Genome Editing: Considerations for Therapeutic Applications Cecilia Fernandez TIDES, 2017 www.editasmedicine.com 1
The Potential to Repair Any Broken Gene Transformative new category of medicines for genetically-defined and genetically-treatable diseases 2
CRISPR Unlocks Genome Editing Editing machinery can be engineered to target nearly any genomic location Targeting Sequence DNA HiFi es PAM Cut Sites Nuclease Guide RNA Complex of nuclease and guide RNA precisely locates and cuts genomic sites Ability to target several sites simultaneously using multiple guide RNAs Nuclease can be engineered to reach more sites and to modulate cutting PAM: Protospacer Adjacent Motif; HiFi: High Fidelity; es: enhanced Specificity 3
Unparalleled Platform for Gene Editing Medicines ~10x Broad Range of Sites SpCas9 SaCas9 SpCas9 Variants SaCas9 Variants Cpf1 Cpf1 Variants Editas Platform Wide Delivery Options Viral Vector Lipid Nanoparticle Electroporation Disrupt Delete Correct Diverse Spectrum of Edits 4
Lead Finding & Specificity to Select grna Identify, Measure, Minimize In silico Selection in silico Prediction of Cutting Sites Testing of On-Target Cutting Targeted Panels for Detection of Sites from Biased & Unbiased Screens Unbiased Detection of Off-Target Cuts (e.g., GUIDE-Seq) 5
Identification of Robust grnas RNP (S. pyogenes) delivery to primary T cells 1 0 0 % % P D Negative 1 N e g a tiv e 8 0 6 0 4 0 2 0 0 No Guide U n tre a te d P D 4 9 P D 5 1 P D 5 2 P D 5 3 P D 5 5 P D 5 6 P D 9 7 grnas P D 9 8 g R N A ID P D 9 9 P D 1 0 0 P D 1 0 1 P D 1 0 2 P D 1 0 3 P D 1 0 4 Several grnas performed well when assessing target expression by FACS Sequencing of gdna confirmed the presence of indels These grnas were analyzed by GUIDE-seq to identify off targets 6
Control and Specificity to Drive Precision Guide-seq allows identification of highly selective candidates GUIDE-Seq Read Count 1000 Reads 500 0 A B C D Guide RNA On-Target Off-Target 1 Off-Target 2 Off-Target 3 7
How Do You Best Measure Editing? A simple question with a complex answer Sequence anchored detection approaches are limited to: What is between the primers and Amplicon size One cannot detect several events and has to build and reassemble answers from disparate technologies (e.g. ddpcr + targeted sequencing): Large insertions Large deletions Inversions Translocations Wanted a size insensitive, multiplex compatible, comparatively easy single tube method that can detect all of the above events 8
Uni-Directional Targeted Sequencing (UDiTaS ) Blending tagmentation and AMP-seq Custom Transposon Custom Transposon + Genomic DNA Tagmentation (~ 2Kb) 5 3 PCR Round 1 (hot start) P5 primer 5 ß UMI Barcode ß Pooling Barcode ß P5 3 ddc PCR Round 2 Sequencing Sequence specific primer UMI P7 P7BCBC BC P7 Picelli, S., et. al (2014). Tn5 transposase and tagmentation procedures for massively scaled sequencing projects. Genome Research, 24(12), 2033 2040. 9
Comparison of Small Indel Measurements UDiTaS correlates well with targeted sequencing and T7E1 assays 100 90 80 70 % Indels 60 50 40 30 20 10 AMP-seq T7E1 UDiTaS 0 0 2 4 6 8 guide RNA # None 10
UDiTaS (Uni-Directional Targeted Sequencing) A simple, robust method for capturing complex editing events in a single reaction 1. Quantitation of editing 2. Quantitation and discovery of large deletions and inversions 3. Translocation discovery and quantitation 4. Multiplexing 11
Ex Vivo Drug Development EngineeredHigh autologous requires multiple components Quality therapy Ribonucleoprotein Particle Delivery RNA Guided Endonuclease Ribonucleoprotein Particle (RNP) + + Engineered Patient Cell RNA Guide Patient Cell 12
Generating Synthetic Covalently-Coupled Dual grna A completely non-enzymatic process for guide production Why make a synthetic guide? Targeted chemistries anywhere in the molecule 5 3 + Unhindered ends and modifications 5 Scale up and purity are more compatible with CMC requirements 5 3 covalently-coupled dual grna (dgrna) 3 13
Cellular Editing Activity In vitro transcribed and synthetic covalently-coupled dgrna are equivalent in cells 80 % Editing (T7E1) 60 40 20 IVT purified by vendor covalently-coupled dgrna IVT purified by collaborator 2-part synthetic 0-11 -10-9 -8-7 -6 Log concentation RNP (M) 14
Assessing grna purity and sequence fidelity Development of an RNA-Seq based method for grna QC synthetic 100mer A synthetic 100mer B covalently-coupled dgrna 15
grna purity and sequence fidelity Covalently-coupled dgrna result in greater sequence fidelity in target region A B Covalently-Coupled dgrna 16
17 grna Purity and Sequence Fidelity A B Covalently-Coupled dgrna Covalently-coupled dgrna result in fewer truncated molecules 0.1502 0.0432 0.5124 1.0384 1.2414 0.8864 1 1 6 0.0055 1.0239 0.0192 0.0028 0.0033 5 0.0056 5 0.011 0.133 0.0244 0.0066 0.0072 2 0.0298 1.0806 0.0914 0.0055 0.0102 0.003 0.016 4 0.0045 0.0123 0.0975 0.0532 0.0172 0.0034 4 0.3896 0.0961 0.0041 0.0528 0.0038 0.3176 0.0413 4 0.002 0.2954 0.0215 5 0.0034 0.005 8e 04 0.0738 0.1507 0.0297 0.0056 1 0.2056 0.1205 0.0059 0.0042 0.0473 0.0139 1 0.0749 0.0777 0.002 0.1023 0.083 8 0.1596 0.1036 0.0067 0.0293 5 0.0061 0.0211 0.019 0.1122 0.0418 0.0129 0.1447 0.0853 0.1403 5 0.004 0.0044 0.2824 0.048 0.0102 0.0107 0.0034 0.0027 0.002 0.3779 0.6421 0.0495 0.2605 1 0.0079 5 8 5 0.0096 0.0038 0.0029 8 0.002 0.0165 6 0.0037 0.0053 0.2478 0.1275 0.0076 0.089 0.0045 0.243 0.029 0.0025 0.0037 0.0029 0.0049 0.0272 0.213 0.1773 0.0037 0.0023 0.3073 0.073 0.0038 0.0087 6 0.0742 0.0772 0.016 0.0849 5 4 3 2 1 1 2 3 4 5 0.2282 0.0221 0.0382 0.0058 0.0037 0.1642 0.0824 0.0056 0.0028 0.0105 0.0279 0.0338 0.0405 6 6 0.0398 0.0813 0.0072 0.0289 4 0.003 0.2042 0.1472 0.0049 2 0.0051 0.0028 0.1094 0.1565 0.0021 0.0037 0.0026 0.0058 0.004 0.0051 0.0047 0.0116 0.0326 0.115 0.0091 0.024 4 9 0.007 0.0088 0.1479 0.1092 0.0179 0.0289 0.0065 0.0403 0.0696 0.1195 0.0035 0.0126 0.6543 0.3407 0.0021 0.0361 6 0.0044 0.004 0.0028 0.0813 0.0263 0.0366 0.0079 0.0226 0.0587 0.099 0.0156 0.0082 0.0161 0.007 0.0917 0.1327 0.1281 0.0026 0.0077 0.3644 0.1726 0.024 0.0424 4 0.0116 0.1693 0.2804 0.0238 9 0.0061 2 0.0119 0.027 9 2 0.0033 0.0026 0.0082 0.2387 0.1481 0.0112 0.2189 4 0.0021 0.0061 0.0256 0.0321 0.0955 0.2361 0.0354 0.0044 6 0.2352 0.4466 0.0023 6 0.0033 0.0144 0.013 0.1982 0.6248 0.0414 0.1614 0.0028 0.0026 0.024 0.0035 4 0.0061 0.0056 0.0109 2 0.0142 0.0093 0.0063 2 0.0263 0.0077 0.0175 0.0608 0.2385 0.0196 0.1896 6 0.003 0.003 0.0033 0.2056 0.0685 0.0154 0.0075 0.0037 0.0144 0.0424 0.0652 0.3991 0.0123 0.0026 9 0.0151 0.0033 0.2603 0.1325 0.0221 0.034 0.0047 6 0.0349 0.1742 0.0366 0.064 0.0525 0.0982 0.0655 0.3007 0.0727 0.0121 0.0086 0.0044 0.0049 0.0114 8e 04 4e 04 9 0.0058 0.0032 5 0.0076 0.0117 0.0084 3 0.0029 0.0054 0.0045 4e 04 0.0089 0.0074 0.0025 0.0148 0.0511 0.0036 0.0065 0.0099 0.0082 0.0054 0.0378 0.0183 0.0192 4 0.0108 0.0043 0.0058 9 0.005 0.0108 0.0331 0.0092 0.0042 0.0079 0.0509 0.2056 0.0023 0.0453 0.0044 0.0397 0.0186 0.0025 0.0034 0.0436 0.0034 0.0172 0.0021 0.0051 0.0022 0.0039 0.0034 5 0.0122 0.031 5 4 3 2 1 1 2 3 4 5 5 4 3 2 1 1 2 3 4 5 5 4 3 2 1 1 2 3 4 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Position Length change 0.26 0.58 1 % error 0.1502 0.0432 0.5124 1.0384 1.2414 0.8864 1 1 6 0.0055 1.0239 0.0192 0.0028 0.0033 5 0.0056 5 0.011 0.133 0.0244 0.0066 0.0072 2 0.0298 1.0806 0.0914 0.0055 0.0102 0.003 0.016 4 0.0045 0.0123 0.0975 0.0532 0.0172 0.0034 4 0.3896 0.0961 0.0041 0.0528 0.0038 0.3176 0.0413 4 0.002 0.2954 0.0215 5 0.0034 0.005 8e 04 0.0738 0.1507 0.0297 0.0056 1 0.2056 0.1205 0.0059 0.0042 0.0473 0.0139 1 0.0749 0.0777 0.002 0.1023 0.083 8 0.1596 0.1036 0.0067 0.0293 5 0.0061 0.0211 0.019 0.1122 0.0418 0.0129 0.1447 0.0853 0.1403 5 0.004 0.0044 0.2824 0.048 0.0102 0.0107 0.0034 0.0027 0.002 0.3779 0.6421 0.0495 0.2605 1 0.0079 5 8 5 0.0096 0.0038 0.0029 8 0.002 0.0165 6 0.0037 0.0053 0.2478 0.1275 0.0076 0.089 0.0045 0.243 0.029 0.0025 0.0037 0.0029 0.0049 0.0272 0.213 0.1773 0.0037 0.0023 0.3073 0.073 0.0038 0.0087 6 0.0742 0.0772 0.016 0.0849 0.2282 0.0221 0.0382 0.0058 0.0037 0.1642 0.0824 0.0056 0.0028 0.0105 0.0279 0.0338 0.0405 6 6 0.0398 0.0813 0.0072 0.0289 4 0.003 0.2042 0.1472 0.0049 2 0.0051 0.0028 0.1094 0.1565 0.0021 0.0037 0.0026 0.0058 0.004 0.0051 0.0047 0.0116 0.0326 0.115 0.0091 0.024 4 9 0.007 0.0088 0.1479 0.1092 0.0179 0.0289 0.0065 0.0403 0.0696 0.1195 0.0035 0.0126 0.6543 0.3407 0.0021 0.0361 6 0.0044 0.004 0.0028 0.0813 0.0263 0.0366 0.0079 0.0226 0.0587 0.099 0.0156 0.0082 0.0161 0.007 0.0917 0.1327 0.1281 0.0026 0.0077 0.3644 0.1726 0.024 0.0424 4 0.0116 0.1693 0.2804 0.0238 9 0.0061 2 0.0119 0.027 9 2 0.0033 0.0026 0.0082 0.2387 0.1481 0.0112 0.2189 4 0.0021 0.0061 0.0256 0.0321 0.0955 0.2361 0.0354 0.0044 6 0.2352 0.4466 0.0023 6 0.0033 0.0144 0.013 0.1982 0.6248 0.0414 0.1614 0.0028 0.0026 0.024 0.0035 4 0.0061 0.0056 0.0109 2 0.0142 0.0093 0.0063 2 0.0263 0.0077 0.0175 0.0608 0.2385 0.0196 0.1896 6 0.003 0.003 0.0033 0.2056 0.0685 0.0154 0.0075 0.0037 0.0144 0.0424 0.0652 0.3991 0.0123 0.0026 9 0.0151 0.0033 0.2603 0.1325 0.0221 0.034 0.0047 6 0.0349 0.1742 0.0366 0.064 0.0525 0.0982 0.0655 0.3007 0.0727 0.0121 0.0086 0.0044 0.0049 0.0114 8e 04 4e 04 9 0.0058 0.0032 5 0.0076 0.0117 0.0084 3 0.0029 0.0054 0.0045 4e 04 0.0089 0.0074 0.0025 0.0148 0.0511 0.0036 0.0065 0.0099 0.0082 0.0054 0.0378 0.0183 0.0192 4 0.0108 0.0043 0.0058 9 0.005 0.0108 0.0331 0.0092 0.0042 0.0079 0.0509 0.2056 0.0023 0.0453 0.0044 0.0397 0.0186 0.0025 0.0034 0.0436 0.0034 0.0172 0.0021 0.0051 0.0022 0.0039 0.0034 5 0.0122 0.031 5 4 3 2 1 1 2 3 4 5 5 4 3 2 1 1 2 3 4 5 5 4 3 2 1 1 2 3 4 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Position Length change 0.26 0.58 1 % error
Orthogonal Analytics Demonstrates Superior Material Intens. x10 7 Total Ion Chromatograph wgs022-a_1-b,1_01_6983.d: TIC -All MS Intens. x10 4 Deconvoluted Mass Spectrum - MS, 1.4-1.5min, Deconvoluted (MaxEnt, 594.37-2004.74, *0.375, 10000) 4 4 A 3 3 32127.1843 2 2 1 1 32310.3252 0 1 2 3 4 5 Time [min] 31146.7258 31781.8774 32455.7245 0 30800 31000 31200 31400 31600 31800 32000 32200 32400 32600 m/z Intens. x10 4 - MS, 1.3-1.7min, Deconvoluted (MaxEnt, 594.35-2004.68, *0.375, 10000) Intens. x10 7 wgs022-c_1-b,3_01_6991.d: TIC-All MS 3 32355.9671 B 6 4 2 2 1 32393.6721 0 1 2 3 4 5 Time [min] 32047.5554 32701.1024 0 31800 32000 32200 32400 32600 32800 m/z Intens. x10 7 wgs022-b_1-b,2_01_6987.d: TIC -All MS Intens. x10 5 2.5 - MS, 1.4-1.6min, Deconvoluted (MaxEnt, 606.93-2003.56, *0.375, 10000) Covalentlycoupled dgrna 6 4 2 2.0 1.5 1.0 32523.0079 0.5 0 1 2 3 4 5 Time [min] 0.0 32707.0850 32780.6450 31000 31250 31500 31750 32000 32250 32500 32750 33000 33250 m/z 18
Unparalleled Platform for Gene Editing Medicines ~10x Broad Range of Sites SpCas9 SaCas9 SpCas9 Variants SaCas9 Variants Cpf1 Cpf1 Variants Editas Platform Wide Delivery Options Viral Vector Lipid Nanoparticle Electroporation Disrupt Delete Correct Diverse Spectrum of Edits 19
Cas9 Stimulates the Endogenous Repair Pathways 5 WT Cas9 5 DSB C-NHEJ Resection 3 Locus Unaltered Small Deletions Small Insertions Alt-NHEJ HDR 3 Blunt EJ MMEJ SD-MMEJ Deletions Deletions Insertions SSA HR Large Deletions Correction 20
Cas9 is a Flexible Tool WT Cas9 N863A Nickases D10A Nickases 5 5 5 Blunt 3 3 Overhang 5 5 Overhang Could we engage different pathways by using these different variants? Could we selectively stimulate HDR? 21
DSBs Generated by D10A are Predominantly Repaired by HDR Modification (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% HDR Insertions Deletions 0% WT N863A D10A WT N863A D10A 3 5 Bothmer et al., Nat Comm 2017 22
Do Gene Conversion and Gene Correction have the same Genetic Requirement? Gene Correction Gene Conversion HBB HBD ssodn Do they both dependent on the HR pathway? 23
Gene Conversion and Gene Correction have Different Genetic Requirements Gene Correction 30 20 10 0 Neg. Cont. K.D. Rad51 Gene Conversion 30 20 10 0 Neg. Cont. K.D. Rad51 Gene Conversion 30 20 10 0 0 or Neg. Cont. or K.D. Rad51 SS ODN donor Endogenous HBD Plasmid donor HR is required for repair from double stranded donors (endogenous homology tracks or plasmids) but not single stranded donors 24
Conclusions from the Dual Nick Analysis Different ends activate different DNA repair pathways 5 WT Cas9 C-NHEJ 5 N863A Nickases Alt-NHEJ 5 D10A Nickases HDR Different donors stimulate different pathways Gene Correction mediated by ssodn is not HR dependent Bothmer et al., Nat Comm 2017 25
Summary We are developing precise and durable therapies by selecting and directing nuclease activity, delivery and the cellular repair response. We can effectively measure off-target and other genomic outcomes. We can make fully synthetic covalently-coupled dgrnas of high quality. Our understanding of how different lesions are repaired and the connection with the nature of the donor allow us to engineer for the desired outcomes. Hayat Abdulkerim Luis Barrera Anne Bothmer Frank Buquicchio Dawn Ciulla Cecilia Cotta-Ramusino Georgia Giannoukos Kiran Gogi Fred Harbinski Hari Jayaram Eugenio Marco Carrie Margulies Vic Myer Tanushree Phadke Terence Ta Chris Wilson i2 Pharmaceuticals team 26