Lymphohematopoietic engineering using cas9 base editors

Abstract

To produce a genetically engineered lymphohematopoietic cell.SOLUTION: Provided herein are methods and systems for targeted gene disruption (knock-out, missense mutation) and targeted gene knock-in in mammalian cells using base editors and guide RNAs (gRNAs) designed to target splice acceptor-splice donor sites. Also provided herein are universally acceptable genetically engineered cells comprising targeted disruptions in immunotherapy-related genes and comprising a CAR / TCR for therapeutic applications.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority from U.S. Provisional Patent Application No. 62 / 642,151, filed March 13, 2018, the disclosure of which is incorporated herein by reference in its entirety for all purposes. [Background technology]

[0002] Precise modulation of primary human cells has multiple applications in the fields of immunotherapy, autoimmunity and enzymopathies. Modulation of a patient's immune cells at the genetic level is an attractive route for therapy due to the permanence of the treatment and the low risk of rejection by the patient. One approach to gene editing of immune cells is to use the clustered regularly interspaced short palindromic repeats (CRISPR) system to induce double-strand breaks (DSBs) within the gene of interest, thereby resulting in the formation of small insertions or deletions (collectively referred to as "indels") created by highly variable repair by the non-homologous end joining (NHEJ) pathway. Alternatively, precise genomic alterations can be achieved by introducing DSBs together with the simultaneous delivery of a DNA template for repair by homology-directed repair (HDR). While this approach is efficient and reliable when simply disrupting a single gene by NHEJ, precise alteration of a single nucleotide by HDR is much less efficient. Furthermore, the induction of multiple DSBs during multiple gene editing procedures may lead to undesirable genotoxicity and the formation of large-scale chromosomal translocations that are potentially oncogenic.Therefore, there remains a need in the field for more controlled and safer methods of multiple genetic manipulation of human immune cells that limit the induction of toxic DSBs. Summary of the Invention

[0003] In a first aspect, a method for generating genetically engineered lymphohematopoietic cells is provided herein, which may comprise or consist essentially of: (a) introducing into lymphohematopoietic cells (i) a plasmid, mRNA, or protein encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain, where the nickase domain comprises a base excision repair inhibitor domain, and (ii) one or more splice acceptor-splice donor (SA-SD) gRNAs having complementarity to a target nucleic acid sequence to be genetically modified, and (b) culturing the transfected cells under conditions promoting the destruction of the splice site targeted by the one or more SA-SD gRNAs, whereby the target nucleic acid sequence is modified by the base editor fusion protein and the one or more splice acceptor-splice donor (SA-SD) gRNAs compared to untransfected lymphohematopoietic cells, and thereby generating genetically engineered lymphohematopoietic cells. In some cases, the method further comprises introducing one or more gRNAs into the lymphohematopoietic cells designed to generate one or more targeted knock-ins or missense mutations, whereby the engineered lymphohematopoietic cells comprise at least one gene knockout and one or more gene knock-ins or missense mutations. In some cases, the method further comprises introducing one or more gRNAs into the lymphohematopoietic cells designed to generate one or more targeted knock-ins and one or more missense mutations, whereby the engineered lymphohematopoietic cells comprise at least one gene knockout, at least one gene knock-in, and at least one missense mutation. The base editor fusion protein can be BE3, BE4, or an adenine base editor (ABE).The lymphoid hematopoietic cells can be T cells, natural killer (NK) cells, B cells, or CD34+ hematopoietic stem progenitor cells (HSPCs). The one or more SA-SD gRNAs can be chemically modified to include a 2'-O-methyl phosphorothioate modification in at least one 5' nucleotide and at least one 3' nucleotide of each gRNA. The base editor fusion protein and the one or more splice acceptor-splice donor (SA-SD) gRNAs can exhibit an efficiency of C to T conversion of about 50% to about 90%. The one or more SA-SD gRNAs can be selected from the sequences shown in Table 1.

[0004] In another aspect, a method for generating genetically modified T cells is provided herein, comprising the steps of: (a) introducing into a human T cell (i) a plasmid, mRNA, or protein encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain, where the nickase domain comprises a base excision repair inhibitor domain, (ii) one or more splice acceptor-splice donor (SA-SD) gRNAs to disrupt expression of TRAC, B2M, and PDCD1, respectively, (iii) a donor DNA template encoding a T cell receptor (TCR) and a chimeric antigen receptor (CAR), and (iv) two gRNAs complementary to a target insertion site; and (b) introducing into the human T cell a plasmid, mRNA, or protein encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain, where the nickase domain comprises a base excision repair inhibitor domain, (ii) one or more splice acceptor-splice donor (SA-SD) gRNAs to disrupt expression of TRAC, B2M, and PDCD1, respectively, (iii) a donor DNA template encoding a T cell receptor (TCR) and a chimeric antigen receptor (CAR), and (iv) two gRNAs complementary to a target insertion site; The method may comprise or consist essentially of (a) culturing the T cells under conditions that promote the disruption of the splice site targeted by the gRNA, whereby expression of TRAC, B2M, and PDCD1 is reduced compared to non-transfected T cells, and (c) culturing the transfected T cells under conditions that promote targeted knock-in of the donor DNA template at the target insertion site. The base editor fusion protein may be BE3, BE4, or an adenine base editor (ABE). The one or more SA-SD gRNAs may be selected from the sequences shown in Table 1. One or more of the SA-SD gRNAs and the gRNAs complementary to the target insertion site may be chemically modified to include a 2'-O-methyl phosphorothioate modification at at least one 5' nucleotide and at least one 3' nucleotide of each gRNA. The donor DNA template may be provided as a rAAV. The TCR may specifically bind to a tumor antigen. The CAR may include a CAR antigen-binding domain that specifically binds to a tumor antigen. The TCR and CAR may bind to different antigens.

[0005] In a further aspect, there is provided herein a genetically modified cell obtained according to the method of this disclosure. The present invention will be better understood, and features, aspects and advantages other than those set forth above will become apparent, when the following detailed description thereof is considered. In such detailed description, reference is made to the following drawings, in which: [Brief description of the drawings]

[0006] [Figure 1] Figure 1 depicts a diagram showing a Cas9 base editor (BE) bound to target DNA (left) and protospacers showing the base editing window achieved with BE3 and BE4 (right). Also shown are logo diagrams showing mammalian splice donor (SD) and splice acceptor (SA) elements and the relative orientation of the protospacers utilized for BE knockout by disruption of splice sites. [Diagram 2] FIG. 1 shows an example workflow for single gene knockout by introduction of a premature STOP (pmSTOP) codon or by disrupting a splice site with a base editor. [Figure 3-1]3A-3L demonstrate the assessment of guide RNA activity for gene disruption in PDCD1, B2M, and TRAC. (a) Diagram of the PDCD1 locus showing the relative position of each sgRNA. The colored portion of the box represents the protein coding region, and the red vertical line indicates the stop codon. (b) Quantification of C to T conversion of the targeted base for each PDCD1 sgRNA following co-delivery with either BE3 mRNA or BE4 mRNA as determined by EditR analysis of Sanger sequencing traces (n=3 independent T cell donors). (c) Frequency of PDCD1 protein knockout following delivery of the indicated sgRNA with either BE3 mRNA or BE4 mRNA as determined by flow cytometry (n=3 independent T cell donors). (d) Quantitation of C to T / A / G conversions at all Cs within the detected editing window (shown in red) of PDCD1 Ex1 SD sgRNA following co-delivery with either BE3 or BE4 mRNA as determined by EditR analysis of Sanger sequencing traces (n=3 independent T cell donors). The underlined C indicates the target nucleotide important for proper splicing. (e) Diagram of the TRAC locus showing the relative position of each sgRNA. (f) Quantitation of C to T conversions at the target base for each TRAC sgRNA following co-delivery with either BE3 or BE4 mRNA as determined by EditR analysis of Sanger sequencing traces (n=3 independent T cell donors). (g) Frequency of TRAC protein knockout following delivery of the indicated sgRNA with either BE3 or BE4 mRNA as determined by flow cytometry for CD3 loss (n=3 independent T cell donors). (h) Quantification of C to T / A / G conversions in all cytosines within the detected editing window (shown in red) of the TRAC Ex3 SA sgRNA following co-delivery with either BE3 or BE4 mRNA as determined by EditR analysis of Sanger sequencing traces (n=3 independent T cell donors). (i) Diagram of the B2M locus showing the relative position of each sgRNA.(j) Quantitation of C to T conversions of the targeted base for each B2M sgRNA following co-delivery with either BE3 or BE4 mRNA as determined by EditR analysis of Sanger sequencing traces (n=3 independent T cell donors). (k) Frequency of B2M protein knockout following delivery of the indicated sgRNA with either BE3 or BE4 mRNA as determined by flow cytometry for B2M loss (n=3 independent T cell donors). (l) Quantitation of C to T / A / G conversions of all cytosines within the detected editing window (shown in red) of B2M Ex1 SD sgRNA following co-delivery with either BE3 or BE4 mRNA as determined by EditR analysis of Sanger sequencing traces (data presented as mean ± SD, n=3 independent biological T cell donors). p values ​​calculated by Student’s paired two-tailed t test between the highest and second highest editing guides (non-significant P>0.05, *P≤0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001). [Figure 3-2] As shown in Figure 3-1 above. [Figure 3-3] As shown in Figure 3-1 above. [Figure 4-1]4A-4F are graphs demonstrating optimization of multiplex editing using optimal sgRNAs (TRAC Ex3 SA, B2M Ex1 SD, and PDCD1 Ex1 SD). (a) Conversion frequency of targeted cytosines to all other bases in TRAC, PDCD1, and B2M analyzed by NGS after co-delivery of the three targeting sgRNAs and first generation BE3 (BE3) mRNA or BE4 (BE4) mRNA delivered at a 3 μg dose; BE4 protein complexed with sgRNAs (BE4 RNP); or codon-optimized BE4 mRNA (coBE4) delivered at a 1.5 μg or 4 μg dose. (b) Frequency of indels in TRAC, PDCD1, and B2M analyzed by NGS after delivery of three targeting sgRNAs and first generation BE3 (BE3) mRNA or BE4 (BE4) mRNA delivered at a 3 μg dose; BE4 protein complexed with sgRNA (BE4 RNP); or codon-optimized BE4 mRNA (coBE4) delivered at a 1.5 μg or 4 μg dose. (c) Frequency of indels in TRAC, PDCD1, and B2M analyzed by NGS after co-delivery of three targeting sgRNAs and SpCas9 nuclease mRNA at a 1.5 μg or 4 μg dose. (d) Frequency of TRAC, PDCD1, and B2M protein loss measured by flow cytometry 7 days after delivery of three targeting sgRNAs and first generation BE3 (BE3) mRNA or BE4 (BE4) mRNA delivered at a 3 μg dose; BE4 protein complexed with sgRNA (BE4 RNP); and codon-optimized BE4 mRNA (coBE4) delivered at a 1.5 μg or 4 μg dose. (e) SPICE display of multiplex flow cytometry analysis performed 7 days after electroporation. (f) Quantification of WT, single gene KO, double gene KO, and triple gene KO fractions. Data presented as mean ± SD, n=2 from two independent biological T cell donors. [Figure 4-2] As shown in Figure 4-1 above. [Figure 4-3] As shown in Figure 4-1 above. [Figure 4-4] As shown in Figure 4-1 above. [Figure 5A]5A-5B demonstrate the frequency of translocations in multiply edited T cells. (a) Circos plots of possible translocation outcomes resulting from double-strand break induction at TRAC, B2M, PDCD1, and PDCD1 OT sites. (b) Quantification by droplet digital PCR of translocation frequency. All assays were performed in technical duplicate across n=2 independent biological T cell donors. [Figure 5B] As shown in Figure 5A above. [Figure 6A] Figures 6A-6D demonstrate the function of multiply edited T cells. (a) Expression of memory markers CD27 and CD45ro after editing and expansion. (b) Individual and (c) combined cytokine production by CD4 and CD8 T cells after activation. (d) Ability of T cells to kill CD19neg K562, CD19pos Raji cells, or CD19pos / PD-L1pos Raji cells after co-culture with T cells, measured by luciferase luminescence assay. Graph titles indicate E:T ratios. Data are presented as mean ± SD for assays performed in triplicate in two independent biological T cell donors. (Non-significant P>0.05, *P≦0.05, **P≦0.01, ***P≦0.001, ****P≦0.0001). [Figure 6B] As shown in Figure 6A above. [Figure 6C] As shown in Figure 6A above. [Figure 6D] As shown in Figure 6A above. [Figure 7] Figures 7A-7C are graphs demonstrating non-target editing for each sgRNA in Figures 3A-3L. Data are analyzed from NGS. Stacked bar heights represent the mean with error bars ± 1 standard deviation. n=3 independent donors. [Figure 8-1] Figures 8A-8C are graphs showing indels for all samples in Figures 3A-3L. Data are analyzed from NGS. Stacked bar heights represent the mean with error bars ± 1 standard deviation. n=3 independent donors. [Figure 8-2] As shown in Figure 8-1 above. [Figure 9]

[0023] Figure 1: Graphs demonstrating multiplex base editing of T cells using first generation, low dose (1.5μg) BE3 mRNA or BE4 mRNA. Bar graphs showing base editor-mediated knockout of TRAC, B2M and PDCD1 at the protein level. Protein expression was assessed by flow cytometry as described in the methods section. n=2 independent donors. [Figure 10]

[0023] Figure 14. Base editor protein levels following electroporation of T cells. Digital Western blot results evaluating protein levels achieved using mRNA encoding Cas9, BE3, BE4, and codon-optimized BE4 24 hours after electroporation of stimulated T cells in two independent donors. Purified BE3 protein was also used as a positive control for antibody detection of BE protein. [Figure 11] Representative flow plots of PDCD1, B2M, and TRAC upon restimulation. Five days after electroporation, T cells were restimulated to induce expression of PDCD1, allowing assessment of the frequency of PD-1 protein knockout. Shown here are representative flow cytometry plots of TRAC, B2M, and PDCD1 expression of donor-matched T cells after multiple co-BE4 mRNA editing and pulse-only controls (left column). [Figure 12] 12A-12B are graphs demonstrating assessment of computationally predicted off-target base editing and indel formation. Frequency of base editing (A) and indels (B) at on-target and top 10 computationally predicted off-target sgRNA binding sites assessed using next-generation sequencing using optimal sgRNAs targeting TRAC, B2M or PDCD1 in combination with Cas9 or BE4 mRNA in T cells. [Figure 13]Graph demonstrating frequency of translocations between three target loci. Quantification by droplet digital PCR of translocation frequency between TRAC, B2M, and PDCD1 following delivery of three sgRNAs and spCas9 protein, spCas9 mRNA, BE4 protein, or coBE4 mRNA. n=2 independent T cell donors were assayed in duplicate. [Figure 14] Graph demonstrating frequency of translocations between target loci and PDCD1 off-target sites. Quantification by droplet digital PCR of frequency of translocations between PDCD1 OT sites and TRAC, B2M, and PDCD1 following delivery of three sgRNAs and spCas9 protein, spCas9 mRNA, BE4 protein, or coBE4 mRNA. n=2 independent T cell donors were assayed in duplicate. [Figure 15] Figures 15A-15B are graphs demonstrating the efficiency of CAR transduction and T cell expansion. (A.) Bar graph showing the frequency of transduced T cells using MND-CD19 CAR-RQR8 lentiviral vector at an MOI of 20 in two independent donors by staining for RQR8. RQR8 is a hybrid molecule that contains domains for staining with antibodies specific for CD34 and CD20 and serves as a surrogate to determine the frequency of CAR-positive T cells. (B.) Bar graph showing the number of viable cells 5 and 12 days after electroporation and transduction. n=2 independent donors. [Figure 16] FIG. 1 illustrates targeted knock-in (KI) using the activity and efficiency of BE nickase compared to Cas9 nuclease and Cas9 nickase. [Figure 17-1]Figures 17A-17B demonstrate sequencing of subcloned PCR products across translocation junctions. (A) Results of translocation PCR performed between the indicated target genes using Cas9 or BE mRNA or protein as described. AAVS1 control PCR was also performed to confirm the quality and functionality of gDNA for PCR. (B) PCR products from (A) were TA cloned into a TOPO plasmid and then analyzed by Sanger sequencing. The resulting chromatograms were then aligned against a hypothetical "perfect" junction sequence between the gRNA cleavage site of the indicated target gene and the gRNA cleavage site of the aligned target gene. The ddPCR probes used to generate the data are also shown. [Figure 17-2] As shown in Figure 17-1 above. [Figure 18-1] Figures 18A-18C demonstrate efficient base editing in expanded, activated UCB CD34+ HSPCs. C to T conversions were calculated using EditR. [Figure 18-2] As shown in Figure 18-1 above. [Figure 19]

[0023] Figure 1 shows a previously identified ADAM17-cleavable region of CD16, with a key serine (red) that, if mutated to a proline, would render CD16 uncleavable by ADAM17. Using CT base editor variants, we were able to target the adjacent valine (blue). Possible amino acid changes to this valine that could be achieved using CT base editor variants are shown. [Figure 20-1]Figures 20A-20B show representative Sanger sequencing chromatograms. (A) shows sequencing chromatograms from control (BE3-VQR alone) and edited (BE3-VQR+CD16 gRNA) samples (top). C to T conversion was calculated using EditR. Results from two separate donors are combined in a bar graph (bottom). (B) shows sequencing chromatograms from control (ABE alone) and edited (ABE+CD16a gRNA) samples. T to C conversion was calculated using EditR. The ABE+CD16a gRNA sample demonstrates conversion to a cleavage-resistant CD16a variant. [Figure 20-2] As shown in Figure 20-1 above. [Figure 21-1] 21A-21D demonstrate that Fanconi Anemia (FA) fibroblasts, which are DNA repair deficient, can undergo base editing using BE3 and BE4. (A) Sequencing chromatogram of FANCA Ex.39 c.3934+2 T>C pos.4 gRNA target site in FA patient-derived fibroblasts treated with BE4. The frequency of base editing was quantified by EditR software. (B) Results from an experiment similar to A, but performed with ABE and PDCD1 Ex.1 SD gRNA. (C) Details of base editing in control (MPS1) fibroblasts and FA patient-derived fibroblasts using BE3. (D) Results of Cas9 nuclease activity in fibroblasts from Sanger sequencing data analyzed using the TIDE algorithm. [Figure 21-2] As shown in Figure 21-1 above. [Figure 21-3] As shown in Figure 21-1 above. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] While the invention is susceptible to various modifications and alternative forms, exemplary embodiments thereof have been shown by way of example in the drawings and are herein described in detail. It should be understood, however, that the description of the exemplary embodiments is not intended to limit the invention to the particular forms disclosed, but on the contrary, is intended to cover all modifications, equivalents and alterations falling within the spirit and scope of the invention as defined in the appended claims. Detailed Description All publications, including but not limited to patents and patent applications, cited in this specification are herein incorporated by reference as if set forth in their entirety in this application.

[0008] The methods, systems, and compositions described herein are based, at least in part, on the inventors' discovery of a protocol for genome manipulation of primary human lymphoid hematopoietic cells using CRISPR-Cas9 base editors. To create point mutations without using homology-directed repair (HDR), researchers have developed CRISPR base editors that fuse Cas9 nickase or dCas9 to a cytidine deaminase-like APOBEC1. Unlike CRISPR, base editing does not cut double-stranded DNA, but instead uses a deaminase enzyme to precisely rearrange some of the atoms in one of the four bases that make up DNA or RNA, converting the base without changing the bases around it. The base editor is targeted to a specific locus by a guide RNA (gRNA) and can convert cytidine to uridine within a small editing window near the protospacer adjacent motif (PAM) site. The uridine is then converted to thymidine by base excision repair, creating a C to T change (or a G to A change on the opposite strand). Third-generation base editors (BE3 series), in which the base excision repair inhibitor UGI is fused to the Cas9 nickase, nick the unmodified DNA strand so that the cell is prompted to use the edited strand as a template for mismatch repair. As a result, the cell repairs the DNA using the U-containing strand (introduced by cytidine deamination) as a template and replicates the base edit. Fourth-generation base editors (BE4 series) use two copies of the base excision repair inhibitor UGI. Adenine base editors (ABEs) have been developed that efficiently convert targeted A·T base pairs to G·C (0–100% efficiency in human cells) in genomic DNA with high product purity (typically at least 99.9%) and low percentage of indels (typically 0.1% or less). See, e.g., Gaudelli et al., Nature 551:464-471 (2017).

[0009] As described in the following paragraphs and examples, our streamlined approach to genome engineering employs base editors (e.g., third and fourth generation base editors, adenine base editors) for targeted gene disruption by knockout and missense mutations and targeted gene knock-in in the presence of a DNA donor template. The advantages of these methods, systems, and compositions are manifold. In particular, this method is useful for creating a trio of gene editing in a single method: targeted gene knockout, targeted missense mutation, and targeted gene knock-in. The methods described herein are highly suitable for studying the biology and gene function of lymphohematopoietic cells, modeling diseases such as primary immunodeficiencies, as well as correcting disease-causing point mutations, and generating novel cell products (e.g., T cell products) for therapeutic applications. Without being bound to a particular theory or mechanism of action, it is believed that the use of predetermined viral integration patterns and limiting the induction of toxic double-strand breaks, the methods, systems, and compositions described herein will enable safer, more controllable engineering of cells.

[0010] Thus, a method for targeted disruption of transcription or translation of a target gene is provided herein. In particular, the method includes targeted disruption of transcription or translation of a target gene by targeted disruption of start codon, introduction of premature stop codon, and / or disruption of intron / exon splice site. In some cases, the present invention provides a method comprising combining a nucleic acid sequence encoding a base editing fusion protein with a guide RNA, thereby obtaining an unexpectedly high rate of base editing, particularly in primary cells such as CD34+ HSPCs, T cells, natural killer cells, and B cells. The method described herein can be used to knock in and / or knock out one or more genes of interest in primary cells with improved efficiency and reduced off-target indel formation rate. In a preferred embodiment, the method is used for multiplexed base editing, including gene knock-in, gene knock-out, and missense mutation.

[0011] In a first aspect, the present invention provides a method for generating genetically engineered lymphocyte hematopoietic cells.In particular, the method includes transfecting base editing components into lymphocyte hematopoietic cells, where the components include (i) a plasmid encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain and a base excision repair inhibitor domain, where the Cas9 nickase domain may be fused to the base excision repair domain; and (ii) one or more gRNAs having complementarity to the target nucleic acid sequence to be genetically modified.When the transfected cells are cultured under conditions that promote the destruction of the splice site targeted by one or more gRNAs, the target nucleic acid sequence is modified by the base editor fusion protein and one or more gRNAs compared to untransfected cells, thereby generating genetically engineered lymphocyte hematopoietic cells. As used herein, the term "lymphocyte hematopoietic cells" refers to T cells, natural killer (NK) cells, B cells, CD34+ hematopoietic stem progenitor cells (HSPCs), and other cells involved in the generation of lymphocytes and cells of the blood, bone marrow, spleen, lymph nodes, and thymus.

[0012] As used herein, a "base editor" (also known as a "nucleobase editor") is a Cas9 fusion protein that includes a Cas9 nickase domain or a dead Cas9 (dCas9) fused to a deaminase. In some embodiments, the fusion protein includes a Cas9 nickase further fused to a UGI domain. In some embodiments, a UGI domain is also provided in the system, but is not fused to the Cas9 domain. In some cases, the base editing fusion protein is a base editor 3 (BE3) or a base editor 4 (BE4), where BE3 and BE4 refer to third and fourth generation base editors, respectively. BE3 and BE4 can generate C>G or A or indel mutations. In other cases, the base editing fusion protein is an adenine base editor (ABE), e.g., an ABE that converts A·T base pairs to G·C base pairs in the DNA of bacteria and human cells. See, e.g., Gaudelli et al., Nature 551:464-471 (2017). It will be understood that other base editors, including those that introduce null mutations at the ATG "start" codon and disrupt expression of the targeted gene, are suitable for use in accordance with the methods described herein.

[0013] In certain embodiments, the method includes knocking out a gene by targeting a splice acceptor-splice donor (SA-SD) site or a premature STOP (pmSTOP) site. In such methods, a CRISPR gRNA molecule is designed to disrupt one or more splice acceptor / donor sites in a target nucleotide sequence. A CRISPR guide RNA molecule (gRNA) comprises a sequence of at least 10 contiguous nucleotides, and often 17-23 contiguous nucleotides that are complementary to a target sequence in the genome of an organism and include a target base pair. The gRNA comprises a nucleotide sequence that is partially or entirely complementary to the gRNA target site. The gRNA target site also comprises a protospacer adjacent motif (PAM) located immediately downstream from the target site. Examples of PAM sequences are known (see, e.g., Shah et al., RNA Biology 10 (5): 891-899, 2013).

[0014] Disruption at SA-SD site is particularly advantageous because it can knock out coding sequences and non-coding RNA (ncRNA) without reading through a stop codon. Exemplary SA-SD gRNAs designed for human long non-coding RNA and human protein-coding genes related to immunotherapy are shown in Table 1. As demonstrated in the following examples, disruption using gRNAs targeting SA-SD site is superior in terms of knockout efficiency to the introduction of premature STOP codons. Base editing efficiency can be determined at genome level by EditR analysis of Sanger sequencing traces or by next generation sequencing (NGS) and at protein level by flow cytometry. Splice acceptor-splice donor base editing gRNAs targeting the splice donor and splice acceptor regions exhibit base conversion efficiencies of at least 5%, and in some cases at least 80% or higher (e.g., 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 99%). In some cases, SA-SD gRNAs are significantly more efficient at converting C to T than gRNAs that introduce disruption of premature stop codons.

[0015] Guide RNAs for targeting SA-SD sites can be designed using an R-based program that identifies gRNAs targeting SA-SD sites in all ncRNA and protein-coding genes. In some cases, the user supplies a reference genome, the Ensembl transcript number of the reference sequence, the protospacer adjacent motif (PAM) site, and the distance to a subset of upstream and downstream exon-intron boundaries. The program extracts 20 base pairs + PAM length upstream and 15 base pairs downstream sequence of the exon-intron boundary, along with splice site motifs. In some embodiments, guide molecules can be 20-120 bases long, or longer. In certain embodiments, guide molecules can be 20-60 bases long, or 20-50 bases, or 30-50 bases, or 39-46 bases long.

[0016] In some cases, it is advantageous to use chemically modified gRNAs that have increased stability when transfected into mammalian cells.For example, gRNAs can be chemically modified to include 2'-O-methyl phosphorothioate modification at least one 5' nucleotide and at least one 3' nucleotide of each gRNA.In some cases, three terminal 5' nucleotides and three terminal 3' nucleotides are chemically modified to include 2'-O-methyl phosphorothioate modification.

[0017] In certain embodiments, the method uses base editing nickase activity to mediate the insertion of donor sequence by (i) homology directed repair (HDR) from template or (ii) integration of viral vector. In such method, Cas9 nickase domain promotes targeted gene knock-in in the presence of DNA donor template using gRNA targeting the insertion site. In some cases, donor sequence is integrated into endogenous safe harbor locus, such as CC motif chemokine receptor 5 (CCR5), adeno-associated virus integration site 1 (AAVS1), ROSAβgeo26 (Rosa26), albumin (ALB), T cell receptor alpha constant (TRAC), and / or hypoxanthine phosphoribosyltransferase 1 (HPRT). For example, guide RNA can be designed to target AAVS1 locus. In such cases, guide RNA has complementarity to DNA target site. Referring to Figure 16 illustrating targeted knock-in (KI), BE nickase activity is highly effective in stimulating HDR when combined with delivery of rAAV DNA donors. In some cases, the efficiency of BE using two guide RNAs complementary to the target insertion site was greatly improved compared to the use of a single gRNA, or the use of Cas9 nuclease or Cas9 nickase.

[0018] The embodiment includes a method for simultaneous multiple gene editing. For example, a method for multiple manipulation in human cells using base editing is provided in the present invention. In some cases, the method includes a base editor fusion protein (e.g., BE3, BE4, ABE) and a gRNA that one or more genes are disrupted (for knockout) and one or more genes are knocked in using a donor DNA template for insertion. With reference to FIG. 12, base editor 3 (BE3) and base editor 4 (BE4) were used with three SA-SD gRNAs to successfully knock out TRAC, B2M, and PDCD1 in human T cells. BE3 and BE4 were also used with these three SA-SD gRNAs, including two AAVS1-targeting gRNAs, to knock in donor DNA templates. These data demonstrate that the base editing method provided in the present invention is useful for multiplex disruption of multiple genes related to immunotherapy in cells (e.g., T cells), with or without target DNA templates or knock-in of missense mutations.

[0019] Also provided herein are methods for genome engineering (e.g., for altering or manipulating the expression of one or more genes or one or more gene products) in prokaryotic or eukaryotic cells in vitro, in vivo, or ex vivo. In particular, the methods provided herein are useful for targeted base editing disruption in mammalian cells, including human T cells, natural killer (NK) cells, CD34+ hematopoietic stem progenitor cells (HSPCs) (e.g., umbilical cord blood HSPCs), and fibroblasts (e.g., MPS1 fibroblasts, Fanconi anemia fibroblasts). Importantly, as shown in Figures 21A-21D, fibroblasts from Fanconi anemia (and thus DNA repair deficient) patients can still undergo base editing, for example, using BE3, BE4, or ABE. Thus, also provided herein are genetically engineered lymphohematopoietic cells, such as T cells, modified according to the methods described herein.

[0020] In some cases, the method is configured to generate engineered T cells suitable as "universally acceptable" cells for therapeutic applications. As used herein, the term "universally acceptable" refers to the general acceptability of cell products in immunological terms, where cross-matching of patients and cells is not required, and immunosuppression is not required. To obtain such cells, the method may include, for example, transfecting into human T cells: (i) a plasmid, mRNA, or protein encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain, where the nickase domain comprises a base excision repair inhibitor domain; (ii) one or more splice acceptor-splice donor (SA-SD) gRNAs for disrupting the expression of TRAC, B2M, and PDCD1, respectively; (iii) a donor DNA template encoding a T cell receptor (TCR) and a chimeric antigen receptor (CAR); and (iv) two gRNAs complementary to the target insertion site. The method further comprises culturing the transfected T cells under conditions that promote the disruption of the splice site targeted by the SA-SD gRNA, thereby reducing the expression of TRAC, B2M, and PDCD1 gene products compared to non-transfected T cells, and culturing the transfected T cells under conditions that promote targeted knock-in of the donor DNA template at the target insertion site. In this example, the resulting genetically modified T cells express CAR / TCR and lack expression of TRAC, PDCD1, and B2M. In some cases, the method further comprises introducing a gRNA designed to disrupt the expression of CTLA-4. Exemplary gRNA sequences for editing target bases in genes TRAC, PDCD1, and B2M are shown in Table 1.

[0021] [Table 1]

[0022] The base editing fusion proteins comprise a deaminase domain fused to a Cas9 nickase domain or an inactivated Cas9 nuclease domain (referred to as dead Cas9 or dCas9), where in some cases the deaminase domain is a deaminase of the apolipoprotein B mRNA editing complex (APOBEC) family. The APOBEC1 cytidine deaminase domain allows for targeted gene disruption with a single base substitution of thymidine for cytidine.

[0023] In another aspect, the present invention provides a method for targeting disease for base editing correction.Target sequence can be any disease-related polynucleotide or gene that is established in the art.Examples of useful applications of mutation or "correction" of endogenous gene sequence include targeting of disease-related gene mutation, modification in sequence that codes for splice sites, modification in regulatory sequence, modification in sequence that causes gain-of-function mutation, and / or modification in sequence that causes loss-of-function mutation, and modification of sequence that codes for structural features of protein.

[0024] In another aspect, the present invention provides a method for using base editing to obtain a cleavage-resistant Fc receptor. For example, in some cases, a mutagenic domain of a base editing enzyme is used to induce a mutation that results in a higher affinity FcγRIIIa (Fc fragment of IgG receptor IIIa) gene product (also known as CD16a). Thus, the method may include introducing components for base editing (e.g., BE3, BE4, gRNA, donor template) into natural killer cells to introduce a mutation in CD16a that enhances antibody-dependent cell-mediated cytotoxicity (ADCC) of NK cells. With reference to Figures 18A-18D, base editing according to the method provided in the present invention alters CD16a into a cleavage-resistant form, thereby enhancing CD3 - CD56 + NK cells were successfully modified. A C to T editing efficiency of 40% was obtained using BE3-VQR and CD16a gRNA.

[0025] In some cases, it may be advantageous to genetically modify cells to express a chimeric antigen receptor (CAR) and / or a T cell receptor (TCR) using the methods described herein. A "chimeric antigen receptor (CAR)" may be referred to as a "chimeric receptor," "T-body," or "chimeric immune receptor (CIR)." As used herein, the term "chimeric antigen receptor (CAR)" refers to an artificially constructed hybrid protein or polypeptide that includes an extracellular antigen-binding domain of an antibody (e.g., a single chain variable domain (scFv)) operably linked to a transmembrane domain and at least one intracellular domain. Generally, the antigen-binding domain of a CAR has specificity for a particular antigen expressed on the surface of a target cell of interest. For example, T cells can be engineered to express a CAR specific for CD19 on B cell lymphoma. For allogeneic antitumor cell therapy that is not limited by donor matching, cells can be engineered to knock out the genes responsible for donor matching (TCR markers and HLA markers) rather than just knocking in the nucleic acid encoding the CAR.

[0026] As used herein, the terms "genetically modified" and "genetically engineered" are used interchangeably and refer to prokaryotic or eukaryotic cells that contain an exogenous polynucleotide, regardless of the method used to insert it. In some cases, the effector cell is modified to contain a non-naturally occurring nucleic acid molecule that has been created or modified by the hand of man (e.g., using recombinant DNA technology) or is derived from such a molecule (e.g., by transcription, translation, etc.). An effector cell that contains an exogenous, recombinant, synthetic, and / or otherwise modified polynucleotide is considered to be an engineered cell.

[0027] In some embodiments, the components including the base editor and guide molecule can be delivered to cells in vitro, ex vivo, or in vivo. In some cases, a viral or plasmid vector system is used to deliver the base editing components described herein. Preferably, the vector is a viral vector, such as a lentivirus or baculovirus, or preferably an adenovirus / adeno-associated virus (AAV) vector, although other means of delivery are known and contemplated (e.g., yeast systems, microvesicles, attaching the vector to a gene gun / gold nanoparticles). In certain embodiments, the nucleic acid encoding the gRNA and the base editor fusion protein is packaged into one or more viral delivery vectors for delivery to cells. Suitable viral delivery vectors include, but are not limited to, adenovirus / adeno-associated virus (AAV) vectors, lentivirus vectors. In some cases, non-viral transfer methods known in the art can be used to introduce the nucleic acid or protein into mammalian cells. The nucleic acid and protein can be delivered with a pharma- ceutically acceptable vehicle or encapsulated, for example, in a liposome. Other delivery means are known and contemplated (e.g., yeast systems, microvesicles, gene guns / attaching vectors to gold nanoparticles). In some cases, cells are electroporated to incorporate the gRNA and base editor (e.g., BE3, BE4, ABE). In some cases, the DNA donor template is delivered as an adeno-associated virus type 6 (AAV6) vector by adding viral supernatant to the culture medium after introduction of the gRNA, base editor, and vector by electroporation.

[0028] The rate of insertion or deletion (indel) formation can be determined by suitable methods.For example, Sanger sequencing or next generation sequencing (NGS) can be used to detect the rate of indel formation.Preferably, contacting causes less than 20% off-target indel formation during base editing.Contacting causes at least 2:1 intended product to unintended product during base editing.

[0029] Cells useful for the methods provided herein can be freshly isolated primary cells or obtained from frozen aliquots of primary cell cultures. In some cases, cells are electroporated for incorporation of gRNA and base-editing fusion proteins. As described in the examples below, electroporation conditions for some assays (e.g., for T cells) can include 1400 volts, 10 millisecond pulse width, 3 pulses. After electroporation, electroporated T cells can be recovered in cell culture medium and then cultured in T cell expansion medium. In some cases, electroporated cells can be recovered in cell culture medium for about 5 to about 30 minutes (e.g., about 5, 10, 15, 20, 25, 30 minutes). Preferably, the recovery cell culture medium does not contain antibiotics or other selection agents. In some cases, the T cell expansion medium is CTS OpTmizer T cell expansion complete medium.

[0030] The terms "nucleic acid" and "nucleic acid molecule" as used herein refer to a compound that includes a nucleic acid base and an acidic moiety, e.g., a nucleoside, a nucleotide, or a polymer of nucleotides. Typically, a polymeric nucleic acid, e.g., a nucleic acid molecule that includes three or more nucleotides, is a linear molecule in which adjacent nucleotides are linked to each other via phosphodiester bonds. In some embodiments, "nucleic acid" refers to an individual nucleic acid residue (e.g., a nucleotide and / or a nucleoside). In some embodiments, "nucleic acid" refers to an oligonucleotide chain that includes three or more individual nucleotide residues. As used herein, the terms "oligonucleotide" and "polynucleotide" may be used interchangeably to refer to a polymer of nucleotides (e.g., a string of at least three nucleotides). In some embodiments, "nucleic acid" encompasses RNA as well as single-stranded and / or double-stranded DNA. A nucleic acid may naturally occur in the context of, for example, a genome, a transcript, an mRNA, a tRNA, an rRNA, an siRNA, an snRNA, a plasmid, a cosmid, a chromosome, a chromatid, or other naturally occurring nucleic acid molecule. On the other hand, a nucleic acid molecule may be a non-naturally occurring molecule, such as a recombinant DNA or RNA, an artificial chromosome, an engineered genome, or a fragment thereof, or a synthetic DNA, RNA, DNA / RNA hybrid, or may contain non-naturally occurring nucleotides or nucleosides. Furthermore, the terms "nucleic acid", "DNA", "RNA", and / or similar terms include nucleic acid analogs, i.e., analogs having other than a phosphodiester backbone. Nucleic acids may be purified from natural sources, purified using recombinant expression systems, and optionally purified and chemically synthesized. Where applicable, for example, in the case of chemically synthesized molecules, nucleic acids may contain nucleoside analogs, such as analogs having chemically modified bases or sugars, and backbone modifications. Nucleic acid sequences are shown in the 5' to 3' direction unless otherwise indicated.In some embodiments, nucleic acids are selected from the group consisting of natural nucleosides (e.g., adenosine, thymidine, guanosine, cytidine, uridine, deoxyadenosine, deoxythymidine, deoxyguanosine, and deoxycytidine); nucleoside analogs (e.g., 2-aminoadenosine, 2-thiothymidine, inosine, pyrrolo-pyrimidine, 3-methyladenosine, 5-methylcytidine, 2-aminoadenosine, C5-bromouridine, C5-fluorour ... the base is or contains: lysine, C5-iodouridine, C5-propynyl-uridine, C5-propynyl-cytidine, C5-methylcytidine, 2-aminoadenosine, 7-deazaadenosine, 7-deazaguanosine, 8-oxoadenosine, 8-oxoguanosine, O(6)-methylguanine, and 2-thiocytidine; chemically modified bases; biologically modified bases (e.g., methylated bases); intercalated bases; modified sugars (e.g., 2'-fluororibose, ribose, 2'-deoxyribose, arabinose, and hexose); and / or modified phosphate groups (e.g., phosphorothioates and 5'-N-phosphoramidite linkages).

[0031] The terms "protein", "peptide" and "polypeptide" are used interchangeably herein and refer to a polymer of amino acid residues linked together by peptide (amide) bonds. The term refers to a protein, peptide, or polypeptide of any size, structure, or function. Typically, a protein, peptide, or polypeptide will be at least three amino acids in length. A protein, peptide, or polypeptide may refer to an individual protein or a collection of proteins. For example, one or more of the amino acids in a protein, peptide, or polypeptide may be modified by the addition of a chemical entity, such as, for example, a carbohydrate group, a hydroxyl group, a phosphate group, a farnesyl group, an isofarnesyl group, a fatty acid group, a linker for conjugation, functionalization, or other modification. A protein, peptide, or polypeptide may be a single molecule or a complex of multiple molecules. A protein, peptide, or polypeptide may be simply a fragment of a naturally occurring protein or peptide. A protein, peptide, or polypeptide may be naturally occurring, recombinant, or synthetic, or any combination thereof. A protein may include various domains, such as a nucleic acid binding domain and a nucleic acid cleavage domain. In some embodiments, a protein comprises a proteinaceous portion, e.g., an amino acid sequence that constitutes a nucleic acid binding domain, and an organic compound, e.g., a compound that can act as a nucleic acid cleaving agent.

[0032] In interpreting this disclosure, all terms should be interpreted in the broadest possible sense consistent with the context. The specific adaptations of the invention described in this disclosure are matters of routine optimization for those skilled in the art and can be made without departing from the spirit of the invention or the scope of the appended claims. As used herein, the terms "synthetic" and "engineered" are used interchangeably and refer to aspects that have been manipulated by the hand of man. So that the compositions and methods provided herein may be more readily understood, certain terms are defined.

[0033] As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Any reference to "or" herein is intended to include "and / or" unless specifically stated otherwise.

[0034] The terms "comprising" and "comprises" and "comprised of," as used herein, are synonymous with "including," "includes," or "containing," "contains," and are inclusive or open-ended and do not exclude additional, unrecited members, elements, or method steps. The phraseology and terminology used herein are for purposes of description and should not be considered limiting. "Including," "comprising," "having," "containing," "involving," and variations thereof are meant to encompass the items listed thereafter as well as additional items. Embodiments referred to as "comprising" certain elements are also contemplated as "consisting essentially of" and "consisting of" those elements. The use of ordinal terms in the claims, e.g., "first," "second," "third," etc., to modify a particular element does not, by itself, imply any priority, precedence, or order of one claim element over another claim element, or the temporal order in which acts of a method are performed. The ordinal terms are used merely as labels to distinguish an element of one claim having a particular name from another element having the same name (absent the use of ordinal terms) to identify elements of the claims.

[0035] As used herein, "modifying" ("modify") one or more target nucleic acid sequences refers to changing all or a portion of the target nucleic acid sequence(s), including truncating, introducing (inserting), substituting, and / or deleting (removing) all or a portion of the target nucleic acid sequence. All or a portion of the target nucleic acid sequence may be fully or partially modified using the methods provided herein. For example, modifying a target nucleic acid sequence includes replacing all or a portion of the target nucleic acid sequence with one or more nucleotides (e.g., an exogenous nucleic acid sequence) or removing or deleting all or a portion of the target nucleic acid sequence (e.g., one or more nucleotides). Modifying one or more target nucleic acid sequences also includes introducing or inserting one or more nucleotides (e.g., an exogenous sequence) into (within) one or more target nucleic acid sequences.

[0036] The terms "about" and "approximately" generally refer to an acceptable degree of error in the measured quantity, taking into account the nature or precision of the measurement. Typically, an exemplary degree of error is within 10%, preferably within 5%, of a given value or range of values. Alternatively, and particularly in biological systems, the terms "about" and "approximately" can refer to values ​​that are within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a given value. Numerical quantities given herein are approximations, unless otherwise stated, meaning that the term "about" or "approximately" can be inferred if not explicitly stated.

[0037] Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. As used in this specification and claims, the singular forms "a", "an" and "the" include singular and plural references unless the context clearly indicates otherwise. Thus, for example, a reference to "an agent" includes a single agent and a plurality of such agents. Any reference to "or" herein is intended to encompass "and / or" unless otherwise indicated. Various exemplary embodiments of the compositions and methods according to this invention are now described in the following non-limiting examples. The examples are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way. Indeed, various modifications in addition to those shown and described herein will become apparent to those skilled in the art from the foregoing description and examples below and are within the scope of the appended claims. EXAMPLES

[0038] This section demonstrates the successful use of third and fourth generation base editors to knock out four different genes in primary human lympho-hematopoietic cells with a high efficiency of 80%. Disruption of splice sites proved to be more effective than premature stop codons for gene knockout. As the Cas9 nickase function of later generation base editors, targeted gene knock-in was also achieved with up to 70% efficiency after administration of AAV6 vectors to deliver donor templates. Taken together, the assays and results described herein demonstrate an improved multiplex gene editing platform that enhances both the safety and efficacy of engineered T cell-based immunotherapy.

[0039] Example 1 Splice site base editing Previously, base editing has been used to induce premature stop (pmSTOP) codons for gene knockout in mouse and mammalian cells. 15~18 However, we determined that disruption of splice sites may have some advantages over induction of pmSTOP codons (Figure 1). For example, readthrough of stop codons has been shown to occur at frequencies up to 31% in some genes and may be enhanced under conditions of cellular stress. 19、20 Splice site editing alleviates this concern because it alters gene processing at the RNA level,21 which is unlikely to be avoided at the translational level.22 Furthermore, current base editors do not produce strict C to T edits, and even recent base editors produce only up to 25% non-targeted edits (C to G / A). 22 In the context of pmSTOP, non-targeted editing is expected to eliminate premature stop codon formation, thereby reducing the efficiency of protein knockout and instead causing undesired amino acid changes.

[0040] To evaluate the performance of both pmSTOP introduction and splice site disruption, we designed a panel of single guide RNAs (sgRNAs) to convert amino acid codons to pmSTOP or disrupt splice donor (SD) and acceptor (SA) sequences within PDCD1, TRAC, and B2M (Figures 2A, 2E, 2I; Table 1). Individual sgRNAs were transfected by electroporation into T cells with first generation BE3 13 mRNA or BE4 22 RNA oligonucleotides chemically modified with mRNA 23 The percentage of targeted C to T edits was analyzed using analytical software developed by our group to streamline and reduce the cost of analyzing base edits at the gene level. 24 The sequences were evaluated by Sanger sequencing and EditR (available on the World Wide Web at baseeditr.com).

[0041] First, we targeted the checkpoint gene PDCD1 (also known as PD-1) by designing eight sgRNAs, three of which were predicted to introduce a pmSTOP codon, two targeted the disruption of an SD site (GT:CA), and three targeted the disruption of an SA site (AG:TC) (Figure 3A). We found that co-delivery of sgRNAs with BE3 or BE4 mRNA mediated measurable editing of the targeted C in all target loci, with some candidate sgRNAs exhibiting significantly higher editing percentages than others (Figure 3B; Figures 7A-7C). Specifically, we found that targeting the SD site in exon 1 of PDCD1 resulted in the highest percentage of targeted C-to-T editing in both BE3 (51.3±7.0%, M±SD) and BE4 (63.7±2.1%) mRNAs (Figure 3B). The next two most efficient sgRNAs targeted a SA site in exon 3 (32.6±5.5% for BE3; 36.0±4.0% for BE4) and a candidate pmSTOP site in exon 2 (37.1±1.2% for BE3; 48.5±3.7% for BE4) (FIG. 3B). To determine whether gene editing resulted in protein loss, we assessed PD-1 protein expression by flow cytometry. Consistent with our genetic analysis, targeting SD of exon 1 of PDCD1 resulted in the highest percentage of protein loss (69.5 ± 7.0% for BE3; 78.6 ± 4.1% for BE4), followed by SA of exon 3 (40.6 ± 7.8% for BE3; 44.7 ± 3.8% for BE4) and pmSTOP of exon 2 (37.9 ± 3.4% for BE3; 51.5 ± 9.0% for BE4) (Figure 3C).

[0042] Using the information of our PDCD1 results, we designed a panel focused on sgRNAs targeting TRAC. Here, we found that C to T conversion was highest at the SD site in exon 1 (47.6±4.6% for BE3; 60.0±11.3% for BE4) and the SA site in exon 3 (40.3±9.7% for BE3; 62.3±11.0% for BE4), with BE4 exhibiting a higher editing percentage than BE3 at each target (Figure 3F). Efficient editing was also observed at two pmSTOP candidate sites in exon 3, albeit with lower efficiency than that of sgRNAs that disrupt either splice site. Both the SD site in exon 1 and the SA site in exon 3 were edited with similar frequency, nevertheless, disruption of the SA site in exon 3 resulted in the highest percentage of TCR disruption, as measured by loss of cell surface CD3 expression (69 ± 15.3% for BE3; 83.7 ± 5.8% for BE4) ( Fig. 3G ).

[0043] Next, we used a similar strategy to target B2M (Figure 3I). We showed that BE4 mRNA delivered with sgRNA targeting the SD site in exon 1 showed the most efficient C to T conversion of the targeted base (58.3 ± 2.5% for BE3; 70.3 ± 3.2% for BE4) (Figure 3J), resulting in effective knockout of B2M protein (79.1 ± 1.3% for BE3; 80.0 ± 3.2% for BE4) (Figure 3K). We also identified a candidate pmSTOP site in exon 2 that resulted in relatively efficient C to T editing (43.3 ± 5.7% for BE3; 55.7 ± 5.0% for BE4) and protein knockout (56.2 ± 5.1% for BE3; 61.5 ± 1.8% for BE4) (Figure 3J, 3K). Notably, targeting the SA site in noncoding exon 3 resulted in efficient C-to-T editing but did not result in a detectable reduction in protein expression (Figures 3J, 3K).

[0044] Non-targeted editing (i.e., C to A or G) was reported for BE3 13 , and is reduced by BE4, which contains a second uracil glycosylase inhibitor (UGI) fused sequentially to the C-terminus. 22 We assessed the non-target editing rate for all Cs within the editing window (mainly bases 4-8 of the protospacer) of our most efficient sgRNAs, including BE3 and BE4. As expected, BE4 showed reduced non-target editing compared to BE3 at all loci (-14% ± 6.6%, P < 2.2e-16, paired one-way t-test) (Figures 3D, 3H, 3L; Figures 7A-7C). Despite only possessing nickase function, low levels of indel formation were observed with both BE3 and BE4. 13、22 Thus, we used next-generation sequencing (NGS) to measure the frequency of indels at all target sites after editing (Figures 8A-8C). Indels were detectable in both BE3 and BE4 at levels that varied based on the target site. Consistent with previous publications, BE4 exhibited an overall reduced frequency of indels (-4.8% ± 6.1%, P < 4.6e-16, paired one-way t-test) (Figures 8A-8C). 22 .

[0045] Towards our goal of evaluating multiplex editing strategies that can be utilized to generate allogeneic "off-the-shelf" T cells with enhanced function, we co-delivered our top sgRNA for each gene with first generation BE3 or BE4 mRNA. Surprisingly, the knockout efficiency at each target was substantially reduced for both BE3 and BE4 when delivered in a multiplex setting (Figure 9). To determine whether the reduced editing efficiency was due to lower protein levels, we delivered equal doses of BE3 mRNA, BE4 mRNA, and nuclease-activated Streptococcus pyogenes Cas9 (SpCas9) mRNA into T cells and measured protein expression 24 hours after electroporation. Strikingly, SpCas9 protein expression was readily detectable at these time points, whereas BE3 and BE4 proteins were undetectable (Figure 10). To address this issue, we first delivered BE3 and BE4 mRNAs at a dose (3 μg) that was twice as high as the dose (1.5 μg) used in our initial multiplex experiments. This strategy improved the editing efficiency at each locus, but the efficiency was still lower than that observed in our single gene targeting experiments (Figure 4A).

[0046] During the course of these experiments, independent reports emerged identifying problems with the use of first-generation BE3 and BE4 expression vectors that significantly reduced both transcription and translation efficiency in human cells. 17、25To circumvent this issue, we delivered purified BE4 protein as a ribonucleoprotein (RNP) complex along with our most efficient sgRNA for each target. By optimizing our electroporation protocol for RNP delivery, we found that BE4 RNP mediated improved editing efficiency over a two-fold dose of first-generation BE4 mRNA (Figure 4C). Next, we codon-optimized the sequence of BE4 (coBE4) and observed increased protein expression when delivered as mRNA (Figure 10). When coBE4 mRNA was delivered along with all three of our optimal sgRNAs at both our standard dose (1.5 μg) and a higher dose (4 μg), we achieved substantially higher percentages of multitarget C-to-T editing at all three loci across multiple independent T cell donors, exceeding 90% in some instances (Figure 4A). Non-targeted editing observed with first generation BE3 mRNA and BE4 mRNA was slightly reduced when using BE4 RNP and further reduced with both doses of coBE4 mRNA (Figure 4A). Next, we evaluated the rate of indel formation at each target site after multiplex base editing according to previous studies and found a lower rate of indel formation at each site with all forms of BE4 compared to BE3 and SpCas9 nuclease (Figure 4B, 4C). Both low dose and high dose coBE4 mRNA exhibited the lowest overall frequency of indel formation at all sites examined (Figure 4B).

[0047] Multiple protein knockouts were analyzed for each target gene by flow cytometry, and the frequency of protein loss correlated well with the frequency of gene editing (Figure 4D, r = 0.90, df = 28, P = 1.3e-11). BE4 RNP demonstrated more efficient protein knockout than the first generation BE3 mRNA and BE4, but coBE4 mRNA was nevertheless the most efficient, with over 90% protein loss for each gene at both low and high doses of mRNA (Figure 4D; Figure 11). An important consideration for multiple editing is the percentage of cells obtained with each possible combination of gene knockouts. To gain a deeper understanding of this phenomenon in our experiments, we simultaneously assessed protein expression of all target genes by flow cytometry and used SPICE analysis to determine the percentage of individual cells with no knockout, single gene knockout, double gene knockout, or triple gene knockout; and combinations of protein loss within each of these fractions (Figure 4E). First generation BE4 mRNA produced end cell populations with various combinations of knockout phenotypes, but the frequency of triple knockout cells was low (21.9 ± 1.1%). The percentage of triple knockout cells was substantially higher using BE4 RNP (68.6 ± 0.37%) and further increased with 1.5 μg (86.6 ± 3.75%) and 4 μg (89.57 ± 4.2%) coBE4 mRNA (Figure 4F).

[0048] Off-target (OT) DSB induction is a key challenge facing nuclease platforms 26To determine the specificity of our optimal sgRNA, we delivered each individually with SpCas9 nuclease or BE4 mRNA and evaluated editing at the top 10 OT sites predicted by NGS (Table 2). No editing was observed at any of the predicted B2M or TRAC OT sites in either SpCas9 nuclease or BE4 treatment conditions (Figure 12). At the predicted PDCD1 OT site, we observed a single OT edit with an indel frequency of 13.0% using SpCas9 mRNA (Figure 12). Strikingly, C to T editing at this site was only 0.9% with respect to BE4 mRNA, and indel formation was close to the lowest detection limit of our assay (0.2%) (Figure 12).

[0049] Table 2. Candidate off-target sites predicted by computation

[0050] [Table 2-1]

[0051] [Table 2-2]

[0052] [Table 2-3]

[0053] [Table 2-4]

[0054] [Table 2-5]

[0055] [Table 2-6]

[0056] [Table 2-7]

[0057] [Table 2-8]

[0058] [Table 2-9]

[0059] [Table 2-10]

[0060] [Table 2-11]

[0061] Nuclease-mediated multiple editing has also been reported to result in unwanted rearrangements in human T cells. 3Because base editing substantially reduces the frequency of DSB formation, we reasoned that translocations should be similarly reduced using our base editing approach. To test our hypothesis, we used droplet digital PCR (ddPCR) to quantitate the frequency of 12 possible translocation outcomes predicted to occur between our three target loci and a single identified OT site (Figure 5A). After co-delivery of our two optimal sgRNAs with either SpCas9 nuclease or mRNA, we were able to detect all 12 predicted translocation outcomes at various frequencies (Figure 5B, Figures 13-14). In all cases, SpCas9 mRNA produced the highest percentage of translocations, and translocations between TRAC and B2M were the most frequent (2.04 ± 0.09%) (Figure 5B). In contrast, the results of translocations between our three target loci were virtually undetectable in cell populations that received BE4 RNP or any dose of coBE4 mRNA together with our optimal sgRNA (Figure 5B, Figure 13). In a single replicate from one donor, the PDCD1:B2M assay yielded two positive droplets for the low dose of coBE4 mRNA (calculated frequency = 0.003 ± 0.006%). These may be artifacts, since no positive droplets were detected for BE4 RNP or the high dose of coBE4 mRNA (Figure 13). Translocations between our three target loci and the single identified OT site were also detected in T cells treated with SpCas9, albeit at a low frequency, but samples treated with BE4 did not show any signal above the detection limit (0.01%) of our assay (Figure 5B, Figure 14).

[0062] Next, we sought to determine whether the multiple knockout T cells generated using our base editing strategy retained cytokine functionality and were capable of mediating target cell killing when equipped with a CAR. We performed a phenotypic evaluation of both electroporation-pulsed control and coBE4 knockout T cells, with or without a CD19-specific CAR, by analyzing differentiation markers. 27 Both untransduced and CAR-transduced T cells exhibited similar differentiation phenotypes, and the fractions of effector and memory populations were similar between control and coBE4 knockout T cells (Figure 6A). CAR transduction and cell expansion were also comparable between the pulse and coBE4 mRNA groups (Figures 15A-15B). After activation, high frequencies of both untransduced and CAR-transduced coBE4 knockout T cells showed robust production of cytokines IL-2, TNFα, and IFNγ (Figure 6B). Cytokine polyfunctionality was similarly retained after the multiple editing process (Figure 6C). Taken together, these data demonstrate that multiple coBE4 editing combined with CAR transduction did not negatively affect T cell phenotype or function. Finally, to determine whether coBE4 knockout T cells equipped with a CD19 CAR retain the ability to kill target cells, we used coBE4 knockout T cells equipped with a CD19 CAR to inhibit killing of non-targeted CD19 T cells, which normally acts to inhibit killing by T cells expressing cell surface PD-1. neg / PD-L1 neg K562; target CD19 pos / PD-L1 neg Raji; and target CD19 engineered to overexpress PD-1 pos / PD-L1 pos We performed an in vitro coculture assay with Raji. Both control and coBE4 knockout T cells expressed CD19 pos mediated specific killing of CD19 neg coBE4 knockout T cells did not mediate specific killing of target cells (Fig. 6D). However, only coBE4 knockout T cells were able to express CD19 pos / PD-L1 posSignificant killing of target cells can be achieved, and CD19 pos / PD-L1 neg The killing efficiency was comparable to that of the target cells (Figure 6D).

[0063] As the requirements for successful cell-based immunotherapy and gene therapy become better understood and as the enthusiasm for generating universal, allogeneic cells increases, advanced multiplex gene editing will likely become more commonplace. However, it has been well documented that DSBs are toxic lesions that can drive genomic instability and cell death. 11、12 While this is a less important issue when engineering cells for research, it can result in transformation or loss of function when gene editing cells for therapeutic applications. Our concerns around DBS are further intensified in the context of multiplex gene editing, where multiple, simultaneous DSBs can add to toxicity. This is highlighted by the parabolic relationship between the number of distinct DSB sites and possible translocation outcomes, such that an editing strategy targeting 10 loci can result in 90 possible translocations, even leaving out other possible genomic alterations such as inversions and large deletions. To overcome these issues, we have implemented the use of base editor technology for multiplex T cell engineering, where the disruption of splice sites by base editing provides an effective and safe approach compared to the use of DSB-induced targeted nucleases.

[0064] Interestingly, we found that rates of both non-targeted editing and indel formation were higher when using BE4 RNP compared to coBE4 expressed from transfected mRNA. This observation is attributable to the extended BE4 residence time achieved when expressed at high levels from stable mRNA as opposed to delivering BE4 protein directly. Even free UGI has been shown to reduce both indel frequency and non-targeted editing in the context of BE3. 28, the extended residence time achieved by the mRNA may allow for the additional ability of the BE4 UGI domain to mitigate DSB formation and non-targeted editing. 28 .

[0065] In our study, we took advantage of lentiviral delivery of CD19-specific CAR, the current industry standard in CAR-T therapy. However, this approach has many drawbacks, including the risk of insertional mutagenesis, variable CAR expression, and gene silencing. 29~31 To overcome these issues, several groups have demonstrated highly efficient, site-specific integration using Cas9 nuclease together with a DNA donor template delivered by rAAV for homologous recombination (HR). This allows the integration of rAAV and Cas9 orthologues, such as Staphylococcus aureus Cas9 (SaCas9) or Francisella novella Cas9 (FnCas9). 32、33 This raises the possibility that BE4 could be employed to simultaneously introduce therapeutic transgenes to safely and efficiently knock out multiple genes in a site-specific manner using Cas9 orthologs. Application of Cas9 orthologs would allow for the simultaneous use of separate sgRNAs specific for BE4 and Cas9 nucleases without concern for cross-utilization. Alternatively, it has been demonstrated that DNA nicks can be used to stimulate HR using naked DNA or rAAV as the DNA donor molecule, albeit with low efficiency. 34 This intriguingly highlights the ability of BE4 to mediate gene knockout through its deaminase activity while simultaneously mediating HR through its nickase function. In this scenario, the sgRNA binding site may require the absence of cytosines within the base editing window to prevent loss of Cas9 binding due to sequence changes due to C to T conversions.

[0066] One notable difference between the use of base editors and targeted nucleases is the number of possible outcomes from editing events. Nuclease-mediated DSBs are repaired via the highly variable non-homologous end joining (NHEJ) pathway, resulting in a spectrum of indels, some of which will not introduce frameshift mutations and therefore have unknown importance to gene expression and function. Instead, our base editing approach has only a limited number of outcomes, all of which, even when considering non-targeted editing, result in loss of function of native splice donors or acceptors. However, it is important to consider that native splice sites do not always result in non-functional products, as alternative or cryptic splicing can maintain the biological function of genes.

[0067] Translocation analysis using small ddPCR amplicons (>200 bp) spanning the sgRNA target site demonstrated that base editing with optimal reagents virtually eliminated detectable translocations, whereas Cas9 nuclease generated numerous translocations, some at frequencies as high as 1.5%. Notably, larger deletions were also identified at the sites of translocations by Sanger sequencing of junction PCR amplicons (~500 bp) subcloned from cells treated with SpCas9 (Figures 17A-17B). These data support the previously reported deletions that are not detected by our current ddPCR assay. 9、10 These results suggest the presence of more complex genomic rearrangements similar to those described in Example 1. Given the variability in the efficiency of nucleic acid delivery between cells by electroporation, it is possible that cells receiving high levels of SpCas9 / sgRNA harbor translocations at a higher frequency.

[0068] Although we demonstrate that BE4 substantially reduces DSB induction and does not produce detectable translocations compared to SpCas9 nuclease, the possibility of undesired events still exists. For example, rAPOBEC1 of BE4 is capable of non-specifically editing cytosines in single-stranded DNA during DNA replication. 35Alternatively, UGI of BE4 could potentially inhibit uracil DNA glycosylase in a non-specific manner, thereby preventing base excision repair of naturally occurring and frequently occurring deaminations of cytosines in normal mammalian cells. 36 Although such unintended events are difficult to definitively document, further studies examining these possible events should be undertaken prior to clinical translation of base-edited cells. Despite uncertainties in these areas, the base editor platform represents a novel approach for highly efficient multiplexed engineering of therapeutic primary cells with an improved safety profile compared to current nuclease technologies.

[0069] method Cloning and virus production: The DNA sequence for the CD19 chimeric antigen receptor linked to RQR8 by T2A was synthesized as a gBlock Gene Fragment (Integrated DNA Technologies [IDT]). The fragment was then inserted into pRRL using the Gibson Assembled 37 (Available on the World Wide Web at www.addgene.org / 36247). The Gibson reaction was transformed into DH10β E. coli and plated on LB agar containing ampicillin. Plasmid DNA was purified from colonies using the GeneJET Plasmid Miniprep Kit (ThermoFisher). After confirmation by Sanger sequencing, clones were sent to the Viral Vector & Cloning Core (VVCC) at the University of Minnesota for viral particle generation and titration.

[0070] Guide RNA design: Guide RNAs (sgRNAs) were designed using SpliceR (available on the World Wide Web at z.umn.edu / splicer) [Kluesner and Lahr et al., in preparation], an sgRNA design program for disrupting splice sites by base editing. SpliceR is written in the R statistical programming language (v.3.4.3). Briefly, SpliceR takes the target Ensembl transcript number, base editor PAM variant, and species as input. Using exon and intron sequences from Ensembl, the program extracts regions around each splice site based on a user-specific window. The N 20 A pattern of -NGG is matched to the antisense strand of the extracted sequence. The matched pattern is then compared to the previously published 13 The target motif is scored based on the location of the predicted editing window based on the sgRNAs. The sgRNAs are then scored based on their location within the transcript, where sgRNAs earlier in the transcript receive higher scores. pmSTOP-inducing gRNAs were designed using the Benchling base editing gRNA design tool (available on the World Wide Web at benchling.com / pub / liu-base-editor).

[0071] Isolation of CD3+ T cells: Peripheral blood mononuclear cells (PBMCs) were isolated from the Trima Accel leukoreduction system (LRS) chamber using ammonium chloride-based red blood cell lysis. CD3+ T cells were isolated from the PBMC population by immunomagnetic negative selection using the EasySep human T cell isolation kit (STEMCELL Technologies). T cells were cultured at 10–20 × 10 per mL of Cryostor CS10 (STEMCELL Technologies). 6 The cells were frozen at 100 cells per well and thawed during culture as needed.

[0072] Culture T cells: 1 × 10 per mL in OpTmizer CTS T Cell Expansion SFM containing 2.5% CTS Immune Cell SR (ThermoFisher), L-glutamine, penicillin / streptomycin, N-acetyl-L-cysteine ​​(10 mM), IL-2 (300 IU), IL-7 (5 ng), and IL-15 (5 ng). 6 T cells were cultured at 37°C and 5% CO2 at 100x the bead:cell ratio. T cells were activated with Dynabead Human T-Activator CD3 / CD28 (ThermoFisher) at a bead:cell ratio of 2:1 for 48–72 h prior to electroporation.

[0073] Electroporation of T cells: After 48 hours, the Dynabeads were magnetically removed and the cells were washed once with PBS before being resuspended in the appropriate electroporation buffer. For singleplex experiments, 3 × 10 5 T cells were electroporated with 1 μg of chemically modified sgRNA (Synthego, Menlo Park, CA) and 1.5 μg of SpCas9, BE3, or BE4 mRNA (TriLink Biotechnologies) in a 10 μL chip using the Neon Transfection System (ThermoFisher) under the following conditions: 1400 volts, 10 ms pulse width, 3 pulses; 1 × 10 per 20 μL. 6For multiplex studies using 100 T cells, 1.5-4 μg BE mRNA as indicated, and Nucleofector program EO-115, the 4D-Nucleofector (Lonza) with the P3 kit was used. RNPs were generated by incubation of 10 μg SpCas9 protein (IDT, Coralville Iowa), or 12 μg BE4 protein (Aldevron, Fargo) with 3 μg of each chemically modified sgRNA (Synthego) for 15 min at room temperature and electroporated using Nucleofector program EH-115. After gene transfer, T cells were recovered in antibiotic-free medium for 20 min at 37°C, 5% CO2, then cultured in complete CTS OpTmizer T cell expansion SFM as described above. Lentiviral transduction: T cells were transduced with pRRL-MND-CAR19-RQR8 lentiviral vector (UMN Viral Vector & Cloning Core) at an MOI of 20 by spinfection on Retronectin (Takara)-coated plates 24 h posttransfection.

[0074] Analysis of genomic DNA: Genomic DNA was isolated from T cells 5 days after electroporation by spin-column-based purification, PCR amplification of CRISPR-targeted loci, Sanger sequencing of PCR amplicons, and analysis by the previously described web application EditR (available on the World Wide Web at baseeditr.com). 24 Base editing efficiency was analyzed at the genome level by subsequent analysis of Sanger sequencing traces using Next Generation Sequencing (NGS) was also performed on the same PCR amplicons.

[0075] Next-generation sequencing and analysis: Primers with Nextera universal primer adaptors (Illumina) were designed to amplify 375-425 bp sites surrounding the region of interest using Primer3Plus (Table 2). Genomic DNA was PCR amplified using AccuPrime Taq DNA polymerase, High Fidelity, following the manufacturer's (Invitrogen) protocol, using 30 cycles of [94°C-2:00]-[94°C-0:30, 55°C-0:30, 68°C-0:30]-[68°C-5:00]-[4°C-hold]. Amplicons were purified from 1% agarose gels using a QIAquick gel extraction kit (Qiagen). Samples were submitted to the University of Minnesota Genomics Center for subsequent amplification with index primers and sequencing on a MiSeq 2x300bp run (Illumina). A minimum of 1,000 aligned read pairs were generated for on-target sites and 10,000 aligned read pairs for off-target sites using the previously described CRISPR / Cas9 editing analysis pipeline CRISPR-DAV. 38 Raw fastq files were analyzed against the reference and sgRNA protospacer sequences using the . The output files, “sample_snp.xlsx” and “sample_len.xlsx”, were compiled and analyzed using custom R markdown scripts (R v3.4.2).

[0076] Flow cytometry: Prior to flow cytometry, singleplex PDCD1 disrupted T cells were restimulated for 48 hours using CD3 / CD28 Dynabeads as described above. For multiplex experiments with TRAC knockout, T cells were activated with phorbol 12-myristate 13-acetate (PMA; 100 ng / mL; Sigma-Aldrich) and ionomycin (250 ng / mL; MilliporeSigma) for 24 hours. PMA / ionomycin treated T cells were washed with PBS, resuspended in culture medium, and incubated for an additional 24 hours before flow cytometry. T cells were stained with fluorophore-conjugated anti-human CD3 (BD ​​Biosciences), B2M (BioLegend), and CD279 (PD-1) (BioLegend) antibodies. Expression of CD19-T2A-RQR8 CAR was detected using anti-human CD34 monoclonal antibody (QBEnd10) (ThermoFisher). 39 Fixable Viability Dye eFluor 780 or LIVE / DEAD Fixable Aqua Dead Cell Stain (ThermoFisher) were used to assess cell viability. T cells were captured on an LSR II or LSRFortessa flow cytometer using FACSDiva software and data were analyzed using FlowJo v10 software. As stimulation did not uniformly upregulate PD-1 expression in all control T cells, the frequency of PD-1+ cells was normalized. The ratio of PD-1+ cells to the PD-1- subpopulation in control samples (rPD1) was used to calculate non-normalized values ​​(F°) for all samples below. pos and F° pos ) to normalized values ​​of PD-1+ and PD-1- subpopulations (F'' pos ' and F' neg ') was calculated. F' pos =F° pos +F° pos (1-rPD1) F' neg =F° neg -Fp°os (1-rPD1)

[0077] Cytokine profiling: Briefly, 2 × 10 cells were cultured in 200 μl of OpTmizer CTS T cell expansion SFM containing 2.5% CTS Immune Cell SR, L-glutamine, penicillin / streptomycin, monensin (0.7 μg / ml; BD Biosciences), and N-acetylglutamine (10 mM) with brefeldin A (10 μg / ml; Sigma-Aldrich) in the absence or presence of K562 or Raji cells, or Raji cells engineered to overexpress PDL1. 5 T cells were cultured for 12 h. After washing, cells were surface stained for CD4, CD8, CD27, and CD45ro; eFluor780 amine-reactive dye was used to exclude dead cells from the analysis. After permeabilization (Cytofix / Cytoperm kit; BD Biosciences), cells were stained for CD3, gamma interferon (IFN-γ), interleukin 2 (IL-2), and tumor necrosis factor (TNF). 5 × 10 4 ~1×10 5 Between 10 and 15 events were collected in each case. Electronic compensation was performed with mAb capture beads (BD Biosciences) stained separately for each mAb used in the test samples. Cells were analyzed using a modified Fortessa flow cytometer (BD Immunocytometry Systems). Data were analyzed using FlowJo version 9.9.3. Forward scatter area versus forward scatter height was used to gate out cell aggregates and dead cells were removed from the analysis to reduce background staining. Background levels of staining and cytokine production were determined using unstimulated T cells.

[0078] Translocation assay: Translocation PCR assays were designed using PrimerQuest software (Integrated DNA Technologies, Coralville IA) using settings for two primer + probe qPCRs. Each sample was run as a duplex assay consisting of an internal reference primer + probe set (HEX) and an experimental primer + probe set (FAM). Primers and probes were ordered from IDT. Reactions were set up using ddPCR Supermix (no dUTP) (Biorad, Hercules, CA) as a probe, with 200 ng of genomic DNA per assay according to the manufacturer's instructions. Droplets were generated and analyzed using a QX200 Droplet-Digital PCR System (Bio-Rad).

[0079] Cytotoxicity assay: Luciferase-expressing K562, Raji, or Raji-PDL1 cells were seeded in 96-well round-bottom plates (3 × 10 4 T cells were counted and added to wells in triplicate at the indicated E:T ratios. Target cells without effectors served as a negative control (spontaneous cell death) and target cells incubated with 1% NP-40 served as a positive control (maximal killing). Co-cultures were incubated at 37°C for 48 h. After incubation, D-luciferin (potassium salt; Gold Biotechnology) was added to each well at a final concentration of 25 μg / mL and incubated 10 min before imaging. Luminescence was read in endpoint mode using a BioTek Synergy microplate reader. Effector-free target cells were set as 100% viable and killing in experimental samples was measured relative to this baseline.

[0080] Immunoblotting assay: 1 × 10 IgG in complete RIPA buffer containing protease and phosphatase inhibitors (Sigma-Aldrich, COEDTAF-RO, P5726, and P0044). 6Protein was isolated from cells. Total protein was quantified using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific Inc., 23225) according to the manufacturer's protocol. Cell lysates at 3 μg / μL were run and analyzed on the Wes platform after denaturation at 95°C for 5 min according to the manufacturer's protocol (ProteinSimple). Primary antibodies against SpCas9 (Cell Signaling, #14697) and actin (Cell Signaling, #8457) were used at dilutions of 1:100 and 1:50, respectively, in the buffer provided with the kit, and platform-optimized secondary antibodies were purchased from Protein Simple.

[0081] Data analysis and visualization: All statistical analyses were performed in R studio. The level of significance was set at α=0.05. Data were subjected to analysis for assumptions of normality and homeodascity prior to statistical testing. Student's pairwise one-tailed or two-tailed t-tests were used as indicated in the text. Data were visualized using Prism 8 (Graphpad) or R studio using various tidyverse (available on the World Wide Web at www.tidyverse.org / ) and Bioconductor (available on the World Wide Web at www.bioconductor.org / ) packages. Data availability: Next generation sequencing reads will be deposited in the NCBI Sequence Read Archive database prior to publication.

[0082] [Table 3]

[0083] Example 2 Base editing in natural killer (NK) cells method Isolation of peripheral blood mononuclear cells (PBMC): Peripheral blood was diluted 3:1 with cold 1x PBS. The diluted blood was added dropwise to 15 mL of Lymphoprep (Stem Cell Technologies). Cells were centrifuged at 400xg for 25 min without brake. Buffy coat was removed, washed with cold 1x PBS, and centrifuged at 400xg for 10 min. Supernatant was removed and cells were either frozen for PBMC or used immediately to purify NK cells. Isolation of CD3-CD56+ NK cells: PBMC density 5 × 10 per mL 7 The total volume was adjusted to 1000 cells and the cells were transferred to a 14 mL polystyrene round-bottom tube. NK cells were isolated using a human NK cell enrichment kit or a human NK cell isolation kit (Stem Cell Technologies) according to the kit instructions. Enriched cells were counted and analyzed for purity (%CD56+, %CD3+) by flow cytometry.

[0084] Stimulation of CD3-CD56+ NK cells: Count CD3-CD56+ NK cells and administer 1.25 x 10 per mL. 5 The NK cells were plated at a density of 100 μg / mL and co-cultured with transgenic mbIL21 K562 feeder cells (clone 9; Denman et al. PLoS One, 2012) at a ratio of 2:1 (feeder:NK). Prior to co-culture, the feeder cells were irradiated at 100 Gray. The feeder and NK cells were resuspended in B0 medium containing 50 IU / mL IL2 (Peprotech). On days 3 and 5, the medium and IL2 were refreshed. On day 7, the cells were counted and used for experiments. Flow cytometry: Cells were washed with cold 1x PBS + 0.5% FBS and stained with anti-human CD3 (eBioscience) and anti-human CD56 (Miltenyi Biotec). Cells were analyzed using an LSR Fortessa (BD Biosciences) and FlowJo (Treestar).

[0085] Neon electroporation of NK cells: Stimulated NK cells were transfected using the Neon transfection kit (Invitrogen). Cells were counted and transfected at 3 × 10 per mL. 7 The cells were resuspended in Buffer T at a density of 1.5 μg of base editor, 1 μg of gRNA, 1 μg of eGFP mRNA. Base editing reagents were added to the cells prior to electroporation: 1.5 μg of base editor, 1 μg of gRNA, 1 μg of eGFP mRNA. The cells were electroporated with two 10 ms pulses of 1850 volts. After electroporation, the cells were plated in B0 medium supplemented with 1 ng / mL of IL15. The medium was refreshed every other day. On day 5, the cells were analyzed for base editing efficiency.

[0086] material

[0087] [Table 4]

[0088] [Table 5]

[0089] B0 culture medium: 60% mL of DMEM 30% mL of Ham's F12 10% mL of human AB serum Penicillin at 100 U / mL; 100 μg / mL streptomycin 20 μM 2-mercaptoethanol 50 μM ethanolamine 10 μg / mL ascorbic acid 1.6ng / mL sodium selenite Cryogen: CryoStor CS10 Cell separation reagents: Human NK cell isolation kit (Stem Cell Technologies), Human NK cell enrichment kit (Stem Cell Technologies) Electroporation reagent: 10 μL of Neon Transfection Kit (Invitrogen)

[0090] Example 3 Base editing in CD34+ hematopoietic stem and progenitor cells (HSPCs) material and method Culture Media: StemSpan Serum Free Expansion Media II (SFEM II) (Stem Cell Technologies Catalog No. 09605); 100ng / ml hSCF (Peprotech); 100ng / ml hTPO (Peprotech); 100ng / ml hFlt-3L (Peprotech); 100ng / ml hIL-6 (Peprotech); StemRegenin1 (0.75μM final concentration) Cayman Chemical Freezing medium: Cryostor CS10 Cell separation reagent: Human UCB CD34+ Enrichment Kit (Stem Cell Technologies; multiple variations available through the above sources). Other Reagents: Neon kit (Thermo Fisher Scientific, multiple options available depending on sharing).

[0091] Thawed samples: Cells were thawed in pre-warmed culture medium (37°C) using the same type of medium used for culture. 1mL of culture medium was added to a 15mL sterile conical tube. Frozen vials were thawed in a 37°C water bath until a single ice crystal remained. Vials were immediately removed to a biosafety cabinet, sprayed with 70% ethanol, and wiped down. Vials were carefully opened. Cell suspension was carefully pipetted from one vial into a 15mL conical tube. An additional 1mL of culture medium was added dropwise and gently swirled. Another 1mL of culture medium was added dropwise and gently swirled. An additional 4mL of culture medium was added and gently mixed. Centrifuged at 175g for 10 minutes. Higher centrifugal forces would result in cell death. Supernatant was aspirated and cell pellet was suspended in culture medium. Cells were counted and tested or placed in culture. It is important not to delay in placing the cells in the culture medium and in the incubator.

[0092] CD34 + Culture of HSPCs: Day 0: 1 x 10 per mL in a 24-well plate in complete medium 6 Cells were plated at a density of 100x100 cells / well. Cells were incubated at 37 °C and 5% CO for 72-96 h and medium was added as needed based on color and cell density. CD34 + Neon transfection of cells: 3e5 viable cells were electroporated in a 10 μl chip. Electroporation parameters: 1450V, 10 ms, 3 pulses. For knockouts using all mRNA: 100 µl chip: 15 µg Cas9 mRNA, 10–20 µg gRNA-RNA; 10 µl chip: 1.5 µg Cas9 mRNA, 1–2 µg gRNA-RNA. After transfection, cells were plated at a density of 3000 cells per μl in antibiotic-free medium containing 5% CO 2 for approximately 20 minutes. After the recovery period, 2x the volume of medium containing antibiotics was added to the wells. Cells were cultured at 37°C in 5% CO2. rAAV transduction of CD34+ cells: rAAV was thawed on ice and mixed in wells before adding cells. The specified MOI was added at the following times post-electroporation: For cells edited with Cas9 mRNA: add virus 4-6 hours after electroporation. For Cas9 protein (RNP): add virus 15 min after electroporation.

[0093] Post-electroporation expansion: observed media color after electroporation as an indication of media addition. Timing will vary depending on cell health for a particular experiment / donor. Once the media color turned orange (in some cases as early as 48 hours), we doubled the volume of culture media using culture media containing 2x the concentration of cytokines ("2x media"). This process continued as necessary for the course of the culture period. In some cases, when cells were growing very rapidly (especially around days 7-9) and medium was quickly consumed, cells were spun down and reconstituted in 2-3 volumes of 1x medium.

[0094] When cell growth was poor and 3-4 days had passed without the need to double the medium, we carefully removed about 50% of the medium from the top by pipetting. We were careful not to disturb the cells that had settled to the bottom of the flask. The removed medium was replaced with an equal volume of twice the medium.

[0095] Example 3 Base editing in primary fibroblasts method Cell culture: 1 x 10 primary fibroblasts per mL from a healthy patient(s) with Fanconi anemia or MPS1 6 Cells were frozen in CryoStor at a concentration of 1000 x g / ml. Cells were thawed in human primary fibroblast (hFib) medium, which consisted of 500 mL of MEM alpha formulated medium (Invitrogen), 5 mL of GlutaMAX (商標) The medium consisted of 20 mL of PBS (100x, Invitrogen), 20 mL of FBS, 2.5 mL of penicillin-streptomycin (10000 U / mL, Invitrogen), 5 mL of non-essential amino acids (100x, Invitrogen), 500 μL of antioxidant supplement (1000x, Sigma-Aldrich), 100 uL of mEGF (50 ng / μL, Sigma-Aldrich), and 100 uL of hFGF (2.5 μg / μL Sigma-Aldrich). Cells were maintained at 70–100% confluency at 37 °C with 5% CO2 and 5% O2 by refreshing the medium every 2 days and subculturing weekly.

[0096] Transfection of primary fibroblasts: One day prior, transfect fibroblasts at 90-100% confluency at 3 × 10 per well of a 12-well plate. 5Cells were plated at a density of 100 x 100 cells / well. On day 0, medium was replaced with 1 mL of 37°C hFib medium. To 100 μL of Opti-MEM, 1.5 μg of BE4max mRNA, 1 μg of cm-sgRNA, and 100 ng of eGFP mRNA were added. 2 μL of mRNA Boost Reagent (Mirus) followed by 2 μL of Trans-IT Reagent (Mirus) were added according to the manufacturer's protocol. After 3 min incubation at room temperature, Opti-MEM solution was dispensed dropwise onto the cells. On day 1, 24 h after transfection, medium was changed and cells were visualized for GFP. On day 3, medium was changed and on day 4, cells were harvested for genomic analysis.

[0097] Genomic Analysis: Genomic DNA analysis: Genomic DNA was isolated from fibroblasts by spin-column-based purification. Base editing efficiency was analyzed at the genome level by PCR amplification of CRISPR-targeted loci, Sanger sequencing of the PCR amplicons, and subsequent analysis of the Sanger sequencing traces using the web application EditR (baseeditr.com) as previously described.

[0098] material

[0099] [Table 6]

[0100] [Table 7]

[0101] Results: To determine whether base editing can be used to correct the pathogenic mutation in the FNACA gene responsible for Fanconi anemia, we introduced BE4max mRNA and cm-sgRNA into primary fibroblasts and attempted to correct the mutation. We found that we were able to achieve 19% correction of the pathogenic mutation to the WT allele (Figure 21A). Further experiments will evaluate whether this correction results in restoration of the WT phenotype, particularly with respect to DNA repair of cross-linked DNA and the ability to be reprogrammed into induced pluripotent stem cells (iPSCs). To determine whether adenosine deaminase base editing is possible in primary Fanconi anemia fibroblasts, we introduced ABE7.10 mRNA together with cm-sgRNA into primary fibroblasts. We found that we were able to achieve 49% editing at the target site, indicating highly efficient base editing in primary cells (Figure 21B). Highly efficient base editing at this site indicates the feasibility of using base editing as an approach to single nucleotide editing, instead of HDR. Further studies will involve the use of adenosine deaminase and cytidine deaminase base editors to correct pathogenic mutations in primary fibroblasts from various patients affected by genetic diseases, with the intention of generating iPSCs for differentiation into hematopoietic stem cells for autologous gene correction and transplantation.

[0102] References TIFF2024113701000019.tif198170 TIFF2024113701000020.tif216170 TIFF2024113701000021.tif218170 TIFF2024113701000022.tif208170 TIFF2024113701000023.tif92170

[0103] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein which equivalents are intended to be encompassed by the following claims. All references disclosed herein, including patent documents, are hereby incorporated by reference in their entirety. Yet another aspect of the present invention may be as follows. [1] A method for producing genetically engineered lymphohematopoietic cells, comprising: (a) In lymphocyte hematopoietic cells, (i) a plasmid, mRNA, or protein encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain, wherein the nickase domain comprises a base excision repair inhibitor domain; and (ii) one or more splice acceptor-splice donor (SA-SD) gRNAs having complementarity to a target nucleic acid sequence to be genetically modified; Introducing (b) culturing the transduced cells under conditions promoting disruption of the splice site targeted by the one or more SA-SD gRNAs, whereby the target nucleic acid sequence is modified by the base editor fusion protein and the one or more splice acceptor-splice donor (SA-SD) gRNAs relative to an untransfected lymphohematopoietic cell, thereby generating a genetically engineered lymphohematopoietic cell. The method includes: [2] The method according to [1], further comprising introducing into the lymphohematopoietic cells one or more gRNAs designed to generate one or more targeted knock-ins or missense mutations, thereby causing the genetically engineered lymphohematopoietic cells to contain at least one gene knockout and one or more gene knock-ins or missense mutations. [3] The method according to [1], further comprising introducing into the lymphohematopoietic cells one or more gRNAs designed to generate one or more targeted knock-ins and one or more missense mutations, thereby causing the genetically engineered lymphohematopoietic cells to contain at least one gene knockout, at least one gene knock-in, and at least one missense mutation. [4] The method of [1], wherein the base editor fusion protein is BE3, BE4, or an adenine base editor (ABE). [5] The method according to [1], wherein the lymphohematopoietic cells are T cells, natural killer (NK) cells, B cells, or CD34+ hematopoietic stem progenitor cells (HSPCs). [6] The method according to [1], wherein the one or more SA-SD gRNAs are chemically modified to contain a 2'-O-methyl phosphorothioate modification in at least one 5' nucleotide and at least one 3' nucleotide of each gRNA. [7] The method of [1], wherein the base editor fusion protein and one or more splice acceptor-splice donor (SA-SD) gRNAs exhibit a C to T conversion efficiency of about 50% to about 90%. [8] The method according to [1], wherein the one or more SA-SD gRNAs are selected from the sequences shown in Table 1. [9] A method for generating genetically modified T cells, comprising: (a) To human T cells (i) a plasmid, mRNA, or protein encoding a base editor fusion protein comprising a deaminase domain fused to a Cas9 nickase domain, wherein the nickase domain comprises a base excision repair inhibitor domain; (ii) one or more splice acceptor-splice donor (SA-SD) gRNAs to disrupt the expression of each of TRAC, B2M, and PDCD1; (iii) a donor DNA template encoding a T cell receptor (TCR) and a chimeric antigen receptor (CAR), and (iv) Two gRNAs complementary to the target insertion site and transfecting (b) culturing the transfected T cells under conditions that promote disruption of a splice site targeted by the SA-SD gRNA, whereby expression of TRAC, B2M, and PDCD1 gene products is reduced compared to untransfected T cells; (c) culturing the transfected T cells under conditions that promote targeted knock-in of the donor DNA template at the targeted insertion site; The method includes:

[10] The method according to [9], wherein the base editor fusion protein is BE3, BE4, or an adenine base editor (ABE).

[11] The method according to [9], wherein the one or more SA-SD gRNAs are selected from the sequences shown in Table 1.

[12] The method according to [9], wherein one or more of the SA-SD gRNA and the gRNA complementary to the target insertion site are chemically modified to contain a 2'-O-methyl phosphorothioate modification in at least one 5' nucleotide and at least one 3' nucleotide of each gRNA.

[13] The method according to [9], wherein the donor DNA template is provided as a rAAV.

[14] The method according to [9], wherein the TCR specifically binds to a tumor antigen.

[15] The method of [9], wherein the CAR comprises a CAR antigen-binding domain that specifically binds to a tumor antigen.

[16] The method according to [9], wherein the TCR and the CAR bind to different antigens.

[17] A genetically modified cell obtained by the method according to any one of [1] to

[16] above.

Claims

1. A genetically modified T cell comprising a point mutation at a splice acceptor (SA) or splice donor (SD) site in at least one target gene, wherein the at least one target gene comprises TRAC, PDCD1 or both, and the at least one target gene comprises one of SEQ ID NOs: 1-5, 9 or 10, and the point mutation is within one of SEQ ID NOs: 1-5, 9 or 10.

2. The genetically modified T cell of claim 1, wherein the at least one target gene includes PDCD1.

3. The genetically modified T cell of claim 1, wherein the at least one target gene includes TRAC.

4. The genetically modified T cell of claim 1, wherein the at least one target gene includes PDCD1 and TRAC.

5. The genetically modified T cell of claim 1, wherein the point mutation includes a C to T mutation or a G to A mutation.

6. The genetically modified T cell of claim 1, wherein the at least one target gene includes PDCD1 and the point mutation is within one of SEQ ID NOs: 1 to 5.

7. The genetically modified T cell of claim 1, wherein the at least one target gene comprises TRAC and the point mutation is within one of SEQ ID NOs: 9 or 10.

8. A genetically modified T cell according to any one of claims 1 to 7, further comprising a DNA template encoding a chimeric antigen receptor (CAR).

9. The genetically modified T cell of claim 8, wherein the CAR comprises a polypeptide comprising an extracellular antigen-binding domain of an antibody linked to a transmembrane domain and at least one intracellular domain.

10. The genetically modified T cell of claim 8, wherein the antigen-binding domain of the CAR comprises a single-chain variable fragment (scFv).

11. A chimeric antigen receptor (CAR) T cell comprising at least one point mutation in one or more splice acceptor (SA) or splice donor (SD) sites in at least one target gene, wherein the at least one target gene comprises TRAC, PDCD1 or both, and the at least one target gene comprises one of SEQ ID NOs: 1-5, 9 or 10, and the point mutation is within one of SEQ ID NOs: 1-5, 9 or 10.

12. The chimeric antigen receptor (CAR) T cell of claim 11, comprising a DNA template encoding a CAR.

13. The CAR T cell of claim 11, wherein the CAR comprises a polypeptide comprising an extracellular antigen-binding domain and at least one intracellular domain of an antibody linked to a transmembrane domain.

14. The CAR T cell of claim 11, wherein the antigen-binding domain of the CAR comprises a single-chain variable fragment (scFv).

15. The CAR T cell of claim 11, wherein the at least one target gene includes PDCD1.

16. The CAR T cell of claim 11, wherein the at least one target gene comprises TRAC.

17. The CAR T cell of claim 11, wherein the at least one target gene includes PDCD1 and TRAC.

18. The CAR T cell of claim 11, wherein the point mutation comprises a C to T mutation or a G to A mutation.

19. The CAR T cell of claim 11, wherein the at least one target gene comprises PDCD1 and the point mutation is within one of SEQ ID NOs: 1-5.

20. The CAR T cell of claim 11, wherein the at least one target gene comprises TRAC and the point mutation is within one of SEQ ID NOs: 9 or 10.

Images