the genome, but a variety of scientists have cleverly fused Cas9 to protein domains that allow them to do
much more than alter DNA sequence. In this section well cover how researchers have engineered CRISPR to
both repress and activate gene expression. See later sections for more great CRISPR applications!
Enabling New Functions with dCas9

The CRISPR/Cas system is a remarkably flexible
tool for genome manipulation. A unique feature of
Cas9 is its ability to bind target DNA independently
of its ability to cleave target DNA. Specifically, both
RuvC- and HNH- nuclease domains can be rendered
inactive by point mutations (D10A and H840A in
SpCas9), resulting in a nuclease dead Cas9 (dCas9)
molecule that cannot cleave target DNA. The dCas9
molecule retains the ability to bind to target DNA
Cas9 can be fused to an activator (A) or repressor (R) to alter expression of a
based on the gRNA targeting sequence. The first
gene whose promoter is targeted by the gRNA.
experiments using dCas9 in bacteria demonstrated
that targeting dCas9 to transcriptional start sites was sufficient to repress or knock-down transcription by
blocking transcription initiation (figure to the right) (1).
Qi et al. showed that targeting dCas9 to transcription factor binding sites in the promoter region of a gene or
to regions downstream of the transcription start site resulted in robust repression in E. coli with few off-target
effects. This is only effective, however, if dCas9 is targeted to the non-template strand and appears to function
by dCas9 physically blocking RNA polymerase. Qi et al. further showed that dCas9 alone can modestly repress
transcription in mammalian cells. Robust repression in mammalian cells and dCas9-based gene activation
requires further engineering.
Further Engineering dCas9 for Gene Repression and Activation

Although dCas9 alone can moderately repress gene expression in mammalian cells, these repressive effects
can be enhanced by fusing dCas9 to the KRAB repressor domain. Gilbert et al. showed that, when targeted to
promoter regions, dCas9-KRAB fusions effectively repress expression of a variety of genes in mammalian and
yeast cells and have few off target effects. Notably however, not all promoter-targeting gRNAs facilitate robust
repression; the rules governing dCas9-based repression in mammalian cells are not well understood yet and
any studies designed using these systems should use multiple gRNAs (2).
Unlike repression, dCas9-based transcription activation requires a dCas9 fusion. In E. coli this has
been accomplished by fusing dCas9 to the omega subunit of RNAP (3). In mammalian cells, a variety of
transcriptional activators have been fused to dCas9 including VP64 and p65. Similarly to the repressors,
the level of activation using these fusions is dependent upon the location targeted as well as the specific
gRNA used and the rules for effective activation are not yet well understood. It is again recommended that
researchers test multiple gRNAs when attempting to activate or repress gene expression using one of these
dCas9 fusions.
70 | Page
REGULATING GENE EXPRESSION WITH CRISPR (CONT)

Recruiting Multiple Transcriptional Activators Simultaneously
Scientists have recently developed a second
generation of activators that alter gene expression
by combining the effects of multiple transcriptional
activators. For instance, the SunTag system recruits
multiple activation domains to the same genetic locus
through co-expression of epitope tagged dCas9
and antibody-activator fusion proteins. Here, the
antibody fusion proteins bind to the epitope-tagged
Cas9 and activate gene expression at the promoter
targeted by a co-expressed gRNA (4). In a similar
vein, the Synergistic Activation Mediator (SAM)
complex consists of a dCas9-VP64 fusion and a
modified gRNA that is capable of interacting with an
additional RNA-binding transcriptional activator. Here
the modified gRNA has a dual role - it both directs
the entire complex to the promoter of interest and
recruits a transcriptional activator resulting in greater
activation of gene expression than with dCas9-VP64
alone (5). The Zhang Lab has developed a SAMbased pooled library that can be used to screen the
effects of activating gene expression genome-wide.
Finally, its possible to simply express Cas9 fused to
multiple transcriptional activators as in dCas9-VPR
from the Church lab (6).
These various methods of altering gene expression
should make it easier than ever for researchers to
study the function of their gene of choice without
knocking it out and should be especially useful to
scientists studying genes with essential functions.
You can find many plasmids with dCas9 based
transcription activators and repressors on the
Addgene website:
Find Activators Here
Some possible ways to use dCas9 to recruit multiple transcriptional

activators domains to a promoter of interest. A) dCas9 is fused to a series of
epitope tags each of which can recruit an activator fusion protein. B) Cas9 is
directly fused to a transcriptional activation domain while a modified gRNA
recruits an additional transcriptional activator. C) Cas9 itself is directly fused
to multiple activators.
Find Repressors Here
Further Reading
1. Qi, Lei S., et al. Repurposing CRISPR as an RNA-guided platform for sequence-specific control of gene
expression. Cell 152.5 (2013): 1173-1183. PubMed PMID: 23452860. PubMed Central PMCID: PMC3664290.
2. Gilbert, Luke A., et al. CRISPR-mediated modular RNA-guided regulation of transcription in
eukaryotes. Cell 154.2 (2013): 442-451. PubMed PMID: 23849981. PubMed Central PMCID: PMC3770145.
71 | Page
REGULATING GENE EXPRESSION WITH CRISPR (CONT)

3. Bikard, David, et al. Programmable repression and activation of bacterial gene expression using an
engineered CRISPR-Cas system. Nucleic Acids Research 41.15 (2013): 7429-7437. PubMed PMID: 23761437.
PubMed Central PMCID: PMC3753641.
4. Tanenbaum, Marvin E., et al. A protein-tagging system for signal amplification in gene expression
and fluorescence imaging. Cell 159.3 (2014): 635-646. PubMed PMID: 25307933. PubMed Central PMCID:
PMC4252608.
5. Konermann, Silvana, et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9
complex. Nature (2014). PubMed PMID: 25494202. PubMed Central PMCID: PMC4420636.
6. Chavez, Alejandro, et al. Highly efficient Cas9-mediated transcriptional programming. Nature Methods
72 | Page
TAGGING GENES WITH CRISPR

By Mary Gearing | December, 2015
First described in the 1980s, protein tags are now one of the most useful items in a scientists toolbox. As
weve covered in Plasmids 101, tags can help you determine localization of a protein of interest, purify it, or
determine its expression level without the need for a custom antibody. CRISPR has made it easier than ever to
tag endogenous proteins, allowing researchers to track how proteins bind to DNA or to other proteins.
The Basics of CRISPR Taggging

Addgene depositors Eric Mendenhall and Richard Myers have developed a method to easily insert FLAG tags
into the C-termini of endogenous loci. To tag your protein of interest, first design a seed sequence targeting
the locus youd like to tag, and then insert it into a plasmid containing Cas9 and the gRNA scaffold. Ideally the
gRNA should cut within -5 to +15 base pairs from the stop codon.
Second, create a repair template according to the guide below. Mendenhall and Myers recommend using IDT
gBlocks to specify the homology arms of the repair template. Plasmid pFETCh_Donor contains the 3X-FLAG
tag and a neomycin resistance gene; the homology arms can be cloned into the vector using Gibson Assembly.
Once you introduce the construct into cells, youll select using neomycin. Isolating clonal populations is not
necessary; Mendenhall and Myers use pooled neomycin-positive cells for experiments.
A schematic for FLAG tagging endogenous proteins using plasmid pFETCh_Donor. Left and right homology arms are cloned into the FLAG-P2A-NeoR
containing destination vector using Gibson Assembly. Homology-directed repair removes the stop codon and adds a 3X-FLAG tag and neomycin
resistance. NeoR is co-transcribed with the FLAG-tagged TF. The P2A self cleaving linker causes the ribosome to skip a peptide bond between the Flagtagged protein and the antibiotic resistance cassette resulting in two separate protein products.
73 | Page
TAGGING GENES WITH CRISPR (CONT)

Application: Multiplexed Transcription Factor Tagging
Mendenhall and Myers are members of the ENCyclopedia of DNA Elements Project (ENCODE), working to
define where the ~1,500 transcription factors of the human genome bind. Since fewer than 10% of antibodies
are suitable for the ChIP-seq analysis commonly used to map transcription factor binding, they created CETChseq (CRISPR epitope tagging ChIP-seq of DNA-binding proteins) to tag transcription factors and analyze their
binding in a scalable, global approach.
To test the universality of CETCh-seq, Savic et al. selected five DNA-binding proteins expressed at different
levels in HepG2 cultured cells. They designed gRNAs to target the 3 UTR and used an EMM0021-based repair
template to add a FLAG tag to the C-terminus of each TF. They screened for homologous recombination using
PCR and subsequently verified tag insertion via Western blotting and Sanger sequencing. From the five TFs
targeted in parallel, Savic et al. successfully tagged four TFs.
Savic et al. subsequently conducted CETCh-seq using a FLAG antibody. In cells that did not contain a tagged
TF, they did not observe binding events, showing that the method has low background. To validate their
results, they compared binding from CETCh-seq to datasets obtained using ChIP-seq with verified antibodies,
finding an average of ~85% overlap in binding sites between the datasets. Technical and biological CETChseq replicates were also highly concordant (rho=0.92-0.98), indicating the robustness and specificity of the
technique. Using RNA-seq, Savic et al. verified that TF tagging does not alter the transcriptome, further
strengthening the case for CETCh-seq as new method to profile TF binding.
After completing the initial experiments in HepG2 cells, Savic et al. turned to MCF7 cells to verify that CETChseq is robust in multiple cell types. For their targeted locus, RAD21, they again found good technical and
biological reproducibility, as well as concordance with validated ChIP results. CETCh-seq was also successful
in murine embryonic stem cells, opening up the possibility of generating transgenic mice through this tagging
approach!
Application: Affinity Purification Tagging for Protein Complex Isolation

Addgene depositor Yannick Doyon is interested in isolating native protein complexes to study biochemical
interactions. Dalvai et al. designed a system to add 3X-FLAG-2X-STREP tags to the N- or C-termini of
endogenous loci, enabling gentle, high-yield purification of protein complexes. Tagging endogenous loci gets
rid of many problems associated with protein overexpression, such as nonphysiological binding. The system
can also be used to insert a cDNA into the AAVS1 locus, a safe harbor locus that does not interrupt the
function of other genes.
To tag genes using the Doyon system, you need a gRNA that cleaves near the terminus of the locus youd
like to tag, as well as a repair template based on Addgene AAVS1_Puro_PGK1_3xFLAG_Twin_Strep. After
constructing the homology arms via PCR or with IDT gBlocks, sequentially digest the vector to insert the left
arm, then the right arm. A basic schematic is shown below, but further details are available in the supplemental
methods of Dalvai et al. If youd like to use this plasmid to insert a cDNA into AAVS1, just use the multiple
cloning site to insert the sequence downstream of the PGK1 promoter.
74 | Page
TAGGING GENES WITH CRISPR (CONT)
A schematic for 3X-FLAG-2X-STREP protein tagging. To use plasmid AAVS1_Puro_PGK1_3xFLAG_Twin_Strep for endogenous protein tagging, digest
with NdeI and NcoI to insert the left homology arm, then digest that construct with BstBI and EcoRI to insert the right arm. If inserting a tagged cDNA into
AAVS1, clone the cDNA using the MCS downstream of hPGK1.
Advantages and Tips

These plasmid systems promise to make endogenous protein tagging much easier and faster than ever before.
Using gBlocks speeds up cloning, and CRISPR greatly increases the frequency of homologous recombination.
Although these systems are FLAG- and STREP tag-based, they can be adapted to other tags, allowing the
tagging of multiple loci within a cell population.
One potential pitfall of CRISPR tagging is that the gRNA must bind close to the targeted terminus. If you cant
find a gRNA that works for your locus using SpCas9, it may make sense to try alternative Cas9s or Cpf1, which
have different PAM requirements.
Ready to start tagging? Plasmids for FLAG and FLAG-STREP tagging are available at Addgene.
Further Reading
1. Savic, Daniel, et al. CETCh-seq: CRISPR epitope tagging ChIP-seq of DNA-binding proteins. Genome
Research 25 (2015): 15811589. PubMed PMID: 26355004.
2. Dalvai, Mathieu, et al. A Scalable Genome-Editing Based Approach for Mapping Multiprotein
Complexes in Human Cells. Cell Reports 13 (2015): 621-633. PubMed PMID: 26456817.
75 | Page
USING CRISPR TO VISUALIZE THE GENOME

The Basics of CRISPR Taggging

CRISPR may be best known for genome editing and transcriptional activation, but its also helping researchers
visualize the genome and its organization within the nucleus, also called the 4D nucleome. Visualizing specific
loci has historically been difficult, as techniques like fluorescent in situ hybridization (FISH) and chromosome
capture suffer from low resolution and cant be used in vivo. Some researchers have used fluorescently tagged
DNA-binding proteins to label certain loci, but this approach is not scalable for every locus...unlike CRISPR!
Using dCas9 to Visualize Genomic Loci In Vivo

The flexibility of CRISPR has improved our ability to target almost every genomic locus, and to do so in vivo.
All you need is fluorescently labeled dCas9 and a gRNA specifying your locus of interest. This technique was
developed in the lab of Addgene depositor Stanley Qi and used in Chen et al. to track telomere dynamics
throughout the cell cycle. To optimize the signal to noise ratio of the system, they modified GFP-dCas9 and the
gRNA scaffold to enhance complex assembly, thereby decreasing the amount of background fluorescence from
unbound GFP-dCas9. With these improvements, labeling efficiency was similar to that of a comparable FISH
method. Importantly, they did not observe labeling when a) no gRNA was present or b) when a gRNA binding a
non-mammalian sequence (GAL4) was supplied.
In addition to labeling repetitive telomeric sequences, Chen et al. successfully labeled protein-coding genes
with both intronic and exonic gRNAs. In fact, the method is specific and sensitive enough to detect gene copy
number based on the number of fluorescent puncta observed. By labeling two genes simultaneously, they
could also monitor the spatial relationship of the two genes over time.
Labeling Multiple, Intrachromosomal Loci

Building on the work of Chen et al., Thoru Pedersons lab has
used CRISPR to label multiple loci in distinct colors. To create
a colorful Cas9 toolbox, Ma et al. turned to SpCas9 and its
orthologs NmCas9 and St1Cas9. Each ortholog was fused to
a different fluorescent protein to create three distinct colors.
The specificity of these orthologs is key: since each ortholog
requires a different PAM sequence, a gRNA designed for one
dCas9 should be specific to that ortholog and not cross-talk
with the other orthologs.
Ma et al. tested their dCas9 variants using gRNAs specific for
telomeric sequences and showed that different fluorescently
labeled dCas9s are efficiently directed to the proper target
sequence. They succeeded in labeling two different pairs
of chromosomes using gRNAs specific to sequences on
chromosomes 9 and 13. They next turned their attention
to mapping pairs of intrachromosomal loci. The technique
successfully resolved loci with physical map distances of
2 and 75 Mbp, with the calculated fluorescent distances
Live cell dual-color CRISPR labeling of two loci on human

chromosome 9
76 | Page
USING CRISPR TO VISUALIZE THE GENOME (CONT)

correlating with the previously established physical map. In comparing pairs of targets ~2 Mbp apart, they
noticed that they could evaluate the degree of chromatin compaction even for this small distance! To the
authors knowledge, this work represents the first mapping of intrachromosomal loci, a major benchmark in
characterizing the 4D nucleome.
In Vitro Application: Genome Structure

In addition to 4D nucleome applications, CRISPR visualization is also useful for studying genome structure.
Shotgun sequencing breaks DNA up into small segments, which are then assembled based on overlapping
sequence. However, in areas with repetitive or duplicated sequences, the sequences may be assembled
incorrectly. Optical mapping can create restriction maps of the genome, better capturing structural variation,
but areas that do not contain restriction sites cannot be mapped.
McCaffrey et al. adapted CRISPR for whole genome mapping by designing gRNAs to specifically target
repetitive sequences. In their technique, DNA strands are incubated with Cas9 nickase, a gRNA, Taq
polymerase and fluorescently labeled nucleotides. At the site of the nick, the polymerase incorporates the
labeled nucleotides. Labeled DNA is subsequently imaged in a nanochannel. When testing their method using a
bacterial artificial chromosome, McCaffrey et al. found a very low false positive rate of 0.6%, indicating limited
labeling in the BAC region not containing the gRNA-specified sequence. The technique can also be used with
traditional sequence motif labeling, which recognizes various restriction sites, to map the location of gRNAspecified repeats.
Using gRNAs binding Alu sequences, which account for up to 10% of the human genome, McCaffrey et al.
created barcode-like patterns that differ among DNA sequences. This technique may become especially useful
for loci of clinical significance, such as the polyglutamine tracts in Huntingtons disease, where certain variants
may prove pathogenic. Importantly, this advance makes genome mapping more flexible and high-throughput,
for both targeted and whole genome approaches.
CRISPR labeling creates information-rich barcodes. Two pieces of human genomic DNA were labeled using an Alu sequence-specific gRNA (red) as well
as traditional nick-labeling with the restriction enzyme Nt.BspQI (blue.) Whereas the Nt.BspQI recognition site is found infrequently in genomic DNA, the
Alu sequences are highly abundant, permitting the generation of genome-specific barcodes. Figure from McCaffrey et al., 2015. Used under a CC-BY-NC
4.0 license.
77 | Page
USING CRISPR TO VISUALIZE THE GENOME (CONT)

Further Reading
1. Chen, Baohui, et al. Dynamic Imaging of Genomic Loci in Living Human Cells by an Optimized
CRISPR/Cas System. Cell 155 (2013): 1379-1491. PubMed PMID: 24360272. PubMed Central PMCID:
PMC3918502.
2. Ma, Hanhui, et al. Multicolor CRISPR labeling of chromosomal loci in human cells. Proc. Natl. Acad.
Sci. USA 112(10) (2015): 3002-7. PubMed PMID: 25713381. PubMed Central PMCID: PMC4364232.
3. McCaffrey, Jennifer, et al. CRISPR-CAS9 D10A nickase target-specific fluorescent labeling of double
strand DNA for whole genome mapping and structural variation analysis. Nucleic Acids Research (2015) pii:
gkv878. PubMed PMID: 26481349.
78 | Page
CRISPR PURIFCATION
Chromatin immunoprecipitation (ChIP) has helped

revolutionize our understanding of chromatin binding
partners. Moving forward, researchers would like to more
broadly understand which proteins and RNA bind specific
genomic loci, and how these binding partners change over
time or in disease states. Although methods such as iChIP
and PICh have been used previously, the CRISPR-based
method enChIP is advantageous due to its simplicity and
multiplexing potential.
Isolating a specific genomic locus is difficult, so researchers
have had to get a little creative. iChIP inserts a recognition
site for a DNA-binding protein, such as LexA, next to the
locus of interest, allowing one to perform ChIP using a
specific antibody. iChIP works well in cell lines, but its use
in model organisms requires making a transgenic organism.
Another technique, PICh, uses a complementary nucleic acid
probe to bind a locus, but the approach has only been used
for high copy loci.
To solve these issues, Addgene Advisory Board member
Hodaka Fujiis lab has created enChIP (engineered DNAbinding molecule-mediated chromatin immunoprecipitation).
As you may have guessed, enChIP uses a gRNA to specify a
given locus, directing FLAG-tagged dCas9 to the appropriate
location. After crosslinking, the sample is immunoprecipitated
using a FLAG antibody, then reverse crosslinked. DNA,
RNA, and proteins bound to the locus can then be identified
through a combination of next-generation sequencing, RNAseq, and mass spectrometry (see figure to the right).
Overview of the enChIP process
To validate the specificity of gRNA targeting and the IP, you can compare the abundance of your locus of
interest in the input and IP DNA using qRT-PCR. Ideally, that locus should be enriched in the IP sample, but
predicted off-target (or randomly selected) loci should not. One way to minimize the effects of off-target
binding is to compare two treatments (e.g. +/- insulin). Off-target binding will likely be seen in both conditions,
eliminating potential false positives. Conducting enChIP with different gRNAs also minimizes the effects of offtarget binding; a molecule identified across multiple experiments is likely a true positive. Fujii has found that offtarget binding decreases when the mismatches are closer to the PAM site, an important consideration in gRNA
design.
Application: Retroviral enChIP System

Fujita and Fujii created a retroviral enChIP system for use in difficult-to-transfect cell lines, as well as primary
cells. Targeting the IRF-1 promoter region, they found that about 10% of input DNA was immunoprecipitated
using the viral enChIP system. Since this promoter is stimulated by IFN, they wanted to see how protein
binding changed with IFN stimulation. To compare the two conditions, they combined enChIP with SILAC
79 | Page
CRISPR PURIFCATION (CONT)

(Stable Isotope Labeling using Amino acids in Cell culture). In SILAC, cells representing one condition are
grown in Light media; cells of the other condition are grown in Heavy media, which contains amino
acids incorporating heavy isotopes. Heavy isotope labeling causes protein mass shifts detectable by mass
spectrometry, so protein abundance in the two conditions can be directly compared.
Using SILAC and enChIP, Fujita and Fujii found many classes of proteins that bound the stimulated IRF-1
promoter, including histone deacetylases previously thought to play a role in promoter activation. To confirm
these interactions, they performed ChIP using antibodies against the proteins found using enChIP. This study
shows that enChIP can shed a great deal of light on transcriptional mechanisms. Fujii also sees potential
applications of enChIP to tough genomic questions like epigenetic regulation, genomic imprinting, and X
chromosome inactivation.
Important Considerations for enChIP gRNA Design

To help you better use enChIP, Fujita and Fujii have created a short list of points to consider when designing
your gRNAs. For a more detailed guide, check out this review.
gRNAs for use with enChIP shouldnt bind to conserved genomic regions. These regions often contain
functional regulatory sequences, including recognition sites for DNA-binding molecules.
When analyzing a promoter: The gRNA should bind several hundred base pairs upstream of the
transcription start site. This location avoids interference with the recruitment of transcription factors and
polymerases. Design gRNAs to bind 100-300 bases from the transciption start site and test to see if gene
expression remains constant (via qRT-PCR.) If gene expression decreases, try designing gRNAs targeting
other areas of the promoter sequence.
When analyzing an enhancer/silencer: gRNAs can be designed to bind directly adjacent to the region of
interest.
enChIP plasmids are available from Addgene. If youd like to learn more about the work in Hodaka Fujiis lab,
check Addgenes interview with him!
Further Reading
1. Fujita, Toshitsugu, and Hodaka Fujii. Efficient isolation of specific genomic regions and identification of
associated proteins by engineered DNA-binding molecule-mediated chromatin immunoprecipitation (enChIP)
using CRISPR. Biochem Biophys Res Commun. 439(1) (2013):132-6. PubMed PMID: 23942116.
2. Fujita, Toshitsugu, and Hodaka Fujii. Identification of Proteins Associated with an IFN-Responsive
Promoter by a Retroviral Expression System for enChIP Using CRISPR. PLoS One 9(7) (2014) :e103084.
3. Fujii, Hodaka, and Toshitsugu Fujita. Isolation of Specific Genomic Regions and Identification of
Their Associated Molecules by Engineered DNA-Binding Molecule-Mediated Chromatin Immunoprecipitation
(enChIP) Using the CRISPR System and TAL Proteins. Int J Mol Sci. 16(9) (2015):21802-12. PubMed PMID:
26370991. PubMed Central PMCID: PMC4613281.
4. Fujita, Toshitsugu, and Hodaka Fujii. Efficient isolation of specific genomic regions retaining molecular
interactions by the iChIP system using recombinant exogenous DNA-binding proteins. BMC Mol Biol. 15
(2014):26. PMID: 25428274. PubMed Central PMCID: 4253623.
80 | Page
CRISPR PURIFCATION (CONT)

5. Dejardin, Jerome, and Robert E. Kingston. Purification of proteins associated with specific genomic
loci. Cell 136(1) (2009):175-186. PubMed PMID: 19135898. PubMed Central PMCID: PMC3395431.
81 | Page
Chapter 4 - CRISPR Expression Systems and Delivery Methods
CRISPR EXPRESSION SYSTEMS AND DELIVERY METHODS
82 | Page
MULTIPLEXIBLE CRISPR EXPRESSION SYSTEMS

CRISPR makes it easy to target multiple loci - a concept called multiplexing. Since CRISPR is such a robust
system, editing or labeling efficiency doesnt usually change when you add multiple gRNAs. Sound good?
Addgene has many tools to help you multiplex - well use mammalian plasmids to introduce you to some
of your potential options and cloning methods, but please scroll down for plasmids suitable for other model
systems, including E. coli, plants, Drosophila, and zebrafish!
One common question Addgene Senior Scientists receive is: can I express more than one gRNA from a single
promoter using a plasmid like pX330? Unfortunately, the short answer is no. Unless you use a system for
processing a continuous multi-gRNA transcript, each gRNA must be expressed from its own promoter. But that
doesnt mean you have to clone and transfect multiple promoter-gRNA constructs in order to target multiple
sites...
Figure 1: Multiplexing allows researchers to express multiple gRNAs from a single construct. DNA and gRNA
sizes are not to scale.
Multiplexing Basics
Lets start with the simplest multiplexing situation: you only need to express two gRNAs at the same time. One
system you could use is pX333 from the Ventura lab. pX333, a modification of pX330, contains humanized
wtCas9 and two U6 promoters. To use this plasmid, you simply order oligonucleotides for your chosen gRNA
target sequences and clone them in just as you would for a single gRNA. Youll clone in the first gRNA using
restriction enzyme BbsI and the second gRNA using restriction enzyme BsaI. If youre working in Drosophila,
a two-gRNA expressing plasmid is available from the Bullock lab, and gRNAs can be inserted using Gibson
Assembly or SLIC cloning methods. A BsaI-based E. coli multiplexing plasmid is available from the Koffas lab.
Golden Gate Assembly Methods

If you want to scale up to 3-7 gRNAs, Addgene has two Golden Gate Assembly-based mammalian systems
from the Gersbach and Yamamoto labs. Golden Gate assembly methods are also available for plants (Qi-Jun
Chen and Yao-Guang Liu labs) and zebrafish (Wenbiao Chen lab).
Golden Gate Assembly uses Type IIS restriction enzymes, which cleave outside of their recognition sequence,
creating flanking overhangs. These overhangs can be customized to link together multiple fragments, allowing
ordered assembly of multiple components into a destination vector.
The first step in CRISPR/Cas9 Golden Gate multiplexing is to clone the oligonucleotides specifying each gRNA
target sequence into distinct expression vectors using the enzyme BbsI. These expression vectors each contain
Type IIS restriction sites flanking the promoter-gRNA construct, but with different sequences adjacent to the
sites. When digested with the appropriate Type IIS enzyme, the unique flanking overhang sequences can link
83 | Page
MULTIPLEXIBLE CRISPR EXPRESSION SYSTEMS (CONT)

together to allow for ordered assembly into a destination vector that expresses Cas9. If youre confused, check
out the schematic below.
Figure 2: gRNA target sequences (colored rectangles) are cloned into various plasmids using oligonucleotides. These plasmids contain
Type IIS restriction sites that flank the promoter-gRNA constructs. When these plasmids are digested, unique overhangs (here, O1-4)
adjacent to the cut sites link fragments together and drive ordered assembly into a Cas9-containing destination vector. Note: depending
on which method you use, the procedure will vary slightly.
Gersbach Lab Multiplexing Plasmids:

This plasmid set allows you to express 2-4 gRNAs, with four being the ideal number. First you generate four
unique kanamycin-resistant plasmids, each containing a different gRNA target sequence downstream of the
7SK, human U6, mouse U6, or human H1 promoters. If you express fewer than four gRNAs, youll clone in a
polyT-termination sequence for each unused promoter. This step is necessary to generate all of the overhangs
needed for the final ligation step. Plasmids are then digested using BsmBI and ligated into one of four Cas9
or dCas9-containing ampicillin-resistant destination vectors. Destination vector options include humanized
wt Cas9, dCas9 (transcriptional repressor), and dCas9-VP64 (transcriptional activator)-containing plasmids. If
youre interested in combining CRISPR/Cas9 with FACS-based screening, these plasmids are for you. Each
destination vector contains GFP, enabling you to select cells with high GFP expression and therefore high
multiplicity of infection (MOI). These high MOI cells have the highest levels of Cas9 and gRNA expression, and
thus the highest frequency of genome editing events.
Yamamoto Lab Multiplex CRISPR/Cas9 Assembly Kit:
This kit is built for serious multiplexing and enables users to express up to 7 gRNAs! The kit contains different
destination vectors depending on the total number of gRNAs you wish to clone, from 2-7. For example, if youre
expressing 4 gRNAs, youd use pX330A-1x4; for 6 gRNAs, youd use pX330A-1x6. This customization means
you dont ever need to clone in filler sequences like with the Gersbach plasmids; the correct flanking regions
are specified by the destination vector. To build your multiplexing construct, you clone all but one of your
gRNAs into spectinomycin-resistant plasmids pX330S-2 to pX330S-(last gRNA number). The 5 most gRNA is
84 | Page

cloned into the Cas9-containing, ampillicin-resistant destination vector. These constructs are digested using
BsaI and assembled to produce a plasmid encoding the gRNAs, Cas9, and ampicillin resistance. As with the
Gersbach lab plasmids, multiple Cas9 variants are available. The standard kit includes wt humanized Cas9
and the D10A nickase mutant (Cas9n). The accessory pack adds two more options: dCas9 (transcriptional
repression) and Fok1-dCas9 (dimeric nuclease.)
Gateway Assembly Method

Frew Lab Multiple Lentiviral Expression Systems (MuLE) Kit:
This kit can be used to create lentiviral vectors expressing wt humanized Cas9 and up to three gRNAs.
Entry vectors containing the U6 promoter and the gRNA scaffold are provided with the kit. Oligonucleotides
specifying the gRNA seed sequence should be compatible with type IIS enzyme BfuAI. Gateway cloning is then
used to combine the multiple gRNAs and Cas9 together into a single plasmid. Although only wt hCas9 entry
vectors are supplied with the kit, you can clone your own entry vectors containing other Cas9 variants to use
with the MuLE system.
Multiplexing from a Sinlge Transcript

You can also multiplex gRNAs via a polycistronic transcript. Rather than being transcribed from different
promoters, the gRNAs are transcribed together and are flanked by specific sites that allow them to be cleaved
and released. These constructs tend to be smaller than constructs with multiple promoter-gRNA cassettes,
making them advantageous for small capacity vectors like AAV. In addition to the mammalian option described
below, plasmids for making polycistronic gRNAs are also available from the Yang lab for use in plants.
The mammalian multiplex systems use the Csy4 RNA nuclease from Pseudomonas aeruginosa. When
overexpressed, Csy4 efficiently cleaves gRNAs sandwiched between 28 base Csy4 recognition sites. If
85 | Page

Csy4 is not expressed, the gRNAs cannot be released, adding temporal and/or spatial control to the system.
pSQT1313 from the Joung lab allows you to express two gRNAs constructed using oligonucleotide assembly.
Unlike some of the plasmids described above, this vector do not contain Cas9, so youll need to supply it with
another plasmid.
Multiplexing in Plants
Qi-Jun Chen Lab Golden Gate/Gibson Assembly Multiplexing Plasmids:
These plasmids allow you to assemble 2-4 gRNAs through Golden Gate or Gibson
Assembly. gRNAs are inserted into the pCBC vectors using BsaI, and promoter-gRNA
fragments are PCR amplified for cloning into one of three Zea mays codon-optimized
Cas9-containing binary vectors. These vectors are based off of pGreen, pSoup, and
pCAMBIA, three of the most popular plant vectors that together are suitable for a wide
Marie-Lan Nguyen /
Wikimedia Commons / CCBY 2.5
variety of applications. The Chen lab system is compatible with both monocot and dicot plants.
Liu Lab Golden Gate/Gibson Assembly Multiplexing Plasmids:
These plasmids can be used to successfully express up to 8 gRNAs after Golden Gate or Gibson Assembly!
Using BsaI, gRNAs are cloned into one of 12 pYLsGRNA plasmids, which contain various promoters and
reporters, and subsequently inserted into a Cas9-containing destination vector based off of pCAMBIA. wt Cas9
is plant-optimized with a high 5 GC-content, and plasmids are available for both monocot and dicot plants,
with a choice of either hygromycin or Basta selection.
Yang Lab Single Transcript Multiplexing Plasmids:
These plasmids are similar to the Csy4 polycistronic system described above, except that they use an
endogenous nuclease system to cleave the gRNAs. gRNAs are flanked by glycine tRNAs to create polycistronic
glycine tRNA-gRNA (PTG) constructs. Eukaryotic RNases P and Z recognize the tRNA sequences, cleave
them, and release the gRNAs. PTGs are assembled into a wt Cas9-containing vector using Golden Gate
assembly, and up to 8 gRNAs may be expressed simultaneously. Vectors for both transient expression and
Agrobacterium-mediated transformation are available. For more information on this system, check out this blog
post.
Multiplexing in Zebrafish
Wenbiao Chen Lab Golden Gate Assembly Multiplex Plasmids:
Attribution: PogrebThese plasmids allow expression of 2-5 gRNAs in zebrafish. As in the Yamamoto system, Image
noj-Alexandroff / Wikimecustom destination vectors are used depending upon the total number of gRNAs you
dia Commons / CC-BY 3.0
wish to clone, so you dont have to clone any filler sequences. You also have the option
of including a previously-designed tyr gRNA, which causes hypopigmentation, thus marking cells that have
undergone genomic modification. In this system, Cas9 must be supplied on a separate plasmid.
86 | Page

Multiplexing in Drosophila
Bullock Lab Multiplex Plasmid:
Two gRNAs can be assembled using Gibson Assembly or SLIC cloning. gRNAs are
expressed from two Drosophila U6 promoters. Cas9 must be supplied on a separate
plasmid.
Image Attribution: Aka /

Wikimedia Commons / CCBY-SA 2.5
Multiplexing in E. coli
Koffas Lab CRISPathBrick Multiplex Plasmid:
This system allows you to assemble type II-A CRISPR arrays for dCas9-based transcriptional repression. The
CRISPathBrick plasmid contains a nontargeting spacer flanked by two CRISPR repeats. The spacer can be
digesting using BsaI, allowing a spacer-repeat brick to be inserted. The BsaI site remains intact, allowing
subsequent bricks to be added one by one. This approach is especially useful for combinatorial analyses. For
example, if you were to develop an array using 3 distinct spacer-repeats (more are possible), you could easily
create 7 unique arrays (e.g. for spacers A, B, and C, you could obtain arrays A, B, C, AB, AC, BC, and ABC).
Check out a curated, up-to-date list of our gRNA multiplexing vectors on our gRNA page!
Further Reading
1. Maddalo, Danilo, et al. In vivo engineering of oncogenic chromosomal rearrangements with the
CRISPR/Cas9 system. Nature 516(7531) (2014): 423-7. PubMed PMID: 25337876. PubMed Central PMCID:
PMC4270925.
2. Kabadi, Ami M., et al. Multiplex CRISPR/Cas9-based genome engineering from a single lentiviral
vector. Nucleic Acids Research 42(19) (2014): e147. PubMed PMID: 25122746. PubMed Central PMCID:
PMC4231726.
3. Sakuma, Tetsushi, et al. Multiplex genome engineering in human cells using all-in-one CRISPR/
Cas9 vector system. Scientific Reports 4 (2014): 5400. PubMed PMID: 24954249. PubMed Central PMCID:
PMC4066266.
4. Albers, Joachim, et al. A versatile modular vector system for rapid combinatorial mammalian genetics.
Journal of Clinical Investigation 125(4) (2015):1603-19. PubMed PMID: 25751063. PubMed Central PMCID:
PMC4396471.
5. Tsai, Shengdar Q., et al. Dimeric CRISPR RNA-guided FokI nucleases for highly specific genome
editing. Nature Biotechnology 32(6) (2014): 569-576. PubMed PMID: 24770325. PubMed Central PMCID:
PMC4090141.
6. Xing, Hui-Li, et al. A CRISPR/Cas9 toolkit for multiplex genome editing in plants. BMC Plant Biology
14 (2014): 327. PubMed PMID: 25432517. PubMed Central PMCID: PMC4262988.
7. Ma, Xingliang, et al. A robust CRISPR/Cas9 system for convenient, high-efficiency multiplex genome
editing in monocot and dicot plants. Molecular Plant 8(8) (2015): 1274-1284. PubMed PMID: 25917172.
8. Xie, Kabin, Minkenberg, Bastian, & Yinong Yang. Boosting CRISPR/Cas9 multiplex editing capability
with the endogenous tRNA-processing system. Proceedings of the National Academy of Sciences USA
112(11) (2015): 3570-5. PubMed PMID: 25733849. PubMed Central PMCID: PMC4371917.
9. Yin, Linlin, et al. Multiplex Conditional Mutagenesis Using Transgenic Expression of Cas9 and sgRNAs.
87 | Page

Genetics 200(2) (2015): 431-441. PubMed PMID: 25855067.
10. Port, Fillip, et al. Optimized CRISPR/Cas tools for efficient germline and somatic genome engineering
in Drosophila. Proceedings of the National Academy of Sciences USA 111(29) (2014): E2967-2976. PubMed
11. Cress, Brady F., et al. CRISPathBrick: Modular Combinatorial Assembly of Type II-A CRISPR Arrays for
dCas9-Mediated Multiplex Transcriptional Repression in E. coli. ACS Synthetic Biology 4(9) (2015):987-1000.
88 | Page
Chapter 4 - CRISPR Expression and Delivery Methods
MAMMALIAN EXPRESSION SYSTEMS AND DELIVERY METHODS

By Nicole Waxmonsky | September 24, 2015
CRISPR technology has been widely adopted for genome editing purposes for numerous reasons including that
its cheaper, faster, and easier than prior editing techniques. With more and more CRISPR tools being published
each month, you may be considering using CRISPR for your next experiment. In this section well provide an
overview of some CRISPR mammalian expression systems, the typical applications for each, and potential
delivery methods.
As with any experiment, there are many factors that
need to be considered during the planning process.
For CRISPR experiments, the following framework can
help get you started:
1.
Determine the type of outcome you are trying
to achieve: Do you want to permanently knock-out or
knock-in a gene? Do you want to enhance or repress
gene expression? Are you trying to create a single
point mutation? Do you want to create a fusion with a
reporter protein such as GFP? All of these outcomes
can be most effectively achieved with different CRISPR
components.
2.
Select the appropriate CRISPR tools for your
application: Wildtype Cas9 or the Cas9 nickase
are appropriate for knocking-in, knocking-out, or
introducing mutations and tags, while a dead or dCas9 can be used in conjunction with activator or repressor
domains to control gene expression.
3.
Choose an appropriate expression system and delivery method: Do you need stable integration or is
transient expression sufficient? Which cell types will you be editing? Do you want to deliver the components as
DNA or would mRNA or protein delivery be more suitable?
4.
Determine how you will evaluate the outcome: Will you be detecting insertions/deletions using a
mismatch repair assay? Or is PCR followed by gel electrophoresis or Next Generation Sequencing more
appropriate?
If you already have an end product in mind, steps 1 and 2 will generally be straightforward. Likewise for step 4,
as this ties directly back to the specific application you have chosen. When thinking about step 3, however, you
may be surprised at the number of options available--how do you choose?
One of the first steps is to identify what CRISPR components you will need to deliver. Minimally, one or more
sgRNAs and Cas9 are required for any application. If you want to include a homology directed repair (HDR)
template to create knock-ins, point mutations, or to add a tag, you will also need to deliver a donor plasmid
or single-stranded DNA oligonucleotide, so you will need to make sure your expression system and delivery
methods are compatible with all your components. Next, consider the best form the CRISPR components
should be in based on your model system. CRISPR reagents can be delivered via transfection, nucleofection,
viral infection, or injection as either protein, RNA, or DNA. Finally, once you have identified the best expression
system, you can then choose the best method for introducing the CRISPR components into your target cells.
89 | Page
MAMMALIAN EXPRESSION SYSTEMS AND DELIVERY METHODS (CONT)

Mammalian CRISPR Expression Systems
Each model system will have its best practices for efficient delivery of CRISPR components. If you are new to
your model system, a good first step would be to consult the literature to see if anyone has published work with
a protocol that would work for your system. Addgene has depositor submitted protocols and links to a CRISPR
forum where you may be able to find information regarding your system of choice. The table below summarizes
the various components included with each expression system as well as suitable applications.
Expression
System
Application
Mammalian
expression
vector
Promoter driving Cas9 expression can be

constitutive or inducible. U6 promoter is typically
used for gRNA. May contain reporter gene (e.g.
GFP) to identify and enrich positive cells or
selection marker to generate stable cell lines.
Transient or stable expression

of Cas9 and/or gRNA in a
mammalian cell line that can be
transfected at high efficiency.
Lentiviral
transduction
Cas9 and gRNA can be present in a single

lentiviral transfer vector or separate transfer
vectors. May contain reporter gene (e.g. GFP)
to identify and enrich positive cells. Packaging
and Envelope plasmids provide the necessary
components to make lentiviral particles.
Stable, tunable expression of

Cas9 and/or gRNA in a wide
variety of mammalian cell lines.
Useful for difficult to transfect
cell types and can be used
in vivo. A common choice
for conducting genome-wide
screens using CRISPR/Cas9.
AAV transduction Only compatible with SaCas9 (packaging limit
Transient or stable expression

of SaCas9 and/or gRNA. Infects
dividing and non-dividing cells.
AAV is least toxic method for in
vivo viral delivery.
Cas9 mRNA
and gRNA
Plasmids containing gRNA and Cas9 are used in

in vitro transcription reactions to generate mature
Cas9 mRNA and gRNA, then delivered to target
cells (e.g. microinjection or electroporation).
Transient expression of CRISPR

components, expression
decreases as RNA is degraded
within the cell. Can be used for
generating transgenic embryos.
Cas9-gRNA
riboprotein
complexes
Purified Cas9 protein and in vitro transcribed

gRNA are combined to form a Cas9-gRNA
complex and delivered to cells using cationic
lipids.
Transient expression of CRISPR

components, expression
decreases as gRNA and Cas9
protein are degraded within the
cell.
~4.5kb). CRISPR elements are inserted into an

AAV transfer vector and used to generate AAV
particles.
90 | Page
MAMMALIAN EXPRESSION SYSTEMS AND DELIVERY METHODS (CONT)

Delivery Methods for Mammalian Cell Lines
As mentioned above, the expression system you choose in many ways dictates the best method for introducing
CRISPR components into your target cells. DNA delivery into mammalian cell lines is quite broad and
includes several different methods, so weve further broken down this category by common cell types and
recommended delivery methods below:
Method
Transfection

Lipid-Mediated

Cationic polymers

Calcium Phosphate
Electroporation

Nucleofection
Viral Delivery

Lentivirus

Retrovirus

Adenovirus

AAV
Transformed cell lines

(HeLa, HEK 293)
Stem cells (hES,

iPS)
Primary cells
(fibroblasts,
epithelial cells)
+
+
This table is not inclusive of all methods, nor are these methods limited exclusively to in vitro cell culture. The
user should review the current literature about their preferred model.
If your expression system is not well characterized in terms of CRISPR use, you will want to invest some time in
optimizing and testing the efficiency of CRISPR delivery. There are a few plasmids at Addgene that have been
published as CRISPR testing tools:
Traffic Light Reporter System: Can evaluate lentiviral component delivery as well as genome repair by nonhomologous end joining (NHEJ) or HDR.
EGFP validation of sgRNAs: Can evaluate component delivery and sgRNA efficacy by cloning in genome target
sequence into EGFP reporter.
Target DNA reporter system: Can evaluate component delivery with validated tools.
Special thanks to Joel McDade for creating the Expression Systems table. Marcy Patrick Contributed to the
writing of this section.
91 | Page
USING SaCAS9 FOR EASIER AAV-BASED CRISPR DELIVERY

By Mary Gearing | July 14, 2015
CRISPR/Cas9 genome editing has quickly become the

most popular system for in vitro and germline genome
editing, but in vivo gene editing approaches have
been limited by problems with Cas9 delivery. Adenoassociated viral vectors (AAV) are commonly used for
in vivo gene delivery due to their low immunogenicity
and range of serotypes allowing preferential infection
of certain tissues. However, packaging Streptococcus
pyogenes (SpCas9) and a chimeric sgRNA together
(~4.2 kb) into an AAV vector is challenging due to the
low packaging capacity of AAV (~4.5 kb.) While this
approach has been proven feasible, it leaves little room
for additional regulatory elements. Feng Zhangs group
previously packaged Cas9 and multiple gRNAs into
separate AAV vectors, increasing overall packaging
capacity but necessitating purification and co-infection
of two AAVs.
Cas9 Orthologs: Shorter, but Just as Potent and Specific?

The previous two AAV strategies described above showed successful target modification, indicating that AAV
is a good delivery vehicle for Cas9. To maximize the genetic capacity of AAV, Gang Baos group has developed
a split-intein Cas9 that can be separated into two AAV cassettes, providing even more room for regulatory
sequences and additional gRNAs in each cassette. However, to fit Cas9 and gRNAs into one AAV construct,
the construct must be made even smaller. Previous attempts to shrink Cas9 include the use of St1Cas9
(~3.3 kb) from Streptococcus thermophilus and a rationally-designed truncated Cas9. Unfortunately, certain
drawbacks limit the utility of these systems: St1Cas9 requires a very specific PAM sequence that limits the
number of targetable loci, and truncated Cas9 has much lower efficiency than its wild-type counterpart.
Ran et al. recently developed a new strategy to overcome these drawbacks. To discover a shorter, but equally
potent Cas9 enzyme, they analyzed over 600 Cas9 orthologs and found that they could be divided into two
groups: one with orthologs of ~1350 amino acids, which includes SpCas9, and one with orthologs of ~1000
amino acids. From the pool of shorter orthologs, only Staphylococcus aureus Cas9 (SaCas9) displayed
cleavage activity in mammalian cells. SaCas9 produced indels at a similar efficiency to SpCas9, leading the
group to focus their efforts on SaCas9 characterization for in vivo studies.
One of the pitfalls of CRISPR/Cas9 genome editing is the potential for off-target effects. To compare the offtarget effects of SpCas9 and SaCas9, Ran et al. used an approach called BLESS (direct in situ breaks labelling,
enrichment on streptavidin and next-generation sequencing). Using this sensitive method, Ran et. al found
that SaCas9 did not display higher levels of off-target activity than SpCas9, confirming its suitability for in vivo
studies.
92 | Page
USING SaCAS9 FOR EASIER AAV-BASED CRISPR DELIVERY (CONT)
Testing AAV-SaCas9 In Vivo

To test the efficiency of AAV-SaCas9 in vivo, Ran et al. created an all-in-one SaCas9 and sgRNA construct
using the liver-specific serotype AAV8. Since the efficiency of CRISPR/Cas9 genome editing varies across
targets, they tested two genes in mice. For both genes, they observed indel formation and phenotypic changes
as early as 1 week post-injection. Livers from these mice were histologically normal and liver injury markers
were not increased compared to a control AAV-GFP. Not only did AAV-SaCas9-sgRNA constructs mediate
genome modification, but they did so without a substantial immune response or toxicity.
The work of Zhangs group illustrates the potential of combining an advantageous vector delivery system (AAV)
with a potent genome modification technique (CRISPR/Cas9). In this best of both worlds scenario, in vivo
genome editing without substantial toxicity or off-target effects will likely become much easier than we could
have imagined.
If youre interested in using SaCas9 in your research, the AAV targeting constructs are available from Addgene.
Further Reading
1.
Ran, F. Ann, et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520.7546 (2015):
2.
Swiech, Lukasz, et al. In vivo interrogation of gene function in the mammalian brain using CRISPRCas9. Nature Biotechnology 33.1 (2015): 102-106. PubMed PMID: 25326897. PubMed Central PMCID:
PMC4492112.
3. Cong, Le, et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339.6121 (2013):
4. Fine, Eli J., et al. Trans-spliced Cas9 allows cleavage of HBB and CCR5 genes in human cells using
compact expression cassettes. Scientific Reports 5 (2015). PubMed PMID: 26126518. PubMed Central
PMCID: PMC4486982.
93 | Page
USING SaCAS9 FOR EASIER AAV-BASED CRISPR DELIVERY (CONT)

5. Sens, Elena, et al. CRISPR/Cas9mediated genome engineering: An adenoassociated viral (AAV)
vector toolbox. Biotechnology Journal 9.11 (2014): 1402-1412. PubMed PMID: 25186301.
6. Garneau, Josiane E., et al. The CRISPR/Cas bacterial immune system cleaves bacteriophage and
plasmid DNA. Nature 468.7320 (2010): 67-71. PubMed PMID: 21048762.
7. Nishimasu, Hiroshi, et al. Crystal structure of Cas9 in complex with guide RNA and target DNA. Cell
94 | Page
ADENOVIRAL DELIVERY OF CRISPR/CAS9 AIMS TO EXPAND GENOME EDITING TO PRIMARY

CELLS
By Kendall Morgan | September 30, 2014
Researchers have shown that it is possible to deliver

RNA-guided CRISPR/Cas9 nuclease complexes
using adenoviral vectors (AdVs), to a wide range of
human cells, including mesenchymal stem cells, and
in a rather straightforward manner. These adenoviral
CRISPR/Cas9 genome editing tools developed
and demonstrated by Manuel Gonalves and his
colleagues at Leiden University Medical Center are
now available at Addgene along with a description of
their experimental protocol. The three plasmids which
have been deposited to Addgene are: pAdSh.PGK.
Cas9, pAdSh.U6.gRNAS1, pAdSh.U6.gRNAGFP.
Although AdVs are being deployed for delivering zincfinger nucleases into human cells, we think they are still
underused in the emerging field of genome editing,
Gonalves said. In contrast, AdVs are extensively
being explored for genetic vaccination and oncolytic
approaches. In genome editing, they are not used
much, but we do think they have a very bright future.
Gonalves says that advantages of AdVs include
their episomal nature and very efficient introduction
of DNA into therapeutically relevant, non-transformed
mammalian cells. These viral vector systems also work equally well in dividing and quiescent, post-mitotic
mammallian cells.
Image courtesy of Manuel Gonalves.
Delivering gRNA and Cas9 Using Adenoviral Vectors

In a Scientific Reports paper introducing the delivery method in May 2014, the researchers report that AdVmediated transduction of gRNA:Cas9 ribonucleoprotein complexes into transformed and non-transformed cells
yielded rates of targeted mutagenesis similar to those achieved by isogenic AdVs encoding TALENs targeting
the same chromosomal region. The CRISPR/Cas9-derived RNA-guided nuclease-induced gene disruption
frequencies in the various cell types ranged from 18% to 65%.
A second paper published online in Nature Methods in August 2014 found that delivering RNA-guided
nucleases or TALENs together with AdV donor DNA leads to a vast majority of AdV-modified human cells being
subjected to scarless homology-directed genome editing. Gonalves said they attribute this phenomenon to
the presence of terminal proteins capping the ends of linear double-stranded AdV genomes. Such proteinDNA structures presumably reduce the likelihood that donor DNA will interact with sporadic double-stranded
chromosomal DNA breaks that always happen naturally.
We think this [most recent] work gives additional rationale for investigating the usefulness of adenoviral vector
technology in the context of genome editing, he said, adding that he hopes others will now begin to make use
of the new AdV delivery tools for a variety of applications.
It would be rewarding if these reagents and protocol are picked up and people start to explore and test this
95 | Page
ADENOVIRAL DELIVERY OF CRISPR/CAS9 AIMS TO EXPAND GENOME EDITING TO PRIMARY

CELLS (CONT)
method of introducing the CRISPR system into a broader range of cells primary cells and cells that are not
transformed and eventually also consider in vivo applications.
Start Your CRISPR/Cas9 Research!

To find more information about the adenoviral delivery of CRISPRS/Cas9 using the Gonalves labs plasmids,
including protocols, check out the plasmids at Addgene: pAdSh.PGK.Cas9 (expresses S. pyogenes Cas9
from the PGK promoter) and U6 promoter-driven guide RNA constructs, pAdSh.U6.gRNAS1 and pAdSh.
U6.gRNAGFP.
Further Reading
1.
Maggio, Ignazio, et al. Adenoviral vector delivery of RNA-guided CRISPR/Cas9 nuclease complexes
induces targeted mutagenesis in a diverse array of human cells. Scientific Reports 4 (2014). PubMed PMID:
2. Holkers, Maarten, et al. Adenoviral vector DNA for accurate genome editing with engineered
nucleases. Nature Methods 11.10 (2014): 1051-1057. PubMed PMID: 25152084.
96 | Page
AN ELEGANS APPROACH TO BETTER CRISPR/CAS9 EDITING EFFICIENCY

By Jordan Ward | January 27, 2015
This section was contributed by Jordan Ward who is a

postdoctoral fellow at UCSF.
Emerging CRISPR/Cas9 editing technologies have
transformed the palette of experiments possible in a
wide range of organisms and cell lines. In C. elegans,
one of the model organisms which I use to study gene
regulation during developmental processes, CRISPR/
Cas9 allows us to knock out sequences and introduce
mutations and epitopes with unprecedented ease. In
the last year, several advances in C. elegans genome
editing using CRISPR/Cas9 have emerged, which I will
describe below. These new C. elegans approaches
rapidly enrich for editing events without the need for
any selective marker to remain in the edited animal. To
my knowledge these approaches have not yet been
extended to other organisms/cell lines, though it is
likely that many aspects will broadly improve editing
efficiency.
Image courtesy of Jordan Ward, UCSF.
Development of CRISPR/Cas9 Editing Strategies in C. elegans

As brilliantly and irreverently illustrated in the parable of the Geneticist vs the Biochemist, we geneticists are
a lazy bunch who love to rely on the awesome power of genetic selection. Knock-in events tend to be rare,
and the challenge with any experiment is inefficient recovery of these edits. Initial work in C. elegans relied on
selective markers such as drug resistance (Chen et al., 2013), fluorescence (GFP; Tzur et al., 2013), or rescue
of mutant phenotypes (unc-119, Dickinson et al., 2013). These approaches allowed effective recovery of
knock-ins, but did result in 1-2 kilobases of additional sequence being introduced. Cre-mediated excision of
the selective cassette minimizes the sequence added, leaving a 34 bp scar, but currently requires additional
experimental manipulation. Including Cre-mediated marker excision, it takes approximately four weeks to
obtain a homozygous, outcrossed knock-in ready for experimentation.
In the last year, several manuscripts published in Genetics detailed approaches in which selection for an editing
event that produces a visible phenotype enriches for knock-outs and knock-ins at other genomic loci. The
first approach co-CRISPR from Craig Mellos lab used inactivation of the unc-22 gene as their selection
marker (Kim et al., 2014). Meanwhile Andy Fires lab used oligo-mediated knock-in of a dominant mutation,
known as co-conversion (Arribere et al., 2014). In both cases, the selected mutation must be removed, which
can be done by isolating animals with particular visible phenotypes. These approaches may become even
more powerful following a report that one can use linear repair templates (ie. PCR-derived dsDNA) with 3060 basepair homology arms to knock in large epitopes, such as GFP (Paix et al., 2014). The co-CRISPR and
co-conversion approaches have the advantage of being used in any genetic background, but require variable
amounts of experimental manipulation and screening. Additionally, they take from 8-14 days to recover knockin homozygotes ready for experimentation, depending on the screening strategy.
97 | Page
AN ELEGANS APPROACH TO BETTER CRISPR/CAS9 EDITING EFFICIENCY (CONT)
To Find a Needle, Remove the Haystack

I was fortunate or unfortunate enough, depending on perspective to work with inefficient sgRNAs in my
initial direct screening efforts, which were similar to the approach detailed by Paix et al. Although I was able
to knock-in a 2xFLAG epitope into my gene of interest, I encountered low efficiency (0.13%) and laborious
handling (screening 768 F1 animals). Recovering rare editing events in a sea of unedited animals struck me
as a needle in the haystack type of problem and led to me to explore alternate approaches. Being a lazy
geneticist, I developed a co-selection approach relying on repair of a conditional-lethal mutation to identify
edits with minimal screening effort. Selecting for repair of a lethal mutation should remove much of the unedited
haystack, facilitating recovery of edited animals.
In the background are 504 restriction digests of PCR products used to

identify a single 2xFLAG knock-in by direct screening. The center images
depict the power of lethal mutation co-selection. Only animals rescued for
the conditional lethal mutation survive the selection; from a representative
co-selection experiment, five knock-ins were recovered from seven rescued
F1 animals (gel image under worm). Image courtesy of Jordan Ward (UCSF).
In my recent Genetics paper (Ward, 2015), I

demonstrated that selection for repair of a temperaturesensitive pha-1 mutation significantly enriches for
knock-in of 2x and 3xFLAG epitopes into other, nonlinked loci; pha-1(e2123) mutant worms are perfectly
viable at 15 C, yet exhibit complete embryonic lethality
at 25 C. This method resulted in efficiencies ranging
from 11-100% of F1 animals carrying precise knockins, and homozygous knock-in animals can be obtained
in eight days. The only animals on a plate, other than
the parental animal, are rescued progeny, which makes
screening extremely rapid. This stringent selection
allowed me to optimize a range of editing parameters:
oligo repair templates with homology arms of 35-80 bp,
and DNA double-strand breaks up to 54 bp from the
desired insertion site result in efficient editing. Repair
oligos do not need to be PAGE purified, although doing
so increases knock-in efficiency. Finally, as shown
in Drosophila S2 cells (Bttcher et al., 2014), inactivation of non-homologous end-joining results in a further
increase in knock-in efficiency, presumably by channeling DNA breaks into the homologous recombination
repair pathway. Reagents required to perform pha-1 co-conversion are available through Addgene.
Future Perspectives
It is interesting to note that recipients of the 2015 Breakthrough Prize included Jennifer Doudna and
Emmanuelle Charpentier for their pioneering CRISPR/Cas9 work, and Victor Ambros and Gary Ruvkun for their
seminal work on micro RNAs in C. elegans. The C. elegans work on small RNAs informed and drove work in
countless other systems, whereas import of CRISPR technology into C. elegans has been a transformative
innovation for our community. Developments in mammalian and yeast cells using modified Cas9 proteins to
regulate gene expression (CRISPRi/CRISPRa), visualize specific genomic loci (CRISPR-imaging), or to find
the proteins associated with a given genomic locus (CRISPR-ChAP-MS) could further transform the range
of experiments imaginable in C. elegans. Equally, some of the elegans methods outlined in this post
particularly the use of co-selection methods could greatly streamline editing in other organisms and systems,
allowing rapid progress in a wide range of fields.
98 | Page
AN ELEGANS APPROACH TO BETTER CRISPR/CAS9 EDITING EFFICIENCY (CONT)

Many thanks to our guest contributor Jordan Ward!
Jordan Ward is a postdoctoral fellow with Keith Yamamoto at UCSF, and is currently
studying gene regulation during nematode molting while preparing to transition to his
own independent group. He received his PhD in Biochemistry from Cancer Research UK/
University of London studying mechanisms of DNA repair in the lab of Professor Simon
Boulton. Learn more at his webpage or follow him on Twitter @Jordan_D_Ward.
Further Reading
1. Chen, Changchun, Lorenz A. Fenk, and Mario de Bono. Efficient genome editing in Caenorhabditis
elegans by CRISPR-targeted homologous recombination. Nucleic Acids Research 41.20 (2013): e193-e193.
2. Tzur, Yonatan B., et al. Heritable custom genomic modifications in Caenorhabditis elegans via a
CRISPRCas9 system. Genetics 195.3 (2013): 1181-1185. PubMed PMID: 23979579. PubMed Central PMCID:
PMC3813848.
3. Dickinson, Daniel J., et al. Engineering the Caenorhabditis elegans genome using Cas9-triggered
homologous recombination. Nature Methods 10.10 (2013): 1028-1034. PubMed PMID: 23995389. PubMed
4. Kim, Heesun, et al. A Co-CRISPR strategy for efficient genome editing in Caenorhabditis elegans.
Genetics 197.4 (2014): 1069-1080. PubMed PMID: 24879462. PubMed Central PMCID: PMC4125384.
5. Arribere, Joshua A., et al. Efficient marker-free recovery of custom genetic modifications with CRISPR/
Cas9 in Caenorhabditis elegans. Genetics 198.3 (2014): 837-846. PubMed PMID: 25161212. PubMed Central
PMCID: PMC4224173.
6. Paix, Alexandre, et al. Scalable and versatile genome editing using linear DNAs with microhomology to
Cas9 sites in Caenorhabditis elegans. Genetics 198.4 (2014): 1347-1356. PubMed PMID: 25249454. PubMed
7. Ward, Jordan D. Rapid and precise engineering of the Caenorhabditis elegans genome with lethal
mutation co-conversion and inactivation of NHEJ repair. Genetics 199.2 (2015): 363-377. PubMed PMID:
8. Bttcher, Romy, et al. Efficient chromosomal gene modification with CRISPR/cas9 and PCR-based
homologous recombination donors in cultured Drosophila cells. Nucleic Acids Research 42.11 (2014): e89-e89.
99 | Page
EVEN MORE ELEGANT: SINGLE INJECTION CRISPR/CAS9 IN C. ELEGANS

By Mary Gearing | July 7, 2015
In the summer of 2013, a remarkable nine papers describing CRISPR/Cas9 genome engineering methods for
C. elegans were released, signaling a new era in C. elegans research. Homology directed repair (HDR), which
enables insertion of custom genomic modifications, is very robust in C. elegans, and the methods for HDRmediated modification continue to be improved. New work from Bob Goldsteins lab at the University of North
Carolina has made using CRISPR in C. elegans even easier - now, one can generate a fluorescent protein
fusion, transcriptional reporter, and loss-of-function allele in just one injection step! The entire protocol takes
about 2-3 weeks but requires less than eight hours worth of hands-on time.
Creating a Modular, Selection-based Gene Tagging Approach

Multiple features of C. elegans have made CRISPR/Cas9 so successful in this species, namely, the robust HDR
mentioned above, an accessible germline, and multiple selectable markers. Bob Goldsteins lab previously used
a transgene selection strategy for their C. elegans/CRISPR experiments, but this strategy required the use of
unhealthy, difficult-to-inject unc-119(ed3) animals, as well as a second injection step to remove the selection
marker. Other labs have used PCR-based screening to detect mutants, but this method limits screening to a
few hundred worms, making the detection of rare events difficult.
Dickinson et al. sought to create a new and efficient selection-based CRISPR/Cas9 method to generate knockins on any genetic background. Their new method uses a selection cassette containing genes for hygromycin
resistance and the Roller movement phenotype (watch the video below to see the Roller phenotype in action!).
To make these markers easy to remove after injection, they added flanking loxP sites and a heat-shock
inducible Cre to generate a self-excising cassette (SEC).
To make it possible to fluorescently label proteins in
C. elegans, they inserted the SEC into a synthetic
intron between a fluorescent protein (FP) and a
3xFLAG tag (see construct schematic below). After
self-excision, the resulting loxP scar is located in this
synthetic intron, and only the fluorescent protein and
3xFLAG-tag sequences are inserted into the genome.
Screen shot of the movie from Dickinson et al. displaying the Rol phenotype.
Fewer animals on the right display the Rol phenotype after the SEC casette
has been removed. Used with permission from the Genetics Society of
America. View the movie here.
To use this SEC-based system, first design and clone

500-700 bp homology arms for your gene of interest
into the FP-SEC vector using Gibson assembly. This
construct will serve as the repair template plasmid to
create fluorescent and 3xFLAG-tag fusion proteins.
The repair template and Cas9-sgRNA plasmid are

then injected into the germline of young adult worms.
Once the worms have laid eggs, hygromycin is added
to the plates to select for candidate knock-ins. These candidate animals can then be further screened based
on their Roller phenotype. If the SEC is inserted at the 5 end of a gene, the SEC separates the targeted gene
from its promoter, creating a loss-of-function allele as well as a promoter-FP fusion for the gene of interest. L1/
L2 larvae from these strains are then heat shocked to remove the SEC and after 5-6 days of growth, wild-type
animals that have undergone SEC excision are screened based on the loss of the Roller phenotype. These
animals contain a fluorescent protein fusion on the targeted gene.
100 | Page
EVEN MORE ELEGANT: SINGLE INJECTION CRISPR/CAS9 IN C. ELEGANS (CONT)
Self-excising cassette schematic from Dickinson et al. Used with permission from the Genetics Society of America.
After initial testing, Dickinson et al. tested this system across 7 genes in parallel; when they injected 60-80
animals per gene, they obtained the desired insertions on the first attempt for 6/7 genes. The data are clear the SEC-based system is both simple to use and highly efficient for C. elegans genome modification.
Schematic from Dickinson et al. illustrating the steps to fluorescently tag a C. elegans protein. The his-72 locus is shown as an example. Used with permission from the Genetics Society of America.
101 | Page
EVEN MORE ELEGANT: SINGLE INJECTION CRISPR/CAS9 IN C. ELEGANS (CONT)
Applications of this System

Dickinson et al. have developed a modular system that requires very little hands-on labor and could facilitate
the tagging of every protein in the C. elegans genome. Although they tested fluorescent protein fusions, this
system could be used to make many other genome modifications, including targeted mutations. SEC-based
systems may prove effective in other species or in cell culture. In addition to CRISPR/Cas9 applications, SEC
selection could replace other, more restrictive C. elegans selectable marker systems. This novel, rapid, and
user-friendly system represents another leap forward for the CRISPR/Cas9 revolution!
Further Reading
1. Dickinson, Daniel J., et al. Streamlined genome engineering with a self-excising drug selection
cassette. Genetics 200.4 (2015): 1035-1049. PubMed PMID: 26044593. PubMed Central PMCID:
PMC4574250.
2. Dickinson, Daniel J., et al. Engineering the Caenorhabditis elegans genome using Cas9-triggered
homologous recombination. Nature Methods 10.10 (2013): 1028-1034. PubMed PMID: 23995389. PubMed
3. Paix, Alexandre, et al. Scalable and versatile genome editing using linear DNAs with microhomology to
Cas9 sites in Caenorhabditis elegans. Genetics 198.4 (2014): 1347-1356. PubMed PMID: 25249454. PubMed
102 | Page
ANOTHER PATHWAY INTO CELLS: ITOP PROTEIN ONLY TRANSDUCTION

By Mary Gearing | June 23, 2015
Primary cells recapitulate the natural biology of a

cell type of interest better than immortalized lines
derived from the same cell type; however, their use
has been limited by technical problems. For instance,
its much more difficult to introduce a gene of interest
into primary cells, so most primary cell lines require
viral infection. A new paper from Niels Geijsens lab
suggests that primary cells may be better transduced
using only protein. Read on for a description of the
labs iTOP protein-only transduction method and
its potential applications to CRISPR/Cas9 genome
editing.
Protein Only Transduction?

Niels Geijsens group in Utrecht wanted to pursue a
Neuronal stem cells transduced with iTOP resulting in GFP expression.
nonviral strategy for transducing primary cells. What
Image courtesy of Diego DAstolfo.
if, instead of supplying a DNA construct, they could
simply induce cells to take up protein directly? Its not a new idea, but it comes with a few caveats. The most
common protein-only transduction strategy is to fuse a cell penetrating peptide (CPP) to a protein of interest
thereby giving it the capacity to enter cells. One such CPP was discovered in the HIV protein Tat. Although
CPP fusions can translocate across a cell membrane, the presence of the CPP may alter protein function or
localization. Thus, the usefulness of CPPs must be evaluated on a case-by-case basis.
While setting up their CPP system to study protein transduction, DAstolfo et al. made the surprising discovery
that both purified CPP-tagged protein and an untagged control could enter cultured cells and activate
expression of a luciferase reporter. CPP was not required for protein entry into cells! After verifying the
specificity of their luciferase assay, they concluded that components in the buffer used for protein purification
must be helping the untagged protein enter cells. Subsequent work found that both NaCl and non-detergent
sulfobetaine 201 (NDSB-201) were required for protein delivery. DAstolfo et al. coined the term iTOP (induced
transduction by osmocytosis and propanebetaine) to describe this method of protein transduction. Later
studies showed that iTOP functions through macropinocytosis.
DAstolfo et al. tested iTOP in a series of primary cell types using Cre and various loxP reporters. In both mouse
and human embryonic stem cells, they found very high recombination percentages (78-79% after two rounds of
transduction). Compared to CCP transduction, iTOP is at least four times more efficient in primary fibroblasts.
iTOP also works in multiple types of primary mouse cells, including neuronal and gut stem cells, dendritic cells,
embryonic fibroblasts, glia cells and neurons, and cell death post-transduction is low.
CRISPR/Cas9: A Potential Application of iTOP

Having tested iTOP using Cre-lox recombination, DAstolfo et al. were eager to see if iTOP would be compatible
with CRISPR/Cas9 genome editing. Although CRISPR/Cas9 is all the rage, they had another reason to test
this system. Its difficult to determine how much protein iTOP causes a cell to take up, and continued protein
expression would require multiple transduction events. For these reasons, iTOP is best suited to binary systems
in which transient protein expression has a long-lasting effect, such as Cre- or Cas9-mediated recombination.
103 | Page
ANOTHER PATHWAY INTO CELLS: ITOP PROTEIN ONLY TRANSDUCTION (CONT)

Both primary hESCs and immortalized cells treated with iTOP efficiently took up both Cas9 protein and a gRNA
(yes, iTOP can also be used for RNA transduction) targeting a gene associated with diphtheria toxin lethality
and were sorted into single wells to generate clonal populations. After iTOP treatment, both primary (hESCs)
and cultured cells showed up to 70% resistance to diphtheria toxin, indicating a 70% frequency of biallelic
mutations. Importantly, hESCs retained their stemness, indicating that iTOP may be valuable for CRISPR/Cas9
stem cell editing.
The small molecule, NSDB, and Na+ hypertonicity are key elements of iTOP. iTOP can be used to transduce Cas9 and a gRNA, which subsequently enter
the nucleus and cleave the specified target. Figure adapted from DAstolfo et al 2015.
In certain instances, the potential benefits of nonviral transduction are numerous. First, you remove the risk of
viral integration causing mutations and gene expression changes elsewhere in the genome. Second, expression
is by nature transient, which may lower nuclease toxicity or the potential for off-target effects. Third, protein
purification (once a protocol has been established) may be less laborious than viral purification, and it does
not raise the safety concerns associated with virus work. Fourth, working directly with proteins removes the
limitations of packaging size associated with viral vectors. For these reasons, iTOP, especially in tandem with
genome editing, may make primary cell work much more efficient and feasible for many labs.
Further Reading
1.
DAstolfo, Diego S., et al. Efficient Intracellular Delivery of Native Proteins. Cell 161.3 (2015): 674-690.
104 | Page
Chapter 5 - Therapeutic Applications of CRISPR
THERAPEUTIC APPLICATIONS OF CRISPR
105 | Page
Chapter 5 - Therapeutic Applications of CRISPR
USING CRISPR/CAS9 TO EDIT DISEASE OUT OF THE GENOME

By Kendall Morgan | January 7, 2014
There can be no doubt that CRISPR/Cas9 technology

has been a breakthrough for the genome-editing
field. Two studies reported in Cell Stem Cell in
December 2013 additionally show that this tool already so useful in the laboratory - might also find its
way to the clinic.
CRISPR cured the mouse on the left of its cataracts. On the right, a control
mouse with cataracts. Image attribution: Jinsong Li.
A team led by Jinsong Li from the Chinese Academy

of Sciences found that mice with a dominant
mutation in a gene that causes cataracts could
be rescued by coinjection of Cas9 mRNA and a
gRNA targeting the mutant allele into zygotes.
An independent team led by Hans Clevers at the
Hubrecht Institute in The Netherlands used the
CRISPR/Cas9 genome editing system to correct the
cystic fibrosis transmembrane conductor receptor
(CFTR) by homologous recombination in cultured
intestinal stem cells of patients with cystic fibrosis.
CRISPR to the Rescue

I was really excited when I saw these two papers, said the University of California Berkeleys Jennifer Doudna,
an Addgene depositor and CRISPR expert who was not involved in either study. They are the first direct
demonstration of use of genome editing to correct point mutations in two diseases with real phenotypic effect.
The difference with CRISPR/Cas9 in comparison to other genome editing techniques is its ease of use. Its a
very simple technology, Doudna explained, giving credit to Addgene for their role in making the technology
so readily available. Thats why were seeing adoption of the technology in different labs for different
applications.
Lis team noted that CRISPR/Cas9 has been used to generate mutations in a wide range of organisms, but its
potential to efficiently correct disease hadnt yet been realized. The researchers chose to explore its potential in
mice with a dominant cataract disorder caused by a single copy of a gene known as Crygc. When all was said
and done, the researchers had cured 24 mice of their disease.
Meanwhile, Hans Clevers and colleagues applied CRISPR to disease correction in adult stem cells isolated
from two patients with cystic fibrosis. They were able to demonstrate functional correction of the gene in
clonally expanded organoids.
CRISPR in adult stem cells has huge potential, because the genomes of adult stem cells in organoid cultures
are stable, said Gerald Schwank, first author of the cystic fibrosis report. This is a big advantage over induced
pluripotent stem cells, which need to go through a phase of crisis where they collect a number of mutations.
Cystic fibrosis may not be the most likely candidate for gene therapy, however, because the disease affects
multiple organ systems. Nevertheless, the findings are a great proof-of-principle for the use of CRISPR to
correct single-gene conditions.
106 | Page
Chapter 5 - Theapeutic Applications of CRISPR
USING CRISPR/CAS9 TO EDIT DISEASE OUT OF THE GENOME (CONT)

Remaining Hurdles
I think it is absolutely possible to use CRISPR to cure genetic disease in the near future, Li says, though there
are some things to sort out first.
The potential for off-target effects is a major limitation. The procedure will also need to be perfected until the
efficiency of repair approaches 100 percent.
We need to be sure its accurate enough and efficient enough for repair, Doudna said. And it cant be too
hard to apply. These papers suggest it might work well enough to consider as a therapeutic.
There are many fundamental questions about CRISPR/Cas biology still left to answer as well, she says, and
those details will continue to be a primary preoccupation of her California lab and many others. Those offtarget effects are important, she said. We also really need to get a handle on how cleavage sites are repaired.
Those insights will be critical for scientists ability to influence editing efficiency.
Further Reading
1. Wu, Yuxuan, et al. Correction of a genetic disease in mouse via use of CRISPR-Cas9. Cell Stem Cell
13.6 (2013): 659-662. PubMed PMID: 24315440.
2.
Schwank, Gerald, et al. Functional repair of CFTR by CRISPR/Cas9 in intestinal stem cell organoids of
cystic fibrosis patients. Cell Stem Cell 13.6 (2013): 653-658. PubMed PMID: 24315439.
107 | Page
TREATING DMD MICE WITH CRISPR GENE EDITING

Having seen CRISPRs success in basic research, researchers are eager to apply it in a clinical setting. CRISPR
is often used for animal germline modification, to repair or add in disease-causing mutations, but, until recently
it hadnt been used to treat disease postnatally. Now, three papers published concurrently in Science have
shown CRISPR can treat a genetic disease in a postnatal mouse model, an important proof of concept for
future preclinical and clinical work.
Duchenne Muscular Dystrophy: The Most Common Inherited Disease

The genetic disease targeted is Duchenne muscular dystrophy (DMD), an X-linked recessive disorder affecting
approximately 1 in 5000 males. DMD is caused primarily by frameshift mutations in dystrophin, a protein
essential for proper muscle function. Without functioning dystrophin, an individual experiences progressive
muscle wasting leading to death at around 30 years of age. Despite the amount of research conducted on
DMD, there is still no good treatment.
The dystrophin gene is very large (79 exons), but much of the sequence appears to be nonessential. Within the
mutational hotspot for dystrophin, exons 45-55, there are multiple common deletions that maintain the proteins
reading frame, leading to the production of a smaller, but at least partially functional protein. Individuals with
these mutations are often asymptomatic, or have only mild symptoms, a condition known as Becker muscular
dystrophy (BMD).
Dystrophins size makes it difficult to deliver via gene therapy, so researchers have set their sights on another
approach. Since shorter forms of dystrophin can still be functional, exon skipping is a good option for DMD
treatment. Clinical trials have used oligonucleotide exon skipping (OEN) to remove mutated exons from the
dystrophin transcript. Unfortunately, the oligonucleotides only modestly improve muscle function, and they
must be injected regularly.
Dystrophin and Genome Editing

Since complex oligonucleotide treatment comes with many challenges, researchers have begun to explore
genome editing approaches for exon skipping. Addgene depositor Charles Gersbachs lab used paired zinc
finger nucleases to remove exon 51 in DMD patient myoblasts. They observed a 13% removal rate of exon 51,
which resulted in appropriately localized dystrophin. In a subsequent study, they used CRISPR with two gRNAs
to delete exon 51 or exons 45-55 in patient myoblasts; when injected into DMD mice, these cells expressed
functional dystrophin.
Editing genes in vivo is of course much more difficult than in cell culture, but Ran et al. have shown that
CRISPR and AAV can be used together for postnatal genome editing in mice. Could this approach work with
dystrophin exon skipping? For such a therapy to be successful, multiple requirements must be met. First,
CRISPR would need to be delivered to both cardiac and skeletal muscle cells, where precision editing of the
dystrophin gene would take place, with minimal risk of off-target editing. In order for the therapy to persist over
time, stem cell editing would be highly desirable. Should stem cell editing occur, the CRISPR components
would only need to be expressed for a short period of time, which would prevent the accumulation of unwanted
mutations over time.
DMD is a good choice for a CRISPR proof of concept treatment, as the disease is especially well suited for
genome editing. The homology-directed repair (HDR) pathway is downregulated in mature tissues - no problem,
108 | Page
TREATING DMD MICE WITH CRISPR GENE EDITING (CONT)

as exon skipping proceeds through the non-homologous end joining (NHEJ) pathway. Its estimated that very
little dystrophin correction (about 4%) is needed to see muscle improvement, with only 30% correction needed
for normal function. Even low-frequency editing could make a huge difference in DMD, and its estimated that
exon skipping therapies would be applicable to 80% of DMD patients.
Exon skipping strategies for Duchenne muscular dystrophy. gRNAs can be designed to flank a single or multiple exons that, when deleted through NHEJ,
will restore the dystrophin reading frame. gRNAs bind in intronic regions, so low-frequency indels will not cause frameshifts. Figure adapted from Osterout
et al. under a CC-BY 4.0 license.
Long et al., Nelson et al., and Tabebordbar et al. each chose to try exon skipping in DMD mice, which have
a mutation in dystrophin exon 23. Long et al. used SpCas9 in conjunction with AAV9, whereas Nelson et al.
and Tabebordbar et al. used the shorter SaCas9 with AAV8 and AAV9, respectively. Although each study
used slightly different methods, they each observed the same positive phenotypes in cardiac and skeletal
muscle. Even when the genome editing frequency is low, the relative abundance of exon-skipped mRNA is
high, likely because this mRNA is not subject to nonsense-mediated decay. Although Nelson et al. observed
only 2% genome editing in one experiment, they found the exon-skipped transcript constituted 59% of total
dystrophin mRNA, similar to the 39% observed by Tabebordbar et al. Long et al. show that the percentage of
dystrophin-expressing muscle cells increases over time, and all three groups confirm dystrophin expression via
Western blot, at <10% of the wild-type level. Muscle histology is improved, with vastly reduced inflammation
and necrosis compared to unedited animals. In grip strength, specific force, and other muscle tests, muscle
function is clearly improved, although not to wild-type levels. With regard to off-target effects, each group finds
very low to no off-target activity at the ten highest-ranking predicted off-target sites.
Each paper characterizes additional, unique facets of CRISPR DMD therapy. Long et al. show that AAV-CRISPR
does not cause obvious germline modification, an important finding given the controversy associated with such
editing. Tabebordbar et al. show that muscle stem cells are modified by this approach, increasing the possibility
that edits will persist long-term. In accordance with that result, Nelson et al. find that dystrophin restoration is
maintained for at least six months.
109 | Page

Future Directions and Obstacles to Clinical CRISPR Editing
Given the success of the CRISPR-mediated exon skipping approach in mice, DMD researchers are very
excited. This approach may also be applicable to a number of rare genetic diseases caused by splicing
defects, including ataxia telangiectasia, congenital disorder of glycosylation, and Niemann-Pick disease type
C. Although the three DMD studies referenced here represent a great step forward for CRISPR gene therapy,
its important to realize that DMD is a simpler case than other genetic diseases wed like to treat with CRISPR.
As seen above, DMD can be treated with one-size-fits-most NHEJ-mediated editing, but most other diseases
will require HDR-mediated precision editing tailored to smaller patient populations.
To bring DMD CRISPR therapy closer to the clinic, much work is still needed. First, CRISPR delivery must be
optimized to:
1.
2.
Reach a high percentage of muscle cells throughout the body, especially stem cells
Remove any immunogenicity of the AAV vector
Once delivery has been optimized, its important to ascertain how long the rescue phenotype will last, and
more importantly, if it does extend lifespan. This work should be done in mouse and large animal models
with mutations in human-relevant exons 45-55, rather than in the traditional exon 23-mutated mouse model.
Potential off-target effects in muscle, as well as unwanted germline editing, must be rigorously examined over a
long period of time. High fidelity Cas9s such as eSpCas9 and SpCas9-HF should be explored to reduce offtarget editing. Short-term CRISPR expression would be especially desirable, as it would reduce the potential
of off-target editing over time, but this approach would require robust stem cell editing to maintain the desired
phenotype.
Precision genome editing faces the challenges above and more. One chief challenge is upregulating HDR in
mature tissues, as this process is necessary to precisely edit point mutations. In addition to upregulating HDR,
NHEJ must be downregulated to prevent the introduction of new mutations by CRISPR; editing exonic regions
comes with much more risk than the intronic editing used in the exon skipping approaches here. In most
genetic diseases, the mutational landscape is broad and varied, necessitating the development of many distinct
CRISPR therapies to match these different mutations. For each therapy, effectiveness and off-target risk must
be evaluated separately, increasing the time to clinical approval.
Even with the challenges precision editing faces, its encouraging to see such progress in the more feasible
case of DMD. If the safety and efficacy of this approach can be optimized, DMD could become one of the first
diseases to be treated clinically with CRISPR. Many obstacles remain for CRISPR gene editing, but given the
speed at which the technology is advancing, precision editing treatments may also be closer than we might
expect.
Further Reading
1.
Kaiser J. CRISPR helps heal mice with muscular dystrophy. Science. 2015 Dec 31. doi:10.1126/science.
aae0169
110 | Page

4. Reardon S. Gene-editing summit supports some research in human embryos. Nature. 2015 Dec 3.
doi:10.1038/nature.2015.18947
5. Siva K, Covello G, Denti MA. Exon-skipping antisense oligonucleotides to correct missplicing in
neurogenetic diseases. Nucleic Acid Ther. 2014 Feb;24(1):69-86. PubMed PMID: 24506781. PubMed Central
PMCID: PMC3922311.
7. Ousterout DG, Kabadi AM, Thakore PI, Perez-Pinera P, Brown MT, Majoros WH, Reddy TE, Gersbach
CA. Correction of dystrophin expression in cells from Duchenne muscular dystrophy patients through genomic
excision of exon 51 by zinc finger nucleases. Mol Ther. 2015 Mar;23(3):523-32. PubMed PMID: 25492562.
8. Ousterout DG, Kabadi AM, Thakore PI, Majoros WH, Reddy TE, Gersbach CA. Multiplex CRISPR/Cas9based genome editing for correction of dystrophin mutations that cause Duchenne muscular dystrophy. Nat
Commun. 2015 Feb 18;6:6244. PubMed PMID: 25692716. PubMed Central PMCID: PMC4335351.
9. Ran FA, Cong L, Yan WX, Scott DA, Gootenberg JS, Kriz AJ, Zetsche B, Shalem O, Wu X, Makarova KS,
Koonin EV, Sharp PA, Zhang F. In vivo genome editing using Staphylococcus aureus Cas9. Nature. 2015 Apr
9;520(7546);186-91. doi: 10.1038/nature14299. Epub 2015 Apr 1. PubMed PMID: 25830891. PubMed Central
PMCID: PMC4393360.
111 | Page
Chapter 6 - CRISPR Safety
CRISPR BIOSAFETY
112 | Page
Chapter 6 - CRISPR Biosafety
GENERAL CRISPR SAFETY CONSIDERATIONS

Addgene loves supplying CRISPR plasmids, but we want to make sure youre aware of potential safety issues.
These include biosafety concerns, as well as biocontainment concerns for researchers working with highly
mobile model organisms like Drosophila. This section provides basic background information to help you use
CRISPR safely, but please be sure to contact your institutions Biosafety Committee before beginning
work.
Viral Vectors
If youre using CRISPR with a viral vector system, please refer to our Viral Vector Biosafety Guide.
Lentiviral CRISPR systems are very common for both single gene and pooled library knockout experiments.
The lentiviral systems at available at Addgene are derived from HIV, but their organization across multiple
plasmids and the deletion of many HIV proteins lowers the probability of generating replication-competent
virus. Lentiviral vectors are handled at BSL-2/2+ safety levels.
Addgene also supplies some CRISPR plasmids for retroviral infection. Retroviral vectors are classified based on
the cell types they infect; vectors that can infect human cells are handled at BSL-2/2+, while other vectors may
be handled at BSL-1 depending on the target gene(s).
One popular CRISPR application uses SaCas9 with an adeno-associated viral vector (AAV) for in vivo gene
editing. Since AAV is replication-incompetent and is not known to cause disease in humans, it is usually
handled at BSL-1, provided that the gRNA(s) used do not have potential oncogenic, apoptotic or toxic effects.
Please note, if you use a helper virus instead of a helper plasmid system to deliver your AAV cargo, your work
should be done in BSL-2 conditions.
Preventing Unintended Editing

Since CRISPR is such a robust editing system, scientists need to be extra cautious when designing
experiments to avoid the possibility of accidental researcher self-editing. When working with model
organisms, the easiest way to reduce this possibility is to design gRNAs that target sequences not conserved
in humans. Once youve used your favorite CRISPR software to design gRNAs, BLAST them against the human
genome to check for potential off-targets.
The method used to deliver Cas9 and gRNA(s) can also affect biosafety risk. Cell culture treatment or animal
injection are contained, relatively low risk methods. Inhalation-based delivery presents a higher risk since it is
more difficult to contain the viral particles. To minimize the risks associated with CRISPR inhalation, Addgene
depositor Andrea Ventura has used an aerosol-based system with a replication-incompetent virus targeting
mouse-specific loci.
As alluded to above, the question of what youre trying to edit is often just as important as how. Work
targeting tumor suppressor or oncogenic genes, like P53 or KRAS, warrants a high level of prudence.
Introducing human disease alleles into model organisms also comes with risk, especially if the gRNA target
sequence is conserved in humans. In general, any editing that promotes oncogenesis or apoptosis, or could be
potentially toxic, should be carefully designed to maximize biosafety and minimize researcher risk.
113 | Page
GENERAL CRISPR SAFETY CONSIDERATIONS (CONT)

Gene Drive Containment
If youve kept up with CRISPR, chances are youve heard of gene drives. If not, check out our section on
gene drives from guest writer Kevin Esvelt. In short, gene drives allow a genetic modification to spread rapidly
through a sexually reproducing population. In standard inheritance, a heterozygous parent has a 50% chance
of passing a modified gene to its offspring. In gene drive inheritance, nearly 100% of offspring inherit the
modification. CRISPR/Cas9-based gene drives consist of Cas9 and a gRNA positioned adjacent to each other
in the locus targeted by the gRNA. If this cassette is present on one chromosome, Cas9 activity can result in
the gene drive being copied to the other chromosome, increasing its inheritance rate and allowing it to spread
rapidly throughout the population.
Gene drives have many potential uses - in fact, a recently published gene drive renders mosquitoes resistant
to the malaria-causing parasite Plasmodium falciparum. If released into the wild, this gene drive could
end malaria as we know it. However, the key advantage of a gene drive is also its main disadvantage - if
accidentally released, a gene drive can spread rapidly, with uncertain consequences. Thus, working with gene
drives, especially in organisms likely to escape confinement, like flies and mosquitoes, requires experimental
confinement strategies.
Avoiding accidental creation of a gene drive is simple: dont use a DNA vector that contains both Cas9 and a
gRNA. This separation prevents Cas9 and a gRNA from integrating together into the genome, and is analogous
to the common strategy of dividing viral vector components across multiple plasmids. For example, Cas9 can
be maintained episomally or integrated into the AAVS1 safe harbor locus.
If you do want to work with gene drives, see the infographic below for some different types of confinement
strategies. Ideally, multiple types of confinement strategies should be used together to prevent accidental
release.
Visual depiction of common containment strategies. For optimal biocontainment, multiple strategies should be used together. Figure adapted from Esvelt
et al. under a CC-BY 4.0 license.
114 | Page

Gene Drive Reversal and Immunization
Stringent confinement is only a part of the best practices for working with gene drives. Ideally, researchers
should also develop tools to rapidly reverse engineered gene drives. The first reversal gene drive was recently
developed in yeast. This reversal drive cuts the previously inserted gene drive to insert a functional version of
the originally targeted gene (see figure below). Although Cas9 and a gRNA remain in the genome, wild-type
gene function is restored.
Gene drive immunization can also prevent unwanted modifications due to accidental release. In this situation,
the immunizing drive recodes a portion of the gene such that the gRNA in the gene drive cannot modify it, thus
preventing a gene drive from affecting this immunized population (see figure below).
Reversal and immunizing drives. Reversal drives reverse previous gene drives by reintroducing previously edited/deleted genes. Immunizing drives
modify a gene of interest to prevent other gene drives from targeting it. Figure adapted from Esvelt et al. under a CC-BY 4.0 license.
Specific questions about how biosafety relates to your research should be directed to your institutions
Biosafety Committee.
Further Reading
1. Maddalo, Danilo, et al. In vivo engineering of oncogenic chromosomal rearrangements with the
CRISPR/Cas9 system. Nature 516(7531) (2014): 423-37. PubMed PMID: 25337876. PubMed Central PMCID:
PMC4270925.
2. Gantz, Valentino M., et al. Highly efficient Cas9-mediated gene drive for population modification of the
malaria vector mosquito Anopheles stephensi. Proc Natl Acad Sci U S A 112(49) (2015): E6736-43. PubMed
115 | Page

3. Akbari, Omar S., et al. BIOSAFETY. Safeguarding gene drive experiments in the laboratory. Science
349(6251) (2015): 927-29. PubMed PMID: 26229113.
4. Esvelt, Kevin M., Andrea L. Smidler, Flaminia Catteruccia & George M. Church. Concerning RNAguided gene drives for the alteration of wild populations. eLife (2014): e03401. PubMed PMID: 25035423.
5. DiCarlo, James E., et al. Safeguarding CRISPR-Cas9 gene drives in yeast. Nat Biotechnology 33(12)
(2015): 1250-55. PubMed PMID: 26571100.
116 | Page
TRANSGENIC ORGANISMS, CAS9 GENE DRIVES, AND APPROPRIATE SAFEGUARDS

By Kevin Esvelt | May 22, 2015
This post was contributed by Kevin Esvelt, a Wyss Technology Development Fellow at the Wyss Institute and
Harvard Medical School.
Scientists making transgenic organisms with Cas9
should be aware of the potential hazards of creating
gene drives capable of spreading through wild
populations. Whereas most genomic changes impose
a fitness cost and are eliminated by natural selection,
gene drives distort inheritance in their favor and
consequently can spread even when costly.
If even a single organism carrying a synthetic gene
drive were to escape the laboratory, the drive could
eventually spread through the entire wild population
with unpredictable ecological effects. Because the
consequences of such a mistake would necessarily
extend far beyond the laboratory and seriously
damage public trust in scientists, experiments
involving potential gene drives should be conducted
with extreme caution.
Image credit: Kevin Esvelt
Why Does Making Transgenic Organisms with Cas9 Risk Creating a

Gene Drive?
The simplest way to make a gene drive is to insert
an endonuclease gene within its own cut site (1). In
heterozygotes, the endonuclease cuts the wild-type
chromosome and its own gene is used as a repair
template, thereby converting the heterozygote into
a homozygote and ensuring that all future offspring
inherit the endonuclease gene. As we pointed out
last year (2), delivering a DNA cassette encoding
the cas9 gene and single guide RNAs (sgRNAs) with
appropriate flanking homology into a germline cell
can create an RNA-guided gene drive. In principle,
Image credit: Kevin Esvelt
this event could happen even if there isnt any
homology available. Consequently, the best way to avoid accidentally creating a gene drive is also the easiest:
dont use a DNA vector that encodes both Cas9 and sgRNA.
What if I Want to Work with Gene Drives?

Then do so! Its a tremendously promising field that could address many pressing global problems in health,
agriculture, and conservation. But please be aware that there are many easy-to-implement confinement
strategies that are robust to human error. Used in combination with standard barrier protocols, they can reduce
the risk of accidental escape to a negligible level.
117 | Page
TRANSGENIC ORGANISMS, CAS9 GENE DRIVES, AND APPROPRIATE SAFEGUARDS (CONT)

One major reason we chose to publish the concept and likely capabilities of RNA-guided gene drives in the
summer of 2014 (after extensive discussions with biosecurity experts, ecologists, and national authorities)
was to detail these robust and easy-to-implement confinement strategies (2). Our hope was that anyone else
who came up with the same idea or wanted to build a gene drive in a model organism would find our work
and consequently use these precautions in their own experiments (2,3). While that didnt work quite as well as
anticipated (4), greater awareness will certainly improve safety.
What Safeguards and Confinement Strategies Are Available?

Molecular confinement involves building gene drives that can spread through populations of transgenic
laboratory organisms but not wild organisms. For example, an sgRNA-only drive will spread exclusively through
populations that already express Cas9 from an unlinked locus, while a Cas9+sgRNA drive targeting a synthetic
sequence will only spread in transgenic laboratory populations with that sequence (2). Both methods are easy
to implement and have been tested in yeast (5).
Ecological confinement involves performing experiments in a geographic area where escaped organisms
wont be able to find mates (2). For example, ongoing experiments attempting to build gene drives in tropical
mosquito vectors of diseases such as malaria and dengue are currently being performed in regions that dont
have resident populations of the relevant mosquito species.
Reproductive confinement involves working with laboratory organisms that cant reproduce with wild ones.
For example, Drosophila lines with compound autosomes are completely infertile when mated to wild fruit flies
(6). Its also worth noting that gene drive experiments are less hazardous in organisms that seldom reproduce
sexually because the drive must be much more efficient and minimally harmful in order to spread.
Barrier confinement seeks to keep the organisms in the laboratory. It varies by organism, but your local
biosafety officer should be familiar with appropriate measures. Barriers should be a component of all gene drive
confinement strategies, but they should not be relied on exclusively because historical studies of pathogen
research have conclusively shown that barrier protocols are vulnerable to human error. And with gene drives,
one mistake can be enough.
Reversal drives are designed to overwrite a previous gene drive and thereby undo the genetic changes driven
by the earlier intervention (2). While an initial reversal drive cannot restore the exact original sequence, it can
restore the original protein-coding sequence using a recoding strategy; a subsequent drive can restore the wildtype sequence (save for the residual sgRNAs and possibly Cas9 gene). An immunizing reversal drive is a variant
that also spreads through the wild population and immunizes it against the first drive. Laboratories interested
in building candidate gene drives intended for eventual release should consider building an appropriate
immunizing reversal drive at the same time to mitigate the potential effects of an accidental release (3).
Which Gene Drive Confinement Strategies Should Be Used?

There are ongoing efforts to develop formal guidelines to answer this question. Until then, using multiple
confinement strategies is strongly advised. The effects are multiplicative, and there are very few gene drive
experiments that cant be performed with two or more confinement methods. Since molecular, ecological, and
reproductive confinement typically require very little effort to implement (depending on the species in question),
why not use them whenever applicable?
Against this, consider the cost of an accidental release. Science relies on popular support, which in turn
118 | Page
TRANSGENIC ORGANISMS, CAS9 GENE DRIVES, AND APPROPRIATE SAFEGUARDS (CONT)

depends on widespread trust in scientists to develop new technologies with care and humility. Nothing would
damage that trust more than an accidental gene drive release, which would represent proof positive that at
least some scientists could not be trusted to handle such a powerful technology responsibly. An accidental
release would be particularly dangerous for molecular biology and genetics because the underlying advance
enabling RNA-guided gene drives is Cas9 genome editing.
Finally, gene drives are an inherently collective technology: because they necessarily alter the shared
environment, they must never be used without popular support (2). An accidental release could delay realworld applications against scourges such as malaria and dengue for many years. Since more than a thousand
children died of malaria today, lets not risk a potential way to dramatically reduce that number for the sake of
laboratory convenience.
Disclaimer: Recommendations concerning the appropriate degree of confinement are those of the author and
do not represent a formal stance taken by Addgene or its staff. Additional information about gene drives can be
found in the referenced publications.
Many thanks to our guest contributor Kevin Esvelt!
An inventor of technologies that harness evolution, Kevin Esvelt studies ways of using
molecular tools to alter populations of organisms and associated ecosystems. Because
the environment is shared by everyone, he is deeply concerned with ensuring that gene
drives and other collective technologies are developed safely and transparently. For more
information, visit www.sculptingevolution.org.
Further Reading
1.
Burt, Austin. Site-specific selfish genes as tools for the control and genetic engineering of natural
populations. Proceedings of the Royal Society of London B: Biological Sciences 270.1518 (2003): 921-928.
2. Esvelt, Kevin M., et al. Concerning RNA-guided gene drives for the alteration of wild populations. elife
3 (2014): e03401. PubMed PMID: 25035423. PubMed Central PMCID: PMC4117217.
3. Oye, Kenneth A., et al. Regulating gene drives. Science 345.6197 (2014): 626-628. PubMed PMID:
25035410.
4. Gantz, Valentino M., and Ethan Bier. The mutagenic chain reaction: A method for converting
heterozygous to homozygous mutations. Science 348.6233 (2015): 442-444. PubMed PMID: 25908821.
5. DiCarlo, James E., et al. RNA-guided gene drives can efficiently and reversibly bias inheritance in wild
yeast. bioRxiv (2015): 013896. bioRxiv.
6. Fitz-Earle, M., D. G. Holm, and D. T. Suzuki. Genetic control of insect populations: I. Cage studies of
chromosome replacement by compound autosomes in Drosophila melanogaster. Genetics 74.3 (1973): 461475. PubMed PMID: 4200686. PubMed Central PMCID: PMC1212962.
119 | Page
CRISPR CONCERNS BEYOND THE LAB BENCH

CRISPR has proven to be a robust and flexible system for genome editing, and the biotechnology sector
has certainly noticed. At least four companies (Caribou Biosciences, Editas Medicine, CRISPR Therapeutics,
and Intellia Therapeutics) are based primarily on CRISPR technology, with numerous others using CRISPR in
combination with other approaches. Since CRISPR technology has developed so quickly, there are still many
attributes of the system that scientists dont understand, and ethical and societal implications that havent
fully explored. CRISPR has the potential to revolutionize agriculture and gene therapy, but before adopting this
system for non-research applications, one must fully consider the potential ramifications of this technology.
CRISPRs Agricultural and Therapeutic Applications

Compared to human editing, we dont hear as much about CRISPR in agriculture, but that doesnt mean theres
not controversy. CRISPR may replace or supplement previous genome editing techniques used in agriculture,
because it makes precise editing much easier. However, these edits are less traceable than traditional
manipulations, which worries some regulators. Regulating CRISPR usage in plants and animals meant for
human consumption will likely not be straightforward, and it remains to be seen how governments will adapt to
this disruptive technology.
The potential therapeutic applications of CRISPR are even more appealing than the agricultural applications.
Given the limited success of gene therapy, researchers have turned to CRISPR as the next great hope. Will
it be possible to use CRISPR for postnatal genome editing to treat genetic disease? Three studies published
concurrently in Science represent the first time CRISPR has been used to treat grown animals with a genetic
disease. Duchenne muscular dystrophy (DMD), caused by various mutations in exons of the dystrophin protein,
can be modeled in mice. Using complementary approaches in DMD mice, the teams were able to edit out
the affected exon, preventing its incorporation into the dystrophin protein and improving muscle function.
Importantly, the teams did not observe high levels of off-target cleavage throughout the genome. Its estimated
that 80% of DMD sufferers could benefit from such a therapy, and these studies represent an important step
toward clinical translation (read more in a previous section). Other monogenic diseases like sickle-cell anemia
and beta-thalassemia are often mentioned as potential targets for CRISPR therapy.
The International Summit on Human Gene Editing

Given the potential and controversy surrounding CRISPR, the U.S. National Academy of Science and National
Academy of Medicine have created a new initiative to help policy makers, researchers/clinicians, and the public
understand human gene editing technology and its implications, with the goal of informing logical decision
making about this technology. One important part of this plan is the International Summit on Human Gene
Editing, held jointly by the U.S. Academies, the Chinese Academy of Science, and the U.K. Royal Society in
December 2015.
Over a three-day program, the International Summit brought together experts to discuss various topics,
including basic research, human germline editing, governmental regulation of editing, and national/international
implications. At the conclusion of the program, the organizing committee issued a statement summarizing their
conclusions. This statement divides human editing into three categories: basic/preclinical research, clinical
somatic editing and clinical germline editing, with different recommendations for each category.
The committee stated that basic/preclinical research on both somatic and germline editing should proceed,
but with appropriate regulations. In the case of embryo/germline editing, these cells should not be used to
120 | Page
CRISPR CONCERNS BEYOND THE LAB BENCH (CONT)

establish a pregnancy. Ideally, this balance will allow researchers to better understand the biology behind such
modifications at a basic level. However, some scientists fear that modifying even embryos not intended for
implantation will be highly controversial, as seen in previous debates over embryonic stem cell research.
Discussing clinical somatic applications, the committee notes that these applications have enormous promise
and potential value. However, the statement emphasizes that risks, such as off-target editing, must be weighed
along with benefits, and that this work should be done within regulatory frameworks for gene therapy. The
recent development of high fidelity Cas9 enzymes eSpCas9 and SpCas9-HF, as well as further research
into improving CRISPR editing, will help bring clinical somatic applications closer to reality. Editas Medicine
is hoping to begin a CRISPR clinical trial as early as 2017, targeting a rare form of blindness called Leber
congenital amaurosis.
Although in vivo, non-heritable gene editing has been relatively well-received, clinical germline modification
remains highly controversial. The International Summit committee writes that it would be irresponsible to
proceed with any clinical use of germline editing unless and until (i) the relevant safety and efficacy issues
have been resolved and (ii) there is broad societal consensus about the appropriateness of the proposed
application. The committee cites six key concerns associated with clinical germline editing (see table below) these concerns are not simply health-related, but also include profound societal, moral, and ethical questions.
The statement does not advocate a complete ban on germline editing, as some scientists had predicted,
but suggests that the possibility of germline editing should be revisited periodically as scientific knowledge
advances and societal views evolve.
Concerns Associated with Clinical Human

Genome Editing
Examples
Risks of inaccurate or incomplete editing
Mosaicism, harmful off-target mutations
Unpredicted effects of genetic edits
Interaction with other variants/the environment

decreases fitness - e.g. sickle-cell disease and
malaria
Irreversibility of genetic edits once introduced into a

population
Edits cannon be tracked once introduced into a

population - may combine unfavorably through
reproduction
Implications for individuals and future

generations carrying genetic alterations
Edited individuals may struggle with feelings of

otherness
Potential for permanent genetic enhancements

affecting social structure
Designer babies, genetic underclass of

unmodified individuals
Moral and ethical considerations of altering

human evolution
Potential conflicts with religion - idea of playing

God
Six key concerns associated with human clinical germline editing raised by the International Summit of Human Gene Editing. Adapted from Section 3 of
the International Summit Statement.
In the closing paragraph of the statement, the committee calls for an ongoing international forum related to
human gene editing. This forum should help inform policymakers and allow nations to coordinate regulation of
editing. In addition to including researchers and policymakers, the forum would seek expertise and opinions
121 | Page
CRISPR CONCERNS BEYOND THE LAB BENCH (CONT)

from many other groups, including health care providers, patients and their families, faith leaders, industry
representatives, and the general public.
With CRISPR technology still in its infancy, we are a long way from societal consensus about how to use these
tools, especially when it comes to clinical germline editing. While some researchers are pro-germline editing,
others worry that the germline controversy will prevent acceptance of clinical somatic applications. As society
struggles with these ethical dilemmas, clear scientific communication will be necessary to ensure that the
public understands both the potential benefits and risks of human CRISPR applications. As the International
Summit has recommended, open dialogue among researchers, government, health care, industry, religious
leaders, and the general public is key to helping society develop a consensus on how to appropriately use this
powerful technology.
Further Reading
1. International Summit on Human Gene Editing. On Human Editing: International Summit Statement. 2015
Dec 03.
2. Kaiser J. CRISPR helps heal mice with muscular dystrophy. Science. 2015 Dec 31. doi:10.1126/science.
aae0169
3. Ledford H. CRISPR, the disruptor. Nature. 2015 June 04;522:20-24. PubMed PMID: 26040877.
6. Reardon S. Gene-editing summit supports some research in human embryos. Nature. 2015 Dec 3.
doi:10.1038/nature.2015.18947
7. Regalado A. Engineering the perfect baby. MIT Technology Review. 2015 Mar 5.
8. Regalado A. CRISPR gene editing to be tested on people as early as 2017, says Editas. MIT Technology
Review. 2015 Nov 5.
9. Regalado A. Patients favor changing the genes of the next generation with CRISPR. MIT Technology
Review. 2015 Dec 2.
122 | Page
ACKNOWLEDGEMENTS AND FINAL WORDS

Special Thanks to Our Guest Contributors!
David Wyatt
Dale Ramsden
John Doench
David Wyatt is a graduate student interested in determining how the structure of broken
DNA ends impacts how they are repaired. He works in Dale Ramsdens lab.
Dale Ramsden is a member of the curriculum in Genetics and Molecular Biology, the
Dept. of Biochemistry and Biophysics and the Lineberger Comprehensive Cancer Center
at UNC Chapel Hill.
John Doench is Associate Director of the Genetic Perturbation Platform at the Broad
Institute and has worked with many Addgenies to help improve the understanding, curation, and explanation of our CRISPR resources. He really likes small RNAs and recently
became a member of Addgenes Advisory Board.
Cameron R. MacPherson
Cameron R. MacPherson is Lead Data Scientist for the Milieu Intrieur project, Institut
Pasteur, Paris. You can follow him on Twitter @CMacPhD.
123 | Page
ACKNOWLEDGEMENTS AND FINAL WORDS (CONT)

Jordan Ward
Kevin Esvelt
Jordan Ward is a postdoctoral fellow with Keith Yamamoto at UCSF, and is currently
studying gene regulation during nematode molting while preparing to transition to his
own independent group. He received his PhD in Biochemistry from Cancer Research UK/
University of London studying mechanisms of DNA repair in the lab of Professor Simon
Boulton. Learn more at his webpage or follow him on Twitter @Jordan_D_Ward.
An inventor of technologies that harness evolution, Kevin Esvelt studies ways of using
molecular tools to alter populations of organisms and associated ecosystems. Because
the environment is shared by everyone, he is deeply concerned with ensuring that gene
drives and other collective technologies are developed safely and transparently. For more
information, visit www.sculptingevolution.org.
Contributions to This eBook Have Been Made by the Following Addgenies Past and Present:
Mary Gearing
Joel McDade
Tyler J. Ford
Mary Gearing is a Science Communications Intern at Addgene currently working on her

Ph.D. in metabolism. You can follow Mary on Twitter at @megearing.
Joel R. McDade is a Senior Scientist at Addgene whose interests include Genome Editing
and Science Education.
Tyler J. Ford is an Outreach Scientist at Addgene. His professional duties include helping
maintain the Addgene blog (blog.addgene.org), talking to people about Addgene, and improving Addgenes services. His non-professional duties include running, biking, drawing,
hiking, playing tennis, reading, and writing. Follow him on Twitter @TyFordFever.
124 | Page
ACKNOWLEDGEMENTS AND FINAL WORDS (CONT)

Chari Cortez
Melina Fan
Marcy Patrick
Nicole Waxmonsky
Kendall Morgan
Chari was previously a Scientist at Addgene. Her primary research experiences are in Immunology, infectious disease, and host cell receptor signaling in the context of initiating
inflammation and host-pathogen interactions. She loves Science, T cells, signaling pathways, and writing about Science!
Melina Fan is a cofounder and Chief Scientific Officer at Addgene. She also cofounded
LabLife, a lab management software company. She loves that her job at Addgene connects her with scientists around the world who are working on everything from cancer to
agriculture.
Marcy is a Senior Scientist at Addgene. She received her Ph.D. in Microbiology and
Immunology from the University of Michigan. She loves that her job allows her to interact
with scientists from all over the world and help them easily share their reagents.
Nicole is an Outreach Scientist and Curator at Addgene. When not riding her bike or
playing broomball, Nicole enjoys using the web to increase scientific sharing and to improve access to basic science information.
Kendall K. Morgan, Ph.D., is a science writer based in North Carolina. She writes about
science, health, and medicine for many organizations and publications with the goal to
help researchers share their work with each other and the world.
125 | Page

CRISPR 101 Ebook Final

Uploaded by

Document Informationclick to expand document informationnh

Document Informationclick to expand document information

Copyright:

Available Formats

CRISPR 101 Ebook Final

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CRISPR 101 Ebook Final

Uploaded by

Copyright:

Available Formats

CRISPR 101:

Created and Compiled by Addgene

CRISPR 101: A Desktop Resource (1st Edition)

CRISPR 101: INTRODUCTION TO ADDGENES DESKTOP RESOURCE

CRISPR 101: A Desktop Resource (1st Edition)

CRISPR 101: Introduction to Addgenes Desktop Resource

Chapter 1: Genome Engineering Overview?

Chapter 2: What is CRISPR?

The PAM Requirement and Expanding Beyond SpCas9

Cpf1: A Cas9 Homolog

Enhancing CRISPR Targeting Specificity with eSpCas9 and SpCas9-HF1

Chapter 3: Using CRISPR in Your Experiments

Which Cas9 Do I Choose for My CRISPR Experiment?

Using Cas9 to Generate Knock-Outs via NHEJ

Editing with Homology Directed Repair

Using Cas9 Nickases for Enhanced Specificity

How to Design Your gRNA for CRISPR Genome Editing

The CRISPR Software Matchmaker

Validating Your Genome Edit

Genome-wide Screening Using CRISPR/Cas9

Regulating Gene Expression with CRISPR

Tagging Genes with CRISPR

Using CRISPR to Visualize the Genome

Chapter 4: CRISPR Expression Systems and Delivery Methods

Mammalian Expression Systems and Delivery Methods

Introduction to Genome Engineering

Planning Your CRISPR Experiment

Multiplexible CRISPR Expression Systems

CRISPR 101: A Desktop Resource (1st Edition)

TABLE OF CONTENTS (CONT)

Using SaCa9 for Easier AAV-Based CRISPR Delivery

Adenoviral Delivery of CRISPR/Cas9 Aims to Expand Genome Engineering to Primary Cells

An Elegans Approach to Better CRISPR/Cas9 Editing Efficiency

Even More Elegant: Single Injection CRISPR/Cas9 in C. elegans

Another Pathway Into Cells: ITOP Protein Only Transduction

Chapter 5: Therapeutic Applications of CRISPR

Treating DMD Mice with CRISPR Gene Editing

Chapter 6: CRISPR Biosafety

CRISPR Concerns Beyond the Lab Bench

Acknowledgements and Final Words

Using CRISPR/Cas9 to Edit Disease Out of the Genome

Transgenic Organisms, Cas9 Gene Drives, and Appropriate Safeguards

Chapter 1 - Genome Engineering Overview

CRISPR 101: A Desktop Resource (1st Edition)

GENOME ENGINEERING OVERVIEW

Chapter 1 - Genome Engineering Overview

CRISPR 101: A Desktop Resource (1st Edition)

INTRODUCTION TO GENOME ENGINEERING

Genome engineering has made it possible to not only

Cre Recombinase Complexed to DNA (1).

Chapter 1 - Genome Engineering Overview

CRISPR 101: A Desktop Resource (1st Edition)

INTRODUCTION TO GENOME ENGINEERING (CONT)

Homologous Recombination and the Journey Towards CRISPR/Cas9

Zinc Finger Nucleases

First reported in 2011, TALENs represented a

Chapter 1 - Genome Engineering Overview