review article the rise of crispr/cas for genome editing in stem...

Review ArticleThe Rise of CRISPRCas for Genome Editing in Stem Cells

Bing Shui1 Liz Hernandez Matias2 Yi Guo34 and Ying Peng3

1Department of Biology Carleton College Northfield MN 55057 USA2Department of Biology University of Puerto Rico Rio Piedras San Juan PR 00931 USA3Department of Biochemistry and Molecular Biology Mayo Clinic Rochester MN 55905 USA4Division of Gastroenterology and Hepatology Mayo Clinic Rochester MN 55905 USA

Correspondence should be addressed to Yi Guo guoyimayoedu and Ying Peng pengyingmayoedu

Received 3 August 2015 Revised 3 November 2015 Accepted 5 November 2015

Academic Editor Jane Synnergren

Copyright copy 2016 Bing Shui et alThis is an open access article distributed under theCreativeCommonsAttribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Genetic manipulation is a powerful tool to establish the causal relationship between a genetic lesion and a particular pathologicalphenotype The rise of CRISPRCas9 genome-engineering tools overcame the traditional technical bottleneck for routine site-specific genetic manipulation in cells To create the perfect in vitro cell model there is significant interest from the stem cellresearch community to adopt this fast evolving technology This review addresses this need directly by providing both the up-to-date biochemical rationale of CRISPR-mediated genome engineering and detailed practical guidelines for the design and executionof CRISPR experiments in cellmodels Ultimately this reviewwill serve as a timely and comprehensive guide for this fast developingtechnology

1 Introduction

Genome-engineering tools facilitate site-specific DNA dele-tions insertions inversions and replacementsThese manip-ulations of the complex eukaryotic genome help researchersunderstand the function of genes in a given cellular contextexplore the mode of gene regulation at the endogenous locusand most importantly model human disease conditionsusing in vitro cellular models or in vivomodel organisms

Since the emergence of designer nucleases based onDNA base recognition by modular protein motifs such asZinc Fingers in Zinc Finger Nucleases (ZFNs) [1ndash3] aswell as TALE domains in transcription activator-like effectornucleases (TALENs) [4 5] site-specific DNA manipula-tions in eukaryotic cells have passed a critical efficiencyand specificity threshold to enable routine applications ina laboratory The recently developed explosively popularCRISPRCas9 (clustered regularly interspaced palindromicrepeatsCRISPR-associated) genome-engineering system hastransformed discovery in this exciting era CRISPRCaswas first discovered in prokaryote adaptive immunity [6ndash8] and has now been more extensively adapted for eukary-otic genome engineering than ZFNs and TALENs [9] The

most widely utilized class the type II CRISPRCas9 systemfrom Streptococcus pyogenes offers users the greatest easeand modularity for design and execution of any genome-engineering experiments [10ndash13] However limitations andcommon practical pitfalls of the CRISPRCas9 system havenot been sufficiently and systematically summarized andemphasized for the emerging population of potential usersin large part due to the great enthusiasm accompanying thesystemrsquos amazing rise in popularity

In this review practical issues associated with the designand execution of a typical CRISPR experiment will be dis-cussed especially in the context of modeling human diseasesusing stem cells Due to the limitation of the current scopethis paper will discuss neither earlier designer nucleases(ZFNs and TALENs) nor applications of CRISPR on modelorganisms although similar rationale and general principlesdiscussed in the following sections would also apply to theseapplications

2 The Discovery of CRISPRCas System

The CRISPR system was first discovered in bacteria as anldquoadaptive immune systemrdquo against plasmids viral DNA or

Hindawi Publishing CorporationStem Cells InternationalVolume 2016 Article ID 8140168 17 pageshttpdxdoiorg10115520168140168

2 Stem Cells International

RNA [6ndash8]This ldquomemory systemrdquo can destroy DNA or RNAif reinfection occurs in the same bacteria or in its descendants[14ndash19]Three types of CRISPR loci exist all of which acquireshort pieces of DNA called spacers from foreign DNA ele-ments [20] Spacers are integrated into the bacterial genomeduring the process of CRISPR adaptation They are usuallyinserted into the CRISPR locus that contains short partiallypalindromic DNA repeats to form loci that alternate repeatedelements (CRISPR repeats)These loci are subsequently tran-scribed and processed into small interfering RNA that guidesnucleases for sequence-specific cleavage of complementarysequencesThrough these stepwise but continuous evolutionsof adaptation CRISPR repeat RNA (crRNA) biogenesis andforeign DNA targeting generated sophisticated CRISPR-based adaptive immune systems in nearly half of the bacterialspecies as well as in most archaea [21]

The sequence in the exogenous nucleic acid element cor-responding to a CRISPR spacer was defined as a protospacer[22] For proper targeting by type I and II CRISPR systemsthe protospacer is usually flanked by a system-specific highlyconserved CRISPR motif namely a protospacer adjacentmotif (PAM) [23] Most PAMs are typically 2 to 5 highlyconserved nucleotides either on the 51015840 end of protospacer(type I system) or on the 31015840 side (most type II systems) Asignificant feature of the PAM for the CRISPR system is todistinguish the foreign DNA against the host genome thusonly the PAM-bearing invading sequence will be targeted fordestruction

3 Different Classes of CRISPRCas

Among the three different types of CRISPR loci type I andIII loci involve a complex panel of multiple Cas proteinsthat form ribonucleoprotein (RNP) complexes with CRISPRRNA to target foreign sequences [15] However the type IICRISPR system uses a much smaller number of Cas proteinsto perform this core function Type II CRISPR loci have threesubdivisions The most commonly used CRISPR system foreukaryotic genome engineering is adopted from a type II Asystem from S pyogenes where a singleCas9 protein (spCas9)is responsible for both forming the CRISPR-RNP complexand subsequent DNA cleavage For the practical reason ofsimplicity most genome-engineering applications use onehybrid RNA (guide RNA gRNA) combining the essentialstructural features of the transactivating RNA (tracrRNA)and crRNA duplex [10] The single-chain gRNA is used herein subsequent discussions

Besides spCas9 a few other orthologous Cas9 proteinsfrom similar type II CRISPR systems share the core featureas the sole protein component for RNA-guided targetingTheCas9 proteins of Streptococcus thermophilusNeisseria menin-gitidis and Treponema denticola demonstrated comparablegenome-editing efficiency to spCas9 (Table 1) [24ndash27] TheseCas9 proteins have different sizes mostly due to their targetrecognition domains (REC) [28] Significantly orthologousCas9 proteins differ in the specific PAM sequences used fortargeting thus they can be used in the same cell when pairedwith their corresponding crRNA to recognize their corre-sponding targets without interfering with each other [29ndash31]

Table 1 Orthogonal type II Cas9 and their optimal PAMpreference

Bacteria PAM CRISPRtype Reference

S thermophiluslowastlowast NNAGAAW(CRISPR1) IIA [28 29]

N meningitidis NNNNGATTNNNNGCTT IIC [28 47

145]T denticola NAAAAN IIA [29]S mutans NGG IIA [47]L innocua NGG IIA [47]L buchneri NAAAAN IIA [47]C jejuni NNNNACA IIC [47]P multocida GNNNCNNA IIC [47]S aureuslowastlowast NNGRRT IIA [31]N cinerea GATlowast IIC [31]C lari GGGlowast IIC [31]P lavamentivorans CATlowast IIC [31]C diphtheriae GGlowast IIC [31]S pasteurianus GTGAlowast IIA [31]

S pyogenes NGG (NAGas minor) IIA [10]

S pyogenes (D1135E)NGG (doesnot recognize

NAG)IIA [35]

S pyogenes VQR(D1135VR1335QT1337R)

NGANNGCG IIA [35]

S pyogenes EQR(D1135ER1335QT1337R) NGAG IIA [35]lowastPutative PAM lowastlowastsignificantly smaller than spCas9 Bottom rows areengineered spCas9 proteins with different PAM preferences

This characteristic enables sequence flexibility of CRISPRexperiments by offering a variety of Cas9 proteins to targetvirtually any particular sequence [25]This orthogonality wasbest demonstrated by recent work that allowed the labelingof distinct genomic regions using different inactivated Cas9-fluorescent fusion proteins simultaneously in a single live cell[32 33] Although most Cas9 proteins from type II CRISPRsystem have one or more optimal PAMs there is also consid-erable flexibility in terms of PAM recognition For examplespCas9 recognizes NGG as its optimal PAM sequence whileNAG can also be recognized with lower frequency ([12]and subsequent) This plasticity might arise from continuousselection pressure on bacterium to target evolving viralsequences [34] In practice this plasticity poses considerablechallenges due to the off-targeted recognition of alternativePAM sequences [12] On the other hand this flexibility allowsfurther engineering of different Cas9 proteins to optimizeor modify PAM preference Initial progress has been madetoward generation of spCas9 with more rigid NGG PAMrecognition and modification of the PAM preferences [35]In a few years further biochemical characterization of nativeorthogonal Cas9 proteins with their PAM preferences andprotein engineering efforts on characterized Cas9 proteins

Stem Cells International 3

will likely generate a full repertoire of Cas9 proteins with highspecificity covering virtually any 2sim5-nucleotide PAMs

A recent important addition to the CRISPR toolbox is thecharacterization of Cpf1 a class II CRISPR effector that isdistinct fromCas9 Cpf1 is a single RNA-guided endonucleasethat uses T-rich PAMs and generates staggered DNA double-stranded breaks instead of blunt ends [36] Its smaller proteinsize and single RNA guide requirement may make futureCRISPR applications simpler and with more precise control

4 Cas9 Enzymology

The Cas9 protein contains two independent endonucleasedomains one is homologous to the HNH endonucleaseand the other one to the RuvC endonuclease (Figure 1)[10] Each domain cleaves one strand of double-strandedDNA (dsDNA) at the target recognition site the HNHdomain cleaves the complementary DNA strand (the strandforming the duplex with gRNA) and the RuvC-like domaincleaves the noncomplementary DNA strand [10] RecentCRISPRCas9 complex structural analysis [37 38] revealeda two-lobed structure for Cas9 a recognition (REC) lobeand a nuclease (NUC) lobe Cas9 interacts with the RNA-DNA duplex using the REC lobe in a largely sequence-independent manner implying that the Cas9 protein itselfdoes not confer significant target sequence preference Onecaveat of the CRISPRCas9 system is that gRNA-loaded Cas9endonuclease cleavage is not completely dependent on alinear guide sequence since some off-target sequences wereshown to be cut with similar or even higher efficiency thanthe designed target sites [12 39ndash42] In general mismatchesbetween the first 12 nucleotides (nts) of the gRNA (seedsequence in gRNA spacer Figure 1) and the DNA target arenot well tolerated suggesting high sequence specificity in thePAM-proximal regionHowevermismatches beyond the first12 nts can be compatible with efficient cleavage (tail regionin gRNA spacer Figure 1) [12] Structural biology insightsinto the Cas9-gRNA RNP complex revealed that the 12-ntsequence is in a fixed ldquoseedrdquo configuration even prior to theDNA substrate binding whereas the 51015840 end of gRNA remainsunstructuredWhile generally true it is an oversimplificationand the sequence recognition specificity of the CRISPR sys-tem is a topic of active investigation [39ndash44] Notably shortergRNA with up to a 5000-fold reduction in off-target effectswas recently described [45] Adding two additional Guanine(G) nucleotides at the 51015840 end of gRNA in some circumstancesmodestly improves the specificity of theCRISPRCas9 system[46] possibly by altering gRNA stability concentration orsecondary structure The relaxation of sequence specificityof the RNA-guided endonuclease system remains the biggestchallenge for its usage in genome engineering A recentbiophysical study [37] for the thermodynamic properties ofCas9 binding provided a likely explanation for the features ofspecificity outlined above and further analyses along theselines will be valuable to further refine design guidelines

A degree of structural flexibility was found from theDNA-gRNA duplex-loaded Cas9 crystallography structure[38] which was substantiated by an independent crystallog-raphy and single-particle electron microscopy study on both

S pyogenes and A naeslundii Cas9 [37] This study demon-strated that a conformational rearrangement is inducedby gRNA binding to Cas9 shaping a central channel toaccommodate the DNA substrate (Figure 1 gRNA binding)[37] Detailed structural information is lacking for howCas9 recognizes targeted sequences within the genome andtriggers the specificDNAcleavage after sequence recognitionHowever the RNA-loaded Cas9 protein reads the PAMin its base-paired configuration (Figure 1 scan for PAM)The recognition of dinucleotide GG in PAM simultaneouslyallows for the local stabilization of the unwound target DNAimmediately upstream of the PAM sequence which mightcompensate for the energy cost of local DNA strand separa-tion starting immediately upstream of PAM (Figure 1 Cas9recognizes PAM) [47] A recent biophysics study for Cas9-mediatedDNA recognition in vitro further revealed that Cas9does not behave as a typical nuclease [48] First gRNA-loadedCas9 enzymatic activity does not follow Michaelis-Mentenkinetics since Cas9 protein stably associates with targetsites on DNA even after inducing a double-strand breakThus the key requirement for successful CRISPR-mediatedgenome engineering is efficient and precise target locatingSecondly gRNA-loaded Cas9 finds the target sequence using3D diffusion without obvious sliding on the DNA substrateCas9 pauses on DNA for interrogation once it recognizes aPAM sequence Many of these reactions are transient and donot lead to DNA cleavage In agreement with this ldquopausingrdquobehavior of the gRNA-loaded Cas9 on the DNA substratein vitro this mode of transient DNA binding on a non-matching target is stable enough in cells to be detected usinggenome-wide CHIP-Seq (Chromatin ImmunoprecipitationSequencing) [43] Besides the highly enriched binding ofCas9 at its on-target site numerous binding events with lowerfrequency can be observed around a short motif of 5sim10nucleotides matching the PAM-proximal region on a gRNAplus NGG PAM sequence [43] Thus these ldquooff-targetedrdquobindings likely involve partial base pairing between gRNAand the PAM-proximal sequence Without intrinsic DNAhelicase activity how Cas9 facilitates the strand replacementon its DNA substrate by the gRNA is not known It issuggested to be a thermodynamically favorable process uponPAM recognition and the unwinding of local DNA basepairing was suggested to be in a directional and sequentialmanner starting at the 31015840 end of the target sequence adjacentto PAM and progressing in the 51015840 direction of the DNAsubstrate (Figure 1 base-pairing extension) [47 48]TheCas9protein likely stabilizes the locally unwound DNA allowingfurther stabilization of the single-stranded DNA chain bycontinuous formation of Watson-Crick base pairing with thegRNA (Figure 1 base-pairing extension) If base pairing isblocked due to a mismatch between the DNA substrate andthe gRNA the thermodynamic energy of the DNA-Cas9interaction might be insufficient to maintain a significantportion of unwound DNA In this case partially unwoundDNA will return to its duplex state and the DNA-Cas9interaction will attenuate simultaneously (Figure 1 mismatchand DNA release) These observations provide an attractivestepwise substrate-unwinding model for target recognitionand cleavage by the gRNA-loaded Cas9 protein This model


Free Cas9

gRNA

Nuclease activated by DNA looping

Double-stranded DNA cleavage

Mismatch between

gRNA and DNA substrate

DNA release

gRNA-loaded Cas9

PAM

DNA

gRNA

Scan for PAM

binding

Cas9 recognizes PAM

Base-pairing extension

PAM

Seed sequence in gRNA spacerTail region in gRNA spacer

HNH domainRuvC-like domain

Figure 1 A proposed model for Cas9 endonuclease to trigger DNA cleavage A conformational change is induced once the Cas9 proteinbinds to gRNA allowing it to search for the DNA substrate The REC lobe of Cas9 scans for the PAM in the genome PAM recognition helpslocal unwinding of dsDNA 51015840 to the PAM region The unwound DNA is transiently stabilized by proteinssDNA interaction Successful basepairing between the ssDNA portion and the gRNA further extends the ssDNA loop A critical loop size may trigger the enzymatic activity ofCas9 to make the double-stranded cut Afterwards Cas9 remains bound to the DNA substrate If the base pairing between ssDNA and gRNAis blocked by mismatches the ssDNA loop collapses to release the Cas9 protein


predicts that only perfectly or nearly perfectly paired DNA-RNA hybrids can lead to significant DNA unwindingupon which Cas9 will cleave both DNA strands (Figure 1nuclease activation and cleavage) This explains the highsequence specificity in the PAM-proximal region observedfor CRISPR-mediated gene editing [49] as well as the recentfinding that off-targeted Cas9 binding through the beginningof the PAM-proximal sequence only rarely leads to off-targeted enzymatic activity in vivo [43] Because unwindingthe DNA duplex across the first-10sim12-nt preconfigured seedsequence might be the critical thermodynamic hurdle toestablish stable Cas9 interaction with DNA and subsequentcleavages a high degree of sequence fidelity in this seedsequence might be both sufficient and necessary via strandreplacement to trigger Cas9 conformational changes andremodeling of the active sites In theory based on this modelthe mismatch of a DNA-gRNA hybrid occurring closest tothe PAM sequence should be the least tolerated and is indeedthe least common amongobserved off-targeted bindings [43]Further thermodynamic modeling based on this model andstructural information will likely improve both the efficiencyand specificity of CRISPR applications

5 On-Target and Off-Target Considerations

Similar to most other engineering applications specificityand efficiency are the main factors ensuring a ratio-nal CRISPR-experiment design In subsequent discussionsspecificity is defined as the probability that Cas9 will targetthe designed locus compared to other undesirable loci (off-target effects) Efficiency is defined as the probability that thelocus of interest will be modified by Cas9 nuclease in thecontext of a pool of available target chromosomes from thecell population In a word vigorous CRISPR design tendsto minimize the off-target effect and maximize the on-targeteffect of the designer nuclease to achieve both high specificityand efficiency

The 18sim20-nt spacer region designed as the protospacersequence in the gRNA is the main determinant for both off-target and on-target effects of CRISPR experiments Togetherwith a given adjacent PAM sequence a gRNA with a 20-ntprotospacer region can achieve in theory unique sequencerecognition in a random sequence space of roughly 17 TB(tera-base pairs) if a perfectly base-paired match is requiredfor targeting While this theoretical upper limit of resolutionexceeds the size of most eukaryotic genomes the practicalspecificity of Cas9 was found to be magnitudes lower thanthe theoretical expectation It was discovered that the ldquoNGGrdquoPAM sequence requirement of spCas9 was not absolutelynecessary since a ldquoNAGrdquo PAM is frequently tolerated with alower efficiency [12] The scientific community also quicklyrealized since the onset of development of CRISPR genomeengineering that mismatches between the protospacer andtargeting DNA are tolerated at a surprisingly high frequencyespecially for the 51015840 sequence of the protospacer [41 4244 50] Further elucidation of Cas9 enzymology revealedthat this bias might be due to the unidirectional (31015840 to 51015840)DNA double-strandmelting coupled with DNA-RNA duplexformation upon PAM recognition by Cas9 nuclease While

the gross 31015840 to 51015840 relaxation gradient of the base-pairingrequirement of Cas9 targeting generally holds true it wasfound that sometimes sequences with mismatches to the12-nt seed sequence in the gRNA spacer can be efficientlytargeted [39 41 42] This suggests that proper base pairingwith the gRNA seed sequence alone does not guaranteespecificity Furthermore targeting efficiency at some off-target sites could be even higher than the desired locus withperfectly matched spacer-protospacer sequences [39 41 42]This phenomenon might be caused by additional factorsbeyond the RNA-based sequence recognition used by Cas9nucleases

Compared to the considerable knowledge for the basisof Cas9 off-target effects relatively little is known abouthow to design a gRNA to make the desired targeting eventmore efficient Multiple factors determine the success of anygiven CRISPR experiments such as the quantity of Cas9proteins and gRNA chromatin accessibility of the targetingloci and cellular response to CRISPR-induced DNA lesionsMost of these issues are beyond experimental controls whena CRISPR experiment is designed A few recent studies [51ndash53] attempted to debug the sequence preference of effectivegRNA by retrieving the successful targeting gRNA sequencesin a large randomly selected gRNA pool This statisticalapproach is limited by current capability to generate a gRNApool with sufficient diversity and the difficulties avoidingartificial bias when selecting the efficiently targeted cell poolsNevertheless a few statistically significant rules have beenrevealed by these pioneering studies on common traits ofefficient gRNA for spCas9 (a)Guanine (G) is strongly favoredat the 31015840 position most proximal to the PAM sequence(especially the minus1 position) This preference might be due toCas9 loading [51] (b) A series of thymine (T) is disfavored atthe four positions (minus1 to minus4) closest to the PAM which mightbe related to the fact that RNA polymerase III recognizesa series of uracil (U) as a pausingtermination signal [54]causing a lower level of gRNA expression [51] (c) Cytosine(C) is preferred at the DNA cleavage site (minus3 position) (d) Inthe PAM region the +1 position favors C while disfavoring T[52] (e) The CRISPR activity correlates with gRNA stabilitywhich can be influenced by the nucleotide composition ofthe spacer G-rich spacers are more stable especially whencomparing with A-rich ones [55]

The emerging gRNA design rationale discussed abovewas continuously incorporated into available bioinformaticstoolboxes as weight matrices for calculating the off-targetor on-target scores for any gRNA [52 55ndash59] Althoughthese scores are informative in facilitating the experimentaldesign process potential CRISPR users should be cautiousabout interpreting gRNA ranking based on these scoressince it does not necessarily indicate superior specificity andefficiency

6 CRISPRCas9 Delivery Methods

As an efficient RNA-guided specific gene-modification toolCRISPR was widely used in many experimental settingsto achieve desired mutations However the delivery of therequired Cas9 protein and gRNA is a long-standing challenge


[60]Three methods of CRISPR delivery including plasmidsviruses and ribonucleoproteins (RNPs) were shown tosuccessfully introduce Cas9 and gRNA into target cells andaccomplish guided gene editing [11 49 61]With their variousmerits and limitations these three delivery methods offerresearchers an opportunity to optimize their gene-editingprocedures based on various experimental needs

61 Delivery Using Plasmid Vectors Delivery using the plas-mid vector system is the conventional and most popularmethod for CRISPR introduction It has the main advantageof being simple to make in vitro In order to introduce afunctional CRISPR system into target cells cells need to betransfected with plasmids encoding the Cas9 protein crRNAand tracrRNA while simultaneously using electroporation orcationic lipid-mediated delivery to achieve assembly of theCRISPR complex in cells [11]

The plasmid system procedure was continually simpli-fied and its application range expanded to in vivo animalstudies Instead of cloning three different plasmids encodingthree different components researchers showed that plasmidencoding gRNA a fusion transcript of crRNA and tracrRNAis sufficient for Cas9 binding andDNA target-site recognition[10] Recently plasmids encoding both Cas9 and gRNAbecame commercially available Therefore transfection ofa single plasmid is the sole requirement for a CRISPRexperiment Multiplex edition of target loci can be accom-plished through simultaneous introduction ofmultiple gRNAspecies by a single plasmid or by cotransfection of multipleplasmids [13] Plasmid delivery was also applied in a tissue-specific CRISPR application inmurine liver [60 62]Throughhydrodynamic tail-vein injection plasmids were efficientlydelivered to sim20 of hepatocytes for transient expressionThis study demonstrated successful gene editing with limitedefficiency in vivo through direct plasmid delivery

However compared to successful delivery in vitro theplasmid delivery system still faces significant challenges for invivo applications such as low delivery efficiency and frequentepigenetic silencing on episomal DNA [63] Converselyplasmid delivery offers the dual possibility of both long-termand transient CRISPR delivery in vitro In a small proportionof transfected cells random but stable integration of all orpart of plasmid DNA into the host genome occurs Thisis possibly due to low levels of spontaneous DNA damagewhich in turn provide continuous Cas9 and gRNA sources[11 49 61 64] When this feature is not desirable deliveredplasmids usually become diluted and gradually lost over a fewcell cycles This limited time window of genome engineeringis critical for obtaining genetic homogenous cell populationsfor downstream functional studies

62 Delivery Using Lenti- Adeno- and Adeno-AssociatedViral Vectors The plasmid system introduces CRISPR intoestablished cell lines efficiently However to expand CRISPRrsquosapplication range viral vectors are used to deliver CRISPRinto primary cells or cells refractory to plasmid transfectionLentiviral vectors stably integrate into the host genomemaking it the preferred means of delivery if the targetinginformation needs to be retrieved after functional selection

processes [51 65ndash67] It is now feasible to carry out genome-wide CRISPR-based functional genomic screens by deliver-ing complex pools of CRISPR reagents into a relevant celltype via lentiviral packaging One significant limitation oflentiviral-based delivery is that the random integration of aviral genome may cause unwanted insertional mutagenesisat undesired host loci Use of nonintegrating viral vectors(NIVVs) including adenoviral vectors and adeno-associatedvectors can efficiently circumvent this problem because theydo not incorporate viral DNA into the host genome [1160] Moreover viral DNA dilutes during mitosis due to thelack of a replication signal [60] Among NIVVs adenoviraland adeno-associated vectors are both potentially suitableCRISPR delivery candidates because of their episomal naturelarge cloning capacity high-titers capability of long-termin vivo expression and ability to transduce many cell lines[39 49 61 62]

While a viral vector encompassing Cas9 and gRNAexpression cassettes can be produced at high-titers thenegative correlation of packaging efficiency versus vectorsize also poses challenges for single-vector delivery of bothCas9 protein and gRNA Successful gene editingwas achievedusing adenovirus-delivered CRISPR in multiple mammaliancells Using different gRNA and Cas9 virus concentrationsresearchers showed that the editing efficiency is dosagedependent [10 61] Besides transfection of stable cell linesadenoviral vector-mediated CRISPR delivery can also beapplied in vivo Through tail-vein injection adenovirusescarrying Cas9 and gRNA expression cassettes can beintroduced into murine liver Resulting Cas9-mediated geneediting is stable even after extensive regeneration of livertissue [13 68] Compared to hydrodynamic tail-vein injectionof plasmids tail-vein injection of adenoviruses achieved 5-to 8-fold greater editing frequency [69] This high efficiencymakes virus-delivered CRISPR an attractive option for invivo genome modification However systematic deliveryusing the adenovirus vector in vivo could induce immuneresponses that eliminate infected cells and eventually impairCRISPR genome-editing efficiency In one recent studyusing adenoviral vector delivery the transduction rate ofliver cells drops from 808 one day after injection to 14fourteen days after injection This is most likely due to theimmune response of the host including elevated expressionof inflammatory cytokines [31 69] In contrast the adeno-associated virus (AAV) induces a mild immune response invivo and can provide long-term expression in nondividingcells The recent study using Staphylococcus aureus Cas9(SaCas9) solved the viral packaging limit problem for spCas9making the AAV-mediated delivery an ideal method for invivo genome editing [31]

63 Delivery Using Cas9-gRNA Ribonucleoproteins (RNPs)In addition to plasmid vector and viral vector deliveryCRISPR delivery using Cas9-gRNA RNPs is another estab-lished method [64] Both plasmid and viral delivery encoun-tered the problem of high off-target editing rates due toprolonged expression of Cas9 and gRNA in cells Using directdelivery of RNPs can effectively circumvent this problemWhen injected directly into cells RNPs induce editing at


target sites immediately after delivery and degrade rapidlyreducing off-target effects [70 71] Additionally using RNPsavoids the possibility of undesired DNA integration into thegenome due to its DNA-free mode of delivery

Application of RNP delivery led to successful genomeediting in multiple human cell lines [64 72] The RNPcomplex can be readily made through incubating in vitropurified Cas9 protein with either a single-chain guide RNA(sgRNA) or dual RNA that consists of crRNA and tracrRNAUnder certain circumstances dual RNA was shown to bemore effective than single gRNA [73] Direct injection ofRNP complexes into cells can lead to efficient CRISPR-mediated genome editing with high specificity and low off-target rates compared to plasmid delivery [64] RNPs aretraditionally delivered by direct microinjection in a low-throughput manner Recently the feasibility of transfectingCRISPR RNPs into cells efficiently using electroporation wasdemonstrated [72] as well as using cationic lipid-mediatedliposome delivery [74] Delivery of RNPs into cell-cycle-synchronized cells also yielded a significantly higher rateof editing compared to delivery in nonsynchronized cellsMore importantly researchers can maximize the utilizationa particular mode of double-strand break (DSB) repairby delivering RNPs into cells arrested at a particular cell-cycle phase [72] Continual improvement of RNP deliverymakes it a prominent method for not only gene editingin an experimental setting but also clinical gene therapydevelopment

7 CRISPR Efficiency Test

71 Test of Indel (Local Point Mutation Insertion and Dele-tion) When assembled with gRNA Cas9 nuclease cleavesdsDNA and induces DSBs DSBs can be repaired by eithernonhomologous end joining (NHEJ) or homologous recom-bination (HR) NHEJ is an error-prone process that generatesrandom insertion or deletion (indel) mutations at the DNArejoining sites Sanger sequencing is the most accurate wayof confirming indel mutations (Figure 2(a)) However dueto the random nature of indels a wide variety of mutatedDNA might be present after a CRISPR-induced NHEJprocess Separating these molecule species using molecularcloning coupled with Sanger sequencing is time-consumingand cost-inefficient [75] Recent progress in bioinformaticstools (such as TIDE Tracking of Indels by DEcomposition)enabled successful digital decoding of Sanger sequencingfrom a mixture of complex indels generated by a uniqueCRISPR-targeting event into separate mutant species [76ndash78] Although this method is still of limited sensitivity andremains to be validated on a larger scale Sanger sequenc-ing of a locally amplified targeted locus offers a quickand reliable readout confirming the efficiency of any givenCRISPR experiment Without sequencing the separation ofDNA with minor differences of length (resulting from someindels) on a Sanger sequencer can be used to quickly accessthe success of a genome-editing experiment IDAA (IndelDetection by Amplicon Analysis) was recently developed tofill this niche [79] Through the use of target-specific primersflanking the target site the different sizes of amplicons can be

detected [79] Furthermore several other methods that takeadvantage of NHEJ-induced indels were developed to effi-ciently assess the cleaving efficiency of CRISPR through thedetection of indel mutations at target loci regardless of DNAlength change these include the Surveyor nuclease assay theT7 Endonuclease I (T7E1) assay the High ResolutionMeltingAnalysis (HRMA) and PAGE electrophoresis [80ndash85]

Surveyor T7E1 and other nuclease-based mutationdetection assays rely on the formation of a locally mis-matched heteroduplex DNA a byproduct of sequence vari-ation caused by NHEJ following the designated nucleasetarget (Figure 2(b)) If CRISPR-mediated cleavage is suc-cessful indels will be generated at the DSB sites throughNHEJ Heteroduplex DNA can be formed after melting andrehybridizing mutant and wild-type alleles The mismatch-recognizing enzymes such as Surveyor and T7E1 nucleasescan detect heteroduplex DNA Bacteriophage resolvase T7E1recognizes and cleaves distorted dsDNA undergoing confor-mational changes [86] Surveyor nuclease is a single-strandednuclease that recognizes a nucleotide mismatch induced byindels It not only cleaves DNA one strand at a time on the31015840 end but also contains 51015840 exonuclease activity [87 88] Bothenzymes recognize indels and induce DSBs at mismatch sitesresulting in shortened DNA fragments of various sizes Thedigested DNA fragments can then be visualized using gelelectrophoresis or DNA fragment analysis [82 88] Howeverboth enzymes exhibit low levels of random single-strandednuclease activity leading to unspecific cleavageThis problemcan be partially resolved through addition of Ampligaseduring the enzyme nuclease reaction [89] which reduces thenonspecific nuclease activity

HRMA is another tool for indel detection utilizing thedifferent denaturation profile of heteroduplex DNA com-pared to that of homoduplex DNA (Figure 2(c)) [90] IfCRISPR-induced indel is present in template DNA het-eroduplex and homoduplex DNA will be formed aftermelting and rehybridizing mutant and WT alleles Differ-ent duplex species exhibit different denaturation patternsHRMA records the temperature-dependent denaturationprofile of the sample and determines the existence of het-eroduplex DNA based on different melting patterns from thesamplemixture Due to its sensitivity HRMA requires properoptimization of PCR conditions to ensure high specificity oftarget amplification

The polyacrylamide gel electrophoresis- (PAGE-) basedmethod was recently proven to be efficient in detectingthe presence of heteroduplex DNA (Figure 2(d)) [85] Thismethod takes advantage of the migration speed differencebetween heteroduplex and homoduplex DNA during nativePAGE Heteroduplex DNA generally migrates at a muchslower rate due to its indel-induced open angle betweenmatched and mismatched DNA strands and therefore can bevisualized using PAGE However whether the PAGE assayprovides sufficient sensitivity across the spectrum of indelmutation variation remains to be verified

72 Sensitivity Issues and Reporter While CRISPR is con-sidered an accurate genome-editing method the efficiencyof CRISPR varies significantly when applied to distinct loci


220210200190

Wild-type sequenceMutant sequencegRNA

PAM

T C C G G A A C A A C C T T A T T A G T A G G A T A G C C C C A G G T

G A T AG C C C C A G G T G C C

(a)

Wild type

Mutant

Mismatch

Denaturation and annealing

Surveyor orT7E1

Full length

Cleaved

Cleaved

6000

3000

1000

700600500

400

300

200

100

Undigested SUR T7E1

(b)

Temperature

Fluo

resc

ence

Wild typeWild typemutantMutant

(c)

Wild type Mutant

Homoduplexes

Heteroduplexes

(d)

Figure 2 Major methodologies for mutation detection (a) Sequence decoding from Sanger sequencing An example of a Sanger sequencingread was shown to illustrate the significant decrease of read quality from the predicted CRISPR cut site (PAM position labeled by magenta)This is due to the inclusion of the mutated DNA (decoded as the bottom sequence) with the wild-type DNA sequence (decoded as thetop sequence) Underlined sequence reveals identical nucleotides between the wild-type and mutant sequences which indicates the majormutation is a 3-nucleotide (TAG) deletion (b) Recognizingmismatched dsDNAusing the single-stranded specific nucleasesMixed sequenceswith local sequence polymorphisms (CRISPR-induced indel mutations) form amismatch when rehybridizingThe result from themismatch-recognizing nuclease assay is visualized using fragment analysis as a digital nucleic acid size profile (c) High Resolution Melting Analysis(d) PAGE electrophoresis of a DNA hybrid


and different cell types In induced pluripotent stem (iPS)cells and human embryonic stem cells (hESCs) for exampleCRISPR-editing efficiency frequently drops below 1 [91 92]This low frequency increases demand for more sensitive raremutation detection methods Sanger sequencing is the goldstandard for determining on-target edition efficiency yet it isa time- and resource-consuming processWhen themutationrate falls below a given threshold (usually sim1) routinemutagenesis detection methodologies (Sanger sequencingnuclease-based heteroduplex cleavage assay HRMA andPAGE) are of limited use due to their sensitivity restraintsHigh-throughput sequencing was developed for accuratemeasurement of rare indels that happen at a frequencyof 001ndash1 However because this method is considerablymore sensitive than traditional methods (such as mismatch-recognizing enzymes) the false-positive frequency is alsoelevated [75]

Single molecule real-time (SMRT) DNA sequencing wasdeveloped as a unique high-throughput sequencing platform[93] It has the advantage of both high sensitivity and longreading length A regular PCR amplified region of interestis ligated with SMRT adaptors to create a single moleculeSMRTbell template to generate sequence reads This methodnot only examines the existence of an editing event butalso quantifies the frequency of editing through either NHEJor HR With an average sequencing length of 3 kb and upto 15 kb SMRT sequencing provides a reliable method forassessing both on-target and off-target rare editing effectsSimilarly other high-throughput sequencing platforms canbe applied to quantitate indels in the targeted amplicon

To further assess CRISPR-editing efficiency using accu-rate quantification for very rare editing events digital dropletPCR (ddPCR) can be applied to CRISPR-edited genometesting [94] Depending on the assay format ddPCR assayhas theoretical mutation detection limits in the range of001sim0001 To achieve individual assessment of the editedgenome sample DNA is partitioned into small dropletsthrough emulsion One set of primers flanking the regionof interest and two competitive fluorescence-tagged probestargeting wild-type and mutant sequences respectively areincluded in the reaction An individual PCR reaction iscarried out in each droplet and fluorescence signals fromeach droplet are subsequently recorded The wild-type andmutant sequences are differentiated and the frequency ofediting can be calculated based on the number of dropletswith different fluorescence signals [91] This method allowsextremely sensitive detection of rare mutations as well asaccurate quantification of CRISPR-editing efficiency NovelddPCR application was explored in other studies includingdifferentiating wild type and mutants based on the size ofamplicons using the nonspecific double-strandDNAbindingdye EvaGreen (EG) [95]

Besides quantifying CRISPR-induced indels live report-ers based on HR can be used to visualize CRISPR activityTypically a reporter plasmid vector can be designed toinclude the identical target-site sequence as the targetinglocus The CRISPR target is flanked by two separate halves ofa fluorescent protein reporter with a stretch of an identicalsequence included in both halves Thus this reporter is

inactive since the fluorescent protein gene is interruptedby the inserted sequence CRISPR components and thereporter plasmid are cotransfected Efficient gRNA loadsCas9 to cleave both the chromosomal targeting locus andthe episomal reporter-targeting site In the reporter the DSBwill be repaired through HR between the two halves ofthe fluorescent protein thus rendering a fully functionalfluorescent protein Hence the ldquoonrdquo status of the reporterplasmid exhibited by the gain of the cellular fluorescencesignal can give a real-time readout of CRISPR efficiency inlive cells independent of additional molecular assays

8 Selection of Mutant Clones

Pure clonal isolation from a single progenitor cell is acritical step in the genetic and functional characterizationof mutations achieved by the CRISPRCas9 system Whileit is usually the most laborious and time-consuming stepin CRISPR-based genome engineering using cell modelsgenerating clonal mutant cell lines is absolutely required todraw any solid conclusions correlating a given mutation andcellular behavior Each single cell upon the introductionof activated Cas9 nuclease is an independent unit thatundergoes stochastic genetic changes dependent on both thenuclease-induced DNA lesion and the subsequent cellularDNA-repair response In the case of transient introduction ofCRISPR agents it is desirable to establish clonogenic culturesby the conclusion of CRISPR action In the stem cell researchfield a clonogenic culture is frequently confused with thesphere generating culture such as formation of embryonicbodies from ES cells or neurospheres from neuronal stemcells [96] While these sphere-forming assays are frequentlyused to estimate the capability of stem cells to self-renewand differentiate the individual spheres formed in standardstem cell culture conditions do not necessarily rise fromsingle cells [97] since sphere aggregation and fusion werefrequently found even at low seeding densities [98ndash100] Therequirement of clonogenity after CRISPR action usually callsfor more rigorous culture conditions to ensure proper clonalseparation of distinct isogeneic pools

There are multiple methods to achieve clonogenity Toprevent sphere fusion single cells can be encapsulated intoa semisolid matrix to form embedded sphere cultures [101]This approach greatly improves the clonogenity of the spheresgenerated and offers greater advantagewhen cell proliferationis strictly dependent on high cell density in the culture[98] However single-cell encapsulation usually requires spe-cific microfluidics devices [102] Furthermore maintainingcapsule integrity and retrieving encapsulated cells remainchallenging Aside from cell encapsulation cells grown insemisolid media such as those containing methylcelluloseor soft agar are less likely to migrate [103] When seeded atlow density single cells in semisolid media can grow intoindividual colonies over time Manual or robotic selectionof these colonies can subsequently establish isogenic clonesThe traditional labor-intensive ways to establish culturesfrom single cells include cloning rings serial dilution andplating and fluorescent-based single-cell sorting [104 105]Regardless of the methodology establishing andmaintaining


a large number of isogenic cell clones are costly and labor-intensive For most genome-engineering experiments theoptimally desired approach should minimize the number ofisogenic cell clones needed to achieve the desired geneticmodification In the following sections the factors to achievethis goal will be discussed

81 Overall Strategy NHEJ or HR DSBs in the eukaryotegenome can be repaired mainly by two different mecha-nisms NHEJ or HR The NHEJ repair mechanism joinsbroken chromosomal ends directly without the guidance of ahomologous sequence Because it lacks a reference templatethis repair pathway is usually error-prone due to local DNAsequence alterations at the repaired junction (the so-calledindels) [106] In contrast the HR repair mechanism is aidedby using a homologous sequence as the repair template Thishomologous sequence can be a sister chromatid duplicatedduring the synthesis (S) phase of cell cycle the homologouschromosome in diploid cells or foreign DNA introducedbearing regions of sequence homology with the targetedlocus Due to the flexibility of donor choice in HR repaira given locus with desirable features (such as restrictionenzyme recognition sites protein fusion tags antibioticselection markers or recombination sites) can be engineeredby incorporating these features with a piece of introducedhomologous DNA Either plasmid construct or synthesizedDNA oligos can be used as the donor template [40] Aplasmid donor can be used when long insertions need tobe introduced [107 108] For small insertions or deletionssingle-stranded DNA containing 80 bp homologous arms at51015840 and 31015840 ends is preferred [107] This method is similarto traditional HR-based gene targeting However since theintroduced DSBs occur in the chromosomal DNA instead ofepichromosomal DNA the HR efficiency is usually severalorders of magnitude higher than traditional HR triggered bybreaking the foreign donor [3 108ndash111]

While the choice of DNA-repair pathways is largelybeyond experimental control the cell-cycle phase uponwhich DSB occurs plays an important role in repair mech-anism determination In general HR takes place in thesynthesis (S) and the premitotic (G2) phases when there aresister chromatids available [112] NHEJ is the predominantrepair mechanism in the growth 1 (G1) and the mitotic (M)phases [113] Although this general guideline holds true inmost cases precautions are warranted for any particular celltype for its capability on HR- or NHEJ-based DNA-repairpathways

Regardless of the preferred DNA-repair mechanisms toget a particular or a range of desired mutations similarclonogenic selection processes are needed Since HR usuallyhappens at a lower frequency than NHEJ for most cell typesit is an efficient strategy to include a selection marker on thedonor construct so that successfully engineered cells can beeasily traced by fluorescence or drug resistanceThemarker isintegrated onto the targeted loci In some cases this feature isnot ideal for downstream functional analysis even when themajority of the selectionmarkers can be subsequently excisedby recombinases

A few seamless genome-engineering applications emergedin the last few years to overcome this hurdle This elegantapproach aims to introduce only the desired genetic modifi-cation without leaving additional footprints at the engineeredloci (including indels at the CRISPR cut sites any selectionmarkers or short residual recombination sites after markerexcision) (Figure 3) [24 114 115] To facilitate clonal selectiona selection marker is included in the DNA donor similarto traditional HR However instead of using a recombinaseto induce flanking recombination sites around the markerwhich would leave behind at least one recombination site(Figure 3(a)) an optimized PiggyBac transposon is used forall exogenous sequences between the homology arms Only aldquoTArdquo dinucleotide sequence is left on each side flanking theexiting PiggyBac (Figure 3(b)) To make this truly seamlessthe left and right homology sequences start with a ldquoTArdquomotif which is abundant in most genomic loci If there is noendogenous ldquoTArdquo around the intended mutation it is usuallyfeasible to introduce one without changing the translatedprotein sequence in exons or make this change in mutation-tolerating introns A negative selection marker is usuallyincluded in the PiggyBac cassette in the designed DNAdonor to facilitate screening the loss of the PiggyBac cassetteby the transposase This method holds great promise forCRISPR-mediated site-specific gene therapy since avoidingany additional sequence modification is highly desirable

Regardless of the choice of methods clonogenic cloneisolation and identification are labor-intensive To design themost effective screening strategy it is crucial to realisticallyestimate the chance of obtaining the desired mutant cells inthe pool undergoing CRISPR-mediated genome engineeringA critical factor is the efficiency of CRISPR targeting thelocus of interest which can be tested by a small-scale pilotexperiment using the mutation detection methodologiesdiscussed in the previous section Depending on the modeof DNA-repair pathway chosen further consideration can bemade regarding whether it is feasible to first reduce the sizeof the cell pool by selection to enrich the targeted cells beforeclonal assay Isolating cells positive for the HR-mediated live-cell cleavage reporter could enrich NHEJ-mediated indelmutations [116] Although these are achieved by differentmechanism of DNA repair the reporter assay may indicatethe subpopulation of cells where CRISPR is more activeSimilarly if the desired mutation was introduced using HRrepair inclusion of the selection marker in the DNA donorcould be an efficient way to reduce the size of clonal selectionpool Frequently the intended mutation might be predictedwith high confidence to cause a specific cellular phenotypein the target-cell type If the specific cellular phenotype canreliably be used for selection target-cell enrichment can beachieved by applying this selection pressure [117] Withouthighly efficient CRISPR reagents a target selection schemeis required to move the mutation frequency above 01 inorder to make clonal single-cell selection feasible

In cases of low mutagenesis frequency and no suitableselection strategy available for mutant enrichment a randomcell partition scheme named sib-selection can be employedto facilitate enrichment of the desired mutation before clonal


3998400 Hom5

998400 Hom

Genome region

Edited genome region with footprint

Crerecombinase

HRDonor plasmid

MarkerloxP loxP

loxP loxPMarker

loxP

(a)

3998400 Hom5

998400 Hom

PiggyBactransposase

Genome region

Edited genome region without footprint

HRDonor plasmid

PiggyBacTA

TA

TA

TA

TA

TA

Marker

Marker

(b)

Figure 3The comparison of seamless genome editing with traditional HR-based marker selection (a) Traditional HR (b) Seamless genomeediting Homology arms (dark grey and light grey boxes) bearing the desired mutation (red bar) are used to flank an excisable selectionmarker cassette This is achieved by using the tandem loxP sites as in (a) and a PiggyBac transposon as in (b) Successful HR will insert theselection marker cassette into the genome (middle panels) Removing the loxP cassette with Cre recombinase will leave one loxP site at thelocus of interest (blue triangle) in (a)The remobilization of the PiggyBac transposon will only leave a ldquoTArdquo dinucleotide in (b) which initiallycan be found in the locus of interest or can be tolerated without any undesired changes to the protein sequence

isolation [91 118] Sib-selection is based on precise measure-ments of mutation frequencies in pools of cells even when therate is extremely low The ddPCR method was used for thispurpose to gain a reliable quantitative mutation rate Whena pool of cell mixtures with a rare mutant is sequentiallypartitioned randomly into smaller pools (such as differentwells in a 96-well plate) the mutation rate in one or afew small pools will increase significantly due to the overallsignificant decrease of cells in a pool following a Poissondistribution The capability to locate these enriched wellsusing a quantitative mutation measurement can facilitateserial pool partition and mutant identification until the rateof desiredmutants surpasses the practical threshold for clonalidentification Although a powerful and quick way to enrichmutation sib-selection is not a clonogenic process per seThus subsequent clonalmutant strain identification is neededto isolate the intended mutant cell

82 Estimation of Off-Target Mutations in Isolated Cell ClonesAcquiring pure cell populations with the desired geneticmodifications should not be considered as the final stepbefore using these cell models for functional studies Nomatter how carefully the experiment was designed it is likelythat some off-target modifications were introduced into thecell pool by CRISPR If any of these are carried on into thefinal selected clones these additional genetic modificationsmight complicate further functional analysis

Whole genome sequencing of the isolated cell clonesremains the most rigorous standard to estimate the off-target lesions [119ndash121] It remains expensive especially forhuman cells since the complete genome requires a significant

sequencing depth to detect the occurrence of low frequencyindels While its costs prohibit routine use to examine all iso-lated cell clones in a typical lab a reasonable approximationcan usually be made by targeted sequencing of predicted off-target sites This can be done in a low-throughput mannerusing PCR and Sanger sequencing of a number of individualpredicted off-target sites with significant targeting prob-ability Alternatively multiplexed next-generation targetedsequencing can be achieved by covering a large number of off-target sites simultaneously from multiple single-cell cloneswith significant sequencing depth [46 122] In the case oftargeted sequencing the choice of examined genomic regionbecomes critical While various in silico platforms give arough estimate of potential off-target sites recent advanceson genome-wide breakpoint sequencing technology (suchas CHIP-Seq [43 122] Digenome-seq [123] and GUIDE-seq[124] and genome-wide translocation sequencing [125]) offera more realistic range of potential off-target sites in any givengenome While these platforms collectively can aid targetedgenome sequencing of the engineered cells precautions arestill warranted since off-target CRISPR targeting can beinfluenced by the different cell types used and minor differ-ences of genome sequence [126] Some additional practicalprecautions should be taken into consideration especiallywhen the undesirable off-target lesions are not sufficientlycharacterized or hard to avoid

83 Correlating Phenotype and Genotype Controls Whena certain phenotype is displayed after CRISPR-mediatedediting in the clonogenically isolated mutant cells the phe-notype is not necessarily caused by the intended target due


to the possibility of poorly characterized off-target lesionsThe genotypephenotype association can be strengthenedby verification using additional clonogenic clones carryingindependentmutations generated by different CRISPR agentstargeting the same locus Because identical off-target lesionsmight be generated by the same gRNA it is not possible tostrictly exclude this possibility by relying on additional clonesgenerated by a single gRNA Therefore additional gRNA isdesired to target the same region of interest to achieve theidentical phenotypic outcome With limited overlapping ofoff-target sites multiple gRNAdesigns ensure that any sharedphenotype exhibited after editing using all gRNA correlateswith the genotype of interest with high confidence Asidefrom establishing proper controls for CRISPR targetinggenetic rescue is considered the gold standard to formallyestablish the causal relationship between phenotype andgenotype For loss of function mutations introducing theintact target genes or gene products into the engineeredcells should serve the purpose Introducing the gene ofinterest back into the endogenous engineered locus is readilyachievable by CRISPR [127ndash129] and is preferable since therescue genetic material is under endogenous transcriptionalcontrol In the case of gain-of-function mutations wheregenetic rescue is difficult to achieve pharmaceutical geneticapproaches are useful in functional validations Fine-tuningthe functionality of a given target or relevant pathways usingwell-characterized specific drugs could provide indepen-dently supported evidence

9 A Much Brighter Future forStem Cell Models

The accumulation of large-scale human genome-sequencingefforts in the past few years greatly accelerated geneticdiscovery by linking genetic variations discovered in humanpopulations or disease-associated somatic tissue to a diseasestate Stem cell models on the other hand are traditionallyextremely powerful in establishing the mechanistic linkagebetween genotype and phenotype The recent explosionof applications of CRISPRCas9 genome-editing techniquesnow establishes the causal relationship between genotypeand cellular behaviors with great flexibility and efficiencyWhile our current review can grasp neither the full extentnor the rapid evolution of these applications a few prominentexamples are highlighted below to demonstrate the range anddepth of these applications

One of the earliest successful applications of CRISPRin stem cell research was to correct the CTCF mutationin cultured intestinal stem cells from cystic fibrosis (CF)patients [130] Besides fixing local sequence errors CRISPRwas recently used to correct a chromosomal structural abnor-mality (a chromosomal inversion over a several-hundred-kilo-base-pair) associated with Hemophilia A [131] Usingstem cell models (especially patient-derived iPSCs) CRISPRwas used to correct more than a dozen disease-associatedgenetic lesions across a wide spectrum [115 130ndash143]including metabolic disorders immunological deficienciesand neuromuscular disorders These genetically corrected

patient-derived stem cells might be the critical vehicle forfuture cell and gene therapies with further improvement onits safety

Regardless of its therapeutic potential CRISPR is aninvaluable tool in establishing the causal relationship betweengenes and stem cell behavior Clevers group recentlymodeledthe occurrence of the 4 most frequent mutations identifiedin human colorectal cancer within the context of a humanintestinal stem cell organoid culture This analysis enabledthem to pinpoint the driver mutations causing extensiveaneuploidy within this cancer stem cell model [117] CRISPRalso helped to pinpoint a specific single-nucleotide polymor-phism (SNP) in the human FTO locus as the critical effectorfor obesity [144] Previous genome-wide association studiesindicated the FTO region harbors the strongest geneticassociation with obesity while no mechanistic associationcould be drawn A SNP in the FTO locus was furthernailed down as the obesity-causing variant Modeling theconversion of this one nucleotide using CRISPR in thecontext of isogenic patient-derived preadipocytes providedthe critical link between this single-nucleotide substitutionand distinct adipocyte differentiation programs thermogenicbeige adipocytes versus fat-storing white adipocytes Thisstem cell model combined with the power of CRISPR-mediated genome editing to change one particular nucleotidein the human genome helped resolve one of the longeststandingmysteries in human geneticsThus we are extremelyenthusiastic for a much brighter future for making and usingstem cell models for similar mechanistic studies

Abbreviations and Acronyms

ZFN Zinc Finger NucleasesTALEN TALE domains in transcription

activator-like effector nucleasesCRISPRCas Clustered regularly interspaced

palindromic repeatsCRISPR-associatedtracrRNA Transactivating CRISPR RNAcrRNA CRISPR repeat RNAPAM Protospacer adjacent motifRNP RibonucleoproteingRNA Guide RNAdsDNA Double-stranded DNADSB Double-strand breakNHEJ Nonhomologous end joiningHR Homologous recombinationPAGE Polyacrylamide gel electrophoresisHRMA High Resolution Melting AnalysisCHIP-Seq Chromatin Immunoprecipitation

Sequencing

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Authorsrsquo Contribution

Bing Shui and Liz Hernandez Matias contributed equally


Acknowledgments

The authors thank Lisa M Anttila Kristi Simons and AlisonSeemann for assistance with paper preparation They thankDr Jeong Heon Lee and the Mayo Clinic Center for Individ-ualized Medicine Epigenetics Development Laboratory forreagents and technical support This work was supported inpart by the Mayo Clinic Center for Individualized MedicineTheworkwas funded by aMayoClinic SummerUndergradu-ate Research Fellowship to Bing Shui a LSAMP Bridge to theDoctorate Cohort XNSFGrant Award (HRD-1400870) to LizHernandez Matias a Mayo Clinic New Investigator StartupFund a Richard F Emslander Career Development Awardand aMayo Clinic Center for Biomedical Discovery PlatformAward to Dr Yi Guo

References

[1] M Bibikova K Beumer J K Trautman and D CarrollldquoEnhancing gene targeting with designed zinc finger nucleasesrdquoScience vol 300 no 5620 p 764 2003

[2] Y-G Kim J Cha and S Chandrasegaran ldquoHybrid restrictionenzymes zinc finger fusions to Fok I cleavage domainrdquo Proceed-ings of the National Academy of Sciences of the United States ofAmerica vol 93 no 3 pp 1156ndash1160 1996

[3] M H Porteus and D Baltimore ldquoChimeric nucleases stimulategene targeting in human cellsrdquo Science vol 300 no 5620 p 7632003

[4] J Boch H Scholze S Schornack et al ldquoBreaking the code ofDNA binding specificity of TAL-type III effectorsrdquo Science vol326 no 5959 pp 1509ndash1512 2009

[5] M J Moscou and A J Bogdanove ldquoA simple cipher governsDNA recognition by TAL effectorsrdquo Science vol 326 no 5959p 1501 2009

[6] A Bolotin B Quinquis A Sorokin and S D Ehrlich ldquoClus-tered regularly interspaced short palindrome repeats (CRISPRs)have spacers of extrachromosomal originrdquo Microbiology vol151 no 8 pp 2551ndash2561 2005

[7] F J M Mojica C Dıez-Villasenor J Garcıa-Martınez and ESoria ldquoIntervening sequences of regularly spaced prokaryoticrepeats derive from foreign genetic elementsrdquo Journal of Molec-ular Evolution vol 60 no 2 pp 174ndash182 2005

[8] C Pourcel G Salvignol andGVergnaud ldquoCRISPR elements inYersinia pestis acquire new repeats by preferential uptake of bac-teriophage DNA and provide additional tools for evolutionarystudiesrdquoMicrobiology vol 151 no 3 pp 653ndash663 2005

[9] M Boettcher and M T McManus ldquoChoosing the right tool forthe job RNAi TALEN or CRISPRrdquoMolecular Cell vol 58 no4 pp 575ndash585 2015

[10] M Jinek K Chylinski I Fonfara M Hauer J A Doudnaand E Charpentier ldquoA programmable dual-RNA-guided DNAendonuclease in adaptive bacterial immunityrdquo Science vol 337no 6096 pp 816ndash821 2012

[11] M Jinek A East A Cheng S Lin EMa and J Doudna ldquoRNA-programmed genome editing in human cellsrdquo eLife vol 2013no 2 Article ID e00471 2013

[12] L Cong F A RanD Cox et al ldquoMultiplex genome engineeringusing CRISPRCas systemsrdquo Science vol 339 no 6121 pp 819ndash823 2013

[13] P Mali L Yang K M Esvelt et al ldquoRNA-guided humangenome engineering via Cas9rdquo Science vol 339 no 6121 pp823ndash826 2013

[14] W Jiang and L A Marraffini ldquoCRISPR-cas new toolsfor genetic manipulations from bacterial immunity systemsrdquoAnnual Review of Microbiology vol 69 no 1 pp 209ndash228 2015

[15] E J Sontheimer and R Barrangou ldquoThe bacterial origins of thecrispr genome-editing revolutionrdquo Human Gene Therapy vol26 no 7 pp 413ndash424 2015

[16] K S Makarova N V Grishin S A Shabalina Y I Wolfand E V Koonin ldquoA putative RNA-interference-based immunesystem in prokaryotes computational analysis of the predictedenzymatic machinery functional analogies with eukaryoticRNAi and hypothetical mechanisms of actionrdquo Biology Directvol 1 article 7 2006

[17] J E Garneau M-E Dupuis M Villion et al ldquoThe CRISPRcasbacterial immune system cleaves bacteriophage and plasmidDNArdquo Nature vol 468 no 7320 pp 67ndash71 2010

[18] R E Haurwitz M Jinek B Wiedenheft K Zhou and J ADoudna ldquoSequence- and structure-specific RNA processing bya CRISPR endonucleaserdquo Science vol 329 no 5997 pp 1355ndash1358 2010

[19] F V Karginov and G J Hannon ldquoThe CRISPR system smallRNA-guided defense in bacteria and archaeardquo Molecular Cellvol 37 no 1 pp 7ndash19 2010

[20] R Barrangou and L A Marraffini ldquoCRISPR-Cas systemsprokaryotes upgrade to adaptive immunityrdquoMolecular Cell vol54 no 2 pp 234ndash244 2014

[21] I Grissa G Vergnaud and C Pourcel ldquoThe CRISPRdbdatabase and tools to display CRISPRs and to generate dictio-naries of spacers and repeatsrdquoBMCBioinformatics vol 8 article172 2007

[22] H Deveau R Barrangou J E Garneau et al ldquoPhage responseto CRISPR-encoded resistance in Streptococcus thermophilusrdquoJournal of Bacteriology vol 190 no 4 pp 1390ndash1400 2008

[23] F J M Mojica C Dıez-Villasenor J Garcıa-Martınez and CAlmendros ldquoShortmotif sequences determine the targets of theprokaryotic CRISPR defence systemrdquoMicrobiology vol 155 no3 pp 733ndash740 2009

[24] K Yusa S T Rashid H Strick-Marchand et al ldquoTargetedgene correction of alpha1-antitrypsin deficiency in inducedpluripotent stem cellsrdquo Nature vol 478 no 7369 pp 391ndash3942011

[25] H Yin W Xue S Chen et al ldquoGenome editing with Cas9 inadult mice corrects a disease mutation and phenotyperdquo NatureBiotechnology vol 32 no 6 pp 551ndash553 2014

[26] K Yoshimi T Kaneko B Voigt and T Mashimo ldquoAllele-specific genome editing and correction of disease-associatedphenotypes in rats using the CRISPR-Cas platformrdquo NatureCommunications vol 5 article 4240 2014

[27] Y Wu D Liang Y Wang et al ldquoCorrection of a genetic diseasein mouse via use of CRISPR-Cas9rdquo Cell Stem Cell vol 13 no 6pp 659ndash662 2013

[28] P D Hsu E S Lander and F Zhang ldquoDevelopment andapplications ofCRISPR-Cas9 for genome engineeringrdquoCell vol157 no 6 pp 1262ndash1278 2014

[29] K M Esvelt P Mali J L Braff M Moosburner S J Yaungand G M Church ldquoOrthogonal Cas9 proteins for RNA-guidedgene regulation and editingrdquoNature Methods vol 10 no 11 pp1116ndash1121 2013


[30] I Fonfara A Le Rhun K Chylinski et al ldquoPhylogeny of Cas9determines functional exchangeability of dual-RNA and Cas9among orthologous type II CRISPR-Cas systemsrdquoNucleic AcidsResearch vol 42 no 4 pp 2577ndash2590 2014

[31] F A Ran L Cong W X Yan et al ldquoIn vivo genome editingusing Staphylococcus aureus Cas9rdquo Nature vol 520 no 7546pp 186ndash191 2015

[32] B Chen L A Gilbert B A Cimini et al ldquoDynamic imaging ofgenomic loci in living human cells by an optimizedCRISPRCassystemrdquo Cell vol 155 no 7 pp 1479ndash1491 2013

[33] H Ma A Naseri P Reyes-Gutierrez S AWolfe S Zhang andT Pederson ldquoMulticolor CRISPR labeling of chromosomal lociin human cellsrdquo Proceedings of the National Academy of Sciencesof the United States of America vol 112 no 10 pp 3002ndash30072015

[34] D Carroll ldquoStaying on target with CRISPR-Casrdquo NatureBiotechnology vol 31 no 9 pp 807ndash809 2013

[35] B P Kleinstiver M S Prew S Q Tsai et al ldquoEngineeredCRISPR-Cas9 nucleases with altered PAM specificitiesrdquoNaturevol 523 no 7561 pp 481ndash485 2015

[36] B Zetsche J S Gootenberg O O Abudayyeh et al ldquoCpf1is a single RNA-guided endonuclease of a class 2 CRISPR-cassystemrdquo Cell vol 163 no 3 pp 759ndash771 2015

[37] M Jinek F Jiang D W Taylor et al ldquoStructures of Cas9endonucleases reveal RNA-mediated conformational activa-tionrdquo Science vol 343 no 6176 Article ID 1247997 2014

[38] H Nishimasu F A Ran P D Hsu et al ldquoCrystal structure ofCas9 in complex with guide RNA and target DNArdquo Cell vol156 no 5 pp 935ndash949 2014

[39] P Mali K M Esvelt and G M Church ldquoCas9 as a versatiletool for engineering biologyrdquo Nature Methods vol 10 no 10pp 957ndash963 2013

[40] F A Ran P D Hsu C-Y Lin et al ldquoDouble nicking by RNA-guided CRISPR cas9 for enhanced genome editing specificityrdquoCell vol 154 no 6 pp 1380ndash1389 2013

[41] P D Hsu D A Scott J A Weinstein et al ldquoDNA targetingspecificity of RNA-guided Cas9 nucleasesrdquo Nature Biotechnol-ogy vol 31 no 9 pp 827ndash832 2013

[42] V Pattanayak S Lin J P Guilinger E Ma J A Doudna and DR Liu ldquoHigh-throughput profiling of off-target DNA cleavagereveals RNA-programmed Cas9 nuclease specificityrdquo NatureBiotechnology vol 31 no 9 pp 839ndash843 2013

[43] X Wu D A Scott A J Kriz et al ldquoGenome-wide binding ofthe CRISPR endonuclease Cas9 in mammalian cellsrdquo NatureBiotechnology vol 32 no 7 pp 670ndash676 2014

[44] Y Fu J A Foden C Khayter et al ldquoHigh-frequency off-targetmutagenesis induced by CRISPR-Cas nucleases in human cellsrdquoNature Biotechnology vol 31 no 9 pp 822ndash826 2013

[45] Y Fu J D Sander D Reyon V M Cascio and J K JoungldquoImproving CRISPR-Cas nuclease specificity using truncatedguide RNAsrdquo Nature Biotechnology vol 32 no 3 pp 279ndash2842014

[46] S W Cho S Kim Y Kim et al ldquoAnalysis of off-targeteffects of CRISPRCas-derived RNA-guided endonucleases andnickasesrdquo Genome Research vol 24 no 1 pp 132ndash141 2014

[47] C Anders O Niewoehner A Duerst andM Jinek ldquoStructuralbasis of PAM-dependent target DNA recognition by the Cas9endonucleaserdquo Nature vol 513 no 7519 pp 569ndash573 2014

[48] S H Sternberg S Redding M Jinek E C Greene and JA Doudna ldquoDNA interrogation by the CRISPR RNA-guidedendonuclease Cas9rdquo Nature vol 507 no 7490 pp 62ndash67 2014

[49] S W Cho J Lee D Carroll J-S Kim and J Lee ldquoHeritablegene knockout in Caenorhabditis elegans by direct injection ofCas9-sgRNA ribonucleoproteinsrdquo Genetics vol 195 no 3 pp1177ndash1180 2013

[50] P Mali J Aach P B Stranges et al ldquoCAS9 transcriptionalactivators for target specificity screening and paired nickases forcooperative genome engineeringrdquoNature Biotechnology vol 31no 9 pp 833ndash838 2013

[51] T Wang J J Wei D M Sabatini and E S Lander ldquoGeneticscreens in human cells using the CRISPR-Cas9 systemrdquo Sciencevol 343 no 6166 pp 80ndash84 2014

[52] J G Doench E Hartenian D B Graham et al ldquoRationaldesign of highly active sgRNAs for CRISPR-Cas9-mediatedgene inactivationrdquo Nature Biotechnology vol 32 no 12 pp1262ndash1267 2014

[53] H Xu T Xiao C-H Chen et al ldquoSequence determinants ofimproved CRISPR sgRNA designrdquo Genome Research 2015

[54] S Nielsen Y Yuzenkova and N Zenkin ldquoMechanism ofeukaryotic RNA polymerase III transcription terminationrdquoScience vol 340 no 6140 pp 1577ndash1580 2013

[55] M A Moreno-Mateos C E Vejnar J Beaudoin et alldquoCRISPRscan designing highly efficient sgRNAs for CRISPR-Cas9 targeting in vivordquoNature Methods vol 12 no 10 pp 982ndash988 2015

[56] C Ronda L E Pedersen H G Hansen et al ldquoAcceleratinggenome editing in CHO cells using CRISPR Cas9 and CRISPya web-based target finding toolrdquo Biotechnology and Bioengineer-ing vol 111 no 8 pp 1604ndash1616 2014

[57] H Xu T Xiao C-H Chen et al ldquoSequence determinants ofimproved CRISPR sgRNA designrdquo Genome Research vol 25no 8 pp 1147ndash1157 2015

[58] M Stemmer TThumberger M del Sol Keyer J Wittbrodt J LMateo and S Maas ldquoCCTop an Intuitive flexible and reliableCRISPRCas9 target prediction toolrdquo PLoS ONE vol 10 no 4Article ID e0124633 p e0124633 2015

[59] S K Upadhyay and S Sharma ldquoSSFinder high throughputCRISPR-Cas target sites prediction toolrdquo BioMed ResearchInternational vol 2014 Article ID 742482 4 pages 2014

[60] Q U Ain J Y Chung and Y-H Kim ldquoCurrent and futuredelivery systems for engineered nucleases ZFN TALEN andRGENrdquo Journal of Controlled Release vol 205 pp 120ndash127 2015

[61] I Maggio M Holkers J Liu J M Janssen X Chen andM A F V Goncalves ldquoAdenoviral vector delivery of RNA-guided CRISPRCas9 nuclease complexes induces targetedmutagenesis in a diverse array of human cellsrdquo Scientific Reportsvol 4 article 5105 2014

[62] W Xue S Chen H Yin et al ldquoCRISPR-mediated directmutation of cancer genes in the mouse liverrdquo Nature vol 514no 7522 pp 380ndash384 2014

[63] Z-Y Chen C-Y He A Ehrhardt and M A Kay ldquoMinicircleDNA vectors devoid of bacterial DNA result in persistent andhigh-level transgene expression in vivordquoMolecularTherapy vol8 no 3 pp 495ndash500 2003

[64] S Kim D Kim S W Cho J Kim and J-S Kim ldquoHighly effi-cient RNA-guided genome editing in human cells via deliveryof purified Cas9 ribonucleoproteinsrdquo Genome Research vol 24no 6 pp 1012ndash1019 2014

[65] H Koike-Yusa Y Li E-P Tan M D C Velasco-Herreraand K Yusa ldquoGenome-wide recessive genetic screening inmammalian cells with a lentiviral CRISPR-guide RNA libraryrdquoNature Biotechnology vol 32 no 3 pp 267ndash273 2014


[66] O Shalem N E Sanjana E Hartenian et al ldquoGenome-scaleCRISPR-Cas9 knockout screening in human cellsrdquo Science vol343 no 6166 pp 84ndash87 2014

[67] Y Zhou S Zhu C Cai et al ldquoHigh-throughput screening of aCRISPRCas9 library for functional genomics in human cellsrdquoNature vol 509 no 7501 pp 487ndash491 2014

[68] R Cheng J Peng Y Yan et al ldquoEfficient gene editing in adultmouse livers via adenoviral delivery of CRISPRCas9rdquo FEBSLetters vol 588 no 21 pp 3954ndash3958 2014

[69] D Wang H Mou S Li et al ldquoAdenovirus-mediated somaticgenome editing of Pten by CRISPRCas9 in mouse liver in spiteof Cas9-specific immune responsesrdquoHuman GeneTherapy vol26 no 7 pp 432ndash442 2015

[70] X Liang J Potter S Kumar et al ldquoRapid and highly efficientmammalian cell engineering via Cas9 protein transfectionrdquoJournal of Biotechnology vol 208 pp 44ndash53 2015

[71] A Hendel R O Bak J T Clark et al ldquoChemically modifiedguide RNAs enhance CRISPR-Cas genome editing in humanprimary cellsrdquoNature Biotechnology vol 33 no 9 pp 985ndash9892015

[72] S Lin B T Staahl R K Alla and J A Doudna ldquoEnhancedhomology-directed human genome engineering by controlledtiming of CRISPRCas9 deliveryrdquo eLife vol 3 Article IDe04766 2014

[73] P K Mandal L M R Ferreira R Collins et al ldquoEfficientablation of genes in human hematopoietic stem and effectorcells using CRISPRCas9rdquo Cell Stem Cell vol 15 no 5 pp 643ndash652 2014

[74] J A Zuris D B Thompson Y Shu et al ldquoCationic lipid-mediated delivery of proteins enables efficient protein-basedgenome editing in vitro and in vivordquo Nature Biotechnology vol33 no 1 pp 73ndash80 2015

[75] T Koo J Lee and J Kim ldquoMeasuring and reducing off-targetactivities of programmable nucleases including CRISPR-Cas9rdquoMolecules and Cells vol 38 no 6 pp 475ndash481 2015

[76] E K Brinkman T Chen M Amendola and B van SteenselldquoEasy quantitative assessment of genome editing by sequencetrace decompositionrdquo Nucleic Acids Research vol 42 no 22article e168 2014

[77] J T Hill B L Demarest B W Bisgrove Y-C Su M Smithand H J Yost ldquoPoly peak parser method and software foridentification of unknown indels using sanger sequencing ofpolymerase chain reaction productsrdquoDevelopmental Dynamicsvol 243 no 12 pp 1632ndash1636 2014

[78] M C Porter K Murray-Leisure and P Dalbey ldquoAeromonashydrophila cellulitis A case reportrdquo Journal of the AmericanPodiatric Medical Association vol 78 no 5 pp 259ndash261 1988

[79] Z Yang C Steentoft C Hauge et al ldquoFast and sensitivedetection of indels induced by precise gene targetingrdquo NucleicAcids Research vol 43 no 9 article e59 2015

[80] D Y Guschin A J Waite G E Katibah J C Miller M CHolmes and E J Rebar ldquoA rapid and general assay for mon-itoring endogenous gene modificationrdquo Methods in MolecularBiology vol 649 pp 247ndash256 2010

[81] J C Miller M C Holmes J Wang et al ldquoAn improved zinc-finger nuclease architecture for highly specific genome editingrdquoNature Biotechnology vol 25 no 7 pp 778ndash785 2007

[82] Y Niu B Shen Y Cui et al ldquoGeneration of gene-modifiedcynomolgus monkey via Cas9RNA-mediated gene targeting inone-cell embryosrdquo Cell vol 156 no 4 pp 836ndash843 2014

[83] T Sakurai S Watanabe A Kamiyoshi M Sato and T ShindoldquoA single blastocyst assay optimized for detecting CRISPRCas9system-induced indel mutations in micerdquo BMC Biotechnologyvol 14 article 69 2014

[84] Y H Sung Y Jin S Kim and H-W Lee ldquoGeneration ofknockout mice using engineered nucleasesrdquoMethods 2014

[85] X Zhu Y Xu S Yu et al ldquoAn efficient genotyping methodfor genome-modified animals and human cells generated withCRISPRCas9 systemrdquo Scientific Reports vol 4 article 64202014

[86] A-CDeclais andDM Lilley ldquoNew insight into the recognitionof branched DNA structure by junction-resolving enzymesrdquoCurrent Opinion in Structural Biology vol 18 no 1 pp 86ndash952008

[87] P Qiu H Shandilya J M DrsquoAlessio K OrsquoConnor J DurocherandG F Gerard ldquoMutation detection using Surveyor nucleaserdquoBioTechniques vol 36 no 4 pp 702ndash707 2004

[88] L Vouillot A Thelie and N Pollet ldquoComparison of T7E1and surveyor mismatch cleavage assays to detect mutationstriggered by engineered nucleasesrdquo G3 GenesmdashGenomesmdashGenetics vol 5 no 3 pp 407ndash415 2015

[89] M C HuangW C Cheong L S Lim andM-H Li ldquoA simplehigh sensitivity mutation screening using Ampligase mediatedT7 endonuclease I and Surveyor nuclease with microfluidiccapillary electrophoresisrdquo Electrophoresis vol 33 no 5 pp 788ndash796 2012

[90] T J DahlemKHoshijimaM J Jurynec et al ldquoSimplemethodsfor generating and detecting locus-specific mutations inducedwith TALENs in the zebrafish genomerdquo PLoS Genetics vol 8no 8 Article ID e1002861 2012

[91] Y Miyaoka A H Chan L M Judge et al ldquoIsolation ofsingle-base genome-edited human iPS cells without antibioticselectionrdquo Nature Methods vol 11 no 3 pp 291ndash293 2014

[92] F Soldner J Laganiere A W Cheng et al ldquoGeneration ofisogenic pluripotent stem cells differing exclusively at two earlyonset Parkinson point mutationsrdquo Cell vol 146 no 2 pp 318ndash331 2011

[93] AHendel E J Kildebeck E J Fine et al ldquoQuantifying genome-editing outcomes at endogenous loci with SMRT sequencingrdquoCell Reports vol 7 no 1 pp 293ndash305 2014

[94] B J Hindson K D Ness D A Masquelier et al ldquoHigh-throughput droplet digital PCR system for absolute quantitationof DNA copy numberrdquo Analytical Chemistry vol 83 no 22 pp8604ndash8610 2011

[95] L Miotke B T Lau R T Rumma andH P Ji ldquoHigh sensitivitydetection and quantitation of DNA copy number and singlenucleotide variants with single color droplet digital PCRrdquoAnalytical Chemistry vol 86 no 5 pp 2618ndash2624 2014

[96] R S Weisman D Price and P H Wald ldquoOutpatient manage-ment of acute and chronic poisoningrdquo Primary Care vol 13 no1 pp 151ndash156 1986

[97] E Pastrana V Silva-Vargas and F Doetsch ldquoEyes wide opena critical review of sphere-formation as an assay for stem cellsrdquoCell Stem Cell vol 8 no 5 pp 486ndash498 2011

[98] B L K Coles-Takabe I Brain K A Purpura et al ldquoDonrsquot lookgrowing clonal versus nonclonal neural stem cell coloniesrdquo StemCells vol 26 no 11 pp 2938ndash2944 2008

[99] I Singec R Knoth R P Meyer et al ldquoDefining the actualsensitivity and specificity of the neurosphere assay in stem cellbiologyrdquo Nature Methods vol 3 no 10 pp 801ndash806 2006


[100] H Mori K Ninomiya M Kino-Oka et al ldquoEffect of neuro-sphere size on the growth rate of human neural stemprogenitorcellsrdquo Journal of Neuroscience Research vol 84 no 8 pp 1682ndash1691 2006

[101] G Orive E Santos J L Pedraz and R M HernandezldquoApplication of cell encapsulation for controlled delivery ofbiological therapeuticsrdquo Advanced Drug Delivery Reviews vol67-68 pp 3ndash14 2014

[102] A Kang J Park J Ju G S Jeong and S-H Lee ldquoCellencapsulation via microtechnologiesrdquo Biomaterials vol 35 no9 pp 2651ndash2663 2014

[103] H C Kluin-Nelemans H W J Hakvoort J H Jansen etal ldquoColony growth of normal and neoplastic cells in variousconcentrations of methylcelluloserdquo Experimental Hematologyvol 16 no 11 pp 922ndash928 1988

[104] P S Hoppe D L Coutu and T Schroeder ldquoSingle-cell tech-nologies sharpen upmammalian stem cell researchrdquoNature cellBiology vol 16 no 10 pp 919ndash927 2014

[105] K Hope and M Bhatia ldquoClonal interrogation of stem cellsrdquoNature Methods vol 8 no 4 supplement pp S36ndashS40 2011

[106] K Rodgers andMMcVey ldquoError-prone repair of DNA double-strand breaksrdquo Journal of Cellular Physiology vol 231 no 1 pp15ndash24 2016

[107] K J Beumer and D Carroll ldquoTargeted genome engineeringtechniques in Drosophilardquo Methods vol 68 no 1 pp 29ndash372014

[108] K J Beumer J K Trautman K Mukherjee and D CarrollldquoDonor DNA utilization during gene targeting with zinc-fingernucleasesrdquo G3 GenesmdashGenomesmdashGenetics vol 3 no 4 pp657ndash664 2013

[109] K J Beumer J K Trautman A Bozas et al ldquoEfficient gene tar-geting inDrosophila by direct embryo injection with zinc-fingernucleasesrdquoProceedings of theNational Academy of Sciences of theUnited States of America vol 105 no 50 pp 19821ndash19826 2008

[110] M Bibikova D Carroll D J Segal et al ldquoStimulation of homol-ogous recombination through targeted cleavage by chimericnucleasesrdquoMolecular andCellular Biology vol 21 no 1 pp 289ndash297 2001

[111] L A Baena-Lopez C Alexandre A Mitchell L Pasakarnisand J-P Vincent ldquoAccelerated homologous recombination andsubsequent genome modification in Drosophilardquo Developmentvol 140 no 23 pp 4818ndash4825 2013

[112] W-D Heyer K T Ehmsen and J Liu ldquoRegulation of homolo-gous recombination in eukaryotesrdquo Annual Review of Geneticsvol 44 pp 113ndash139 2010

[113] JMDaley and P Sung ldquo53BP1 BRCA1 and the choice betweenrecombination and end joining at DNA double-strand breaksrdquoMolecular and Cellular Biology vol 34 no 8 pp 1380ndash13882014

[114] L Ye J Wang A I Beyer et al ldquoSeamless modification of wild-type induced pluripotent stem cells to the natural CCR5Delta32mutation confers resistance toHIV infectionrdquoProceedings of theNational Academy of Sciences of theUnited States of America vol111 no 26 pp 9591ndash9596 2014

[115] F Xie L Ye J C Chang et al ldquoSeamless gene correctionof 120573-thalassemia mutations in patient-specific iPSCs usingCRISPRCas9 and piggyBacrdquo Genome Research vol 24 no 9pp 1526ndash1533 2014

[116] S Ramakrishna ldquoSurrogate reporter-based enrichment of cellscontaining RNA-guided Cas9 nuclease-induced mutationsrdquoNature Communications vol 5 article 3378 2014

[117] J Drost R H van Jaarsveld B Ponsioen et al ldquoSequentialcancer mutations in cultured human intestinal stem cellsrdquoNature vol 521 no 7550 pp 43ndash47 2015

[118] M McCormick ldquoSib selectionrdquo inMethods in Enzymology vol151 chapter 33 pp 445ndash449 Elsevier 1987

[119] C Smith A Gore W Yan et al ldquoWhole-genome sequencinganalysis reveals high specificity of CRISPRCas9 and TALEN-based genome editing in human iPSCsrdquo Cell Stem Cell vol 15no 1 pp 12ndash13 2014

[120] K Suzuki C Yu J Qu et al ldquoTargeted gene correctionminimally impacts whole-genome mutational load in human-disease-specific induced pluripotent stem cell clonesrdquo Cell StemCell vol 15 no 1 pp 31ndash36 2014

[121] A Veres B S Gosis Q Ding et al ldquoLow incidence of off-targetmutations in individual CRISPR-Cas9 and TALEN targetedhuman stem cell clones detected bywhole-genome sequencingrdquoCell Stem Cell vol 15 no 1 pp 27ndash30 2014

[122] H OrsquoGeen I M Henry M S Bhakta J F Meckler and DJ Segal ldquoA genome-wide analysis of Cas9 binding specificityusing ChIP-seq and targeted sequence capturerdquo Nucleic AcidsResearch vol 43 no 6 pp 3389ndash3404 2015

[123] D Kim S Bae J Park et al ldquoDigenome-seq genome-wideprofiling of CRISPR-Cas9 off-target effects in human cellsrdquoNature Methods vol 12 no 3 pp 237ndash243 2015

[124] S Q Tsai Z Zheng N T Nguyen et al ldquoGUIDE-seq enablesgenome-wide profiling of off-target cleavage by CRISPR-CasnucleasesrdquoNature Biotechnology vol 33 no 2 pp 187ndash197 2015

[125] R L Frock J Hu R M Meyers Y Ho E Kii and F WAlt ldquoGenome-wide detection of DNA double-stranded breaksinduced by engineered nucleasesrdquoNature Biotechnology vol 33no 2 pp 179ndash186 2015

[126] L Yang D Grishin GWang et al ldquoTargeted and genome-widesequencing reveal single nucleotide variations impacting speci-ficity of Cas9 in human stem cellsrdquoNature Communications vol5 article 5507 2014

[127] Z Zhu N Verma F Gonzalez Z Shi and D HuangfuldquoA CRISPRCas-mediated selection-free knockin strategy inhuman embryonic stem cellsrdquo Stem Cell Reports vol 4 no 6pp 1103ndash1111 2015

[128] F T Merkle W Neuhausser D Santos et al ldquoEfficient CRISPR-Cas9-mediated generation of knockin human pluripotent stemcells lacking undesired mutations at the targeted locusrdquo CellReports vol 11 no 6 pp 875ndash883 2015

[129] V M Bedell and S C Ekker ldquoUsing engineered endonucleasesto create knockout and knockin zebrafish modelsrdquo Methods inMolecular Biology vol 1239 pp 291ndash305 2015

[130] G Schwank B-K Koo V Sasselli et al ldquoFunctional repairof CFTR by CRISPRCas9 in intestinal stem cell organoids ofcystic fibrosis patientsrdquo Cell Stem Cell vol 13 no 6 pp 653ndash658 2013

[131] C Y Park D Kim J Son et al ldquoFunctional correction oflarge factor VIII Gene chromosomal inversions in hemophilia apatient-derived iPSCs Using CRISPR-Cas9rdquo Cell Stem Cell vol17 no 2 pp 213ndash220 2015

[132] C Y Park T Halevy D Lee et al ldquoReversion of FMR1methylation and silencing by editing the triplet repeats in fragileX iPSC-derived neuronsrdquo Cell Reports vol 13 no 2 pp 234ndash241 2015

[133] L Xu K H Park L Zhao et al ldquoCRISPR-mediated genomeediting restores dystrophin expression and function in mdxmicerdquoMolecular Therapy 2015


[134] R Flynn A Grundmann P Renz et al ldquoCRISPR-mediatedgenotypic and phenotypic correction of a chronic granulo-matous disease mutation in human iPS cellsrdquo ExperimentalHematology vol 43 no 10 pp 838ndash848e3 2015

[135] C W Chang Y Lai E Westin et al ldquoModeling human severecombined immunodeficiency and correction by CRISPRCas9-enhanced gene targetingrdquo Cell Reports vol 12 no 10 pp 1668ndash1677 2015

[136] A L Firth T Menon G Parker et al ldquoFunctional genecorrection for cystic fibrosis in lung epithelial cells generatedfrom patient iPSCsrdquo Cell Reports vol 12 no 9 pp 1385ndash13902015

[137] P Xu Y Tong X-z Liu et al ldquoBoth TALENs and CRISPRCas9directly target the HBB IVS2-654 (C gt T) mutation in 120573-thalassemia-derived iPSCsrdquo Scientific Reports vol 5 Article ID12065 2015

[138] B Song Y Fan W He et al ldquoImproved hematopoietic differ-entiation efficiency of gene-corrected beta-thalassemia inducedpluripotent stem cells by CRISPRCas9 systemrdquo Stem Cells andDevelopment vol 24 no 9 pp 1053ndash1065 2015

[139] D G Ousterout A M Kabadi P I Thakore W H Majoros TE Reddy and C A Gersbach ldquoMultiplex CRISPRCas9-basedgenome editing for correction of dystrophin mutations thatcause Duchennemuscular dystrophyrdquoNature Communicationsvol 6 article 6244 2015

[140] M J Osborn R Gabriel B R Webber et al ldquoFanconianemia gene editing by the CRISPRCas9 systemrdquoHumanGeneTherapy vol 26 no 2 pp 114ndash126 2015

[141] H L Li N Fujimoto N Sasakawa et al ldquoPrecise correction ofthe dystrophin gene in duchenne muscular dystrophy patientinduced pluripotent stem cells by TALEN and CRISPR-Cas9rdquoStem Cell Reports vol 4 no 1 pp 143ndash154 2015

[142] YWu H Zhou X Fan et al ldquoCorrection of a genetic disease byCRISPR-Cas9-mediated gene editing in mouse spermatogonialstem cellsrdquo Cell Research vol 25 no 1 pp 67ndash79 2015

[143] C Long J R McAnally J M Shelton A A Mireault R Bassel-Duby and E N Olson ldquoPrevention of muscular dystrophyin mice by CRISPRCas9-mediated editing of germline DNArdquoScience no 6201 pp 1184ndash1188 2014

[144] M Claussnitzer S N Dankel K Kim et al ldquoFTO obesityvariant circuitry and adipocyte browning in humansrdquoThe NewEngland Journal of Medicine vol 373 no 10 pp 895ndash907 2015

[145] R M Walsh and K Hochedlinger ldquoA variant CRISPR-Cas9system adds versatility to genome engineeringrdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 110 no 39 pp 15514ndash15515 2013

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation httpwwwhindawicom

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttpwwwhindawicom

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research



Microbiology


RNA [6ndash8]This ldquomemory systemrdquo can destroy DNA or RNAif reinfection occurs in the same bacteria or in its descendants[14ndash19]Three types of CRISPR loci exist all of which acquireshort pieces of DNA called spacers from foreign DNA ele-ments [20] Spacers are integrated into the bacterial genomeduring the process of CRISPR adaptation They are usuallyinserted into the CRISPR locus that contains short partiallypalindromic DNA repeats to form loci that alternate repeatedelements (CRISPR repeats)These loci are subsequently tran-scribed and processed into small interfering RNA that guidesnucleases for sequence-specific cleavage of complementarysequencesThrough these stepwise but continuous evolutionsof adaptation CRISPR repeat RNA (crRNA) biogenesis andforeign DNA targeting generated sophisticated CRISPR-based adaptive immune systems in nearly half of the bacterialspecies as well as in most archaea [21]

The sequence in the exogenous nucleic acid element cor-responding to a CRISPR spacer was defined as a protospacer[22] For proper targeting by type I and II CRISPR systemsthe protospacer is usually flanked by a system-specific highlyconserved CRISPR motif namely a protospacer adjacentmotif (PAM) [23] Most PAMs are typically 2 to 5 highlyconserved nucleotides either on the 51015840 end of protospacer(type I system) or on the 31015840 side (most type II systems) Asignificant feature of the PAM for the CRISPR system is todistinguish the foreign DNA against the host genome thusonly the PAM-bearing invading sequence will be targeted fordestruction

3 Different Classes of CRISPRCas

Among the three different types of CRISPR loci type I andIII loci involve a complex panel of multiple Cas proteinsthat form ribonucleoprotein (RNP) complexes with CRISPRRNA to target foreign sequences [15] However the type IICRISPR system uses a much smaller number of Cas proteinsto perform this core function Type II CRISPR loci have threesubdivisions The most commonly used CRISPR system foreukaryotic genome engineering is adopted from a type II Asystem from S pyogenes where a singleCas9 protein (spCas9)is responsible for both forming the CRISPR-RNP complexand subsequent DNA cleavage For the practical reason ofsimplicity most genome-engineering applications use onehybrid RNA (guide RNA gRNA) combining the essentialstructural features of the transactivating RNA (tracrRNA)and crRNA duplex [10] The single-chain gRNA is used herein subsequent discussions

Besides spCas9 a few other orthologous Cas9 proteinsfrom similar type II CRISPR systems share the core featureas the sole protein component for RNA-guided targetingTheCas9 proteins of Streptococcus thermophilusNeisseria menin-gitidis and Treponema denticola demonstrated comparablegenome-editing efficiency to spCas9 (Table 1) [24ndash27] TheseCas9 proteins have different sizes mostly due to their targetrecognition domains (REC) [28] Significantly orthologousCas9 proteins differ in the specific PAM sequences used fortargeting thus they can be used in the same cell when pairedwith their corresponding crRNA to recognize their corre-sponding targets without interfering with each other [29ndash31]

Table 1 Orthogonal type II Cas9 and their optimal PAMpreference

Bacteria PAM CRISPRtype Reference

S thermophiluslowastlowast NNAGAAW(CRISPR1) IIA [28 29]

N meningitidis NNNNGATTNNNNGCTT IIC [28 47

145]T denticola NAAAAN IIA [29]S mutans NGG IIA [47]L innocua NGG IIA [47]L buchneri NAAAAN IIA [47]C jejuni NNNNACA IIC [47]P multocida GNNNCNNA IIC [47]S aureuslowastlowast NNGRRT IIA [31]N cinerea GATlowast IIC [31]C lari GGGlowast IIC [31]P lavamentivorans CATlowast IIC [31]C diphtheriae GGlowast IIC [31]S pasteurianus GTGAlowast IIA [31]

S pyogenes NGG (NAGas minor) IIA [10]

S pyogenes (D1135E)NGG (doesnot recognize

NAG)IIA [35]

S pyogenes VQR(D1135VR1335QT1337R)

NGANNGCG IIA [35]

S pyogenes EQR(D1135ER1335QT1337R) NGAG IIA [35]lowastPutative PAM lowastlowastsignificantly smaller than spCas9 Bottom rows areengineered spCas9 proteins with different PAM preferences

This characteristic enables sequence flexibility of CRISPRexperiments by offering a variety of Cas9 proteins to targetvirtually any particular sequence [25]This orthogonality wasbest demonstrated by recent work that allowed the labelingof distinct genomic regions using different inactivated Cas9-fluorescent fusion proteins simultaneously in a single live cell[32 33] Although most Cas9 proteins from type II CRISPRsystem have one or more optimal PAMs there is also consid-erable flexibility in terms of PAM recognition For examplespCas9 recognizes NGG as its optimal PAM sequence whileNAG can also be recognized with lower frequency ([12]and subsequent) This plasticity might arise from continuousselection pressure on bacterium to target evolving viralsequences [34] In practice this plasticity poses considerablechallenges due to the off-targeted recognition of alternativePAM sequences [12] On the other hand this flexibility allowsfurther engineering of different Cas9 proteins to optimizeor modify PAM preference Initial progress has been madetoward generation of spCas9 with more rigid NGG PAMrecognition and modification of the PAM preferences [35]In a few years further biochemical characterization of nativeorthogonal Cas9 proteins with their PAM preferences andprotein engineering efforts on characterized Cas9 proteins




4 Cas9 Enzymology





Free Cas9

gRNA



Mismatch between


DNA release

gRNA-loaded Cas9

PAM

DNA

gRNA

Scan for PAM

binding

Cas9 recognizes PAM


PAM


































220210200190


PAM



(a)

Wild type

Mutant

Mismatch


Surveyor orT7E1

Full length

Cleaved

Cleaved

6000

3000

1000

700600500

400

300

200

100

Undigested SUR T7E1

(b)

Temperature

Fluo

resc

ence


(c)

Wild type Mutant

Homoduplexes

Heteroduplexes

(d)




















3998400 Hom5

998400 Hom

Genome region


Crerecombinase

HRDonor plasmid

MarkerloxP loxP

loxP loxPMarker

loxP

(a)

3998400 Hom5

998400 Hom

PiggyBactransposase

Genome region


HRDonor plasmid

PiggyBacTA

TA

TA

TA

TA

TA

Marker

Marker

(b)


















Sequencing






Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology




4 Cas9 Enzymology





Free Cas9

gRNA



Mismatch between


DNA release

gRNA-loaded Cas9

PAM

DNA

gRNA

Scan for PAM

binding

Cas9 recognizes PAM


PAM


































220210200190


PAM



(a)

Wild type

Mutant

Mismatch


Surveyor orT7E1

Full length

Cleaved

Cleaved

6000

3000

1000

700600500

400

300

200

100

Undigested SUR T7E1

(b)

Temperature

Fluo

resc

ence


(c)

Wild type Mutant

Homoduplexes

Heteroduplexes

(d)




















3998400 Hom5

998400 Hom

Genome region


Crerecombinase

HRDonor plasmid

MarkerloxP loxP

loxP loxPMarker

loxP

(a)

3998400 Hom5

998400 Hom

PiggyBactransposase

Genome region


HRDonor plasmid

PiggyBacTA

TA

TA

TA

TA

TA

Marker

Marker

(b)


















Sequencing






Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


Free Cas9

gRNA



Mismatch between


DNA release

gRNA-loaded Cas9

PAM

DNA

gRNA

Scan for PAM

binding

Cas9 recognizes PAM


PAM


































220210200190


PAM



(a)

Wild type

Mutant

Mismatch


Surveyor orT7E1

Full length

Cleaved

Cleaved

6000

3000

1000

700600500

400

300

200

100

Undigested SUR T7E1

(b)

Temperature

Fluo

resc

ence


(c)

Wild type Mutant

Homoduplexes

Heteroduplexes

(d)




















3998400 Hom5

998400 Hom

Genome region


Crerecombinase

HRDonor plasmid

MarkerloxP loxP

loxP loxPMarker

loxP

(a)

3998400 Hom5

998400 Hom

PiggyBactransposase

Genome region


HRDonor plasmid

PiggyBacTA

TA

TA

TA

TA

TA

Marker

Marker

(b)


















Sequencing






Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology































220210200190


PAM



(a)

Wild type

Mutant

Mismatch


Surveyor orT7E1

Full length

Cleaved

Cleaved

6000

3000

1000

700600500

400

300

200

100

Undigested SUR T7E1

(b)

Temperature

Fluo

resc

ence


(c)

Wild type Mutant

Homoduplexes

Heteroduplexes

(d)




















3998400 Hom5

998400 Hom

Genome region


Crerecombinase

HRDonor plasmid

MarkerloxP loxP

loxP loxPMarker

loxP

(a)

3998400 Hom5

998400 Hom

PiggyBactransposase

Genome region


HRDonor plasmid

PiggyBacTA

TA

TA

TA

TA

TA

Marker

Marker

(b)


















Sequencing






Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology



















3998400 Hom5

998400 Hom

Genome region


Crerecombinase

HRDonor plasmid

MarkerloxP loxP

loxP loxPMarker

loxP

(a)

3998400 Hom5

998400 Hom

PiggyBactransposase

Genome region


HRDonor plasmid

PiggyBacTA

TA

TA

TA

TA

TA

Marker

Marker

(b)


















Sequencing






Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology












Sequencing






Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


Acknowledgments


References





























































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology
































































































































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology

review article the rise of crispr/cas for genome editing in stem...

Documents