CRISPR/Cas9  as  a  Promising  Gene  Editing  Tool  for  Fanconi Anemia  TreatmentVineeta Sharma10, Mark  J.  Osborn1-­3,  Richard  Gabriel4,5,  Beau  R.  Webber1,  Anthony  P.  DeFeo1,  Amber  N.  McElroy1,  Jordan  Jarjour6,  Colby  G.  Starker2,3,  John  E.  Wagner1,3,  

J.  Keith  Joung7,8, Daniel  F.  Voytas2,9,  Christof von  Kalle4,5,  Manfred  Schmidt4,5,  Bruce  R.  Blazar1,3,  Mark  J.  Ahn10,  Timothy  J.  Miller10,  Jakub  Tolar 1,31Department  of  Pediatrics,  Division  of  Blood  and  Marrow  Transplantation,   University  of  Minnesota,  Minneapolis,   MN  55455.,   2Center  for  Genome  Engineering,  University  of  Minnesota,  Minneapolis,   MN  55455.

3Stem  Cell  Institute,  University  of  Minnesota,   Minneapolis,  MN  55455., 4Department  of  Translational  Oncology,   National  Center  for  Tumor  Diseases,  Heidelberg  69120,  Germany.  5German  Cancer  Research  Center  (DKFZ),  Heidelberg  69120,  Germany.,6Pregenen,  Inc.,  Seattle,  WA  98103.,  7Molecular  Pathology  Unit,  Center  for  Computational   &  Integrative  Biology,  and  Center  for  Cancer  Research,  Massachusetts   General  Hospital,  Charlestown,  MA  02114.,

8Program  in  Biological  and  Biomedical  Sciences,  Harvard  Medical  School,  Boston,  MA  02115.,  9Department  of  Genetics,  Cell  Biology  &  Development,  University  of  Minnesota,  Minneapolis,   MN  55455.,10Abeona   Therapeutics,   Inc.,  Cleveland,  OH  44103

Introduction: Fanconi anemia (FA) is a rare inherited diseasemanifested by bone marrow failure and increased risk of malignancy.The c.456 + 4A>T (IVS4 + 4A>T) point mutation in FAcomplementation group C (FA-­C) gene results in a cryptic splice siteand causes aberrant splicing and in-­frame deletion of FANCC exon 4.Gene editing is highly desirable alternative to allogeneichematopoietic cell transplantation (HCT) for FA. In the present study,we have generated a CRISPR/Cas9 system for FANCC locus anddemonstrated its usefulness in repairing the FANCC c.456+4A>Tmutation.

Methods and Results: To test the ability of our custom-­designedCRISPR/Cas9 reagents to mediate FANCC gene HDR, a transformedskin fibroblast culture was derived from an FA-­C patient homozygousfor the c.456+4A>T mutation [1]. The cells were treated with a donorplasmid and either the CRISPR/Cas9 nuclease or nickase. Todetermine whether genome editing by CRISPR/Cas9 resulted inrestoration of exon 4 expression, HDR-­specific PCR was performedusing an allele specific RT-­PCR. Interestingly, CRISPR/Cas9 nucleaseand nickase clones each identified correction of c.456+4A>Tcompared to untreated and WT controls. Furthermore, to evaluatefunctional capability of our gene editing method, H2AX staining clearlydemonstrated inability of untreated FA-­C cells to phosphorylate γ-­H2AX, however, the clones that were corrected by the nickase or thenuclease showed clear evidence to phosphorylate γ-­H2AX. Thesefindings confirm correction of the c.456+4A>T mutation at DNA, RNAand protein level.

An important safety concern of gene editing based correction strategyis potential for off target (OT) effects. To assess this important safetyissue, a surveyor assay and an integration deficient lenti virus (IDLV)reporter gene trapping assay was performed and no OT activity for thenuclease or nickase was observed. Moreover, to identify the sites ofintegration of the IDLV, the samples were tested using LAM PCR andnonrestrictive (nr)LAM PCR, these results documented only on targetevents. In total, the data suggests highly specific CRISPR/Cas9reagents.

Conclusions: To summarize, this data show that CRISPR/Cas9mediated direct c.456+4A>T mutation repair resulted in normalizationof the FANCC gene. This study also demonstrates that nickase wasmore efficient and reliable compared to nuclease. Furthermore, thegene editing model system established here provides support for afavorable safety profile using these synthetic molecules for correctionof FA and other genetic disease in human cells.

ØWe have generated CRISPR/Cas9 nuclease and nickase reagents for targeting c.456+A>T mutation at FANCC locus.

ØOur data demonstrates both nuclease and nickase mediated c.456+A>T repair, however, the nickase was more efficient due to itspreference towards error free HDR pathway over NHEJ.

ØIn silico and genome wide LAM PCR methodologies confirmed highly specific on-­target HDR activity of our CAS9 reagents resulting inphenotypic rescue of FANCC in an ex vivo disease model where fibroblasts were derived from a patient with FA.

ØAside from bone marrow transplantation that carries a risk of significant side effects, there is no treatment available that can halt orreverse the symptoms of FA. Using the CRISPR-­Cas9 gene-­editing system to repair the FANCC gene in human fibroblasts from a FApatient, our study has demonstrated significant and promising results.

ØOur study provides proof of principle that CRISPR/Cas9 system has the potential to allow safe and precise gene modification for FA andother blood disorders in human cells.

CellsHuman

NHEJInDel  resulting   in  premature

Stop   codon

HRHomology  directed   recombination   for  

precise  gene   editing

Ex-­vivo

Cas9  RuvCdomain

Cas9  HNHdomain

Fig.1) The core components of CRISPR-­Cas9 are a nuclease Cas9comprising two catalytic active domains RuvC and HNH, and aguide RNA (gRNA). gRNA directs Cas9 to the target site by base-­pairing, resulting in Cas9-­generated site-­specific DNA double-­strand breaks (DSBs) that are subsequently repaired byhomologous directed repair (HDR) or by non-­homologous end-­joining (NHEJ). Additionally, Cas9 can be reprogrammed intonickase by inactivating either RuvC or HNH. Nickase makes singlestranded breaks and favors HDR over error prone NHEJ.

CRISPR can be used in seemingly for ex-­vivo or in-­vivo genomeediting.

Figure  1:  Mechanism  of  CRISPR-­CAS9  mediated  genome  editing

Fig. 2-­A) CRISPR Design Tool identified a target site within 15 bp of c.456+4A>Tlocus. B) CRISPR architecture and FANCC gene target recognition. C) Nuclease ornickase were expressed form a plasmid containing the CMV promoter and BGH pA,gRNA gene expression was mediated by U6 Poly IIIa promoter and a transcriptionalterminator (pT). D) The FANCC locus in cells that received CRISPR/Cas9 nucleaseor nickase with corresponding gRNA (target site shown as a green box), or a GFP-­treated control group (labeled “C”), were amplified with primers (red arrows).Nuclease-­ or nickase-­generated insertions/deletions result in heteroduplexformation with unmodified amplicons that are cleaved by the Surveyor nucleaseresulting in 228 & 189 bp products. Surveyor assay indicated higher rates ofactivity of nuclease compared to nickase. E) 293T cells F) FA-­C fibroblasts.

Figure  2:  FANCC  c.456+4A>T  gene  targeting

Figure  3:  Assesment of  DNA  repair  fates  

Fig. 3) Quantification of NHEJ and HDR using TLR: A) At its basal state, the TLRconstruct does not express a functional fluorescent protein, however, followingclevage of the target sequence in context to an exogenous GFP donor repair template,GFP expression can be restored by HDR repair. Conversely, target site cleavage andrepair by the error-­prone NHEJ results in an in-­frame mCherry expression. B, C) Basalrate of green or red fluresence were minute for either untransfected cells or cellsreceiving donor only. D) Nuclease delievery resulted in both mCherry and GFPfluorescence, indicating both NHEJ and HDR events, however, nickase version ofCas9 promotes HDR and minimizes NHEJ.

Figure  4:  Homology-­directed  repair  and  phenotypic  restoration

Fig. 4-­ i) To test the ability of our custom designed Cas9 reagents to mediate FANCC gene HDR, transformed skin fibroblastculture from an FA-­C patient homozygous for the c.456+6A>T was used. i-­A) The FANCC locus, red arrow indicatesc.456+4A>T locus, blue arrow indiactes primers used for HDR screening. B) Gene correction Donor. C) Gel image of PCRscreening approach for HDR using donor and locus specific primers. D) Number of gene-­corrected clones obtained. E)Sanger sequencing data of untreated cells and gene corrected clones confirms correction of c.456+4A>T mutation. B) Cas9mediated HDR restores FANCC expression at mRNA and protein level in ex-­vivo culture system where fibroblasts were takenfrom an FA-­C patient homozygous for c.456 A>T . ii, A-­E) CRISPR-­Cas9 mediated HDR of c.456 A>T restores FANCCexpression at DNA,mRNA and protein levels in patient fibroblast.

Figure  5:  CRISPR  off-­target  and  on-­target    analysis

Fig. 5-­ i-­A) CRISPR Design Tool revelead five intragenic OT sites. i-­B) Surveyor analysis indicated no off-­target activity forany of the five intragenic OT sites. ii-­ A) Tandam delivery of CRISPR/Cas9 nuclease or nickase with GFP IDLV resulted in GFPexpression. ii-­B) GFP expressed cells were sorted and expanded. ii-­C, D) PCR analysis using a 3′ LTR forward primer and aFANCC locus reverse primer yielded a PCR product for the Cas9 nuclease and nickase treated cells but not the IDLV-­onlycontrol cells. Sequencing of these products showed an LTR:FANCC genomic junction immediately upstream of the CRISPRPAM (data not shown) suggesting high specificity of CAS9 reagents. Iii-­A, B) Genome-­wide screen for off-­target loci reportedCLIS frequency as 7-­31 at intended target loci, while no CLIS activity was reported at loci containing partial target sitehomology.

1: Osborn, M.J., Gabriel, R., Webber, B.R., DeFeo, A.P., McElroy, A.N., Jarjour, J., Starker, C.G., Wagner, J.E., Joung, J.K., Voytas, D.F., von, Kalle. C., Schmidt, M., Blazar, B.R.,Tolar, J. Fanconi anemia gene editing by the CRISPR/Cas9 system. Hum Gene Ther. 2015, 26(2):114-­26.

Technology  is  licensed  by  Abeona  Therapeutics  Inc.Funded  by  the  National  Center  for  Advancing  Translational  Sciences  of  the  National  Institute  of  Health  Award  Number  UL1TR000114  (MJO).

i ii iii

iii

BACKGROUND

ABSTRACT RESULTS

CONCLUSIONS

REFRENCESMATERIAL  &  METHODS

CRISPR  &  donor  construction

Gene  Transfer

Surveyor  Nuclease

Selection &  transgene  excision

Traffic light  reporter  cell  line  generation  

Cas9  nuclease  &  nickasescreening

Molecular  &  protein  screening

Off-­target  analysis

Genome-­widescreening