A high-efficiency and versatile CRISPR/Cas9-mediated HDR-based biallelic editing system

A high-efficiency and versatile CRISPR/Cas9-mediated HDR-based biallelic editing system 基于同源介导修复（HDR）的CRISPR/Cas9双等位编辑系统的构建 1 2 first-author 1 1 1 2 1 corresp Zhiying ZHANG, zhangzhy@nwafu.edu.cn Zhiying ZHANG, zhangzhy@nwafu.edu.cn zhangzhy@nwafu.edu.cn https://orcid.org/0000-0001-6012-0889 College of Animal Science and Technology, Northwest A&F University, Yangling 712100, China College of Animal Science and Technology, Northwest A&F University, Yangling 712100, China Karolinska Institute, Department of Medicine Solna, Center for Molecular Medicine, Karolinska University Hospital Solna, 17176 Stockholm, Sweden Karolinska Institute, Department of Medicine Solna, Center for Molecular Medicine, Karolinska University Hospital Solna, 17176 Stockholm, Sweden Clustered regulatory interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 nuclease (Cas9), the third-generation genome editing tool, has been favored because of its high efficiency and clear system composition. In this technology, the introduced double-strand breaks (DSBs) are mainly repaired by non-homologous end joining (NHEJ) or homology-directed repair (HDR) pathways. The high-fidelity HDR pathway is used for genome modification, which can introduce artificially controllable insertions, deletions, or substitutions carried by the donor templates. Although high-level knock-out can be easily achieved by NHEJ, accurate HDR-mediated knock-in remains a technical challenge. In most circumstances, although both alleles are broken by endonucleases, only one can be repaired by HDR, and the other one is usually recombined by NHEJ. For gene function studies or disease model establishment, biallelic editing to generate homozygous cell lines and homozygotes is needed to ensure consistent phenotypes. Thus, there is an urgent need for an efficient biallelic editing system. Here, we developed three pairs of integrated selection systems, where each of the two selection cassettes contained one drug-screening gene and one fluorescent marker. Flanked by homologous arms containing the mutated sequences, the selection cassettes were integrated into the target site, mediated by CRISPR/Cas9-induced HDR. Positively targeted cell clones were massively enriched by fluorescent microscopy after screening for drug resistance. We tested this novel method on the amyloid precursor protein ( APP ) and presenilin 1 ( PSEN1 ) loci and demonstrated up to 82.0% biallelic editing efficiency after optimization. Our results indicate that this strategy can provide a new efficient approach for biallelic editing and lay a foundation for establishment of an easier and more efficient disease model. 目的： 2 本研究拟基于同源介导修复（HDR）原理，创建一种高效的双等位基因编辑系统 创新点： 2 通过引入双药物和双荧光筛选系统，提高双等位基因编辑效率的基础上，进一步通过剪切肽缩短表达盒长度，提高了CRISPR/Cas9双等位编辑的效率。 方法： 2 将嘌呤霉素抗性基因及博莱霉素抗性基因同绿色荧光蛋白及红色荧光蛋白分别配对组成筛选表达盒，插入带有基因组同源臂区域和突变序列的供体载体中组成双等位编辑供体组合。利用CRISPR/Cas9打靶系统在基因组靶位点处造成双链断裂，经HDR方式将突变序列引入基因组中，并经过荧光显微镜和双药物抗性筛选富集细胞克隆。采用聚合酶链式反应（PCR）、限制性酶切和DNA测序对细胞基因组编辑阳性率进行鉴定和统计。 结论： 2 成功构建了三套高效、普适的基因组双等位敲入系统，为基因组双等位精确编辑提供了新的思路。此方法的建立可为通过CRISPR/Cas9双等位编辑鉴定疾病发生过程中的纯合基因功能奠定基础。 Biallelic editing CRISPR/Cas9 Homology-directed repair (HDR) Homozygote 双等位编辑 CRISPR/Cas9 同源介导的修复（HDR） 纯合子 1 Introduction 1 The discovery and application of artificial endonucleases, including zinc finger nucleases ( Bibikova et al., 2003 Bibikova et al., 2003 ), transcription activator-like effector nucleases ( Boch et al., 2009 Boch et al., 2009 ), and clustered regulatory interspaced short palindromic repeats (CRISPR)/CRISPR-associated (Cas) systems ( Jinek et al., 2012 Jinek et al., 2012 ; Cong et al., 2013 Cong et al., 2013 ), have made arbitrary genomic modification a reality. To accurately modify the target sequence by genome editing, CRISPR-mediated base editing and prime-editing technologies were recently generated ( Li et al., 2021 Li et al., 2021 ); they can achieve targeted DNA insertions, deletions, and base conversions without generating double-strand breaks (DSBs) or needing donor DNA templates (as in most traditional genome-editing methods). However, for accuracy and efficiency, extensive knowledge of the precise design is needed to avoid off-target base editing and produce appropriate prime-editing guide RNA (pegRNA) designs ( Anzalone et al., 2019 Anzalone et al., 2019 ; Zuo et al., 2019 Zuo et al., 2019 ). Currently, homology-directed repair (HDR)‍-based CRISPR/Cas9 methods are still the most used due to their high efficiency and simple design principle ( Mali et al., 2013 Mali et al., 2013 ; Hsu et al., 2014 Hsu et al., 2014 ). Plasmids ( Koch et al., 2018 Koch et al., 2018 ), polymerase chain reaction (PCR) fragments ( Chu et al., 2015 Chu et al., 2015 ; Shao et al., 2017 Shao et al., 2017 ), and single-stranded oligonucleotides (ssODNs) ( Yang et al., 2013 Yang et al., 2013 ; Paquet et al., 2016 Paquet et al., 2016 ) are frequently used as donors, to introduce base insertions, deletions, or substitutions. Without selection markers, HDR efficiency only reaches a 15% positive rate at the cell line level and 18% for zygotes ( Paquet et al., 2016 Paquet et al., 2016 ; Codner et al., 2018 Codner et al., 2018 ). Its efficiency further decreases as the distance between the target site and introduced mutation increases ( Paquet et al., 2016 Paquet et al., 2016 ). Moreover, detection of positive clones is in general time-consuming and labor-intensive. To efficiently improve the acquisition of positive cell clones, plasmid donors containing genes resistant to drug screening can be used. However, the efficiency of correct and simultaneous modification of both alleles remains a challenge. Biallelic editing of homozygous cell lines and model animals is a reliable tool to study gene functions and species characteristics, as they have stable phenotypes and do not produce trait segregation. Thus, efficient biallelic gene editing is essential for gene function research. The introduction of drug-screening genes or fluorescent protein genes into reporter vectors or genomes can significantly improve the positive rate of targeted cells ( Ren et al., 2015 Ren et al., 2015 ; Wang et al., 2017 Wang et al., 2017 ; Wu et al., 2017 Wu et al., 2017 ; Yan et al., 2020 Yan et al., 2020 ). However, single selection systems cannot discriminate between monoallelic and biallelic targeting. Therefore, we hypothesized that using two donor vectors, each containing a different fluorescent protein as well as a drug-resistant gene, could lead to higher biallelic editing efficiency. Here, we have developed three pairs of such donors for targeted allele integration mediated by CRISPR/Cas9. All three were proved to be efficient and versatile, allowing a highly efficient selection of biallelic targeted clones. 2 Materials and methods 1 2.1 Preparation for genome targeting 2 The targeting plasmid was constructed by inserting the single-guide RNA (sgRNA) fragment under the human U6 promoter on the pLL3.7 backbone, which also contained the Streptococcus pyogenes Cas9 protein expression sequence. The efficiency of the targeting plasmids was detected using the single-strand annealing (SSA)‍-DsRed-puromycin resistance gene (Puro)-enhanced green fluorescent protein (eGFP) (RPG) surrogate reporter, with the insertion of the target sequence amplified by target-F/target-R ( Ren et al., 2015 Ren et al., 2015 ). The primers used for genome targeting preparation can be found in Table 1 Table 1 . Lower case: protective base for enzyme site; Italics: enzyme site; Bold: CRISPR/Cas9 target site. PCR: polymerase chain reaction; CRISPR: clustered regulatory interspaced short palindromic repeats; Cas9: CRISPR-associated protein 9 nuclease. Primer name Sequence (5'→3') U6-F ( Xba I) cgc TCTAGA TTTCCCATGATTCCTTCATAT U6-R ( Not I) acc GCGGCCGC AAAAAAGCACCGACTCGGTGCC APP-Left arm-F ( Eco RI) ccg GAATTC GGGTTGACAAATATCAAGACGG CAAAGGATGGAAGTTACAGG APP-Left arm-R ( Not I) aat GCGGCCGC CGCTAAAGAGGAAGAGGACAGGGTCAGTCTTGATATTTGTCAACCCAG APP-SSA-F ( Xho I) ccg CTCGAG CGCTAAAGAGGAAGAGGACAGGGAAAGCCCCAAAGTAGCAGTT APP-Right arm-R ( Cla I) ccc ATCGAT GGGTTGACAAATATCAAGACGG TTAACCCTCTTACTGCACCT PSEN1.target-F aat GCGGCCGC TCACAGAAGATACCGAGACT PSEN1.target-R cgc GGATCC GACAAGAATACCCAACCATAAG PSEN1.sgRNA-F TGTTGTCATGACTATCCTCCGTTTTAGAGCTAGAAATAGCAAG PSEN1.sgRNA-R GGAGGATAGTCATGACAACAGGTGTTTCGTCCTTTCCACAAG PSEN1.LA-F ( Bam HI) cgc GGATCC TGTTGTCATGACTATCCTCCTGG TCTGACAGTAGGAAGGAGTAGG PSEN1.LA-R ( Not I) aat GCGGCCGC CGCTAAAGAGGAAGAGGACAGGGCACGACAACAATGACACTGATCATG PSEN1.SSA-F ( Xho I) ccg CTCGAG CGCTAAAGAGGAAGAGGACAGGGACCTTTCTTCTGACAACCACTT PSEN1.SSA-R TCACGACAACAATGACACTGATCATG PSEN1.RA-F ATCAGTGTCATTGTTGTCGTGAGGATCCTCCTTGTGGTTCTGTATAAATACAG PSEN1.RA-R ( Hin dIII) ccc AAGCTT TGTTGTCATGACTATCCTCCTGG TAGGCAAGGTGTGACCAGTC pGK-F cgc GGATCC CCGGTAGGCGCCAA pGK-R TCAACTTGGCCATATTGGCTGCAGGTCGAA Zeo-F AGCCAATATGGCCAAGTTGACCAGTGCCGT Zeo-T2A-R1 ACCGCATGTTAGCAGACTTCCTCTGCCCTCGTCCTGCTCCTCGGCCACGA Zeo-T2A-R2 cac GCTAGC TGGGCCAGGATTCTCCTCGACGTCACCGCATGTTAGCAGACTTC P1 GGCTTGCTCTGTGTTGAATC P2 TCCCTATTGGCGTTACTATGG P3 TGAGCAAAGACCCCAACGAG P3-2 CTACGACGCCGAGGTCAAGA P4 GCAAATCCAAATCCCAAGTCAG P5 CCAGATTAGTACGGTGGCTTCA P6 ACAGTGGAGATGGTGAGAGGAT P2A-F1 ( Eco RI) ccg GAATTC GCAACAAACTTCTCACTA P2A-R ACGAGGCCATAGGCCCGGGATTCTCCTCCA TK-F TCCCGGGCCTATGGCCTCGTACCCCGGCCA TK-R ( Xho I) CCGCTCGAGCAGACATGATAAGATACA P2A-F2 (RFP) CACCGGCACCGCAACAAACTTCTCACTACT RFP-F ( Nhe I) CTAGCTAGCATGGCCTCCTCCGAGGAC RFP-R AGTTTGTTGCGGTGCCGGTGGAGTGGCGGC TK-F2 ( Nhe I) CTAGCTAGCATGGCCTCGTACCCCGGCCA TK-R2 ( Eco RI, Pme I) CCG GAATTCGTTTAAAC CTA ACCGGT GTTAGCCTCCC P7 GCTTGTACTCGGTCATATTGG P7-2 AACGGCACTGGTCAACTTGG P8 CCAACGGCGACCTGTATAAC To ensure that both alleles could be targeted by CRISPR/Cas9, sequences of the target sites were amplified by a target-F/target-R primer pair and confirmed in advance by Sanger sequencing. The primers used in homozygosity detection are listed in Table 1 Table 1 . 2.2 Construction of the selection cassette and donor vectors of the TK-Puro-eGFP/TK-Zeo-mRFP (TPG/TZR) system 2 Firstly, the eGFP-bovine growth hormone (BGH) polyA module in the JMB81-thymidine kinase (TK)-Puro-T2A-eGFP backbone (used for the SSA-mediated precise gene-editing system ( Li et al., 2018) Li et al., 2018) ) was replaced by a monomeric red fluorescent protein (mRFP)-BGH polyA fragment and it was cloned from plasmid pB-Puro-mRFP following Nhe I and Xho I enzymatic digestion. Next, the phosphoglycerate kinase (PGK) promoter-zeocin resistance gene (Zeo)‍-T2A module was cloned by overlap PCR. The PGK promoter was amplified from the JMB81-TK-Puro-T2A-eGFP backbone using PGK-F/R primer pair. The first half of the T2A sequence was attached to the end of Zeo by amplification using Zeo-F and Zeo-T2A-R1 primers. The PGK promoter and Zeo-T2A were then linked using PGK-F and Zeo-T2A-R2 primers, with the rest half of the T2A sequence added simultaneously. We chose the amyloid precursor protein ( APP ) gene locus to detect the feasibility of our TPG/TZR biallelic editing system because the APP .Donor (TPG) vector containing the APP homologous arms was available in our lab from previous research ( Li et al., 2018 Li et al., 2018 ). To generate the double-cut donor, the APP target sequence (GGGTTGACAAATATCAAGA CGG ) was added to the 5' and 3' of the homologous arms by PCR, using the following primers: APP-Left arm-F ( Eco RI) and APP-Right arm-R ( Cla I). Finally, the TPG selection cassette in the APP .Donor (TPG) (Fig. S1a) was replaced by the TZR selection cassette cut from the JMB81-TK-Zeo-T2A-mRFP backbone to construct the APP .Donor (TZR) (Fig. S1b). The homologous-arm plasmids for targeting the presenilin 1 ( PSEN1 ) locus were constructed according to the same strategy as the APP locus ( Li et al., 2018 Li et al., 2018 ). The PSEN1 target sequence (TGTTGTCATGACTATC‍CTCC TGG ) was added to the 5' and 3' of the homologous arms by PCR primers: PSEN1.LA-F ( Bam HI) and PSEN1.RA-R ( Hin dIII). Then, the TPG or TZR cassette was inserted into the homologous plasmid to produce PSEN1 .Donor (TPG) (Fig. S1c) or PSEN1 .Donor (TZR) vectors (Fig. S1d). All primers used to construct the donor plasmids are listed in Table 1 Table 1 . 2.3 Cell culture and transfection 2 HEK293T cells were used for all transfection in this study and were maintained in Dulbecco's modified Eagle's medium (DMEM; Gibco, USA) supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin, and 10% (volume fraction) fetal bovine serum (FBS; HyClone, USA), at 37 ℃ and 5% CO 2 . HEK293T cells were seeded into 24-well plates at a density of 40% and were transfected using Lipofectamine TM 2000 (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's manual, when cell confluence reached 90%. 2.4 Drug screening, expansion, and detection of cell clones 2 One day (24 h) after transfection, one quarter of the cells were transferred into a 100-mm plate and allowed to grow for an additional 5 d. Six days after transfection, the medium was supplemented with puromycin (3 μg/mL; Invitrogen) and zeocin (600 μg/mL; Invitrogen). After 3 d of drug treatment, the concentration of puromycin (Invitrogen) was reduced to 1 μg/mL; the concentration of zeocin remained unchanged. After another 7 d of selection, and upon control confirmation that all negative clones had died, we removed drugs from the media to minimize their negative effects on the growth of positive colonies and prevent random insertion caused by drug selection pressure. A number of cell colonies (about 50) with fluorescence were picked and seeded into 24-well plates for proliferation. Three-quarters of the cells for each clone were used for genomic DNA preparation using the E.Z.N.A. DNA Kit from OMEGA Bio-Tek (GA, USA) along with PCR analysis to confirm the correct integration. The other quarter were left in the well to grow until confluency and then frozen. To verify the insertion of the selection cassette, we designed primers located beside the homologous arm sequences. For the application of the 5' homologous arm, the forward primer was located upstream of the 5' homologous arm, and the reverse primer was located just upstream of the selection cassette (TPG/TZR, Puro-TK/Zeo-TK (PT/ZT), Puro-eGFP-TK/Zeo-mRFP-TK (PGT/ZRT)). For the application of the 3' homologous arm, the forward primer was located downstream of the selection cassette, whereas the reverse primer was located downstream of the 3' homologous arm. Primers flanking the homologous arms were used to detect the biallelic targeting clones. After integration of the selection cassette, the distance between the two flanking primers became too long to be amplified, so only the wild-type genome could produce the amplicons. All primer sequences are shown in Table 1 Table 1 . 2.5 Construction of PGT/ZRT selection cassette and donor vectors 2 The TK gene was cut off from the TPG/TZR selection-cassette plasmids (JMB81-TK-Puro-T2A-eGFP and JMB81-TK-Zeo-T2A-mRFP) by Cla I/ Bam HI restriction enzyme digestion. Later, the JMB81-pPGK-Puro-T2A-eGFP-BGH polyA and JMB81-pPGK-Zeo-T2A-mRFP-BGH polyA vectors were produced after sticky end fill-in with EasyPfu DNA polymerase (Trangen Biotech, China) and T4 ligase ligation (NEB, USA). The P2A shear peptide was cloned from the pU6-CBh-Cas9-T2A-mcherry-P2A-Ad4E4orf6 plasmid (Addgene #64222) using the primer pair P2A-F1 ( Eco RI) and P2A-R. The TK-SV40T polyA fragment was obtained by amplification with the primer pair TK-F and TK-R ( Xho I). P2A and TK-SV40T polyA were linked by overlap PCR using primer pair P2A-F1 ( Eco RI) and TK-R2 ( Xho I). Then the P2A-TK-SV40T polyA fragment was inserted into the intermediate JMB81-pPGK-Puro-T2A-eGFP-BGH polyA plasmid to generate the PGT selection cassette plasmid JMB81-pPGK-Puro-T2A-eGFP-P2A-TK-SV40T polyA. The mRFP fragment was amplified by primer pair RFP-F ( Nhe I) and RFP-R and later added to the 5' of the P2A-TK-SV40T polyA fragment, which was itself amplified by the primer pair P2A-F2 (RFP) and TK-R ( Xho I). Later, the mRFP-P2A-TK-SV40T polyA fragment was inserted into the intermediate JMB81-pPGK-Zeo-T2A-mRFP-BGH polyA plasmid to produce the ZRT selection-cassette plasmid JMB81-pPGK-Zeo-T2A-mRFP-P2A-TK-SV40T polyA. Lastly, the generated PGT/ZRT selection cassettes were transferred to the APP .Donor vectors by Not I and Xho I enzymatic restrictions, producing the final APP .Donor (PGT) and APP .Donor (ZRT) (Figs. S1e and S1f). 2.6 Construction of PT/ZT selection cassette and donor vectors 2 The TK gene fragment amplified by the primer pair TK-F2 ( Nhe I) and TK-R2 ( Eco RI, Pme I) and the eGFP gene of the JMB81-TK-Puro-T2A-eGFP plasmid replaced the TK gene by Nhe I/ Eco RI digestion and ligase reaction. Simultaneously, the mRFP gene on the JMB81-TK-Zeo-T2A-mRFP plasmid was replaced by the TK gene using Nhe I/ Pme I digestion. The newly constructed PT/ZT selection-cassette plasmids were then inserted into the APP .Donor (PT) (Fig. S1g) and APP .Donor (ZT) (Fig. S1h) by enzyme restriction with Not I and Xho I. 3 Results 1 3.1 Design and construction of the TPG/TZR for HDR-based biallelic editing of the APP locus 2 We chose the TK-Puro-eGFP backbone as one of the selection cassette systems ( Fig. 1 Fig. 1 a), because we had used it effectively in previous studies ( Li et al., 2018 Li et al., 2018 ). Zeocin and mRFP, which were learned from previous plasmid structures, were introduced (for fluorescence microscopy and drug screening, respectively) into a second selection cassette, TK-Zeo-mRFP, constructed from the TK-Puro-eGFP backbone ( Fig. 1 Fig. 1 a). For both APP. Donors, a double-cut donor design was applied to reach higher HDR efficiency ( Chen et al., 2017 Chen et al., 2017 ; Zhang et al., 2017 Zhang et al., 2017 ), and synonymous mutations were introduced to create Xba I restriction sites for later detection and prevention of secondary targeting. The TK gene was included to allow removal of the selection cassette from the genome after specific modification. To achieve biallelic editing, we transfected the biallelic editing system, consisting of the three vectors APP .CRISPR/Cas9, APP .Donor (TPG), and APP .Donor (TZR) (Figs. S1a and S1b), into HEK293T cells ( Fig. 1 Fig. 1 b). After 10-d dual drug screening for puromycin and zeocin resistance, 51 cell clones with both green and red fluorescence were successively picked, expanded, and analyzed by PCR and Sanger sequencing (Figs. 1c 1c – 1e 1e ). We used PCR with specific primer pairs for the 5' homologous arm (1315 bp, P1/P2), TPG 3' homologous arm (2225 bp, P3/P4), and TZR 3' homologous arm (2360 bp, P3-2/P4) to select the clones in which biallelic editing was successful (Figs. 1b 1b and 1d 1d ). The PCR amplicon for the 3' homologous arm could be digested by the Xba I restriction endonuclease, confirming successful introduction of the point mutation in the APP locus ( Fig. 1 Fig. 1 d). The insertion of the selection cassettes into both alleles was further confirmed by the fact that there was no band amplification as a result of primer pair P1/P4 flanking the homologous arm region; as the amplicon was too long to be amplified after the insertion of the selection cassette, only the wild-type allele can produce amplified products (Figs. 1b 1b and 1d 1d ). Finally, all three homologous arm detection fragments were detected by Sanger sequencing, confirming effective editing on both alleles ( Fig. 1 Fig. 1 e). Out of the 51 selected clones, 27 revealed biallelic editing at the APP locus; this represented a 52.94% positive biallelic targeting rate. 3.2 HDR-based biallelic editing of the PSEN1 locus 2 To test the universality of our biallelic editing system, we adapted the TPG/TZR selection cassettes to edit the PSEN1 locus, following the same design strategy as for the APP locus. Instead of Xba I, we introduced a Bam HI restriction site in the 3' homologous arm for detecting the point mutation after HDR ( Fig. 2 Fig. 2 a). After the transfection of the PSEN1 .CRISPR/Cas9, PSEN1 .Donor (TPG), and PSEN1 .Donor (TZR) plasmid mixtures (Figs. S1c and S1d), HEK293T cell clones with dual drug resistance and dual fluorescence were expanded for analysis. PCR amplification was used to detect the three homologous arms: 5' (1138 bp, P5/P2), TPG 3' (1947 bp, P3/P6), and TZR 3' (2082 bp, P3-2/P6). Primers flanking the TPG 3' arm and TZR 3' arm were digested by the Bam HI restriction endonuclease, verifying the success of biallelic editing ( Fig. 2 Fig. 2 b). No evident amplification with the flanking primer pair (P5/P6), as well as the Sanger sequencing results, conclusively proved the biallelic integration of the selection cassette ( Fig. 2 Fig. 2 c). In summary, 28 out of 51 picked clones had the correct biallelic editing on the PSEN1 locus, representing a 54.90% positive rate. 3.3 Optimization of the biallelic editing system 2 Since our 5 kb-length TPG/TZR system gave us only approximately 50% positive rates at two different loci, we next tested to find out whether a shorter cassette would improve the efficiency of our biallelic editing system. For comparison, we designed two sets of shorter selection cassettes to target the APP locus again ( Fig. 3 Fig. 3 a). We shortened the vectors to 3.2 and 2.9 kb (for PGT and ZRT, respectively) by using self-cleaving peptides for the expression of the Puro / Zeo and eGFP / mRFP screening genes simultaneously, with the same promoter and terminal sequences. Subsequently, to test the efficiency of the optimized biallelic editing systems, we cloned the newly constructed pairs of selection cassettes into the APP .Donor vectors (Figs. S1e and S1f) and co-transfected them with the targeting plasmid APP .CRISPR/Cas9 ( Fig. 3 Fig. 3 b). 1;2; After drug screening, clones from the PGT/ZRT-targeted cells were expanded and detected by PCR amplification (Figs. 3c 3c and 3d 3d ). To detect the two 5' homologous arms after biallelic editing, we combined reverse primers P7 and P7-2 (located in the 5' end of the Puro / Zeo sequences, respectively) with primer P1. The 3' homologous arm was amplified with the forward primer P8 (located in the 3' end of the TK gene) and primer P4, and by subsequent digestion of the amplicon with the Xba I endonuclease (Figs. 3b 3b and 3d 3d ). Sanger sequencing results for the two 5' homologous arms and the one 3' homologous arm revealed the success of biallelic integration (Fig. S2a). Notably, our new PGT/ZRT system increased its biallelic targeting rate to 82.00% (41 of the 50 clones). We finally constructed a third and even shorter pair of selection cassettes (PT/ZT) by further removing the fluorescent markers, which could also be used without the need for fluorescent microscopy or flow cytometry ( Fig. 3 Fig. 3 e). APP .Donor (PT) and APP .Donor (ZT) (Figs. S1g and S1h) flanking the same homologous arm sequences of APP locus and were transfected together with APP .CRISPR/Cas9 into HEK293T cells ( Fig. 3 Fig. 3 f). Without eGFP and mRFP fluorescent genes, we examined cell clones under a bright-field microscope ( Fig. 3 Fig. 3 g). The detection primers located in Puro , Zeo , and TK gene sequences were not influenced by the deletion of the fluorescent protein gene located in the middle of the selection cassette; thus, we used the same primer pairs to detect biallelic targeted clones (Figs. 3b 3b , 3f 3f , and 3h). Through the same screening process of PCR amplification, restriction enzyme digestion, and Sanger sequencing, the PT/ZT system led to 70.00% positive clones (biallelic targeting in 35 out of 50 clones). 4 Discussion 1 Eukaryotic cells have evolved efficient systems to repair hazardous DSBs for cell survival. The non-homologous end joining (NHEJ) pathway, which introduces indels or alterations, can perform the DNA repair in most cell cycles ( Bétermier et al., 2014 Bétermier et al., 2014 ; Zhao et al., 2017 Zhao et al., 2017 ; Scully et al., 2019 Scully et al., 2019 ). In contrast, the HDR pathway, which induces high-fidelity repair, is only active in the S/G2 phase, and besides, its repair efficiency is highly dependent on the concentration of the donors ( Saleh-Gohari and Helleday, 2004 Saleh-Gohari and Helleday, 2004 ; Savic et al., 2018 Savic et al., 2018 ). DSBs can be introduced simultaneously on both alleles by the efficient CRISPR/Cas9 system. However, even in the presence of enough donors, in most cases, only one of the alleles is repaired by HDR; the ends of the other allele are joined by NHEJ ( Xi et al., 2015 Xi et al., 2015 ; Paquet et al., 2016 Paquet et al., 2016 ; Li et al., 2018 Li et al., 2018 ). An ideal biallelic genome editing system should achieve high HDR efficiencies on both alleles in the same round of the repair process. Here, we designed and constructed an efficient, versatile, and programmable system for biallelic HDR-dependent gene editing that relies on dual screening donors. This TPG/TZR biallelic editing system achieved a positive rate of 52.94% and 54.90% at the APP and PSEN1 loci, respectively, when targeted separately. A multicistronic structure using the self-cleaving 2A peptides allowed for even higher biallelic targeting efficiency of the APP locus: 82.00% for the PGT/ZRT system and 70.00% for the PT/ZT system (which was further edited to remove the fluorescent reporter genes). These results reveal that the shorter the inserted fragment, the higher the positive rate of biallelic targeted clones. Moreover, the use of fluorescent selection genes appears to strengthen the screening of cell clones, but only to a certain extent. Of note, we used the TPG selection cassette to efficiently target IGF2 and RELA loci in the pig kidney cell line (PK15), obtaining 30.0% and 53.3% efficiency, respectively (unpublished data). This further emphasizes the versatility of our selection systems. Interestingly, although the TK gene and single-strand annealing arm were introduced in the donor sequences to allow for potential deletion of the selection cassettes from the genome after the first round of HDR-induced repair, we failed to obtain colonies containing both marker-free alleles. This is possibly a result of introducing four DSBs into the genome simultaneously, which could be too stressful for chromosome repair. Further research and optimization of our gene editing system will be needed if deletion of the selection cassette is required after gene editing. To put our novel strategy into perspective, we have previously tried to use dual surrogate reporter-integrated donors to screen biallelic gene-modified cells and demonstrated 34.09% efficiency at the C-C motif chemokine receptor 5 ( CCR5 ) locus in HEK293T cells ( Wu et al., 2017 Wu et al., 2017 ). Others have also reported low efficiency of biallelic gene editing; for instance, Eggenschwiler et al. (2016) Eggenschwiler et al. (2016) achieved biallelic editing efficiency of only 8.70% at the SERPINA locus using a dual-fluorescence selection strategy, and ssODN has led to 1.2% to 8.3% accurate biallelic editing ( Paquet et al., 2016 Paquet et al., 2016 ). Relatively higher homozygously targeted efficiency (70%) has been reported by Koch et al. (2018) Koch et al. (2018) using CRISPR/Cas9 D10A nickase to introduce a GFP fusion protein that was smaller than 1 kb and flanked with relatively short homologous arms (500–800 bp). In this study, we were able to significantly improve those results, obtaining a high biallelic editing efficiency of up to 50%–82%, despite using over 2 kb inserts and 800–1600 bp homologous arms ( Table 2 Table 2 ). TGP: TK-Puro-eGFP; TZR: TK-Zeo-mRFP; PGT: Puro-eGFP-TK; ZRT: Zeo-mRFP-TK; PT: Puro-TK; ZT: Zeo-TK; TK: thymidine kinase; Puro: puromycin resistance gene; eGFP: enhanced green fluorescent protein; Zeo: zeocin resistance gene; mRFP: monomeric red fluorescent protein. Biallelic system Targeted locus Drug resistance Fluorescent reporter Selection cassette (kb) Homology arms (kb) Ratio of positive clones Biallelic targeting efficiency (%) TPG/TZR APP Puromycin/zeocin eGFP/mRFP 5.1/4.9 1.0/1.6 27/51 52.94 TPG/TZR PSEN1 Puromycin/zeocin eGFP/mRFP 5.1/4.9 0.8/1.0 28/51 54.90 PGT/ZRT APP Puromycin/zeocin eGFP/mRFP 3.2/2.9 1.0/1.6 41/50 82.00 PT/ZT APP Puromycin/zeocin None 2.4/2.2 1.0/1.6 35/50 70.00 Dosage compensation effects can minimize protein expression differences caused by genomic equivalence. In this case, both alleles need to be manipulated for a proper study of gene function or point mutation effects in disease models. When generating in vivo disease models, targeting of both alleles is of the essence. As an example, homozygous mutations of the APP (APP Swe ) or PSEN1 (PSEN1 M146V ) genes showed higher Alzheimer's disease-causing protein expression than heterozygous counterparts ( Paquet et al., 2016 Paquet et al., 2016 ). Furthermore, in regards to gene therapy, a homozygous genotype is often necessary to overcome genetic diseases ( Evan, 2012 Evan, 2012 ; Sur et al., 2012 Sur et al., 2012 ), and heterozygous phenotypes are still not sufficient for restoring health ( Ye et al., 2016 Ye et al., 2016 ; Frangoul et al., 2021 Frangoul et al., 2021 ). Our universal selection cassettes overcome the limitations imposed by the relatively low efficiency of HDR in biallelic genome editing. Although recent advances in base editing and prime editing techniques allow editing of the genome without a donor DNA supply, the traditional HDR pathway is still advantageous in biallelic editing with dual donors ( Zhang, 2021 Zhang, 2021 ). 5 Conclusions 1 We generated three pairs of selection cassettes to achieve high efficiency of biallelic editing at different loci, allowing the insertion of 2–5 kb fragments into the genome flanked with 0.8‍–‍1.6 kb homology arms for HDR. APP and PSEN1 loci were chosen to be tested in HEK293T cells, and reached up to 82.00% positive efficiency. Our proof-of-concept studies strongly suggest that the three integrated selection systems we generated could be easily adapted to improve the targeting efficiency of virtually any other target locus. Regardless of the genome modification (knock-out or knock-in), homozygous cell lines not only provide more rigorous isogenic controls but also exhibit more typical genome-dependent disease-associated phenotypes. Consequently, based on extremely high biallelic knock-in efficiency, our biallelic editing system offers a good choice for disease model generation. References Anzalone AV , Randolph PB , Davis JR , et al. , 2019 . Search-and-replace genome editing without double-strand breaks or donor DNA . Nature , 576 ( 7785 ): 149 - 157 . https://doi.org/10.1038/s41586-019-1711-4 https://doi.org/10.1038/s41586-019-1711-4 Search-and-replace genome editing without double-strand breaks or donor DNA Bétermier M , Bertrand P , Lopez BS , 2014 . Is non-homologous end-joining really an inherently error-prone process? PLoS Genet , 10 ( 1 ): e1004086 . https://doi.org/10.1371/journal.pgen.1004086 https://doi.org/10.1371/journal.pgen.1004086 Is non-homologous end-joining really an inherently error-prone process? Bibikova M , Beumer K , Trautman JK , et al. , 2003 . Enhancing gene targeting with designed zinc finger nucleases . Science , 300 ( 5620 ): 764 . https://doi.org/10.1126/science.1079512 https://doi.org/10.1126/science.1079512 Enhancing gene targeting with designed zinc finger nucleases Boch J , Scholze H , Schornack S , et al. , 2009 . Breaking the code of DNA binding specificity of TAL-type III effectors . Science , 326 ( 5959 ): 1509 - 1512 . https://doi.org/10.1126/science.1178811 https://doi.org/10.1126/science.1178811 Breaking the code of DNA binding specificity of TAL-type III effectors Chen XY , Janssen JM , Liu J , et al. , 2017 . In trans paired nicking triggers seamless genome editing without double-stranded DNA cutting . Nat Commun , 8 : 657 . https://doi.org/10.1038/s41467-017-00687-1 https://doi.org/10.1038/s41467-017-00687-1 In trans paired nicking triggers seamless genome editing without double-stranded DNA cutting Chu VT , Weber T , Wefers B , et al. , 2015 . Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells . Nat Biotechnol , 33 ( 5 ): 543 - 548 . https://doi.org/10.1038/nbt.3198 https://doi.org/10.1038/nbt.3198 Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells Codner GF , Mianné J , Caulder A , et al. , 2018 . Application of long single-stranded DNA donors in genome editing: generation and validation of mouse mutants . BMC Biol , 16 : 70 . https://doi.org/10.1186/s12915-018-0530-7 https://doi.org/10.1186/s12915-018-0530-7 Application of long single-stranded DNA donors in genome editing: generation and validation of mouse mutants Cong L , Ran FA , Cox D , et al. , 2013 . Multiplex genome engineering using CRISPR/Cas systems . Science , 339 ( 6121 ): 819 - 823 . https://doi.org/10.1126/science.1231143 https://doi.org/10.1126/science.1231143 Multiplex genome engineering using CRISPR/Cas systems Eggenschwiler R , Moslem M , Fráguas MS , et al. , 2016 . Improved bi-allelic modification of a transcriptionally silent locus in patient-derived iPSC by Cas9 nickase . Sci Rep , 6 : 38198 . https://doi.org/10.1038/srep38198 https://doi.org/10.1038/srep38198 Improved bi-allelic modification of a transcriptionally silent locus in patient-derived iPSC by Cas9 nickase Evan G , 2012 . Taking a back door to target Myc . Science , 335 ( 6066 ): 293 - 294 . https://doi.org/10.1126/science.1217819 https://doi.org/10.1126/science.1217819 Taking a back door to target Myc Frangoul H , Altshuler D , Cappellini MD , et al. , 2021 . CRISPR-Cas9 gene editing for sickle cell disease and β-thalassemia . N Engl J Med , 384 ( 3 ): 252 - 260 . https://doi.org/10.1056/NEJMoa2031054 https://doi.org/10.1056/NEJMoa2031054 CRISPR-Cas9 gene editing for sickle cell disease and β-thalassemia Hsu PD , Lander ES , Zhang F , 2014 . Development and applications of CRISPR-Cas9 for genome engineering . Cell , 157 ( 6 ): 1262 - 1278 . https://doi.org/10.1016/j.cell.2014.05.010 https://doi.org/10.1016/j.cell.2014.05.010 Development and applications of CRISPR-Cas9 for genome engineering Jinek M , Chylinski K , Fonfara I , et al. , 2012 . A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity . Science , 337 ( 6096 ): 816 - 821 . https://doi.org/10.1126/science.1225829 https://doi.org/10.1126/science.1225829 A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity Koch B , Nijmeijer B , Kueblbeck M , et al. , 2018 . Generation and validation of homozygous fluorescent knock-in cells using CRISPR-Cas9 genome editing . Nat Protoc , 13 ( 6 ): 1465 - 1487 . https://doi.org/10.1038/nprot.2018.042 https://doi.org/10.1038/nprot.2018.042 Generation and validation of homozygous fluorescent knock-in cells using CRISPR-Cas9 genome editing Li C , Brant E , Budak H , et al. , 2021 . CRISPR/Cas: a Nobel Prize award-winning precise genome editing technology for gene therapy and crop improvement . J Zhejiang Univ-Sci B (Biomed & Biotechnol) , 22 ( 4 ): 253 - 284 . https://doi.org/10.1631/jzus.B2100009 https://doi.org/10.1631/jzus.B2100009 CRISPR/Cas: a Nobel Prize award-winning precise genome editing technology for gene therapy and crop improvement Li XY , Bai YC , Cheng XZ , et al. , 2018 . Efficient SSA-mediated precise genome editing using CRISPR/Cas9 . FEBS J , 285 ( 18 ): 3362 - 3375 . https://doi.org/10.1111/febs.14626 https://doi.org/10.1111/febs.14626 Efficient SSA-mediated precise genome editing using CRISPR/Cas9 Mali P , Yang LH , Esvelt KM , et al. , 2013 . RNA-guided human genome engineering via Cas9 . Science , 339 ( 6121 ): 823 - 826 . https://doi.org/10.1126/science.1232033 https://doi.org/10.1126/science.1232033 RNA-guided human genome engineering via Cas9 Paquet D , Kwart D , Chen A , et al. , 2016 . Efficient introduction of specific homozygous and heterozygous mutations using CRISPR/Cas9 . Nature , 533 ( 7601 ): 125 - 129 . https://doi.org/10.1038/nature17664 https://doi.org/10.1038/nature17664 Efficient introduction of specific homozygous and heterozygous mutations using CRISPR/Cas9 Ren CH , Xu K , Liu ZT , et al. , 2015 . Dual-reporter surrogate systems for efficient enrichment of genetically modified cells . Cell Mol Life Sci , 72 ( 14 ): 2763 - 2772 . https://doi.org/10.1007/s00018-015-1874-6 https://doi.org/10.1007/s00018-015-1874-6 Dual-reporter surrogate systems for efficient enrichment of genetically modified cells Saleh-Gohari N , Helleday T , 2004 . Conservative homologous recombination preferentially repairs DNA double-strand breaks in the S phase of the cell cycle in human cells . Nucleic Acids Res , 32 ( 12 ): 3683 - 3688 . https://doi.org/10.1093/nar/gkh703 https://doi.org/10.1093/nar/gkh703 Conservative homologous recombination preferentially repairs DNA double-strand breaks in the S phase of the cell cycle in human cells Savic N , Ringnalda FC , Lindsay H , et al. , 2018 . Covalent linkage of the DNA repair template to the CRISPR-Cas9 nuclease enhances homology-directed repair . eLife , 7 : e33761 . https://doi.org/10.7554/eLife.33761 https://doi.org/10.7554/eLife.33761 Covalent linkage of the DNA repair template to the CRISPR-Cas9 nuclease enhances homology-directed repair Scully R , Panday A , Elango R , et al. , 2019 . DNA double-strand break repair-pathway choice in somatic mammalian cells . Nat Rev Mol Cell Biol , 20 ( 11 ): 698 - 714 . https://doi.org/10.1038/s41580-019-0152-0 https://doi.org/10.1038/s41580-019-0152-0 DNA double-strand break repair-pathway choice in somatic mammalian cells Shao SM , Ren CH , Liu ZT , et al. , 2017 . Enhancing CRISPR/Cas9-mediated homology-directed repair in mammalian cells by expressing Saccharomyces cerevisiae Rad52 . Int J Biochem Cell Biol , 92 : 43 - 52 . https://doi.org/10.1016/j.biocel.2017.09.012 https://doi.org/10.1016/j.biocel.2017.09.012 Enhancing CRISPR/Cas9-mediated homology-directed repair in mammalian cells by expressing Saccharomyces cerevisiae Rad52 Sur IK , Hallikas O , Vähärautio A , et al. , 2012 . Mice lacking a Myc enhancer that includes human SNP rs6983267 are resistant to intestinal tumors . Science , 338 ( 6112 ):‍ 1360 - 1363 . https://doi.org/10.1126/science.1228606 https://doi.org/10.1126/science.1228606 Mice lacking a Myc enhancer that includes human SNP rs6983267 are resistant to intestinal tumors Wang G , Yang LH , Grishin D , et al. , 2017 . Efficient, footprint-free human iPSC genome editing by consolidation of Cas9/CRISPR and piggyBac technologies . Nat Protoc , 12 ( 1 ): 88 - 103 . https://doi.org/10.1038/nprot.2016.152 https://doi.org/10.1038/nprot.2016.152 Efficient, footprint-free human iPSC genome editing by consolidation of Cas9/CRISPR and piggyBac technologies Wu Y , Xu K , Ren CH , et al. , 2017 . Enhanced CRISPR/Cas9-mediated biallelic genome targeting with dual surrogate reporter-integrated donors . FEBS Lett , 591 ( 6 ): 903 - 913 . https://doi.org/10.1002/1873-3468.12599 https://doi.org/10.1002/1873-3468.12599 Enhanced CRISPR/Cas9-mediated biallelic genome targeting with dual surrogate reporter-integrated donors Xi LH , Schmidt JC , Zaug AJ , et al. , 2015 . A novel two-step genome editing strategy with CRISPR-Cas9 provides new insights into telomerase action and TERT gene expression . Genome Biol , 16 : 231 . https://doi.org/10.1186/s13059-015-0791-1 https://doi.org/10.1186/s13059-015-0791-1 A novel two-step genome editing strategy with CRISPR-Cas9 provides new insights into telomerase action and TERT gene expression Yan NN , Sun YS , Fang YY , et al. , 2020 . A universal surrogate reporter for efficient enrichment of CRISPR/Cas9-mediated homology-directed repair in mammalian cells . Mol Ther Nucleic Acids , 19 : 775 - 789 . https://doi.org/10.1016/j.omtn.2019.12.021 https://doi.org/10.1016/j.omtn.2019.12.021 A universal surrogate reporter for efficient enrichment of CRISPR/Cas9-mediated homology-directed repair in mammalian cells Yang LH , Guell M , Byrne S , et al. , 2013 . Optimization of scarless human stem cell genome editing . Nucleic Acids Res , 41 ( 19 ): 9049 - 9061 . https://doi.org/10.1093/nar/gkt555 https://doi.org/10.1093/nar/gkt555 Optimization of scarless human stem cell genome editing Ye L , Wang JM , Tan YT , et al. , 2016 . Genome editing using CRISPR-Cas9 to create the HPFH genotype in HSPCs: an approach for treating sickle cell disease and β‍-thalassemia . Proc Natl Acad Sci USA , 113 ( 38 ): 10661 - 10665 . https://doi.org/10.1073/pnas.1612075113 https://doi.org/10.1073/pnas.1612075113 Genome editing using CRISPR-Cas9 to create the HPFH genotype in HSPCs: an approach for treating sickle cell disease and β‍-thalassemia Zhang BH , 2021 . CRISPR/Cas gene therapy . J Cell Physiol , 236 ( 4 ): 2459 - 2481 . https://doi.org/10.1002/jcp.30064 https://doi.org/10.1002/jcp.30064 CRISPR/Cas gene therapy Zhang JP , Li XL , Li GH , et al. , 2017 . Efficient precise knockin with a double cut HDR donor after CRISPR/Cas9-mediated double-stranded DNA cleavage . Genome Biol , 18 : 35 . https://doi.org/10.1186/s13059-017-1164-8 https://doi.org/10.1186/s13059-017-1164-8 Efficient precise knockin with a double cut HDR donor after CRISPR/Cas9-mediated double-stranded DNA cleavage Zhao X , Wei CW , Li JJ , et al. , 2017 . Cell cycle-dependent control of homologous recombination . Acta Biochim Biophys Sin , 49 ( 8 ): 655 - 668 . https://doi.org/10.1093/abbs/gmx055 https://doi.org/10.1093/abbs/gmx055 Cell cycle-dependent control of homologous recombination Zuo EW , Sun YD , Wei W , et al. , 2019 . Cytosine base editor generates substantial off-target single-nucleotide variants in mouse embryos . Science , 364 ( 6437 ): 289 - 292 . https://doi.org/10.1126/science.aav9973 https://doi.org/10.1126/science.aav9973 Cytosine base editor generates substantial off-target single-nucleotide variants in mouse embryos Xinyi LI and Zhiying ZHANG conceived and supervised the project. Xinyi LI, Bing SUN, Hongrun QIAN, and Jinrong MA conducted experiments and data analysis. Xinyi LI wrote the manuscript and edited it in collaboration with Magdalena PAOLINO. All authors have read and approved the final manuscript, and therefore, have full access to all the data in the study and take responsibility for the integrity and security of the data. This work was supported by the National Science and Technology Major Project of China (No. 2018ZX08010-09B), the Swedish Research Council (No. NE 2016-04458), and the Ragnar Söderberg Foundation (No. M21/17), Sweden. We would like to thank all the members of Professor ZHANG's lab for their excellent technical assistance and helpful discussions. 10.1631/jzus.B2100196 1673-1581（2022）02-0141-12 This work was supported by the National Science and Technology Major Project of China (No. 2018ZX08010-09B), the Swedish Research Council (No. NE 2016-04458), and the Ragnar Söderberg Foundation (No. M21/17), Sweden. 2021-02-27 2021-06-20 2022-02-15 2022-04-19 2021-12-15 Copyright © Zhejiang University and Springer-Verlag GmbH Germany, part of Springer Nature 2022 2022 浙江大学出版社 Xinyi LI, Bing SUN, Hongrun QIAN, Jinrong MA, Magdalena PAOLINO, and Zhiying ZHANG declare that they have no conflict of interests. This article does not contain any studies with human or animal subjects performed by any of the authors. Compliance with ethics guidelines Supplementary information 1 Figs. S1 and S2 浙江大学学报（英文版）（B辑：生物医学和生物技术） 2 23