Epitope engineering of cell surface receptors

Abstract

Genetically engineered hematopoietic cells, such as hematopoietic stem cells, having one or more gene-edited genes for cell surface proteins, and their therapeutic uses alone or in combination with immunotherapy targeting the cell surface protein(s).

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS The present invention claims the benefit of priority to U.S. Provisional Application No. 63 / 311,707, filed February 18, 2022, and U.S. Provisional Application No. 63 / 426,138, filed November 17, 2022, the contents of which are incorporated by reference in their entireties herein.

[0002] Provided herein is the use of genetically engineered hematopoietic cells, such as hematopoietic stem cells, that have one or more gene-edited genes for cell surface proteins, which can be used in combination with immunotherapy (i.e., cytotoxic agents such as chimeric antigen receptor T cells). [Background technology]

[0003] Innovations in gene transfer have made it possible to reprogram immune cells to target molecules expressed in cancer cells. With very promising results, the FDA approved the first adoptive cellular immunotherapy, known as CD19 CAR-T cells, for the treatment of B-lymphoblastic leukemia. Although these successes are expected to revolutionize the oncology field, their application has been hindered because most suitable candidates are often shared with healthy bone marrow cells, causing immunosuppression and severe hematopoietic toxicity. Anti-myeloid / stem cell CAR-T-induced toxicity limits their applicability for salvage therapy in a limited time window before HSCT and may be insufficient for eradicating the disease. Thus, there remains an unmet need to effectively target cells of interest, e.g., cancer cells, without targeting or damaging normal cell populations. Summary of the Invention

[0004] The present disclosure relates generally to genetically engineered hematopoietic cells, such as hematopoietic stem cells, that have one or more gene-edited genes for cell surface proteins, and chimeric antigen receptors that can target the same cell surface proteins.

[0005] In one embodiment, a genetically engineered hematopoietic stem cell (HSPC) is provided, comprising a genetically engineered FLT3 gene, the genetically engineered FLT3 gene being engineered to reduce binding of its encoded protein to a therapeutic anti-FLT3 antibody. In an embodiment, the genetically engineered FLT3 gene comprises at least one mutation in exon 9 of the FLT3 gene. In an embodiment, the at least one mutation in exon 9 of the genetically engineered FLT3 gene results in a polypeptide having a mutation at position N399. In an embodiment, the mutation at position N399 is N399D or N399G. In an embodiment, the therapeutic anti-FLT3 antibody is an anti-FLT3 clone 4G8 antibody. In an embodiment, the therapeutic anti-FLT3 antibody is an antibody that has the same six CDRs as the 4G8 antibody or competes with it. In an embodiment, the genetically engineered HSPC is genetically engineered using a CRISPR system comprising a guide nucleic acid and a nuclease. In one embodiment, the nuclease is any of Streptococcus pyogenes Cas9 (SpCas9), Staphylococcus aureus (SaCas9), Lachnospiraceae bacteria Cas12a (LbCas12a), or Acidaminococcus sp. BV3L6 (AsCas12a). In one embodiment, the CRISPR system comprises SpCas9. In one embodiment, the guide nucleic acid is selected from the group consisting of SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, and SEQ ID NO: 16. In one embodiment, the CRISPR system further comprises a template DNA. In one embodiment, the template DNA is selected from the group consisting of SEQ ID NO: 40, SEQ ID NO: 41, SEQ ID NO: 42, and SEQ ID NO: 43. In one embodiment, the nuclease is a catalytically impaired SpCas9 linked to a base editor enzyme. In one embodiment, the base editor enzyme is a nucleotide deaminase. In one embodiment, the nucleotide deaminase is either cytosine deaminase or adenosine deaminase.In another embodiment, the catalytically impaired SpCas9 is NG-SpCas9 or SpRY-SpCas9.In one embodiment, the catalytically impaired SpCas9 comprises a mutation at position D10A.In one embodiment, the catalytically impaired SpCas9 further comprises a mutation at position K918N. In one embodiment, the guide RNA is selected from the group consisting of SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, and SEQ ID NO: 23. In one embodiment, the engineered FLT3 gene encodes a polypeptide comprising the amino acid sequence of SEQ ID NO: 51 or SEQ ID NO: 52. Also provided is a population of engineered hematopoietic stem cells (HSPCs) comprising the engineered hematopoietic stem cells (HSPCs) described above. In some embodiments, a method of treating a hematopoietic malignancy is provided, the method comprising administering to a human subject (a) the population of engineered hematopoietic stem cells described above, and (b) a therapeutically effective amount of at least one agent comprising an anti-FLT3 antibody binding domain or an antibody or antibody fragment comprising an anti-FLT3 binding domain. In one embodiment, the at least one agent comprises a chimeric antigen receptor-T (CAR-T) cell comprising an anti-FLT3 antibody binding domain. In one embodiment, the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), or T-cell acute lymphoblastic leukemia (T-ALL). In one embodiment, the method further comprises obtaining HSPCs from a biological sample from the human subject and genetically engineering the HSPCs from the biological sample from the human subject, thereby forming a population of genetically engineered HSPCs. In one embodiment, the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood derived CD34+ hematopoietic stem and progenitor cells.

[0006] Also provided in one embodiment is an engineered hematopoietic stem cell (HSPC) comprising an engineered CD123 gene, the engineered CD123 gene being engineered to reduce binding of its encoded protein to a therapeutic anti-CD123 antibody. In one embodiment, the therapeutic anti-CD123 antibody is the clone 7G3 antibody or its humanized counterpart CSL362. In an embodiment, the engineered CD123 gene comprises at least one mutation in exon 2 of the CD123 gene. In one embodiment, the at least one mutation in exon 2 of the engineered CD123 gene results in a polypeptide having a mutation at position S59. In one embodiment, the mutation at S59 is S59P or S59F. In one embodiment, the therapeutic anti-CD123 antibody is the anti-CD123 clone 6H6 antibody or the anti-CD123 clone S18016F antibody. In one embodiment, the engineered CD123 gene comprises at least one mutation in exon 3 of the CD123 gene. In one embodiment, at least one mutation in exon 3 of the engineered CD123 gene results in a polypeptide having a mutation at position P88. In one embodiment, the mutation at P88 is P88L or P88S. In one embodiment, the engineered HSPCs are engineered using a CRISPR system comprising a guide nucleic acid and a nuclease. In one embodiment, the guide nucleic acid is selected from the group consisting of SEQ ID NO:24, SEQ ID NO:27, SEQ ID NO:30, SEQ ID NO:32, SEQ ID NO:33, and SEQ ID NO:34. In one embodiment, the nuclease is a catalytically impaired SpCas9 linked to a base editor enzyme. In one embodiment, the base editor enzyme is a nucleotide deaminase. In one embodiment, the base editor enzyme is either a cytosine deaminase or an adenosine deaminase. In one embodiment, the catalytically impaired SpCas9 is NG-SpCas9 or SpRY-SpCas9. In one embodiment, the catalytically impaired SpCas9 comprises a mutation at position D10A. In one embodiment, the catalytically impaired SpCas9 further comprises a mutation at position K918N.In one embodiment, the engineered CD123 gene encodes a polypeptide comprising the amino acid sequence of SEQ ID NO:54, SEQ ID NO:55, SEQ ID NO:56, SEQ ID NO:57, or SEQ ID NO:58. Also provided herein is a population of engineered hematopoietic stem cells (HSPCs) comprising the engineered hematopoietic stem cells (HSPCs) described above. Also provided is a method of treating hematopoietic malignancies, comprising administering to a human subject (a) the engineered hematopoietic stem cell population described above, and (b) a therapeutically effective amount of at least one agent comprising an anti-CD123 antibody binding domain or an antibody or antibody fragment comprising an anti-CD123 binding domain. In one embodiment, the at least one agent comprises a chimeric antigen receptor-T (CAR-T) cell comprising an anti-CD123 antibody binding domain. In one embodiment, the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL), or blastic plasmacytoid dendritic cell leukemia (BPCDN). In one embodiment, the method further comprises obtaining HSPCs from a biological sample from the human subject and genetically engineering the HSPCs from the biological sample from the human subject, thereby forming a population of genetically engineered HSPCs. In an embodiment, the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood derived CD34+ hematopoietic stem and progenitor cells.

[0007] In some embodiments, a population of genetically engineered hematopoietic stem cells (HSPCs) is provided, comprising: (i) a genetically engineered FLT3 gene, wherein the genetically engineered FLT3 gene encodes a protein with reduced binding to a therapeutic anti-FLT3 antibody; and (ii) a genetically engineered CD123 gene, wherein the genetically engineered CD123 gene encodes a protein with reduced binding to a therapeutic anti-CD123 antibody. In an embodiment, the genetically engineered FLT3 gene comprises at least one mutation in exon 9 of the FLT3 gene. In one embodiment, the at least one mutation in exon 9 of the genetically engineered FLT3 gene results in a polypeptide having a mutation at position N399. In an embodiment, the genetically engineered CD123 gene comprises at least one mutation in exon 2 of the CD123 gene. In one embodiment, the at least one mutation in exon 2 of the genetically engineered CD123 gene results in a polypeptide having a mutation at position S59. In an embodiment, the therapeutic anti-FLT3 antibody is anti-FLT3 clone 4G8 antibody. In an embodiment, the therapeutic anti-CD123 antibody is anti-CD123 clone 7G3 antibody or CSL362 antibody. In an embodiment, the population of HSPCs is engineered using a CRISPR system comprising at least two guide nucleic acids and a nuclease. In an embodiment, the at least two guide nucleic acids are 1) SEQ ID NO: 18 or SEQ ID NO: 20, and 2) SEQ ID NO: 24 or SEQ ID NO: 27. In an embodiment, the at least two guide nucleic acids are SEQ ID NO: 20 and SEQ ID NO: 27. In an embodiment, the nuclease is a catalytically impaired SpCas9 linked to a base editor enzyme. In an embodiment, the base editor enzyme is a nucleotide deaminase. In an embodiment, the base editor enzyme is either a cytosine deaminase or an adenosine deaminase. In an embodiment, the catalytically impaired SpCas9 is NG-SpCas9 or SpRY-SpCas9. In one embodiment, the catalytically impaired SpCas9 comprises a mutation at position D10A, hi one embodiment, the SpCas9 further comprises a mutation at position K918N.Also provided is a method of treating a hematopoietic malignancy, comprising administering to a human subject (a) a population of HSPCs as described above, and (b) a therapeutically effective amount of at least one agent comprising one or both of (1) an anti-FLT3 antibody binding domain or an antibody or antibody fragment comprising an anti-FLT3 binding domain, and / or (2) an anti-CD123 antibody binding domain or an antibody or antibody fragment comprising an anti-CD123 binding domain. In one embodiment, the at least one agent comprises a chimeric antigen receptor-T (CAR-T) cell comprising an anti-FLT3 antibody binding domain and / or an anti-CD123 antibody binding domain. In one embodiment, the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL), or blastic plasmacytoid dendritic cell leukemia (BPCDN). In one embodiment, the method further comprises obtaining HSPCs from a biological sample from the human subject and genetically engineering the HSPCs from the biological sample from the human subject, thereby forming a population of genetically engineered HSPCs. In one embodiment, the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood derived CD34+ hematopoietic stem and progenitor cells.

[0008] Also provided in one embodiment is an engineered hematopoietic stem cell (HSPC) comprising an engineered KIT gene, the engineered KIT gene engineered to reduce binding of its encoded protein to a therapeutic anti-KIT antibody. In one embodiment, the engineered KIT gene comprises at least one mutation in exon 7 of the KIT gene. In one embodiment, the at least one mutation in exon 7 of the engineered KIT gene results in a polypeptide having a mutation at position H378. In one embodiment, the mutation at position H378 is H378R. In one embodiment, the therapeutic anti-KIT antibody is an anti-KIT clone Fab79D antibody. In one embodiment, the engineered HSPC is engineered using a CRISPR system comprising a guide nucleic acid and a nuclease. In one embodiment, the guide nucleic acid is selected from the group consisting of SEQ ID NO: 36, SEQ ID NO: 37, SEQ ID NO: 38, and SEQ ID NO: 39. In one embodiment, the nuclease is a catalytically impaired SpCas9 linked to a base editor enzyme. In one embodiment, the base editor enzyme is a nucleotide deaminase. In one embodiment, the base editor enzyme is either a cytosine deaminase or an adenosine deaminase. In one embodiment, the catalytically impaired SpCas9 is NG-SpCas9 or SpRY-SpCas9. In one embodiment, the catalytically impaired SpCas9 comprises a mutation at position D10A. In one embodiment, the SpCas9 further comprises a mutation at position K918N. Also provided is a population of genetically engineered hematopoietic stem cells (HSPCs) comprising the genetically engineered hematopoietic stem cells (HSPCs) described above. Also provided is a method of treating a hematopoietic malignancy, comprising administering to a human subject (a) the genetically engineered hematopoietic stem cell population described above, and (b) a therapeutically effective amount of at least one agent comprising an anti-KIT antibody binding domain or an antibody or antibody fragment comprising an anti-KIT antibody binding domain. In one embodiment, the at least one agent comprises a chimeric antigen receptor-T (CAR-T) cell comprising an anti-KIT antibody binding domain.In one embodiment, the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), or T-acute lymphoblastic leukemia (T-ALL). In one embodiment, the method further comprises obtaining HSPCs from a biological sample from the human subject and genetically engineering the HSPCs from the biological sample from the human subject, thereby forming a population of genetically engineered HSPCs. In an embodiment, the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood derived CD34+ hematopoietic stem and progenitor cells.

[0009] Also provided in some embodiments is a genetically engineered population of hematopoietic stem cells (HSPCs), comprising: (i) a genetically engineered KIT gene, wherein the genetically engineered KIT gene encodes a protein with reduced binding to a therapeutic anti-KIT antibody; and (ii) a genetically engineered CD123 gene, wherein the genetically engineered CD123 gene encodes a protein with reduced binding to a therapeutic anti-CD123 antibody. In one embodiment, the genetically engineered KIT gene comprises at least one mutation in exon 7 of the KIT gene. In one embodiment, the at least one mutation in exon 7 of the genetically engineered KIT gene results in a polypeptide having a mutation at position H378. In one embodiment, the genetically engineered CD123 gene comprises at least one mutation in exon 2 of the CD123 gene. In one embodiment, the at least one mutation in exon 2 of the genetically engineered CD123 gene results in a polypeptide having a mutation at position S59. In one embodiment, the therapeutic anti-KIT antibody is an anti-KIT clone Fab79D antibody. In one embodiment, the therapeutic anti-CD123 antibody is an anti-CD123 clone 7G3 antibody or a CSL362 antibody. In one embodiment, the population of HSPCs is engineered using a CRISPR system comprising at least two guide nucleic acids and a nuclease. In one embodiment, the at least two guide nucleic acids are 1) SEQ ID NO:36, SEQ ID NO:37, SEQ ID NO:38, or SEQ ID NO:39, and 2) SEQ ID NO:24 or SEQ ID NO:27. In one embodiment, the at least two guide nucleic acids are SEQ ID NO:37 and SEQ ID NO:27. In one embodiment, the nuclease is a catalytically impaired SpCas9 linked to a base editor enzyme. In one embodiment, the base editor enzyme is a nucleotide deaminase. In one embodiment, the base editor enzyme is either a cytosine deaminase or an adenosine deaminase. In one embodiment, the catalytically impaired SpCas9 is NG-SpCas9 or SpRY-SpCas9. In one embodiment, the catalytically impaired SpCas9 comprises a mutation at position D10A.In one embodiment, the SpCas9 further comprises a mutation at position K918N. Also provided is a method of treating a hematopoietic malignancy, comprising administering to a human subject (a) a population of HSPCs as described above, and (b) a therapeutically effective amount of at least one agent comprising one or both of (1) an anti-KIT antibody binding domain or an antibody or antibody fragment comprising an anti-KIT binding domain, and / or (2) an anti-CD123 antibody binding domain or an antibody or antibody fragment comprising an anti-CD123 binding domain. In one embodiment, the at least one agent comprises a chimeric antigen receptor-T (CAR-T) cell comprising an anti-KIT antibody binding domain and / or an anti-CD123 antibody binding domain. In one embodiment, the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL), or blastic plasmacytoid dendritic cell leukemia (BPCDN). In one embodiment, the method further comprises obtaining HSPCs from a biological sample from the human subject and genetically engineering the HSPCs from the biological sample from the human subject, thereby forming a population of genetically engineered HSPCs. In an embodiment, the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood derived CD34+ hematopoietic stem and progenitor cells.

[0010] Also provided is a population of genetically engineered hematopoietic stem cells (HSPCs), comprising: (i) a genetically engineered FLT3 gene, wherein the genetically engineered FLT3 gene encodes a protein with reduced binding to a therapeutic anti-FLT3 antibody; and (ii) a genetically engineered KIT gene, wherein the genetically engineered KIT gene encodes a protein with reduced binding to a therapeutic anti-KIT antibody. In an embodiment, the genetically engineered FLT3 gene comprises at least one mutation in exon 9 of the FLT3 gene. In one embodiment, the at least one mutation in exon 9 of the genetically engineered FLT3 gene results in a polypeptide with a mutation at position N399. In one embodiment, the genetically engineered KIT gene comprises at least one mutation in exon 7 of the KIT gene. In one embodiment, the at least one mutation in exon 7 of the genetically engineered KIT gene results in a polypeptide with a mutation at position H378. In an embodiment, the therapeutic anti-FLT3 antibody is the anti-FLT3 clone 4G8 antibody. In one embodiment, the therapeutic anti-KIT antibody is an anti-KIT clone Fab79D antibody. In one embodiment, the population of HSPCs is engineered using a CRISPR system comprising at least two guide nucleic acids and a nuclease. In one embodiment, the at least two guide nucleic acids are 1) SEQ ID NO: 18 or SEQ ID NO: 20, and 2) SEQ ID NO: 36, SEQ ID NO: 37, SEQ ID NO: 38, or SEQ ID NO: 39. In one embodiment, the at least two guide nucleic acids are SEQ ID NO: 20 and SEQ ID NO: 37. In one embodiment, the nuclease is a catalytically impaired SpCas9 linked to a base editor enzyme. In one embodiment, the base editor enzyme is a nucleotide deaminase. In one embodiment, the base editor enzyme is either a cytosine deaminase or an adenosine deaminase. In one embodiment, the catalytically impaired SpCas9 is NG-SpCas9 or SpRY-SpCas9. In one embodiment, the catalytically impaired SpCas9 comprises a mutation at position D10A. In one embodiment, the SpCas9 further comprises a mutation at position K918N.Also provided is a method of treating a hematopoietic malignancy, comprising administering to a human subject (a) a population of HSPCs as described above, and (b) a therapeutically effective amount of at least one agent comprising one or both of (1) an antibody or antibody fragment comprising an anti-FLT3 antibody binding domain or an antibody fragment comprising an anti-FLT3 antibody binding domain, and / or (2) an antibody or antibody fragment comprising an anti-KIT antibody binding domain or an antibody fragment comprising an anti-KIT antibody binding domain. In one embodiment, the at least one agent comprises a chimeric antigen receptor-T (CAR-T) cell comprising an anti-FLT3 antibody binding domain and / or an anti-KIT antibody binding domain. In one embodiment, the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL), or blastic plasmacytoid dendritic cell leukemia (BPCDN). In one embodiment, the method further comprises obtaining HSPCs from a biological sample from the human subject and genetically engineering the HSPCs from the biological sample from the human subject, thereby forming a population of genetically engineered HSPCs. In an embodiment, the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood derived CD34+ hematopoietic stem and progenitor cells.

[0011] Also provided in some embodiments is a chimeric antigen receptor (CAR) comprising a polypeptide comprising: (a) one or more epitope-binding fragments that bind to an epitope of one or more cell surface lineage-specific proteins; (b) a hinge domain; (c) a transmembrane domain; (d) a costimulatory domain; and (e) a cytoplasmic signaling domain, wherein the one or more cell surface lineage-specific proteins are selected from FLT3, CD123, and / or KIT.

[0012] In some embodiments, the cell surface lineage specific protein is FLT3 and the CAR comprises the amino acid sequence of SEQ ID NO: 73. In some embodiments, the cell surface lineage specific protein is FLT3 and the one or more epitope binding fragments comprise one or more epitope binding fragments from SEQ ID NO: 73. In some embodiments, the cell surface lineage specific protein is FLT3 and the one or more epitope binding fragments comprise the following CDR sequences: GYTFTSYWMH (SEQ ID NO: 96), EIDPSDSYKDYNQKFK (SEQ ID NO: 97), RAITTTPFDF (SEQ ID NO: 98), RASQSISNNLH (SEQ ID NO: 99), YASQSIS (SEQ ID NO: 100), and QQSNTWPYT (SEQ ID NO: 101). In some embodiments, the cell surface lineage specific protein is CD123 and the CAR comprises the amino acid sequence of SEQ ID NO: 75, SEQ ID NO: 86, or SEQ ID NO: 87. In some embodiments, the cell surface lineage specific protein is CD123 and the one or more epitope binding fragments comprise one or more epitope binding fragments from SEQ ID NO: 75, SEQ ID NO: 86, or SEQ ID NO: 87. In some embodiments, the cell surface lineage specific protein is CD123 and the one or more epitope binding fragments comprise the following CDR sequences: GYSFTDYYMK (SEQ ID NO: 104), DIIPSNGATFYNQKFKG (SEQ ID NO: 105), ARSHLLRASWFAY (SEQ ID NO: 106), SQSLLNSGNQKNYLT (SEQ ID NO: 107), WASTRES (SEQ ID NO: 108), and QNDYSYPYT (SEQ ID NO: 109). In some embodiments, the cell surface lineage specific protein is CD123 and the one or more epitope binding fragments comprise the following CDR sequences: DIIPSNGATFYNQKFKG (SEQ ID NO: 105), SQSLLNSGNQKNYLT (SEQ ID NO: 107), WASTRES (SEQ ID NO: 108), and QNDYSYPYT (SEQ ID NO: 109). In some embodiments, the cell surface lineage specific protein is KIT and the CAR comprises the amino acid sequence of SEQ ID NO: 69 or SEQ ID NO: 71. In some embodiments, the cell surface lineage specific protein is KIT and the one or more epitope binding fragments comprise one or more epitope binding fragments from SEQ ID NO: 69 or SEQ ID NO: 71.In some embodiments, the cell surface lineage specific protein is KIT and the one or more epitope binding fragments comprise the following CDR sequences: GFNISVYMMH (SEQ ID NO: 88), SIYPYSGYTYYADSVKG (SEQ ID NO: 89), ARYVYHALDY (SEQ ID NO: 90), RASQRGLRNVAVA (SEQ ID NO: 91), SASSLYS (SEQ ID NO: 92), and QQWAVHSLIT (SEQ ID NO: 93). In some embodiments, the one or more cell surface lineage specific proteins are FLT3 and CD123 and the CAR comprises the amino acid sequence of SEQ ID NO: 77 or SEQ ID NO: 79. In some embodiments, the one or more cell surface lineage specific proteins are FLT3 and CD123 and the one or more epitope binding fragments comprise one or more epitope binding fragments from SEQ ID NO: 77 or SEQ ID NO: 79. In some embodiments, the one or more cell surface lineage specific proteins are FLT3 and CD123, and the one or more epitope-binding fragments comprise the following CDR sequences: GYTFTSYWMH (SEQ ID NO: 96), EIDPSDSYKDYNQKFK (SEQ ID NO: 97), RAITTTPFDF (SEQ ID NO: 98), RASQSISNNLH (SEQ ID NO: 99), YASQSIS (SEQ ID NO: 100), QQSNTWPYT (SEQ ID NO: 101), GYSFTDYYMK (SEQ ID NO: 104), DIIPSNGATFYNQKFKG (SEQ ID NO: 105), ARSHLLRASWFAY (SEQ ID NO: 106), SQSLLNSGNQKNYLT (SEQ ID NO: 107), WASTRES (SEQ ID NO: 108), and QNDYSYPYT (SEQ ID NO: 109). In some embodiments of any of the above CARs, the hinge domain is a CD28 hinge, an IgG4 hinge, or a CD8α hinge. In some embodiments of any of the above CARs, the transmembrane domain is CD28 TM, CD8α TM, or 4-1BB TM. In some embodiments of any of the above CARs, the costimulatory domain is CD28z, 4-1BB, ICOS, or OX40. In some embodiments of any of the above CARs, the cytoplasmic signaling domain is CD3z.

[0013] Also provided herein is a cell expressing any of the above-mentioned CARs. In some embodiments, the cell is an immune cell. In some embodiments, the immune cell is a T cell.

[0014] Also provided herein is a method of treating a hematopoietic malignancy, the method comprising administering to a human subject: (a) a population of genetically engineered hematopoietic stem cells (e.g., any of those described above); and (b) a cell (e.g., an immune cell) that expresses any of the CARs described above.

[0015] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:51, wherein the polypeptide sequence comprises a mutation at N399D, and the polypeptide sequence has reduced binding to a therapeutic anti-FLT3 antibody.

[0016] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:52, wherein the polypeptide sequence comprises a mutation at N399G, and the polypeptide sequence has reduced binding to a therapeutic anti-FLT3 antibody.

[0017] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:54, wherein the polypeptide sequence comprises a mutation at S59P, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody.

[0018] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:55, wherein the polypeptide sequence comprises mutations at Y58H and S59P, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody.

[0019] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:56, wherein the polypeptide sequence comprises a mutation at S59F, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody.

[0020] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:57, wherein the polypeptide sequence comprises a mutation at P88S, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody.

[0021] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:58, wherein the polypeptide sequence comprises a mutation at P88L, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody.

[0022] Also provided herein is a polypeptide sequence comprising a polypeptide sequence at least 80% identical to the sequence set forth in SEQ ID NO:67, wherein the polypeptide sequence comprises mutations at F316S, M318V, I319K, V323I, I334V, E360K, P363V, E366D, E376Q, and H378R, and wherein the polypeptide sequence has reduced binding to a therapeutic anti-KIT antibody.

[0023] Also provided herein is a polypeptide sequence comprising a polypeptide sequence that is at least 80% identical to the sequence set forth in SEQ ID NO:68, wherein the polypeptide sequence comprises a mutation at H378R, and the polypeptide sequence has reduced binding to a therapeutic anti-KIT antibody.

[0024] Also provided herein are nucleic acids encoding any of the above-mentioned polypeptides.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. Methods and materials for use in the present invention are described herein. Other suitable methods and materials known in the art can also be used. The materials, methods, and examples are merely illustrative and are not intended to be limiting. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

[0026] Other features and advantages of the invention will be apparent from the following detailed description and drawings, and from the claims. [Brief description of the drawings]

[0027] [Figure 1] Diagram of the structure of cytokine receptors encoded by genes such as FLT3 or KIT. Figure 1 (left) illustrates an epitope engineering approach to abolish binding of therapeutic antibodies by introducing mutations in the extracellular domain. Figure 1 (right) shows the main features of epitope engineered surface proteins: loss of antibody recognition with maintenance of ligand affinity, protein function, and intracellular signaling. [Figure 2-1]Figure 2 (left) is a schematic of Sleeping Beauty transposase experiments including transfer vector design to introduce cDNA sequences of desired receptor variants into human or mouse cell lines to assess recognition of such variants by different antibody clones or to measure ligand affinity by flow cytometry. K562 cells were electroporated with 100 ng of transfer plasmid and 500 ng of pSB100x transposase using a Lonza 4D-Nucleofector. Figure 2 (right) shows flow cytometry plots of K562 cells transduced with human wild-type FLT3 (hFLT3), mouse FLT3 (mFLT3), and the epitope-modified variant eFLT3-01 with amino acid substitutions within extracellular domain 4 by Sleeping Beauty transposase. Transduced cells are identified by mCherry fluorescence (y-axis). Cells were stained with either the FLT3 BV10A4 clone (control antibody, binds to ECD2), the FLT3 4G8 therapeutic clone (binds to ECD4), an anti-mouse FLT3 antibody, and AF488-conjugated human FLT3L to assess binding affinity. [Figure 2-2] As mentioned above. [Figure 2-3] As mentioned above. [Diagram 3] Figure 3 (top) is a sequence alignment of human FLT3 extracellular domain 4 with orthologous genes of several animal species. The black rectangles highlight the less conserved residues that were mutated in eFLT3-01 and included in the combinatorial library shown in Figure 5. Figure 3 (bottom) is a sequence alignment of human KIT extracellular domain 4 with orthologous genes of several animal species. The black rectangles highlight the less conserved residues that were mutated in eKIT-01 and predicted to contribute to the epitope to which clone Fab79D binds. [Figure 4A]1 is a flow cytometry plot showing co-staining of Sleeping Beauty-transduced K562 cells with control (clone BV10A4) and therapeutic (clone 4G8) antibodies. Overexpression of the eFLT3-01 variant shows lack of recognition by the 4G8 clone, whereas wild-type FLT3 is bound by both therapeutic and control Abs. [Figure 4B] 1 is a flow cytometry plot showing fluorescent FLT3-ligand binding by Sleeping Beauty transduced K562 cells. Cells were incubated with AF488-conjugated human FLT3L and control antibody BV10A4, which recognizes ECD2. The MFI ratio between the control antibody and AF488-conjugated human FLT3L is reported in the plot. The eFLT3-01 variant shows ligand binding affinity comparable to wild-type FLT3. [Figure 4C] Western blot of Sleeping Beauty transduced K562 cell protein extracts. Cells were serum starved for 16 hours and then stimulated or not with FLT3L 100ng / mL and lysed to obtain protein extracts. K562 (untransduced), K562 overexpressing wild-type FLT3, and K562 overexpressing eFLT3-01 were probed with antibodies recognizing the FLT3 extracellular domain, phospho-FLT3 Y589-591, or actin (control). The eFLT3-01 variant, like wild-type FLT3, shows phosphorylation of the kinase domain in response to FLT3L stimulation. [Figure 5-1]Figure 5 (top left) shows the design of the FLT3 combinatorial library cloned into the Sleeping Beauty plasmid. Figure 5 (bottom left) shows flow cytometry plots of wild-type FLT3 or FLT3 combinatorial library transduced K562 cells stained with control antibody (clone BV10A4) and therapeutic antibody (clone 4G8). NGS sequencing of sorted single- and double-positive cells highlighted the presence of the N399D mutation only in the sorted single-positive samples. Figure 5 (right) is a validation of overexpression of the FLT3 N399D mutation in K562 cells via Sleeping Beauty transduction. Co-staining with BV10A4 and 4G8 clones shows a lack of binding by the latter, while ligand affinity is maintained as assessed with AF488-conjugated FLT3L. [Figure 5-2] As mentioned above. [Figure 6A] 1 is a flow cytometry plot showing co-staining of Sleeping Beauty-transduced NIH3T3 cells with control antibody (clone 104D2) and therapeutic antibody (clone Fab79D). Overexpression of the eKIT-01 variant shows a lack of recognition by the Fab79D clone, whereas wild-type KIT is bound by both therapeutic and control Abs. [Figure 6B] 1 is a flow cytometry plot showing fluorescent SCF binding by Sleeping Beauty transduced NIH3T3 cells. Cells were incubated with AF488-conjugated human SCF and control antibody 104D2. The MFI ratio between the control antibody and AF488-conjugated human SCF is reported in the plot. eKIT-01 variants show comparable ligand binding affinity to wild-type KIT. [Figure 6C]Western blot of Sleeping Beauty transduced NIH3T3 cell protein extracts. Cells were serum starved for 16 hours and then stimulated or not with SCF 100 ng / mL and lysed to obtain protein extracts. NIH3T3 (untransduced), NIH3T3 overexpressing wild type KIT, and NIH3T3 overexpressing eKIT-01 were probed with antibodies recognizing the KIT extracellular domain, phospho-KIT Y719, or actin (control). The eKIT-01 variants show phosphorylation of the kinase domain in response to SCF stimulation similar to wild type KIT. [Figure 6D-1] We report flow cytometry plots of HEK-293T cells transduced with wild-type KIT as well as several mutated KIT variants in the Sleeping Beauty system. Cells were stained with a control antibody (clone 104D2) and a therapeutic antibody (Fab79D) to identify the best candidate mutations to abolish Fab79D binding while preserving surface expression. [Figure 6D-2] As mentioned above. [Figure 7-1] Figure 7 (top left) is the experimental design for the KIT extracellular domain 4 degenerate library screening performed to map the epitope of clone Fab79D. Each amino acid residue within ECD4 was replaced with a fully degenerate codon (NNN) to allow any amino acid substitution and cloned into the Sleeping Beauty plasmid along with mTagBFP2 as a co-expressed marker for transduced cells. The library was electroporated into HEK-293T cells. Figure 7 (bottom left) is the gating strategy for FACS sorting of the KIT extracellular domain 4 degenerate library to isolate single positive cells (for KIT control antibody clone 104D2) for NGS sequencing. Figure 7 (right) is a heatmap matrix showing each amino acid position within the extracellular domain 4 library region as a column, and the amino acid substitutions that are enriched in FACS sorted and sequenced single positive vs. double positive cells. The most frequently mutated amino acid positions represent residues involved in Fab79D epitope recognition (reported at the top of the plot). [Figure 7-2] As mentioned above. [Figure 7-3] As mentioned above. [Figure 7-4] As mentioned above. [Figure 7-5] As mentioned above. [Figure 7-6] As mentioned above. [Figure 7-7] As mentioned above. [Figure 8-1] Experimental layout of targeted EF1-alpha promoter insertion upstream of FLT3 and IL3RA (CD123) reading frames mediated by CRISPR-Cas9 or Cas12a homology-directed repair. Since unmodified K562 cells do not express either gene, this experiment was performed to generate a reporter cell line that allows enhanced and rapid evaluation of editing results by flow cytometry. The same strategy has also been applied to the KIT gene to overexpress it from the endogenous locus. K562 was electroporated with RNP complexes together with dsDNA donor templates with homology arms for the FLT3 or IL3RA promoter regions. Cells expressing the desired genes were then sorted by FACS, single-cell cloned, and screened to identify the clones with the highest expression. [Figure 8-2] As mentioned above. [Figure 9] Mapping of the anti-CD123 clone 7G3 epitope on the N-terminal domain of CD123. Each amino acid position was substituted with either alanine or an evolutionarily conserved amino acid and screened for 7G3 binding. White boxes identify residues that do not affect 7G3 binding, whereas progressively darker grey boxes are associated with a 25-50-75% loss of clone 7G3 binding. Similarly, the bottom part of Figure 9 shows the binding site of IL3 on CD123. [Figure 10A]The locations of several sgRNAs relative to IL3RA exons 2 and 3 that were used in base editing screening experiments to identify gRNAs and mutations capable of abolishing clone 7G3 binding while preserving CD123 surface expression are shown. The type of Cas protein associated with each sgRNA is reported in the legend. [Figure 10B-1] Representative flow cytometry plots of K562 reporter cells expressing CD123 from the endogenous locus edited with different sgRNA and base editor pairs are shown (indicated above and to the left of each plot). Cells are stained with a control antibody, CD123 clone 9F5, and three different therapeutic antibodies, CD123 clones 7G3, 6H6, and S18016F. Doses of sgRNA and plasmid electroporated into cells are reported in the figures. [Figure 10B-2] As mentioned above. [Figure 11-1] Base editing screening experiments in CD123 reporter cells edited with several sgRNA-base editor pairs (reported on the left or top of the plot, respectively). Cells were electroporated with 360 pmol of sgRNA and 500 ng of base editor plasmids using a Lonza 4D-Nucleofector and assessed by flow cytometry after 72 hours. Cells are stained with a control antibody, CD123 clone 9F5, and three different therapeutic antibodies, CD123 clone 7G3. [Figure 11-2] As mentioned above. [Figure 11-3] As mentioned above. [Figure 11-4] As mentioned above. [Figure 11-5] As mentioned above. [Figure 11-6] As mentioned above. [Figure 11-7] As mentioned above. [Figure 12A]We report the design of lentiviral vectors encoding FLT3 4G8 CAR and a truncated EGFR transduction (LV) marker / safety switch. Third-generation LV vectors were generated and fresh PBMC-derived T cells were transduced with CD3-CD28 Dynabeads on days 2–3 of stimulation and cultured for 14 days in medium containing IL7 and IL15. For co-culture killing experiments, K562 cells expressing either no FLT3, wild-type FLT3, or an epitope-engineered variant of FLT3 were plated in 96-well plates at different effector:target ratios (ET) with 4G8-CAR T cells or untransduced (UT) T cells. [Figure 12B-1] Figure 12B (top row) is a flow cytometry plot of plated K562 cells showing expression of FLT3 by co-staining with control antibody (clone BV10A4) and therapeutic antibody (clone 4G8). Figure 12B (middle and bottom rows) are flow cytometry plots of live cells (LiveDead yellow - AnnexinV-) from a co-culture killing experiment 4 and 48 hours after plating. Effector cells are identified by CellTrace fluorescence or CD4 / CD8 expression, while target cells are CellTrace and CD4 / CD8 negative and FSC-A high. [Figure 12B-2] As mentioned above. [Figure 12C] 1 is a plot showing the percentage of viable target cells (LiveDead-AnnexinV-) at 4 hours in all test conditions. [Figure 12D] 1 is a plot showing CD107a surface staining on T lymphocytes at 4 hours as a marker of degranulation during co-culture with target cells. [Figure 12E] Plot reporting the MFI of CellTrace yellow after 48 hours of co-culture with target cells to assess T cell proliferation by dye dilution. [Figure 13A]Schematic showing the location of three gRNAs relative to FLT3 exon 9 for the CRISPR-Cas homology-directed repair (HDR) editing strategy. Two gRNAs are designed for use in combination with AsCas12a, and one gRNA is designed for use in combination with SpCas9 nuclease. Using single-stranded oligo-deoxynucleotides (ssODNs) as template donors, we report the design of four different variants. Black boxes within the ssODN sequence identify mutated codons that differ from the wild-type sequence (additional mutations are included to reduce the rate of Cas nuclease re-cutting after HDR-mediated repair). The reverse complement of each ssODN was also tested. [Figure 13B-1] 13A-13C are flow cytometry plots reporting the results of CRISPR-Cas HDR editing experiments performed in reporter K562 cells expressing FLT3 using the gRNAs and ssODNs reported in FIG. 13A. Cells were co-stained with control (clone BV10A4) and therapeutic (clone 4G8) antibodies. Filled boxes highlight edited cells. [Figure 13B-2] As mentioned above. [Figure 13B-3] As mentioned above. [Figure 14A] FIG. 1 is a schematic showing the location of two sgRNAs relative to FLT3 exon 9 and the N399 codon used in combination with an adenine base editor (ABE). [Figure 14B] The sequence of each gRNA is reported in relation to the N399 codon and the PAM sequence. [Figure 14C] FIG. 1 is a schematic diagram of three adenine base editor variants with mutated Cas9 to enable their use with alternative PAM sequences. [Figure 14D-1] 14A and 14B and the ABE reported in FIG. 14C. Cells were co-stained with a control antibody (clone BV10A4) and a therapeutic antibody (clone 4G8). Each plot reports the gating of edited cells and the percentage of edited cells. [Figure 14D-2] As mentioned above. [Figure 15A] Schematic diagram showing the location of five sgRNAs relative to FLT3 exon 9 and the N399 codon used in combination with an adenine base editor (ABE) to introduce the N399D mutation. The legend indicates the different PAM and SpCas9 requirements for each sgRNA. [Figure 15B] The sequence of each gRNA is reported in relation to the N399 codon and the PAM sequence. [Figure 15C-1] 15A and 15B are flow cytometry plots showing the results of base editing experiments in FLT3-expressing K562 reporter cells using the sgRNAs shown in FIG. 15A and FIG. 15B and two adenine base editors (NG-ABE8e and SpRY-ABE8e-V106W 3xNLS). Cells were co-stained with a control antibody (clone BV10A4) and a therapeutic antibody (clone 4G8). Each plot reports the gating of edited cells and the percentage of edited cells. [Figure 15C-2] As mentioned above. [Figure 16A] Flow cytometry plots of FLT3 reporter K562 cells overexpressing different FLT3 variants by Sleeping Beauty transduction and co-stained with anti-FLT3 control antibody (clone BV10A4) and AF488-conjugated human FLT3L. The slope of the double positive population is proportional to the FLT3L affinity (MFI ratio between FLT3 control antibody and FLT3L staining is reported in each plot). [Figure 16B] FIG. 11 is a histogram showing the distribution of MFI ratios between FLT3 control antibody staining and AF488-conjugated FLT3L staining compared to wild-type FLT3 (grey overlay). [Figure 17A] 1 shows the design of a custom cloned pmRNA plasmid to produce base editor mRNA for editing of human CD34+ cells. [Figure 17B] We describe the workflow of the in vitro transcription protocol used for the production of BE mRNA. [Figure 17C]Agilent Fragment Analyzer profile of in vitro transcribed SpRY-ABE8e-V106W mRNA. The x-axis is nucleotide size and the y-axis is relative fluorescence unit (RFU) signal. [Figure 18-1] Panel A is a flow cytometry plot showing the results of a base editing experiment in FLT3-expressing K562 reporter cells using FLT3-18-NRN sgRNA and IVT adenine base editor mRNA (SpRY-ABE8e-V106W 3xNLS) shown in Figure 15 (Panel B). Cells were co-stained with a control antibody (clone BV10A4) and a therapeutic antibody (clone 4G8). Each plot reports the gating of edited cells and the percentage of edited cells. Experiments were performed with different mRNA and sgRNA doses (reported at the top and left of the plots, respectively). Figure 18B is a dot plot showing the dose-effect correlation between mRNA, gRNA and editing efficiency by flow cytometry. Figure 18 (Panel C) is a dot plot showing the dose-effect correlation between mRNA, gRNA and cell viability after editing. [Figure 18-2] As mentioned above. [Figure 18-3] As mentioned above. [Figure 19A] Heatmap reporting the editing efficiency at each denine base position of the FLT3-18-NRN sgRNA at different mRNA and sgRNA doses (reported on the left) performed in FLT3 reporter K562 cells (same as FIG. 18). The gRNA sequence and predicted editing window are reported on top of the heatmap. [Figure 19B] FIG. 11 is a dot plot showing the dose-effect correlation between mRNA, gRNA and editing efficiency by gDNA sequencing (only editing at the A6 and A7 base positions is reported). [Figure 20A] Experimental design for FLT3 base editing experiments in human mobilized peripheral blood-derived CD34+ HSPCs using IVT mRNA and FLT3-18-NRN sgRNA. Media composition is reported on the right and a timeline reports the time points of flow cytometry and gDNA collection. [Figure 20B] We report a gating strategy for flow cytometric assessment of the stem cell phenotype of cultured CD34+ HSPCs. [Figure 20C] 13 is a bar plot reporting the fold expansion of cultured CD34+ HSPCs on days 0, 3, and 6 after editing. [Figure 20D] Bar plot showing the composition of cultured CD34+ HSPCs at days 3 and 6 after editing by flow cytometry (gating is reported in FIG. 20B). [Figure 21A] 20 is a heatmap reporting the editing efficiency at each adenine base position of the FLT3-18-NRN sgRNA at different mRNA and sgRNA doses performed in mobilized peripheral blood-derived CD34+ HSPCs (same as FIG. 20). [Figure 21B] FIG. 13 is a dot plot showing dose-effect correlation between mRNA, gRNA and editing efficiency by gDNA sequencing in CD34+ HSPCs. [Figure 22A-1] Flow cytometry plots showing a CAR-T co-culture killing assay in which target cells were either unmodified or base-edited FLT3 expressing K562 cells. Target cells were plated with 4G8-CAR T cells or untransduced T cells at different effector:target ratios (reported above) and assessed at 6 hours by flow cytometry. Live cells (AnnexinV-LiveDead yellow-) are plotted and relative % is reported. [Figure 22A-2] As mentioned above. [Figure 22B] Target cell viability (AnnexinV-LiveDead yellow-) at each E:T ratio after 6 h of co-culture is reported. [Figure 22C] T cell degranulation by CD107a surface staining at each E:T ratio is reported 6 h after coculture. [Figure 23A]Flow cytometry plots showing a CAR-T co-culture killing assay in which target cells were either unmodified or base-edited human CD34+ HSPCs (editing efficiency 46%). Target cells were plated with 4G8-CAR T cells or untransduced T cells at different effector:target ratios (reported above) and assessed at 48 hours by flow cytometry. [Figure 23B] We report specific killing of CD34+ cells by 4G8 CAR-T cells at 48 hours. [Figure 23C] We report specific killing of CD34+CD90+ stem cell enriched subsets by 4G8 CAR-T cells at 48 hours. [Figure 24A] We report the experimental design of a pilot in vivo experiment to evaluate the resistance of FLT3 epitope-engineered CD34+ HSPCs to 4G8 CAR-T cells. The experimental timeline and procedures, as well as the number of treatment groups, are reported. [Figure 24B] 1 is a bar plot reporting the relative abundance and absolute numbers of human CD45+ engraftments in bone marrow (BM) at the time of sacrifice for each treatment group. [Figure 24C] FIG. 11 is a bar plot illustrating the lineage composition of human engraftments derived from xenotransplanted CD34+ HSPCs with either unmodified or base-edited FLT3 genes. [Figure 25A] Bar plots reporting the relative abundance and absolute numbers of human CD34+CD38- progenitor cells within human CD45+ engraftments in bone marrow at the time of sacrifice (same experiment as in Figure 24). Mice xenografted with unmodified CD34+ HSPCs show a significant reduction in CD34+CD38- progenitor cells upon 4G8 CAR-T administration. [Figure 25B] Bar plot reporting the percentage of CD69+ cells (T cell activation marker) on CAR-T cells in the spleens of treated mice (same experiment as in Figure 24). [Figure 25C]Bar plots reporting T cell phenotype (assessed by surface expression of CD62L and CD45RA) divided into central memory (CD45RA-CD62L+, CM), effector memory (CD45RA-CD62L-, EM), naive (CD45RA+CD62L+) and effector memory cells (CD45RA+CD62L-, TEMRA) that re-express CD45RA. [Figure 25D] FIG. 26 shows the gating strategy for identifying lineage-negative human progenitor cells (viable / hCD45+ / CD3- / CD19- / CD33- / CD34+ / CD38-) in flow cytometry of bone marrow samples (same as in FIG. 24) from in vivo xenotransplantation experiments. [Figure 25E] Cumulative flow cytometry plots obtained from pooled events of mice under the same conditions (same experiment as in FIG. 24) are reported showing the percentage of human CD34+CD38- progenitor cells in the bone marrow at the time of sacrifice. [Figure 26A] Experimental design for FLT3 and CD123 dual base editing experiments in human mobilized peripheral blood-derived CD34+ HSPCs using IVT mRNA and FLT3-18-NRN and CD123-N sgRNAs. Media composition is reported on the right and a timeline reports the time points of flow cytometry and gDNA collection. [Figure 26B] 13 is a bar plot reporting the fold expansion of cultured CD34+ HSPCs on days 0, 3, and 7 after editing. [Figure 26C] Flow cytometry plots of edited CD34+ HSPCs highlighting the loss of CD123 clone 7G3 recognition in the CD123 base editing condition. Cells were co-stained with a control CD123 antibody (clone 9F5) and a therapeutic antibody (7G3). Cells were pre-gated on CD34+CD90+CD45RA-. [Figure 27] Heatmap reporting the editing efficiency at each adenine base position of FLT3-18-NRN and CD123-N sgRNAs at different mRNA and sgRNA doses performed in mobilized peripheral blood-derived CD34+ HSPCs (same as FIG. 26). [Figure 28A]Flow cytometry plots of K562 cells base edited with FLT3-18-NRN sgRNA or CD123-N sgRNA in combination with either SpRY-ABE8e-V106W or SpRY-K918-ABE8e-V106W to show improved CD123 base editing efficiency. [Figure 28B] We report the sequences of three different gRNAs targeting the CD123 S59 residue. [Figure 28C] We present the design of several different adenine base editors, including modified SpCas9, including higher fidelity variants (HF1, Sniper), the K918N variant, the BlackJack variant (more tolerant of longer gRNAs), and combinations of these mutations. [Figure 28D] FIG. 28C is a heatmap reporting the editing efficiency using the three gRNAs from FIG. 28B (plus FLT3-18-NRN as a control) and the base editor variants from FIG. 28C. [Figure 29A] Experimental design for FLT3 and CD123 dual base editing experiments in human mobilized peripheral blood-derived CD34+ HSPCs using IVT mRNA and FLT3-18-NRN and CD123-R sgRNAs. Media composition is reported on the right and a timeline reports the time points of flow cytometry and gDNA collection. [Figure 29B] 13A-C are flow cytometry plots showing the gating strategy for CD123 base editing efficiency in cultured CD34+ HSPCs. [Figure 29C] 13 is a bar plot reporting the fold expansion of cultured CD34+ HSPCs on days 0, 3, and 6 after editing. [Figure 30-1] Heatmap reporting base editing efficiency at each adenine position of FLT3-18-NRN and CD123-R sgRNA (same experiment as in Figure 29). Targeted bases are reported at the bottom of the heatmap. On the right, the heatmap reports the percentage of CD123 bases edited by flow cytometry. [Figure 30-2] As mentioned above. [Diagram 31] FIG. 29 is a heatmap reporting base editing efficiency on CD34+ HSPCs at each adenine position of KIT-gRNA-Y targeted residue H378 from the same experiment as FIG. [Figure 32A] We report the experimental design for an in vivo xenograft experiment to confirm tolerance of FLT3 epitope engineered human CD34+ HSPCs to 4G8 CAR-T cells. We report the experimental timeline and treatment groups and numbers. [Figure 32B] We report FLT3 base editing efficiencies in peripheral blood samples 8 weeks post-transplant for all mice. [Figure 33A] The percentage of polymorphonuclear granulocytes (CD3-CD19-CD33+SSCl, PMN) and granulomonocytic progenitors (lineage-CD34+CD38+CD45RA+FLT3+, GMP) in human CD45+ engraftment in bone marrow at the time of sacrifice is reported (same experiment as in Figure 32). [Figure 33B] Shows FLT3 editing efficiency on genomic DNA at several time points for 4G8 CAR-T cell treated and untreated mice. LC: liquid culture (pre-transplant samples), W8, W12: 8 and 12 week bleeding, BM: bone marrow, CFU: colony forming unit derived cells (plated from bone marrow samples at time of sacrifice). [Figure 34-1] Representative flow cytometry plots showing the relative abundance of polymorphonuclear granulocytes (CD3-CD19-CD33+SSCl, PMN), granulomonocyte progenitors (lineage-CD34+CD38+CD45RA+FLT3+, GMP), and hematopoietic stem cells (lineage-CD34+CD38-CD90+CD45RA-, HSC) in the bone marrow of mice xenografted with FLT3 or AAVS1-edited CD34+ HSPC. [Figure 34-2] As mentioned above. [Diagram 35] Several alternative designs for single or bispecific chimeric antigen receptors (CARs) are shown, expressed either as two separate constructs or as tandem CARs (two scFvs on the same molecule separated by a linker). [Figure 36A]Schematic diagram of type III receptor tyrosine kinase (FLT3, KIT) with extracellular domains (ECDs) recognized by control or therapeutic (magenta) monoclonal antibodies. Binding affinity of mutant receptors was assessed using AF488-conjugated FLT3L or SCF ligands. [Figure 36B] Flow cytometry plots (top) showing loss of therapeutic mAbs against chimeric receptors with 16 or 10 amino acid substitutions (4G8 and Fab79D for FLT3 and KIT, respectively) and fluorescent ligand binding assays (bottom) for wild-type (WT) and mutated receptor variants. [Figure 36C] Western blot of pFLT3 Y589-591 and pKIT Y719 in cell lines expressing FLT3 and KIT variants unstimulated or incubated with 100 ng / mL FLT3L or SCF. [Figure 36D] Schematic diagram of the Sleeping Beauty plasmid containing the FLT3 cDNA with either human or mouse codons at the 16 amino acid position of ECD4 (top left), as well as K562 cells transduced with the FLT3 library and FACS sorted (4G8- and 4G8+) for NGS sequencing of the library region (top right), and sequence logo showing the relative amino acid frequency at positions 384-413 (bottom). [Figure 36E] Schematic diagram of the Sleeping Beauty plasmid (top left) containing a degenerate codon (NNN) at each position in ECD4; K562 cells transduced with the KIT library and FACS sorted (Fab79D- and Fab79D+) for NGS sequencing of the library regions (top right); as well as a sequence logo showing the log fold change of amino acid substitutions enriched in Fab-79D low positive cells versus double positive cells (aa.314 to 381) (bottom; positions with multiple enriched amino acid substitutions coincide with previously predicted contact points). [Figure 36F-1]Schematic diagram of tested gRNAs targeting FLT3 codon N399 (left) and representative plots of K562 FLT3 reporter cells electroporated with base editor expression plasmids (NG-ABE8e or SpRY-ABE8e) and sgRNAs and assessed by flow cytometry 72 hours after editing (right; the percentage of cells positive for control mAb BV10A4 and negative for clone 4G8 is reported in each plot; unedited condition indicates gating strategy). [Figure 36F-2] As mentioned above. [Figure 36G] Screening of CD123 epitopes by base editing. Top: sgRNAs for targeted base editing of 7G3 contact residues used in screening. Dark blue: NGG PAM, grey: NGN PAM, light blue: NRN PAM. PAM positions are indicated by arrowheads. Bottom: FACS plots of CD123 reporter K562 cells treated with SpRY-cytidine base editor (top) or adenine base editor (bottom). The % of cells positive for control mAb 9F5 and negative for therapeutic clone 7G3 are reported at the bottom right of each plot. [Fig. 36H] Affinity of therapeutic antibodies against stealth receptor variants measured on K562 (for FLT3 and CD123) or NIH-3T3 (for KIT) cells expressing the receptor variants. Mean ± SD (N=3) MFI of therapeutic mAbs normalized to control mAb after background subtraction. Comparison by two-way ANOVA. [Figure 36I] Affinity assays for FLT3, SCF, and IL3. Cell lines expressing Sleeping Beauty transposase-mediated receptor variants were incubated with fluorescent ligands and assessed by flow cytometry. Mean MFI of ligands ± SD (N=3 FLT3 / CD123, N=4 KIT). Comparisons by two-way ANOVA. [Figure 37A] Full-length sequence logo of the FLT3 EC4 combinatorial library showing the amino acid frequency at each position (357-421) of ECD4 in FACS-sorted 4G8- and 4G8+ cells. [Figure 37B] Design of Sleeping Beauty transposons encoding FLT3 variants with mCherry and puromycin N-acetyltransferase (PAC) reporter / resistance cassettes (top) and flow cytometry plots showing loss of 4G8 recognition for the N399D and N399G variants expressed in K562 cells (bottom). [Figure 37C] Generation of FLT3, KIT and CD123 reporter K562 cells via targeted homology-directed repair-mediated integration of the EF1α promoter upstream of the gene transcription start site (TSS) is shown. dsDNA donors with 50 bp long homology arms were generated by PCR on a plasmid template encoding the complete EF1α promoter. K562 cells were electroporated with SpCas9 (FLT3, KIT) or AsCas12a nuclease (CD123) and gRNAs recognizing the upstream region of the coding sequence of each gene. 0.5-10ug of dsDNA donor template was co-electroporated with Cas RNP in an electroporation volume of 20μL. Representative flow cytometry plots show populations of cells positive for overexpressed genes that were sorted and expanded by FACS. For FLT3 and CD123, single cell cloning of sorted cells was performed to isolate clones with the highest surface expression. Unless otherwise stated, all epitope editing tests and optimizations were performed in K562 reporter cells. FLT3 / CD123 dual reporters were obtained by a second round of CD123-targeted RNP+ donor electroporation in FLT3-expressing K562 cells (data shown in Figure 49B). [Figure 37D]Introduction of the FLT3 N399D mutation via CRISPR-Cas-mediated homology-directed repair is shown. K562 reporter cells were electroporated with SpCas9 or AsCas12a nuclease, gRNA, and a 200 bp ssODN template donor encoding the N399D mutation (or their reverse complements, rev.comp.). An additional silent mutation was included to reduce the risk of nuclease re-cutting after HDR repair. 72 hours after editing, cells were assessed by flow cytometry. The percentage of FLT3+ cells (with control mAb BV10A4) (but 4G8-) is reported in the lower right corner. [Figure 37E] Characterization of KIT mutations derived from epitope mapping. For amino acid positions resulting from KIT epitope mapping, substitutions that can be obtained with adenine BE (ABE, red) or cytidine BE (CBE, blue) were cloned individually in the Sleeping Beauty transposon and electroporated into HEK-293T cells. After selection with puromycin, cells were stained with both Fab-79D and the control Ab 104D2. The ratio of Fab-79D MFI to 104D2 MFI is reported for each mutation. To exclude variants affecting SCF binding to KIT, the same variants were incubated with AF488-conjugated SCF and with the control mAb 104D2 (which does not impair SCF binding). The ratio of SCF to 104D2 median fluorescence is reported in a bar plot. The horizontal line indicates the reference mutation H378R. [Figure 37F] Optimization of KIT H378R adenine base editing. sgRNA targeting codon H378 in exon 7 was co-electroporated with SpRY-ABE8e into K562 cells. Editing efficiency on gDNA is reported for each adenine in the protospacer (including position number relative to the KIT-Y sgRNA). [Figure 37G-1]Design of Sleeping Beauty transposons encoding KIT variants with mTagBFP2 and a puromycin N-acetyltransferase (PAC) reporter / resistance cassette (top) and flow cytometry plots showing loss of Fab79D recognition for KIT H378R expressed in HEK-293T cells (bottom). [Figure 37G-2] As mentioned above. [Fig. 37H-1] Screening of CD123 epitopes by base editing. sgRNAs for targeted base editing of 7G3 contact residues reported in Figure 1G were co-electroporated with adenine base editor (ABE) or cytidine base editor (CBE) expression plasmids into CD123 reporter K562 cells. The percentage of cells positive for the control mAb 9F5 and negative for the therapeutic clone 7G3 is reported in each plot. The unedited condition indicates the gating strategy. BE4, evoApobec-1 BE4max [Fig. 37H-2] As mentioned above. [Fig. 37I] Figure 37 shows that CD123 CBE with sgRNA-F results in loss of binding of clones 6H6 and S18016F. The same conditions from the BE screen reported in Figure 37H were stained with CD123-targeting clones 6H6 and S18016F, which have different epitopes than clone 7G3. [Figure 37J] 1 shows the design of Sleeping Beauty transposons encoding CD123 variants with co-expression of the common beta chain CSFR2B to enable intracellular signaling. [Figure 38A]Figure 2 shows that FLT3 epitope engineered variants maintain kinase activation. Western blot of protein extracts from K562 cells expressing FLT3 variants with Sleeping Beauty transposase. Cells were serum starved overnight and stimulated with different concentrations of human FLT3L for 10 min at 37°C. pFLT3 Y589-591, total FLT3, and actin were probed on the same lysates. Total FLT3 was probed after pFLT3 membrane stripping. Normalized pFLT3 signal intensity (above actin) is reported on the right. Comparison by two-way ANOVA. [Figure 38B] Figure 1 shows that KIT epitope engineered variants maintain kinase activation. Western blot of protein extracts from NIH-3T3 cells expressing KIT variants with Sleeping Beauty transposase. Cells were serum starved overnight and stimulated with different concentrations of human SCF for 10 min at 37°C. pFLT3 Y719, total KIT, and actin were probed on the same lysates. Normalized pKIT signal intensity (relative to total KIT) is reported in the right plot. Comparison by two-way ANOVA. [Figure 38C] We show that CD123 epitope engineered variants maintain STAT5 activation. BaF3 cells expressing Sleeping Beauty transposase CD123 variants were starved of mouse IL-3 and stimulated with different concentrations of human IL-3. After 48 hours, cells were assessed for STAT5 phosphorylation by intracellular flow cytometry (left, representative FACS plots showing CD123 S59P conditions at different hIL-3 doses; right, pSTAT5 PE MFI). [Figure 38D]We show that epitope engineered variants of FLT3, KIT, and CD123 induce proliferative responses similar to the WT receptor. BaF3 cells expressing FLT3, KIT, and CD123 variants by Sleeping Beauty transposase were starved for mouse IL-3 overnight and stimulated with different concentrations of human FLT3, SCF, and IL-3, respectively. Cells were cultured for 5 days and analyzed by flow cytometry to obtain absolute counts (CountBeads). Plots report absolute counts normalized to unstimulated conditions. [Figure 39A] A schematic diagram of the bidirectional lentiviral vector (LV) co-expressing a second-generation CAR and a truncated human epidermal growth factor receptor (EGFRt) (top) and a schematic diagram of T cell culture, transduction, and analysis for the generation of CAR-T cells (bottom) are shown. [Figure 39B] The percentage of EGFRt surface expression on T cells (left) and fold expansion (right) are shown at the indicated days (D) after transduction with LV 4G8-CAR at different multiplicities of infection (MOI). [Figure 39C] Figure 2 shows that FLT3 N399D or N399G avoid 4G8 CAR killing. K562 cells expressing FLT3 variants by Sleeping Beauty transposase were cultured with 4G8 CAR-T cells at different effector:target ratios (E:T). Left: Representative flow cytometry plots of K562 cells expressing either FLT3 WT or N399D after 48 h of coculture with 4G8 CAR. T and target cells are identified by CD3 and FLT3 staining, respectively. From left to right: plots reporting i) percentage of persisting live target cells (absolute number of AnnexinV-7AAD- cells) relative to E:T=0, ii) T cell activation by CD69 staining (%), and iii) surface expression of FLT3 on remaining live target cells by BV10A4 staining normalized at E:T=0. Mean ± SD (N=4). Comparison by two-way ANOVA. [Figure 39D]KIT H378R avoids 79D CAR killing. Left: Representative flow cytometry plots of K562 cells expressing either KIT WT or H378R with Sleeping Beauty transposase after 48 h co-culture with 79D CAR. T cells and target cells are identified by CD3 and KIT staining, respectively. From left to right: plots reporting i) percentage of persisting live target cells (absolute number of AnnexinV-7AAD- cells) relative to E:T=0, ii) T cell activation by CD69 staining (%), and iii) surface expression of KIT on remaining live target cells by 104D2 staining normalized to E:T=0. K562: untransduced. Mean ± SD (N=4). Comparison by two-way ANOVA. [Figure 39E] CD123S59 BE cells are resistant to CSL362 CAR. Left: Representative flow cytometry plots of CD123-reporter K562 base edited or unmodified after 48 h co-culture with CSL362 CAR. T cells are identified by CD3 / 4 / 8 staining. From left to right, E: plots reporting the percentage of persisting live target cells (absolute number of AnnexinV-7AAD- cells) relative to T=0, ii) T cell activation by CD69 staining (%), and iii) surface expression of CD123 on remaining live target cells by 9F5 staining. Mean ± SD (N=4). Comparison by two-way ANOVA. [Figure 40A] CD4 / CD8 composition of CAR-T cells during in vitro culture. Fresh healthy donor-derived PBMCs were cultured with CD3 / CD28 Dynabeads (beads:cell ratio = 3:1), IL-7 5ng / mL and IL-15 5ng / mL and transduced with lentivirus on day 2 (D) containing a vector (LV) encoding the 4G8 CAR. Culture composition was assessed by flow cytometry on days 2, 4, 6 and 12. Plots report N=5 conditions transduced with LV at different multiplicities of infection (MOI). Mean ± SD. [Figure 40B]CAR-T cell phenotype by flow cytometry. T cell subsets were assessed by CD62L and CD45RA staining (CD45RA+62L+, naive / T stem cells; CD45RA-62L+, central memory, CM; CD45RA-62L-, effector memory, EM; CD45RA+62L-, terminally differentiated EM cells; CD45RA, EMRA re-expression). Representative FACS plots (left) and culture composition according to CD4+ and CD8+ subsets (right) are reported. D0 refers to uncultured peripheral blood T cells after Ficoll separation. Mean ± SD (N=5). [Figure 40C] Figure 1 shows that FLT3WT cells are eliminated by 4G8 CAR-T cells while FLT3N399 BE cells are protected. Co-culture assay of 4G8 CAR-T cells with FLT3 reporter K562 cells with either unmodified or base-edited FLT3N399. (Left) Representative flow cytometry plots at early time point (6 hours) gated on live cells (AnnexinV-7AAD-). T cells are identified by CellTrace marking, while K562 targets are FLT3+. (Left to right) Target cell viability at 6 hours (%), T cell degranulation by CD107a surface staining at 6 hours (%), and FLT3 expression in viable target cells at 48 hours (MFI, normalized to E:T=0). N=2. Comparison by two-way ANOVA. [Figure 40D-1] Figure 3 shows that epitope engineered receptors provide protection from CAR-T cells. Each row reports additional plots from co-culture with K562 cells expressing FLT3, CD123, and KIT (same experiment reported in Figure 39C,D,E). From left to right per column: CellTrace MFI of CAR-T cells at 48 h of co-culture, % target cell viability after 48 h of co-culture with CAR-T cells, % target cell viability after 48 h of co-culture with untransduced. T cells (%), absolute number of live cells after 48 h of co-culture with untransduced T cells (AnnexinV-7AAD-, E: normalized to T=0), % CD69+ untransduced T cells after 48 h of co-culture, FLT3, CD123 and KIT MFI (non-normalized) on target cells after 48 h of co-culture with untransduced T cells. Mean ± SD (N=4). [Figure 40D-2] As mentioned above. [Figure 40D-3] As mentioned above. [Figure 40E-1] The experimental layout of the co-culture assay with two target cell populations, one expressing FLT3 and the other expressing CD123, is shown. Unmodified or epitope-edited FLT3 and CD123 K562 reporter cells were mixed at approximately 1:1 ratio and co-cultured with either 4G8 CAR, CSL362 CAR or untransduced T cells. The FLT3+ / CD123+% composition of live target cells (pre-gated on FLT3+ or CD123+) is reported as a bar plot for each combination at different effector:target (E:T) ratios. Mean ± SD (N=4). [Figure 40E-2] As mentioned above. [Figure 41A] Schematic diagram of CD34+ HSPC culture, base editing, and analysis. mPB: mobilized peripheral blood, HSPC: hematopoietic stem and progenitor cells, FLT3L: FLT3 ligand, SCF: stem cell factor, TPO: thrombopoietin, SR-1: StemRegenin. [Figure 41B] Representative plots reporting the editing window and editing efficiency of CD34+ HSPCs at each adenine within the gRNA sequences of FLT3, KIT, and CD123 measured after electroporation with different doses of adenine BE mRNA. Mean ± SD. [Figure 41C] Editing efficiency in bulk CD34+ cells or FACS-sorted stem-enriched CD90+ or ​​stem-depleted CD90- subsets is shown. Mean ± SD (N=2 donors). [Figure 41D] Immunophenotype of epitope-edited CD34+ HSPCs. Left: Representative flow cytometry plots showing CD90 / CD45RA subsets within CD34+133+ HSPCs in in vitro culture on day 5 after base editing. Right: Bar plots showing CD90 / CD45RA subset composition of FLT3N399, KITH378R, and CD123S59 epitope-edited CD34+ cells. Mean ± SD. Sample sizes are reported within the bar graphs. [Figure 41E] In vitro 4G8 CAR killing assay on FLT3N399 epitope-edited HSPCs. Persistent viable cell fraction (left) and viability (right) of CD34+45RA+ cells by AnnexinV-7AAD staining, either FLT3 or AAVS1 base edited, 48 hours after co-culture with 4G8 CAR or untransduced T cells. Mean ± SD (N=4). Comparison by two-way ANOVA. [Fig.41F] In vitro CSL362 CAR killing assay on CD123S59 epitope-edited HSPCs. Persistent live CD90+ cell fraction (left) and CD34+ cell viability (right) by AnnexinV-7AAD staining of CD123S59 or AAVS1-BE HSPCs after 48 hours of co-culture with CSL362 CAR or untransduced T cells. N = 8 from two different HSPC donors. Comparison by two-way ANOVA. [Figure 41G] Percentage of absolute live CD34+ cells versus no mAb control for KITH378 BE or AAVS1 BE HSPCs plated with increasing concentrations of Fab-79D mAb and cultured for 48 hours. Mean ± SD. N = 6 from two different HSPC donors. Comparison by two-way ANOVA. [Fig. 41H] Absolute numbers of total CD34+ and CD90+45RA- cells (top left) and myeloid and erythroid colonies (N=2, bottom left) of base-edited HSPCs are shown. Right: Representative image of a CFU plate (one of two replicates) 14 days after plating. [Fig.41I] FIG. 1 is a schematic diagram of primary and secondary xenografts and analysis of FLT3N399 or AAVS1 BE HSPCs. [Figure 41J] Human engraftment (hCD45+ cells) by flow cytometry at different time points in primary recipients (left) and BM lineage composition as a percentage of total human (hCD45+) cells by flow cytometry (right). PMN: polymorphonuclear granulocytes, mono: monocytes, lin-: lineage negative. Mean ± SD. FLT3N399 N=7, AAVS1 BE N=4. Comparison by two-way ANOVA. [Figure 41K] FLT3 editing levels measured in total circulating cells, hematopoietic organs, or CFUs at different time points in primary and secondary transplanted mice are shown. Mean ± SD. LC: liquid culture, W8: week 8, W12: week 12, SP: spleen, BM: bone marrow, CFU: colony forming units. [Figure 42A] Schematic diagram of the plasmid templates used for in vitro transcription (IVT) of base editor mRNA. The templates were linearized using the IIS restriction enzyme BbsI. T7, T7 RNA polymerase promoter; UTR, untranslated region; HBB, hemoglobin β gene; polyA, polyadenine sequence (approximately 110-120 nt). [Figure 42B] Representative plot of IVT SpRY-ABE8e mRNA analyzed on an Agilent Fragment Analyzer RNA for quality control. >90% of the IVT mRNA corresponds to the predicted size. [Figure 42C] SpRY-ABE8e V106W mRNA dose-finding study on FLT3 reporter K562 cells base-edited for FLT3N399 with sgRNA-18. Right: Correlation of FLT3 editing efficiency with mRNA x sgRNA dose by flow cytometry and correlation of FLT3 editing efficiency by flow cytometry and gDNA analysis. Spearman r2 and p-values ​​are reported. [Fig.42D]Optimization of CD34+ HSPC base-edited mRNA is shown. Several SpRY-ABE8e mRNA variants were examined in dual FLT3 / CD123 editing experiments. Variables tested included mRNA purification method (beads, sparQ PureMag magnetic beads; columns, NEB Monarch RNA columns), dephosphorylation, replacement of UTP with N1-methyl-pseudo-uridine (N1m-U) or 5-methoxy-uridine (5me- U), capping technology (CleanCap, Trilink CleanCap AG; ARCA, NEB 3'-O-Me-m7G(5')ppp(5')G RNA cap structure analog), and the K918N SpCas9 mutation (a reported method to improve nuclease efficiency). 64 Bar plots showing FLT3 and CD123 editing efficiency by genomic DNA (gDNA) analysis (%) and absolute numbers of bulk (CD34+) and stem-enriched (CD90+45RA-) cells at the end of in vitro culture. [Figure 42E] Optimization of culture conditions for base editing. CD34+ HSPCs were base edited with SpRY-ABE8e mRNA and FLT3N399 sgRNA. Different culture conditions were used, including the addition of supplements during electroporation (RNAsin, Promega RNAsin RNAse inhibitor; glycerol) or adjustment of cytokine concentrations (standard: 100 ng / mL FLT3L, SCF, and 50 ng / mL TPO; 1.5x: 150 ng / mL FLT3L, SCF, and 75 ng / mL TPO; 3x: 300 ng / mL FLT3L, SCF, and 150 ng / mL TPO; +IL-3: standard supplemented with hIL-3 20 ng / mL), different stem cell preserving compounds (standard: SR-1 0.75 μM, UM171 35 nM; SR-1 only 0.75 μM; UM171 only 35 nM; no SR-1 / UM171), and anti-inflammatory compounds (PGE2, prostaglandin E2 10 μM; DEX, dexamethasone 1 μM). Bar plot showing FLT3 editing efficiency (%) by gDNA analysis and absolute numbers of bulk (CD34+) and stem-enriched (CD90+45RA-) cells at the end of in vitro culture. [Fig.42F]Bar plots. CD34+ HSPCs were electroporated at different time points (24h, 48h, 72h) after thawing. Each condition was edited for all combinations of two of the selected targets (FLT3, CD123, KIT). Bar plots showing editing efficiency by gDNA analysis (%) and absolute numbers of bulk (CD34+) and stem-enriched (CD90+45RA-) cells at the end of in vitro culture. [Fig.42G] Representative flow cytometry plots showing the gating strategy used for the analysis of edited CD34+ HSPCs. From left to right, cells were gated for singlets (FSC-H / FSC-A plot), viability (PI / FSC-A plot; PI, propidium iodide), physical parameters (SSC-A / FSC-A plot), CD34+ (CD34 / CD90 plot), CD133+ (CD34 / CD133 plot), and CD45RA-90+, CD45RA-90-, CD45RA+90- (CD45RA / 90 plot). After pre-gating on singlets (S), the beads (B) gate identifies CountBeads (FSC-A low PI high). This is further gated on two additional fluorescent parameters to exclude debris (not shown). [Fig. 42H] We show that epitope-edited HSPCs retain their proliferative response to cytokine stimulation. FLT3-, KIT-, and CD123-edited CD34+ HSPCs were plated with different concentrations of the respective ligands and cultured for 4 days. Absolute counts were obtained by flow cytometry using CountBeads. [Fig.42I] Editing efficiency at the experimental endpoint is shown (technical triplicates were pooled together for gDNA analysis) (see FIG. 42H). Mean ± SD. N=4 from 2 healthy donors. Multiple comparisons by 2-way ANOVA. [Figure 43A]Human engraftment (%hCD45+) by flow cytometry in peripheral blood and bone marrow (BM) at endpoint week 12 (W12) in secondary recipient NBSGW mice xenografted with BM cells from the experiment shown in Figure 3I (each was primary transplanted into one secondary recipient). Mean ± SD. Comparison by two-way ANOVA. [Figure 43B] Absolute numbers of myeloid (left) and lymphoid (right) lineages in the BM of secondary xenografted mice are shown. AAVS1BE N=4, FLT3N399 N=7. Mean ± SD. Comparison by two-way ANOVA. HSC: hematopoietic stem cell, MPP: multipotent progenitor, LMPP: lymphoid-primed multipotent progenitor, CMP: common myeloid progenitor, GMP: granulocyte-monocyte progenitor, myeloblasts: defined as CD33 / 66b+19-14-11c-34-SSC low, mono: monocyte, iPMN: immature polymorphonuclear granulocyte, mature PMN: mature granulocyte. [Figure 43C] Human engraftment (hCD45+%) by flow cytometry in peripheral blood and BM of NBSGW xenotransplanted with 1M CD34+ HSPCs (either AAVS1 or KITH378 edited) at weeks 9 and 11 (W9, W11). Mean ± SD. Comparison by two-way ANOVA. [Fig. 43D] Absolute counts of myeloid (left) and lymphoid (right) lineages in the BM of mice from C. Comparisons by two-way ANOVA. AAVS1 N=5, KITH378 N=4. [Figure 43E] KIT editing efficiency measured on liquid culture (LC), total blood cells 9 weeks post-transplant (W9), and FACS-sorted B (CD19) and bone marrow (CD33) BM cells at the end of the experiment are shown. Mean ± SD. [Figure 43F]Schematic diagram of lentiviral vector encoding mNeonGreen fluorescent protein under the hPGK promoter used to transduce human PDX (left) and representative flow cytometry plots (right) showing transduction efficiency of patient-derived AML xenografts versus bone marrow (PDX-1) or spleen (PDX-2) samples. PDX cells were transduced ex vivo overnight, transplanted into NBSGW mice for expansion, FACS sorted for mNeonGreen+ cells, and injected into secondary recipients. [Figure 43G] Genetic signature (left) and surface immunophenotype (right) of AML PDXs used for in vivo experiments at thaw time are shown. Genetic alterations and % positive cells for each marker are reported in heatmaps. ITD, internal tandem duplication; TKD, tyrosine kinase domain mutation. [Figure 44A] Schematic of xenograft and analysis of FLT3N399 or AAVS1 BE HSPCs co-engrafted with AML PDX-1 and treated with 4G8 CAR-T cells. [Figure 44B] FLT3 base editing measured in liquid culture (LC), total blood cells (weeks 8 and 9), and sorted CD33+ and CD19+ bone marrow (BM) cells at endpoint in mice treated with 4G8 CAR-T cells or untreated. Multiple unpaired t-test. Mean ± SD. [Figure 44C] Representative flow cytometry plots of BM samples from mice engrafted with CD34+ HSPCs, CD34+ HSPCs + AML PDX-1, or CD34+ HSPCs + AML PDX-1 and treated with 4G8 CAR-T. Plots are pre-gated on human CD45+. CAR T cells were identified by CD3 staining and AML PDX cells are mNeonGreen+. Mean ± SD. Comparison by one-way ANOVA. [Fig.44D] Bar plot showing percentage of CD3+ cells in hCD45+mNeonGreen- BM cells. Mean ± SD. Comparison by one-way ANOVA. [Figure 44E]Bar plot showing the percentage of AML cells within hCD45+CD3- BM cells. Mean±SD. [Fig.44F] Bar plot showing the percent of AML (mNeonGreen+) cells as measured by flow cytometry relative to total BM-derived CFU. Mean ± SD. Statistical comparison of FLT3N399 vs. AAVS1 BE conditions by one-way ANOVA. [Figure 44G] Representative FACS plots showing depletion of BM B cells (CD19+) by 4G8 CAR-T in mice engrafted with FLT3N399 or AAVS1 BE HSPCs. Plots were pre-gated on hCD45+CD3-mNeonGreen-. [Fig.44H] Bar plots showing percentages of pre-B cells (CD19+10+34-) (FIG. 44H), pro-B cells (CD19+10+34+) (FIG. 44I) within hCD45+CD3-mNeonGreen-, and percentages of immature granulocytes (CD33 / 66b+14-10-11c-SSC high) (FIG. 44I) within myeloid cells (CD33 / 66b+). Mean ± SD. Statistical comparison of FLT3N399 vs. AAVS1 BE conditions by one-way ANOVA. [Fig.44I] Bar plots showing percentages of pre-B cells (CD19+10+34-) (FIG. 44H), pro-B cells (CD19+10+34+) (FIG. 44I) within hCD45+CD3-mNeonGreen-, and percentages of immature granulocytes (CD33 / 66b+14-10-11c-SSC high) (FIG. 44I) within myeloid cells (CD33 / 66b+). Mean ± SD. Statistical comparison of FLT3N399 vs. AAVS1 BE conditions by one-way ANOVA. [Fig.44J] Bar plots showing percentages of pre-B cells (CD19+10+34-) (FIG. 44H), pro-B cells (CD19+10+34+) (FIG. 44I) within hCD45+CD3-mNeonGreen-, and percentages of immature granulocytes (CD33 / 66b+14-10-11c-SSC high) (FIG. 44I) within myeloid cells (CD33 / 66b+). Mean ± SD. Statistical comparison of FLT3N399 vs. AAVS1 BE conditions by one-way ANOVA. [Figure 44K] Representative FACS plots showing the composition of lineage-CD34+38+10- progenitors. Granulocyte progenitors (GMPs) are defined as CD45RA+FLT3+, common myeloid progenitors (CMPs) as CD45RA-FLT3+, and megaerythroid progenitors (MEPs) as CD45RA-FLT3-. [Figure 44L] Bar blot showing % GMP in lin-CD34+38+, mean ± SD. [Figure 44M] Absolute numbers of myeloid subsets in BM, from HSC to differentiated leukocytes, are shown. Untreated mice were pooled together (grey bars). 4G8-treated FLT3N399 and AAVS1 BE mice are reported in pink and yellow, respectively. Mean ± SD. Fold change in absolute numbers (FLT3N399 / AAVS1) for CAR-treated groups is reported above the bar plots for each population. One-way ANOVA with multiple comparisons. FDR-adjusted p-values ​​for comparisons between FLT3N399 vs. AAVS1 BE conditions treated with 4G8 CAR are reported (p<0.05 in bold). [Figure 44N] Representative FACS plots showing the composition of lineage-CD34+38-10- progenitor cells. Stem cells (HSCs) are defined as CD45RA-90+, multipotent progenitor cells (MPPs) are defined as CD45RA-90-, and lymphoid-primed MPPs (LMPPs) are defined as CD45RA+90-. [Fig.44O] Bar plot showing LMPP (%) in lin-CD34+38-. Mean±SD. [Figure 44P]Absolute numbers of lymphoid lineage subsets in BM, from HSC to differentiated leukocytes, are shown. Pre-B / NK are defined as CD33 / 66b-19-56-34+38+10+, pro-B lymphocytes are defined as CD33 / 66b-19-56-34-10+, pro-B cells are defined as CD33 / 66b-19+10+34+, pre-B cells are defined as CD33 / 66b-19+10+34-, and mature B cells are defined as CD33 / 66b-19+10-34-. Untreated mice were pooled together (gray bars). 4G8-treated FLT3N399 and AAVS1 BE mice are reported in pink and yellow, respectively. The fold change in absolute numbers of CAR-treated groups (FLT3N399 / AAVS1) is reported above the bar plots of each population. Mean ± SD. One-way ANOVA with multiple comparisons. FDR-adjusted p-values ​​are reported for comparisons between FLT3N399 vs. AAVS1 BE conditions treated with 4G8 CAR (p<0.05 in bold). [Figure 45A] Peripheral blood lineage composition of NBSGW mice xenografted with AAVS1BE and FLT3N399 HSPCs at week 8. Mean ± SD. AAVS1BE N=13, FLT3N399 N=18. Statistical comparison by unpaired t-test. [Figure 45B] Showing that lineage negative CD34+ progenitor cells are depleted by 4G8 CAR-T in vivo and protected by FLT3N399 editing (experiment from Figure 44). Representative flow cytometry plot of lineage-neg cells (mNeonGreen-CD3-19-14-11c-56-) with gating on CD34+ progenitor cells (% reported within gate). [Figure 45C] The percentage of lin-CD34+ cells in hCD45+3-mNeonGreen- BM cells is shown. Mean ± SD. Comparison by one-way ANOVA. [Figure 45D]Relative composition of BM lin-CD34+ in mice from Figure 44. Mean ± SD. Two-way ANOVA with multiple comparisons (only AAVS1 vs. FLT3N399 comparisons are reported). HSC: hematopoietic stem cells, MPP: multipotent progenitors, LMPP: lymphoid-primed multipotent progenitors, CMP: common myeloid progenitors, GMP: granulocyte-monocyte progenitors, MEP: megaerythroid progenitors. [Figure 45E] FLT3 expression (MFI) on myeloid (left) and lymphoid (right) BM subsets at endpoint is shown. LMPP, MPP, and HSC from 4G8 CAR-treated AAVS1BE condition were not evaluable (NE) due to low cell numbers. Mean ± SD. Statistical differences are reported by multiple unpaired t-test. [Fig.45F] CAR cell phenotype by flow cytometry in the BM of mice from Figure 44. CD45RA+62L+: naive, CD45RA-62L+: central memory (CM), CD45RA-62L-: effector memory (EM), CD45RA+62L: terminally differentiated EM cells re-expressing CD45RA (EMRA). [Figure 45G] Bar plots reporting (from left to right) the percentage of EGFRt+ in BM CD3+ cells, the MFI of PD1 (CD279) on BM CD8+ CAR T cells, and the MFI of PD1 (CD279) on BM CD4+ CAR T cells. Mean ± SD. Comparisons between AAVS1BE and FLT3N399 conditions by one-way ANOVA are reported. [Fig. 45H] Peripheral blood lineage composition of NBSGW mice xenografted with AAVS1BE and CD123S59 HSPCs at week 10 (experiment from FIG. 46). Mean±SD (N=11). Statistical comparisons are reported by unpaired t-test. [Fig.45I]Lineage-negative CD34+ progenitor cells are depleted by CSL362 CAR-T in vivo (experiment from Figure 46). Left, representative flow cytometry plot of lineage-negative cells (mNeonGreen-CD3-19-14-11c-56-) with gating on CD34+ progenitor cells. Right, representative flow cytometry plot of lin-CD34+38-10- cells with gating on HSC (CD45RA-90+), MPP (CD45RA-90-), LMPP (CD45RA+90-) subsets. [Figure 45J] Relative composition of BM lin-CD34+ of mice from Figure 46. Mean ± SD. Two-way ANOVA with multiple comparisons (only AAVS1 vs. CD123S59 comparisons are reported). HSC: hematopoietic stem cells, MPP: multipotent progenitors, LMPP: lymphoid-primed multipotent progenitors, CMP: common myeloid progenitors, GMP: granulocyte-monocyte progenitors, MEP: megaerythroid progenitors. [Figure 46A] Schematic of xenograft and analysis of CD123S59 or AAVS1 BE HSPCs co-engrafted with AML PDX-1 and treated with 5M CSL362 CAR-T cells. [Figure 46B] CD123 base editing measured on total blood cells (weeks 8, 10) and sorted CD33+ and CD19+ bone marrow (BM) cells at endpoint in mice treated or untreated with CSL362 CAR-T cells. CD123S59 N=6, AAVS1 BE N=5. Mean ± SD. Statistical comparison by multiple unpaired t-test. [Figure 46C] Representative flow cytometry plots of BM samples from CD34+ HSPC + AML PDX-1 engrafted mice either treated with CSL362 CAR-T or untreated, and mice engrafted with AML PDX-1 alone are shown. Plots are pre-gated on total human CD45+. CAR T cells are identified by CD3 staining, while AML PDX cells are mNeonGreen+. [Figure 46D]Bar plot showing percentage of AML PDX cells within hCD45+CD3− BM cells. Mean±SD. [Figure 46E] Bar plot showing percentage of CD3+ cells within hCD45+mNeonGreen- BM cells. Mean±SD. [Figure 46F] Bar plot showing absolute numbers of total hCD45+3-mNeonGreen- cells in the BM. Mean±SD. [Figure 46G] Bar plot showing the percentage of pro-B cells (CD19+10-34-) in human CD45+3-mNeonGreen- BM cells. Mean ± SD. Statistical comparison of CD123S59 vs. AAVS1 BE conditions by one-way ANOVA. [Fig. 46H] Representative FACS plots showing depletion of BM myeloid cells (CD33 / 66b+19-, highlighted in orange gating) by CSL362 CAR-T (left) and granulocytes (PMN, CD33 / 66b+19-14-SSC high, orange gating) in mice engrafted with CD123S59 or AAVS1 BE HSPCs (right). [Figure 46I] Bar plots showing total myeloid cells (CD33 / 66b+19- in hCD45+ cells) (Figure 46I), PMN (CD33 / 66b+19- in CD33 / 66b+19- cells) (Figure 46J), and immature PMN (CD33 / 66b+19-14-10-11c-SSC high in CD33 / 66b+19- cells) (Figure 46K). Mean ± SD. Statistical comparison of CD123S59 vs. AAVS1 BE conditions by one-way ANOVA. [Figure 46J] Bar plots showing total myeloid cells (CD33 / 66b+19- in hCD45+ cells) (Figure 46I), PMN (CD33 / 66b+19- in CD33 / 66b+19- cells) (Figure 46J), and immature PMN (CD33 / 66b+19-14-10-11c-SSC high in CD33 / 66b+19- cells) (Figure 46K). Mean ± SD. Statistical comparison of CD123S59 vs. AAVS1 BE conditions by one-way ANOVA. [Figure 46K]Bar plots showing total myeloid cells (CD33 / 66b+19- in hCD45+ cells) (Figure 46I), PMN (CD33 / 66b+19- in CD33 / 66b+19- cells) (Figure 46J), and immature PMN (CD33 / 66b+19-14-10-11c-SSC high in CD33 / 66b+19- cells) (Figure 46K). Mean ± SD. Statistical comparison of CD123S59 vs. AAVS1 BE conditions by one-way ANOVA. [Figure 46L] Representative flow cytometry plots showing loss of dendritic cell (DC) subsets by CSL362 CAR-T in mice engrafted with CD123S59 or AAVS1 BE HSPCs. Left: conventional DC (cDC, CD33 / 66b+14-11c+FLT3+SSC low), plot is gated on CD33 / 66b+14-SSC low cells. Right: plasmacytoid DC (pDC, CD33 / 66b+14-11c-FLT3+CD123 high SSC low), plot is gated on CD33 / 66b+14-11c-SSC low cells. [Figure 46M] Bar blots showing the percentage of cDC and pDC within hCD45+3-mNeonGreen- cells, respectively. [Figure 46N] Bar blots showing the percentage of cDC and pDC within hCD45+3-mNeonGreen- cells, respectively. [Fig. 46O] Percentage of lineage-CD34+ progenitor cells within hCD45+3-mNeonGreen- cells is shown. Mean ± SD. Statistical comparison of CD123S59 vs. AAVS1 BE conditions by one-way ANOVA. [Figure 46P]Absolute numbers of myeloid (P) and lymphoid (Q) lineage subsets in BM, from HSC to differentiated leukocytes, are shown. Untreated mice were pooled together (grey bars), while CSL362-treated CD123S59 and AAVS1 BE mice are reported in pink and blue, respectively. Fold change in absolute numbers (CD123S59 / AAVS1) for CAR-treated groups is reported above the bar plots for each population. Mean ± SD. One-way ANOVA with multiple comparisons. FDR-adjusted p-values ​​for comparisons between CD123S59 vs. AAVS1 BE conditions treated with 4G8 CAR are reported (p<0.05 in bold). [Figure 46Q] Absolute numbers of myeloid (P) and lymphoid (Q) lineage subsets in BM, from HSC to differentiated leukocytes, are shown. Untreated mice were pooled together (grey bars), while CSL362-treated CD123S59 and AAVS1 BE mice are reported in pink and blue, respectively. Fold change in absolute numbers (CD123S59 / AAVS1) for CAR-treated groups is reported above the bar plots for each population. Mean ± SD. One-way ANOVA with multiple comparisons. FDR-adjusted p-values ​​for comparisons between CD123S59 vs. AAVS1 BE conditions treated with 4G8 CAR are reported (p<0.05 in bold). [Figure 47A] CD123 expression (MFI) on myeloid (left) and lymphoid (right) BM subsets at endpoint is shown. Mean ± SD. Statistical differences are reported by multiple unpaired t-test. [Figure 47B] Figure 1 shows that 4G8 CAR T cells deplete PDX-1 but not PDX-2 in vivo. NSBGW female mice were xenografted with AML PDX cells and treated with 4G8 CAR-T cells 10 days later. Experimental endpoint was 14 days after CAR-T administration. From left to right, bar plots showing % of AML PDX cells among total CD45+ cells in bone marrow (BM), % of AML PDX cells among total CD45+ cells in spleen (SP), absolute number of BM T cells, FLT3. MFI on viable AML cells in BM and FLT3 MFI on viable AML cells in SP. Mean ± SD. Comparison by one-way ANOVA. [Figure 47C]Figure 1 shows that 4G8 CAR T cells deplete PDX-1 but not PDX-2 in vivo in mice pre-engrafted with AAVS1BE and FLT3N399 HSPCs. NBSGW mice were xenografted with edited HSPCs and injected with PDX-1 or PDX-2 cells 11 weeks later. Ten days later, mice were treated with 2.5M 4G8 CAR-T cells and outcomes were assessed 2 weeks later. Percentage of AML cells among total BM hCD45+ cells is reported in bar plots. Mean ± SD. Comparison by one-way ANOVA. [Figure 47D] Figure 1 shows that CAR T cell combinations have improved efficacy against PDX-2 in vivo. NBSGW mice were xenografted with 0.75M PDX-2 cells and 10 days later treated with 2.5M 4G8 CAR or 4G8+CSL362, 4G8+Fab79D or CSL362+Fab79D CAR T cell combinations (2.5M each). Results were assessed after 2 weeks. Bar plots show the percentage of AML cells within total BM hCD45+ cells. Mean ± SD. One-way ANOVA multiple comparisons versus untreated condition. [Figure 47E] Peripheral blood lineage composition of NBSGW mice xenografted with AAVS1BE and double-edited FLT3N399 / CD123S59 HSPCs at week 9 (experiment from FIG. 49D). Statistical comparison by unpaired t-test. Mean±SD. [Figure 48A-1]Representative flow cytometry gating strategy for bone marrow analysis of mice xenografted with edited HSPCs and / or AML PDXs (experiment from Figure 46). Cells are pre-gated with single, live, and physical parameter gating. Populations: 1. Human CD45+; 2. T cells (CAR); 3. mNeonGreen+ (AML PDX); 4. Human CD45+ without CAR or AML; 5. Total CD19+; 6. Total myeloid cells (CD33 / 66b+); 7. Granulocytes+ mast cells; 8. Monocytes; 9. Mast cells; 10. Total granulocytes; 11. Mature granulocytes (CD10+ / 11c+); 12. Immature granulocytes (CD10- / 11c-); 13. Classical dendritic cells, cDC; 14. Pro-B cells; 15. Pre-B cells; 16. Mature B cells; 17. Natural killer cells, NK; 18. Pro-lymphocytes; 19. Myeloblasts; 20. lin-CD34+ (CD33 / 66+ 1. lin-CD34+ and CD33 / 66-lin-CD34+ are pooled for downstream gating); 21. Lin-CD34+38+; 22. Lin-CD34+38-; 23. Pre-B / NK; 24. Granulocyte-monocyte, GMP; 25. Common myeloid progenitor, CMP; 26. Megaloblastoid progenitor, MEP; 27. Hematopoietic stem cell, HSC; 28. Multipotent progenitor, MPP; 29. ​​Lymphoid primed multipotent progenitor, LMPP; 30. Bead gating; 31. Mouse CD45+; 32. Plasmacytoid DC; 33. Multipotent lymphoid progenitor, MLP. CountBeads are first gated on singlets as FSC-A low PI high and then gated on two additional fluorescence parameters (bottom left corner plots). [Figure 48A-2] As mentioned above. [Figure 48A-3] As mentioned above. [Figure 48A-4] As mentioned above. [Figure 48B-1]Representative flow cytometry gating strategy for T cell panels on bone marrow and spleen. Cells are pre-gated with single, live, and physical parameter gates. Human CD45+ cells are separated into CD3+ (T cells) and CD3-, in which residual AML is gated as mNeonGreen+. CAR+ cells are identified by EGFR staining. T cell phenotype is assessed by CD45RA and CD62L in gated CD8+ and CD4+ cells. [Figure 48B-2] As mentioned above. [Figure 49A] Figure 41 shows the efficiency of CD34+ HSPC base editing with FLT3N399 alone or in combination with KITH378R, measured on bulk cells or FACS-sorted CD90+ primitive progenitor cells. For comparison, KITH378R single editing from Figure 41C is reported. Mean ± SD. [Figure 49B] Representative flow cytometry plots of dual FLT3 / CD123 reporter K562 cells showing loss of recognition by 4G8 and 7G3 mAbs after multiple epitope editing. [Figure 49C] Edited cells in each quadrant in B were FACS sorted and then co-cultured with bispecific FLT3 / CD123 CAR-T cells. Plots showing percentage of live target cells (left), % of T cell activation (CD69 expression, middle) and degranulation (CD107a surface expression, right), median fluorescence intensity (MFI) of CellTrace marking on T cells and FLT3. CD123 surface expression (MFI) on target cells is reported at different E:T ratios. Mean ± SD (N=4). Statistical comparison by two-way ANOVA. [Figure 49D] Schematic of xenograft and analysis of dual FLT3N399 / CD123S59 epitope edited or AAVS1 BE HSPCs co-engrafted with AML PDX-2 and treated with a 1:1 pool of 4G8 and CSL362 CAR-T cells. [Figure 49E]Representative flow cytometry plots of BM samples from mice engrafted with CD34+ HSPCs, CD34+ HSPCs + AML PDX-2, or CD34+ HSPCs + AML PDX-2 treated with 4G8 / CSL362 CAR are shown. Plots are pre-gated on total human CD45+. CAR T cells were identified by CD3 staining and AML PDX cells are mNeonGreen+. [Fig.49F] Bar plot showing percentage of AML PDX cells within hCD45+CD3- BM cells. Mean ± SD. Statistical comparison by one-way ANOVA. [Figure 49G] Shown are FLT3 (left) and CD123 (right) base editing measured in total blood cells (9, 10 weeks) and sorted CD33+ and CD19+ BM cells at endpoint in mice from D. Mean ± SD. Statistical comparison by multiple unpaired t-test. [Fig. 49H] Absolute numbers of myeloid (left) and lymphoid (right) lineage subsets in BM from HSC to differentiated leukocytes are shown. Untreated mice are pooled (gray bars), CAR-treated FLT3N399+CD123S59 mice are reported in pink, and CAR-treated AAVS1BE mice are reported in blue. Fold change in absolute numbers in CAR-treated groups (FLT3N399+CD123S59 / AAVS1) is reported above the bar plots for each population. Mean ± SD. One-way ANOVA with multiple comparisons. FDR-adjusted p-values ​​for comparisons between 4G8 CAR-treated FLT3N399 / CD123S59 vs. AAVS1 BE conditions are reported (p<0.05 in bold). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] Identifying suitable proteins for targeted cancer therapy is very challenging. Many potential target proteins are present on both the cell surface of cancer cells and the cell surface of normal non-cancerous cells, which may be necessary or critically involved in the development and / or survival of a subject. Many of the target proteins contribute to the functionality of such essential cells. Thus, treatments targeting these proteins may result in adverse effects in subjects, such as significant toxicity and / or other side effects. Furthermore, resistance to CAR-T therapy remains a challenge in the treatment of hematopoietic malignancies, such as acute myeloid leukemia (AML), due to the switch of cancer antigens on cancer cells that evade CAR-T therapy.

[0029] Thus, the present disclosure provides methods, cells, compositions, and kits aimed at addressing at least the above problems. The methods, cells, compositions, and kits described herein provide safe and effective treatment of hematological malignancies, allowing targeting of one or more cell surface proteins present not only on cancer cells, but also on cells essential for their development, and / or survival of a subject. In some cases, described herein are engineered hematopoietic cells, such as hematopoietic stem cells (HSPCs), with gene editing in one or more genes encoding cell surface proteins, e.g., FLT3, CD123, and / or KIT, methods of producing same, e.g., by a CRISPR approach using specific guide RNAs, methods of treating hematopoietic malignancies using the engineered hematopoietic cells, alone or in combination with one or more cytotoxic agents (e.g., CAR-T cells) that target wild-type cell surface antigens but not those encoded by the edited genes in the engineered hematopoietic cells, and kits including the engineered hematopoietic cells.

[0030] I. Genetically engineered hematopoietic cells In some embodiments, the genetically engineered hematopoietic cells have edited FLT3, CD123, or KIT genes. In some, one or more of these genes are mutated. In some cases, the mutated FLT3, CD123, or KIT genes contain mutations or deletions in one or more non-essential epitopes so as to retain (in whole or in part) the biological activity of the FLT3, CD123, or KIT genes.

[0031] i.Hematopoietic stem cells In some embodiments, the hematopoietic cells described herein are hematopoietic stem cells. Hematopoietic stem / progenitor cells (HSPCs) can give rise to both myeloid and lymphoid progenitor cells, which further give rise to myeloid (e.g., monocytes, macrophages, neutrophils, basophils, dendritic cells, erythrocytes, platelets, etc.) and lymphoid. cells (e.g., T cells, B cells, NK cells), respectively. HSPCs are characterized by expression of the cell surface marker CD34 (e.g., CD34+), which can be used to identify and / or isolate HSPCs.

[0032] In some embodiments, the HSPCs are obtained from a subject, such as a mammalian subject. In some embodiments, the mammalian subject is a non-human primate, rodent (e.g., mouse or rat), cow, pig, horse, or livestock. In some embodiments, the HSPCs are obtained from a human patient, such as a human patient suffering from a hematopoietic malignancy. In some embodiments, the HSPCs are obtained from a healthy donor. In some embodiments, the HSPCs are obtained from a subject to which the genetically engineered HSPCs are subsequently administered. HSPCs administered to the same subject from which the cells were obtained are referred to as autologous cells, whereas HSPCs obtained from a subject other than the subject to which the cells are administered are referred to as allogeneic cells (methods: reducing the incidence of rejection are standard and well known in the art).

[0033] HSPCs can be obtained from any suitable source using conventional means known in the art. In some embodiments, HSPCs are obtained from a sample from a subject (or donor), for example, from a bone marrow sample or a blood sample. Alternatively, or in addition, HSPCs can be obtained from the umbilical cord (i.e., cord blood cells). In some embodiments, HSPCs are derived from bone marrow, cord blood cells, or peripheral blood mononuclear cells (PBMCs). In general, bone marrow cells can be obtained from the iliac eminence, femur, tibia, spine, ribs, or other medullary cavity of a subject (or donor). Bone marrow can be taken from a patient and isolated through various separation and washing procedures known in the art. An exemplary procedure for isolating bone marrow cells includes the following steps: a) extracting a bone marrow sample; b) centrifuging the bone marrow suspension into three fractions and recovering the intermediate fraction or buffy coat; c) centrifuging the buffy coat fraction from step (b) once more in a separation fluid (typically Ficoll™) to recover the intermediate fraction containing the bone marrow cells; and d) washing the fraction recovered in step (c) to recover reinjectable bone marrow cells.

[0034] HSPCs are usually present in bone marrow, but can be mobilized into the circulating blood by administering a mobilization agent to harvest HSPCs from peripheral blood. In some embodiments, the subject (or donor) from which HSPCs are obtained is administered a mobilization agent, such as granulocyte colony-stimulating factor (G-CSF). The number of HSPCs collected after mobilization using a mobilization agent is typically greater than the number of cells obtained without the use of a mobilization agent.

[0035] In some embodiments, a sample is obtained from a subject (or donor) and then enriched for the desired cell type (e.g., CD34+, CD34+CD38-, CD133+, CD90+, CD49f+). For example, PBMCs and / or CD34+ hematopoietic cells can be isolated from blood as described herein. Cells can also be isolated from other cells, for example, by isolation and / or activation with an antibody that binds to an epitope on the cell surface of the desired cell type. Another method that can be used includes negative selection using antibodies against cell surface markers that selectively enrich for a particular cell type without activating the cells by receptor binding.

[0036] ii. Mutant cell surface antigens In some embodiments, hematopoietic stem cells (HSPCs) as described herein may contain an edited gene in a mutated form (mutant or variant, used interchangeably herein) encoding one or more cell surface proteins of interest (e.g., FLT3, CD123, KIT), the mutated form having reduced or no binding to a cytotoxic agent as described herein (e.g., anti-FLT3 antibody, anti-CD123 antibody, anti-KIT antibody). The mutant may possess one or more mutations in the epitope to which the cytotoxic agent binds, such that binding to the cytotoxic agent is reduced or abolished compared to the native or wild-type cell surface protein counterpart. Such mutants are preferred to maintain substantially similar biological activity as the wild-type counterpart.

[0037] As used herein, the term "reduced binding" refers to binding that is reduced by at least 25%. The level of binding may refer to the amount of binding of a cytotoxic agent to a hematopoietic stem cell or the amount of binding of a cytotoxic agent to a cell surface protein when compared to a wild-type (i.e., unengineered, unmutated) protein. In some embodiments, binding is reduced by at least 25%, 30%, 40%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%. In some embodiments, binding is reduced such that there is substantially no detectable binding in a conventional assay. As used herein, "no binding" refers to substantially no binding, e.g., no detectable binding, or only baseline binding as determined in conventional binding assays.

[0038] In some cases, the variant contains one or more amino acid residue substitutions (e.g., 2, 3, 4, 5, or more) within the epitope of interest, such that the cytotoxic agent does not bind to the mutated epitope or has reduced binding to the mutated epitope. Such variants may have substantially reduced binding affinity to the cytotoxic agent (e.g., at least 40%, 50%, 60%, 70%, 80%, or 90% lower binding affinity than its wild-type counterpart). In some examples, such variants may have abolished binding activity to the cytotoxic agent. In other cases, the variant contains a deletion of a region that includes the epitope of interest. Such a region may be encoded by an exon. In some embodiments, the region is a domain of the cell surface protein of interest that encodes the epitope. In one example, the variant has only the epitope deleted. The length of the deleted region can be in the range of 3 to 60 amino acids, for example, 5 to 50, 5 to 40, 10 to 30, 10 to 20, etc.

[0039] The mutation(s) or deletion(s) in the variant of the cell surface antigen may be within or surrounding a non-essential epitope such that the mutation(s) or deletion(s) do not substantially affect the biological activity of the protein.

[0040] As used herein, the term "epitope" refers to the amino acid sequence (linear or conformational) of a protein, such as a cell surface antigen, that is bound by a CDR of an antibody. In some embodiments, a cytotoxic agent binds to one or more (e.g., at least 2, 3, 4, 5 or more) epitopes of a cell surface antigen. In some embodiments, a cytotoxic agent binds to multiple epitopes of a cell surface antigen, and the hematopoietic cell is engineered such that each of the epitopes is absent and / or unavailable for binding by the cytotoxic agent.

[0041] In some embodiments, the genetically engineered HSPCs described herein have one or more edited genes for a cell surface antigen, such that the edited genes express a mutant cell surface antigen with a mutation in one or more non-essential epitopes. A "non-essential epitope" (or a fragment containing it) refers to a domain within a cell surface protein / antigen, the mutation of which is unlikely to substantially affect the biological activity of the cell surface protein. For example, hematopoietic cells that contain a deletion or mutation of a non-essential epitope of a cell surface antigen, such hematopoietic cells can proliferate and / or undergo erythropoietic differentiation to a similar level as hematopoietic cells expressing wild-type cell surface antigens.

[0042] Methods for identifying and / or validating non-essential epitopes in cell surface antigens are known and appreciated by those of skill in the art and are also within the scope of the present disclosure. Additionally, methods for assessing the functionality of cell surface antigens and hematopoietic cells are known in the art, such as proliferation assays, differentiation assays, colony formation, expression analysis (e.g., gene and / or protein), protein localization, intracellular signaling, functional assays, and in vivo humanized mouse models.

[0043] iii. Preparation of genetically engineered hematopoietic cells Any of the genetically engineered hematopoietic cells, such as HSPCs, with edited genes for one or more cell surface antigens can be prepared by conventional methods or by the methods described herein. In some embodiments, the genetic engineering is performed using genome editing. As used herein, "genome editing" refers to a method of modifying a genome, including any protein-coding or non-coding nucleotide sequence of an organism, to modify the expression of a target gene. In general, genome editing methods include the use of endonucleases that can cleave the nucleic acid of the genome, for example, at a target nucleotide sequence. In some cases, genome editing methods include the use of nucleases that are killer nucleases or nickases. Repair of double-strand breaks in the genome can be repaired by introducing mutations and / or exogenous nucleic acids can be inserted at the target site. In some cases, genome editing methods include the use of catalytically inactive or partially inactive endonucleases fused to functional domains, for example, in the case of base editors, adenine or cytidine deaminase domains. Other functional domains include prime editors, CRISPR-Cas activators or repressors, and the like.

[0044] Genome editing methods are generally classified based on the type of endonuclease involved in generating double-stranded breaks in the target nucleic acid. These methods include the use of zinc finger nucleases (ZFNs), transcription activator-like effector-based nucleases (TALENs), meganucleases, and CRISPR / Cas systems.

[0045] In one embodiment of the present disclosure, the replacement of cancer cells with a modified population of normal cells is performed using engineered normal cells such that the cells do not bind to cytotoxic drugs. Such modifications can include deletion or mutation of specific cell surface protein epitopes using a CRISPR-Cas system, where the clustered regularly interspaced short palindromic repeats (CRISPR)-Cas system is an engineered, non-naturally occurring CRISPR-Cas system. System.

[0046] The CRISPR system encodes an RNA-guided endonuclease that is essential for bacterial adaptive immunity. CRISPR-associated (Cas) nucleases can be easily programmed to cleave a variety of target DNA sequences for genome editing. One class of these nucleases, called Cas9 proteins, complex with two short RNAs: crRNA and trans-activating crRNA (tracrRNA). SpCas9, the most commonly used Cas9 orthologue, uses a crRNA that has 20 nucleotides (nt) at its 5' end that are complementary to the "protospacer" region of the target DNA site. Efficient cleavage also requires SpCas9 to recognize a protospacer adjacent motif (PAM). The crRNA and tracrRNA are usually combined into a single ~100 nt guide RNA (gRNA) that directs the DNA cleavage activity of SpCas9. A Cas protein called Cpf1 has been identified that can also be programmed to cleave target DNA sequences. Unlike SpCas9, Cpf1 requires only a single 42-nt crRNA, which has 23 nt at its 3' end that is complementary to the protospacer of the target DNA sequence.

[0047] In some embodiments, the Cas endonuclease is Cas9 nuclease or its variant. Cas9 endonuclease cuts the double-stranded DNA of the target nucleic acid to produce blunt ends. In some embodiments, the Cas endonuclease is Cpf1 nuclease or its variant. Cleavage by Cpf1 nuclease produces terminal arrangement of nucleic acid.

[0048] Cas9 In some embodiments, the Cas endonuclease is a Cas9 enzyme or a variant thereof. In some embodiments, the Cas9 endonuclease is derived from Streptococcus pyogenes (SpCas9) or Staphylococcus aureus (SaCas9).

[0049] The SpCas9 wild type sequence is as follows (SEQ ID NO:1): [ka] [ka]

[0050] The SaCas9 wild type sequence is as follows (SEQ ID NO:2): [ka] [ka]

[0051] Generally, the target nucleic acid is flanked at the 3' or 5' side by a protospacer motif adjacent (PAM) which can interact with the endonuclease and further participate in targeting the endonuclease activity to the target nucleic acid. Generally, the PAM sequence flanking the target nucleic acid is believed to depend on the endonuclease and the source from which the endonuclease is derived. For example, for the Cas9 endonuclease from Streptococcus pyogenes, the PAM sequence is NGG, but the PAM sequences NAG and NGA can be recognized with lower efficiency. For the Cas9 endonuclease from Staphylococcus aureus, the PAM sequence is NNGRRT (SEQ ID NO: 3).

[0052] Thus, in some cases, the endonuclease is engineered / modified so that it can recognize one or more PAM sequences. In some embodiments, the endonuclease is engineered / modified to recognize one or more PAM sequences that are different from the PAM sequence that the endonuclease recognizes without engineering / modification. In some embodiments, the endonuclease is engineered / modified to reduce the off-target activity of the enzyme. In some embodiments, the nucleotide sequence encoding the endonuclease is modified to change the PAM recognition of the endonuclease. For example, a Cas endonuclease (e.g., SpCas9) has a mutation at one or more of the following positions: A61, L1111, D1135, S1136, G1218, E1219, N1317, A1322, R1333, R1335, T1337. See, for example, International Patent Application Publication Nos. WO2016 / 141224 and WO2017 / 040348, and U.S. Patent Application Publication No. 2021 / 0284978A1 (all of which are incorporated by reference into this specification).

[0053] In some embodiments, the Cas9 endonuclease is a wild type version of the nuclease. For example, the Cas9 endonuclease is a SpCas9 endonuclease having the sequence shown in SEQ ID NO: 1 above. In some embodiments, the SpCas9 endonuclease is, for example, at least 80% identical, for example, 85%, 90%, or 95% identical to the amino acid sequence of SEQ ID NO: 1, for example, with up to 5%, 10%, 15%, or 20% difference from the residues of SEQ ID NO: 1, for example, replaced by conservative mutations. In preferred embodiments, the endonuclease retains the desired activity of the parent, for example, nuclease activity (unless the parent is a nickase or death Cas9), and / or the ability to interact with guide RNA and target DNA. In other cases, the Cas9 endonuclease is a SaCas9 endonuclease having the sequence shown in SEQ ID NO: 2 above. In some embodiments, the SaCas9 endonuclease is, e.g., at least 80% identical, e.g., 85%, 90%, or 95% identical, to the amino acid sequence of SEQ ID NO:2, e.g., has up to 5%, 10%, 15%, or 20% differences from the residues of SEQ ID NO:2, e.g., replaced by conservative mutations. In preferred embodiments, the endonuclease retains the desired activity of the parent, e.g., nuclease activity (unless the parent is a nickase or death Cas9), and / or the ability to interact with guide RNA and target DNA.

[0054] In some embodiments, the Cas9 endonuclease is a catalytically inactive Cas9. For example, dCas9 contains mutations at catalytically active residues (D10, E762, D839, H983, or D986, and / or H840 or N863) and does not have nuclease activity. For example, the mutations are: (i) D10A or D10N, and / or (ii) H840A, H840N, or H840Y.

[0055] In some embodiments, the Cas9 endonuclease comprises a mutation at K918. For example, the mutation is K918N.

[0056] In some embodiments, the nucleotide sequence encoding Cas9 endonuclease is further modified to change the activity of the protein. In some embodiments, Cas9 endonuclease is modified to inactivate one or more catalytic residues of the endonuclease. In some embodiments, Cas9 endonuclease is modified to inactivate one of the catalytic residues of the endonuclease, which is called "nickase" or "Cas9n". Cas9 nickase endonuclease cuts one DNA strand of the target nucleic acid.

[0057] In some cases, the endonuclease is an NG-SpCas9 nickase and has the following mutations: D10A, L1111R, D1135V, G1218R, E1219F, A1322R, R1335V, T1337R (relative to wild-type SpCas9). In some cases, the endonuclease is an SpRY-Cas9 nickase and has the following mutations: D10A, A61R, L1111R, D1135L, S1136W, G1218K, E1219Q, N1317R, A1322R, R1333P, R1335Q, T1337R (relative to wild-type SpCas9).

[0058] Cpf1(Cas12a) In some embodiments, the Cas endonuclease is Cpf1 nuclease or a variant thereof. Cpf1 endonuclease generally recognizes a PAM sequence located at the 5' end of the target nucleic acid. For Cpf1 nuclease, the PAM sequence is TTTN. As will be understood by those skilled in the art, the Cas endonuclease Cpf1 nuclease may also be referred to as Cas12a. In some embodiments, the host cell expresses Cpf1 nuclease from Lachnospiraceae bacteria (LbCpf1), Acidaminococcus sp. (AsCpf1), or Francisella tularensis (FnCpf1). The wild-type sequences of each are shown below.

[0059] V-type CRISPR-associated protein Cpf1 [Lachnospiraceae bacteria ND2006], GenBank accession number WP_051666128.1 (SEQ ID NO: 4) [ka] [ka]

[0060] V-type CRISPR-associated protein Cpf1 [Acidaminococcus sp. BV3L6], NCBI reference sequence: WP_021736722.1 (SEQ ID NO: 5) [ka] [ka]

[0061] Type V CRISPR associated protein Cpf1 [Francisella tularensis], GenBank accession number WP_003040289.1 (SEQ ID NO: 6) [ka] [ka]

[0062] In some embodiments, the Cpf1 endonuclease is a wild type version of the nuclease. For example, the Cpf1 endonuclease is a Cpf1 endonuclease having the sequence shown in SEQ ID NO: 4, SEQ ID NO: 5, or SEQ ID NO: 6 above. In some embodiments, the Cpf1 endonuclease is, for example, at least 80% identical, for example, 85%, 90%, or 95% identical to the amino acid sequence of SEQ ID NO: 4, SEQ ID NO: 5, or SEQ ID NO: 6, and has, for example, up to 5%, 10%, 15%, or 20% difference with the residues of SEQ ID NO: 4, SEQ ID NO: 5, or SEQ ID NO: 6, for example, replaced by conservative mutations. In a preferred embodiment, the endonuclease retains the desired activity of the parent, for example, nuclease activity (unless the parent is a nickase or death Cas9), and / or the ability to interact with guide RNA and target DNA.

[0063] Catalytically inactive variants of Cpf1 (Cas12a) can be referred to as dCas12a.Thus, in some embodiments, for AsCpf1, catalytic activity disruptive mutations are made at D908 and E993, for example, D908A and E993A.And for LbCpf1 catalytic activity disruptive mutations at D832 and E925, for example, D832A and E925A, and for FnCpf1 catalytic activity disruptive mutations at D917A and E1006A.

[0064] Functional domains Alternatively, or in addition, the Cas endonuclease (i.e., Cas9 or Cas12a) can be fused to another protein or portion thereof, e.g., a heterologous functional domain. In some embodiments, the heterologous functional domain is a transcriptional activation domain (e.g., VP64 or NF-KB p65). In some embodiments, the heterologous functional domain is a transcriptional silencer or transcriptional repression domain (e.g., a Krupel-associated box (KRAB) domain, an ERF repressor domain (ERD), or an mSin3A interacting domain (SID), and a transcriptional silencer is heterochromatin protein 1 (HP1)). In some embodiments, the heterologous functional domain is an enzyme that modifies the methylation state of DNA (e.g., a DNA methyltransferase (DNMT) or a TET protein (such as TET1)). In some embodiments, the heterologous functional domain is an enzyme that modifies a histone subunit (e.g., a histone acetyltransferase (HAT), histone deacetylase (HDAC), histone methyltransferase (HMT), or histone demethylase). In some embodiments, the heterologous functional domain is a biological tether (e.g., MS2, Csy4, or lambda N). In some embodiments, the heterologous functional domain is FokI.

[0065] In some embodiments, the heterologous functional domain comprises a deaminase that modifies cytosine DNA bases, such as a cytidine deaminase from apolipoprotein B mRNA editing enzyme, base editors such as the catalytic polypeptide-like (APOBEC) family of deaminases including APOBEC1, APOBEC2, APOBEC3A, APOBEC3B, APOBEC3C, APOBEC3D / E, APOBEC3F, APOBEC3G, APOBEC3H, APOBEC4, activation-induced cytidine deaminase (AID), cytosine deaminase 1 (CDA1), and CDA2, and cytosine deaminases acting on tRNA (CDAT). Specifically, evoAPOBEC1-BE4max, eA3A-BE5, and EA-BE4max are included.

[0066] In some embodiments, the heterologous functional domain is a deaminase that modifies adenosine DNA bases, for example, the deaminase includes adenosine deaminase 1 (ADA1), ADA2, adenosine deaminase acting on RNA1 (ADAR1), ADAR2, ADAR3; adenosine deaminase acting on tRNA1 (ADAT1), ADAT2, ADAT3, and naturally occurring or engineered tRNA-specific adenosine deaminase (TadA). For example, ABE8e-TadA-8e. In some embodiments, the TadA adenosine deaminase domain comprises a V106W substitution.

[0067] In some embodiments, the heterologous functional domain is an enzyme, domain, or peptide that inhibits or enhances endogenous DNA repair or base excision repair (BER) pathways, such as uracil DNA glycosylase inhibitor (UGI), which inhibits uracil DNA glycosylase (UDG), also known as uracil N-glycosylase, or UNG, which initiates BER by mediated excision of uracil, or DNA end-binding proteins such as Gam from bacteriophage Mu.

[0068] In some embodiments, the endonuclease is a base editor. The base editor endonuclease generally comprises a catalytically inactive Cas endonuclease fused to a base editor. For example, the endonuclease is SpCas9 with a mutation at D10, E762, D839, H983, or D986, and / or fused to a base editor as described above at H840 or N863.

[0069] In some cases, the endonuclease (Cas9 or Cas12a) is fused to one or more of a nuclear localization sequence, a cell penetrating peptide sequence, an affinity tag, and / or a fluorescent protein. For example, the nuclear localization sequence is an SV40 large T antigen nuclear localization sequence (PKKKRKV, SEQ ID NO: 82), a nucleoplasmin nuclear localization sequence (KRPAATKKAGQAKKKK, SEQ ID NO: 83), or a c-Myc nuclear localization sequence (PAAKRVKLD; SEQ ID NO: 84). For example, the nuclear localization sequence(s) are fused to the N-terminus and / or C-terminus of the Cas9 or Cas12a protein. In some embodiments, when a heterologous functional domain is fused to the N-terminus and / or C-terminus of the Cas9 or Cas12a protein, the nuclear localization sequence(s) are inserted at the N-terminus and / or C-terminus of the heterologous functional domain-Cas protein complex, or between the heterologous functional domain and the Cas protein.

[0070] Exemplary Cas endonuclease sequences are provided below: SEQ ID NO:7 (Amino acid sequence of SpRY-ABE8e-V106W 3xNLS adenine base editor): [ka] [ka]

[0071] SEQ ID NO:85 (amino acid sequence of SpRY-ABE8e 3xNLS adenine base editor): [ka] [ka]

[0072] SEQ ID NO:8 - Amino acid sequence of SpRY-evoAPOBEC1-BE4 3xNLS adenine base editor [ka] [ka]

[0073] SEQ ID NO:9 - Amino acid sequence of SpRY-K918N-ABE8e-V106W 3xNLS adenine base editor [ka] [ka]

[0074] SEQ ID NO:10 - Nucleotide sequence of SpRY-ABE8e-V106W 3xNLS adenine base editor: [ka] [ka] [ka] [ka]

[0075] SEQ ID NO:11 - Nucleotide sequence of SpRY-evoAPOBEC1-BE4 3xNLS adenine base editor: [ka] [ka] [ka] [ka] [ka]

[0076] SEQ ID NO:12 - Nucleotide sequence of SpRY-K918N-ABE8e-V106W 3xNLS adenine base editor [ka] [ka] [ka] [ka]

[0077] gRNA The terms "gRNA", "guide RNA", and "CRISPR guide sequence" may be used interchangeably throughout and refer to a nucleic acid that contains a sequence that determines the specificity of the Cas DNA-binding protein of the CRISPR / Cas system. The gRNA hybridizes (partially or fully complementary) to a target nucleic acid sequence in the genome of a host cell. In some cases, gRNA refers to the crRNA and tracrRNA together (e.g., when Cas9 nuclease is used - in those cases, the guide RNA may be referred to as a single guide RNA, i.e., sgRNA). In other cases, gRNA refers to the crRNA only (e.g., when Cpf1 endonuclease is used). The gRNA or portion thereof that hybridizes to the target nucleic acid may be 15-25 nucleotides, 18-22 nucleotides, or 19-21 nucleotides in length. In some embodiments, the gRNA sequence that hybridizes to the target nucleic acid is 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or 25 nucleotides in length. In some embodiments, the gRNA sequence that hybridizes to the target nucleic acid is 10-30 nucleotides in length, or 15-25 nucleotides in length. In some embodiments, the gRNA sequence is at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or at least 100% complementary to the target nucleic acid.

[0078] Exemplary guide RNAs for editing FLT3, CD123, and kits are shown in Table 1 below. As will be apparent to one of skill in the art, the selection of gRNA sequences may depend on factors such as the number of predicted on-target and / or off-target binding sites. In some embodiments, gRNA sequences are selected to maximize potential on-target sites and minimize potential off-target sites. [Table 1-1] [Table 1-2]

[0079] In some embodiments, multiple gRNAs are introduced into the cell (e.g., one for FLT3 and one for CD123). In some embodiments, two or more guide RNAs are transfected into the cell in equimolar amounts. In some embodiments, two or more guide RNAs are provided in non-equimolar amounts. In some embodiments, two or more guide RNAs are provided in amounts optimized to ensure that editing of each target occurs at equal frequency. In some embodiments, two or more guide RNAs are provided in amounts optimized to ensure that editing of each target occurs at an optimal frequency.

[0080] Template donor sequence In some embodiments, provided herein are "template" donor sequences. The template donor sequence comprises a 100-500 nucleotide (e.g., 200 nucleotide) long single-stranded oligodeoxynucleotide (ssODN), which serves as a donor template for homology-directed repair (HDR), thereby allowing the desired mutation to be inserted into the gene. Each donor template further comprises silent mutations selected for bystander amino acids to reduce the risk of re-cutting by the CRISPR-Cas ribonucleoprotein nuclease complex after successful DNA repair.

[0081] An exemplary template donor sequence for introducing the mutation N399D in FLT is shown in Table 2 below. [Table 2]

[0082] iv. Genetically engineered hematopoietic cells Provided herein are methods of producing engineered hematopoietic cells as described herein, carrying edited genes for expressing one or more cell surface antigens in a mutant form. Such methods may include providing a cell and introducing into the cell cellular components of a CRISPR Cas system for genome editing. In some embodiments, a nucleic acid comprising a CRISPR-Cas guide RNA (gRNA) that hybridizes or is predicted to hybridize to a portion of a nucleotide sequence encoding a cell surface antigen is introduced into the cell. In some embodiments, a vector is introduced into the cell by electroporation. In some embodiments, a Cas endonuclease is introduced into the cell. In some embodiments, the Cas endonuclease is introduced into the cell as a nucleic acid encoding the Cas endonuclease. In some embodiments, the nucleotide sequence encoding the gRNA and the Cas endonuclease are introduced into the cell on the same nucleic acid (e.g., the same vector). In some embodiments, the Cas endonuclease is introduced into the cell in the form of a protein. In some embodiments, the Cas endonuclease and gRNA are preformed in vitro and introduced into the cell as a ribonucleoprotein complex.

[0083] v. Mutant FLT3 In some embodiments, the cell surface protein is FLT3. The amino acid sequence of wild-type FLT3 is shown below. SEQ ID NO: 48 (FLT3 wild-type amino acid sequence) [ka] [ka]

[0084] In some embodiments, the methods described herein comprise genetically engineering a population of hematopoietic cells using a Cas nuclease or variant thereof. In some embodiments, the methods described herein comprise genetically engineering a gene encoding a cell surface antigen in a population of hematopoietic cells using a Cas nuclease or variant thereof (e.g., SpCas9 or AsCpf1). In some embodiments, the methods described herein comprise genetically modifying or editing the FLT3 gene, or genetically modifying or editing the CD123 gene, or genetically modifying or editing the Kit gene, or genetically modifying or editing the FLT3 gene and the CD123 gene in the population. Reduction of hematopoietic cells using Cas nuclease.

[0085] In some embodiments, the methods described herein include engineering a mutant FLT3 gene in a population of hematopoietic cells using a Cas nuclease or variant thereof. In some embodiments, the methods described herein include engineering a mutation in exon 9 of FLT3 (e.g., resulting in a mutation at position N399 of the polypeptide encoded thereby) in a population of hematopoietic cells using a Cas nuclease or variant thereof. In some embodiments, the methods described herein include engineering a mutant FLT3 gene in a population of hematopoietic cells using a Cas nuclease or variant thereof and a guide sequence provided by any one of SEQ ID NOs: 13-23. In some embodiments, a template donor DNA sequence is also provided. For example, the template donor DNA sequence is provided by any one of SEQ ID NOs: 40-43.

[0086] In some embodiments, the engineered FLT3 gene encodes a protein that has reduced binding to therapeutic anti-FLT3 antibody. In some cases, the engineered FLT3 gene comprises at least one mutation in exon 9 of the FLT3 gene. In some cases, the at least one mutation in exon 9 of the engineered FLT3 gene results in a polypeptide having a mutation at position N399. In some cases, the mutation at position N399 is N399D or N399G. An example of the amino acid sequence of engineered FLT3 is shown below. SEQ ID NO: 49 FLT3 N354S, S356Q, D358E, Q363P, E366K, Q378R, T384I, R387Q, K389A, K395R, D398E, N399D, N408D, H411N, Q412K, H419Y, FLT3-ECD4 variant (having 16 amino acid substitutions in extracellular domain 4, which reduces binding of clone 4G8, including N399D) [ka]

[0087] SEQ ID NO: 50 FLT3 N354S, S356Q, D358E, Q363P, E366K, Q378R, T384I, R387Q, K389A, K395R, D398E, N399D (FLT3 exon 9 mutation) [ka]

[0088] SEQ ID NO:51 (FLT3 N399D variant amino acid sequence) [ka] [ka]

[0089] SEQ ID NO:52 (FLT3 N399G variant amino acid sequence) [ka]

[0090] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO:51, wherein the polypeptide sequence comprises a mutation at N399D, and the polypeptide sequence has reduced binding to a therapeutic anti-FLT3 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0091] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO:52, wherein the polypeptide sequence comprises a mutation at N399G, and the polypeptide sequence has reduced binding to a therapeutic anti-FLT3 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0092] vi. Mutant CD123 In some embodiments, the cell surface protein is CD 123. The amino acid sequence of wild-type CD123 is shown below. SEQ ID NO: 53 (CD123 wild type amino acid sequence) [ka]

[0093] In some embodiments, the methods described herein comprise engineering a mutant CD123 gene in a population of hematopoietic cells using a Cas nuclease or variant thereof. In some embodiments, the methods described herein comprise engineering a mutant CD123 gene in a population of hematopoietic cells using a Cas nuclease or variant thereof to engineer a mutation in exon 2 of CD123 (e.g., resulting in a mutation at position S59 of the encoded polypeptide) or a mutation in exon 3 of CD123 (e.g., P88 of the encoded polypeptide), a Cas nuclease or variant thereof, in a population of hematopoietic cells. In some embodiments, the methods described herein comprise engineering a mutant CD123 gene in a population of hematopoietic cells using a Cas nuclease or variant thereof and a guide sequence provided by any one of SEQ ID NOs: 24-35.

[0094] In some embodiments, the engineered HSPCs comprise an engineered CD123 gene, and the engineered CD123 gene encodes a protein with reduced binding to a therapeutic anti-CD123 antibody. In some cases, the engineered FLT3 gene comprises at least one mutation in exon 2 and / or exon 3 of the CD123 gene. In some cases, the at least one mutation in exon 2 of the engineered CD123 gene results in a polypeptide with a mutation at position S59. In some cases, the mutation at position S59 is S59P or S59F. In some cases, the at least one mutation in exon 3 of the engineered CD123 gene results in a polypeptide with a mutation at position P88. In some cases, the mutation at position P88 is P88L or P88S. An example of an amino acid sequence of engineered CD123 is shown below.

[0095] SEQ ID NO:54 (CD123 S59P variant amino acid sequence) [ka]

[0096] SEQ ID NO: 55 (CD123 Y58H S59P variant amino acid sequence) [ka]

[0097] SEQ ID NO: 56 (CD123 S59F variant amino acid sequence) [ka]

[0098] SEQ ID NO: 57 (CD123 P88S variant amino acid sequence) [ka]

[0099] SEQ ID NO: 58 (CD123 P88L variant amino acid sequence) [ka]

[0100] The nucleotide sequence is as follows: CD123 exon 2 wild type sequence (SEQ ID NO: 59) [ka]

[0101] CD123 exon 2 sequence with base editing induced by IL3RA_gRNA_N or IL3RA_gRNA_R and SpRY_ABE8e_V106W (I variant) (SEQ ID NO: 60) [ka]

[0102] CD123 exon 2 sequence with base editing induced by IL3RA_gRNA_N or IL3RA_gRNA_R and SpRY_ABE8e_V106W (II variant) (SEQ ID NO: 61) [ka]

[0103] CD123 exon 2 sequence with base editing induced by IL3RA_gRNA_N or IL3RA_gRNA_R and SpRY_ABE8e_V106W (III variant) (SEQ ID NO: 62) [ka]

[0104] CD123 exon 2 sequence with base editing induced by IL3RA_gRNA_N or IL3RA_gRNA_R and SpRY_ABE8e_V106W (IV variant) (SEQ ID NO: 63) [ka]

[0105] CD123 exon 2 sequence with base editing induced by IL3RA_gRNA_N or IL3RA_gRNA_R and SpRY_ABE8e_V106W (V variant) (SEQ ID NO: 64) [ka]

[0106] CD123 exon 2 sequence with base editing induced by IL3RA_gRNA_N or IL3RA_gRNA_R and SpRY_ABE8e_V106W (VI variant) (SEQ ID NO: 65) [ka]

[0107] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO: 54, wherein the polypeptide sequence comprises a mutation at S59P, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0108] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO:55, wherein the polypeptide sequence comprises mutations at Y58H and S59P, and wherein the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0109] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO: 56, wherein the polypeptide sequence comprises a mutation at S59F, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0110] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO: 57, wherein the polypeptide sequence comprises a mutation at P88S, and the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0111] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO: 58, wherein the polypeptide sequence comprises a mutation at P88L, and wherein the polypeptide sequence has reduced binding to a therapeutic anti-CD123 antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0112] vii. Mutant KIT In some embodiments, the cell surface protein is KIT. The amino acid sequence of wild-type KIT is shown below.

[0113] SEQ ID NO: 66 (KIT wild type amino acid sequence) [ka]

[0114] In some embodiments, the methods described herein comprise engineering a mutant KIT gene in a population of hematopoietic cells using a Cas nuclease or variant thereof. In some embodiments, the methods described herein comprise engineering a mutation in exon 7 of KIT (e.g., mutation position H378) in a population of hematopoietic cells using a Cas nuclease or variant thereof. In some embodiments, the methods described herein comprise engineering a mutant KIT gene in a population of hematopoietic cells using a Cas nuclease or variant thereof and a guide sequence provided by any one of SEQ ID NOs: 36-47.

[0115] In some embodiments, the engineered HSPCs comprise an engineered KIT gene, the engineered KIT gene encoding a protein with reduced binding to a therapeutic anti-KIT antibody. In some cases, the engineered KIT gene comprises at least one mutation in exon 6 and / or exon 7 of the KIT gene. In some cases, the at least one mutation in exon 7 of the engineered KIT gene results in a polypeptide having a mutation at position H378. In some cases, the mutation at position H378 is H378R or H378S or H378P or H378A or H378F or H378K or H378G or H378L or H378M. In some cases, the mutation in exon 6 of KIT results in the encoded polypeptide having one or more of the following mutations: F316S, M318V, I319K, V323I, I334V, E360K, P363V, E366D. In some cases, the Kit exon 7 mutation results in a polypeptide having a mutation at E376Q and / or H378R.

[0116] An exemplary amino acid sequence of an engineered KIT is shown below: SEQ ID NO: 67 KIT F316S, M318V, I319K, V323I, I334V, E360K, P363V, E366D, E376Q, H378R (KIT-ECD4 variant amino acid sequence) [ka] [ka]

[0117] SEQ ID NO: 68 (KIT H378R variant amino acid sequence) [ka]

[0118] In some embodiments, provided herein are polypeptide sequences that are at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO:67, wherein the polypeptide sequence comprises mutations at F316S, M318V, I319K, V323I, I334V, E360K, P363V, E366D, E376Q, and H378R, and wherein the polypeptide sequence has reduced binding to a therapeutic anti-KIT antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0119] In some embodiments, provided herein is a polypeptide sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the sequence set forth in SEQ ID NO: 68, wherein the polypeptide sequence comprises a mutation at H378R, and wherein the polypeptide sequence has reduced binding to a therapeutic anti-KIT antibody. Nucleic acids encoding nutritive polypeptides or proteins are also provided herein.

[0120] viii. Genetically engineered hematopoietic cells expressing mutant FLT3, CD123, and / or Kit genes In some embodiments, the cell surface proteins are FLT3 and CD123. In some embodiments, the cell surface proteins are FLT3 and KIT. In some embodiments, the cell surface proteins are KIT and CD123. In some embodiments, two or more guides are transfected simultaneously with each other. In some embodiments, two or more guides are provided sequentially or consecutively, i.e., in two or more separate transfections. For example, FLT3 guide RNA (any one of SEQ ID NOs: 13-23), CD123 guide RNA (any one of SEQ ID NOs: 24-35), KIT guide RNA (any one of SEQ ID NOs: 36-47).

[0121] In some embodiments, the engineered HSPC comprises an engineered FLT3 gene, and the engineered FLT3 gene encodes a protein with reduced binding to a therapeutic anti-FLT3 antibody. In some cases, the engineered FLT3 gene comprises at least one mutation in exon 9 of the FLT3 gene. In some cases, the at least one mutation in exon 9 of the engineered FLT3 gene results in a polypeptide having a mutation at position N399. In some cases, the mutation at position N399 is N399D or N399G. Exemplary amino acid sequences of engineered FLT3 are shown above (see SEQ ID NOs: 49-52).

[0122] In some embodiments, the engineered HSPCs comprise an engineered CD123 gene, the engineered CD123 gene encoding a protein with reduced binding to a therapeutic anti-CD123 antibody. In some cases, the engineered CD123 gene comprises at least one mutation in exon 2 and / or exon 3 of the CD123 gene. In some cases, the mutation in exon 2 of the engineered CD123 gene results in a polypeptide having a mutation at position S59. In some cases, the mutation at position S59 is S59P or S59F. In some cases, the mutation is S59P and Y58H. In some cases, the at least one mutation in exon 3 of the engineered CD123 gene results in a polypeptide having a mutation at position P88. In some cases, the mutation at position P88 is P88L or P88S. Exemplary amino acid sequences of engineered CD123 are shown above (see SEQ ID NOs: 54-58).

[0123] In some embodiments, the engineered HSPCs comprise an engineered KIT gene, wherein the engineered KIT gene encodes a protein with reduced binding to a therapeutic anti-KIT antibody. In some cases, the engineered KIT gene comprises at least one mutation in exon 6 and / or exon 7 of the KIT gene. In some cases, the at least one mutation in exon 7 of the engineered KIT gene results in a polypeptide having a mutation at position H378. In some cases, the mutation at position H378 is H378R or H378S or H378P or H378A or H378F or H378K or H378G or H378L or H378M. In some cases, a mutation in exon 6 of KIT results in an encoded polypeptide having one or more of the following mutations: F316S, M318V, I319K, V323I, I334V, E360K, P363V, E366D. In some cases, a mutation in exon 7 in KIT results in a polypeptide having a mutation in one or more of E376Q and H378R. Exemplary amino acid sequences of engineered KIT are shown above (see SEQ ID NOs: 67-68).

[0124] II. Immunotherapeutic agents specific to cell surface antigens Cytotoxic agents that target cells expressing a cell surface antigen (e.g., cancer cells) can be used in combination with the genetically engineered hematopoietic cells as described herein. As used herein, the term "cytotoxic agent" refers to any agent that can directly or indirectly induce cytotoxicity of a target cell expressing a particular cell surface antigen (e.g., a target cancer cell). Such cytotoxic agents can include protein-binding fragments that bind to and target epitopes of a particular cell surface antigen.

[0125] i. Therapeutic antibodies / antibody-drug conjugates In some cases, the cytotoxic agent includes a therapeutic antibody, which can be conjugated to a drug (e.g., an anti-cancer drug) to form an antibody-drug conjugate (ADC). In some embodiments, the agent is an antibody-drug conjugate. In some embodiments, the antibody-drug conjugate includes an epitope-binding fragment and a toxin or drug that induces cytotoxicity in a target cell.

[0126] In some embodiments, the therapeutic anti-FLT3 antibody is anti-FLT3 clone 4G8 antibody. In some embodiments, the therapeutic anti-CD123 antibody is clone 7G3 antibody or its humanized counterpart CSL362 ("talatuzumab"). In some embodiments, the therapeutic anti-CD123 antibody is anti-CD123 clone 6H6 antibody or anti-CD123 clone S18016F antibody. In some embodiments, the therapeutic anti-KIT antibody is anti-KIT clone Fab79D antibody.

[0127] Toxins or drugs suitable for use in antibody-drug conjugates are well known in the art and will be apparent to those skilled in the art. See, for example, Peters et al. Biosci. Rep. (2015) 35(4): e00225, Beck et al. Nature Reviews Drug Discovery (2017) 16: 315-337, Marin-Acevedo et al. J. Hematol. Oncol. (2018) 11: 8, Elgundi et al. Advanced Drug Delivery Reviews (2017) 122: 2-19. In some embodiments, the antibody-drug conjugate may further comprise a linker (e.g., a peptide linker, such as a cleavable or non-cleavable linker) that connects the antibody and the drug molecule. Examples of antibody-drug conjugates include, but are not limited to, brentuximab vedotin, glembatumumab vedotin / CDX-011, depatuxizumab mafodotin / ABT-414, PSMA ADC, polatuzumab vedotin / RG7596 / DCDS4501A, denintuzumab mafodotin / SGN-CD19A, AGS-16C3F, CDX-014, RG7841 / DLYE5953A, RG7882 / DMUC406A, RG7986 / DCDS0780A, SGN-LIV1A, enfortumab vedotin / ASG-22ME, AG-15ME, AGS67E, terisotuzumab vedotin / ABBV-399, ABBV-221, ABBV-085, GSK-2857916, tisotuzumab vedotin / HuMax-TF-ADC, HuMax-Axl-ADC, pinotuzumab vedotin / HuMax-TF ... stuzumab vedotin / RG7593 / DCDT2980S, rifastuzumab vedotin / RG7599 / DNIB0600A, indusatumab vedotin / MLN-0264 / TAK-264, bundletuzumab vedotin / RG7450 / DSTP3086S, sofituzumab vedotin / RG7458 / DMUC5754A, RG7600 / DMOT4039A, RG7336 / DEDN6526A, ME1547, PF-06263507 / ADC5T4, trastuzumab emtansine / T-DM1, mirvetuximab soravtansine / IMGN853, coltuximab ravtansine / SAR3419,Naratuximab emtansine / IMGN529, indatuximab labtansine / BT-062, anetumab labtansine / BAY94-9343, SAR408701, SAR428926, AMG224, PCA062, HKT288, LY3076226, SAR566658, lorvotuzumab mertansine / IMGN901, cantuzumab mertansine / SB-408075, cantuzumab labtansine / IMGN242, laprituximab emtansine / IMGN289, IMGN388, bivatuzumab mertansine, AVE9633, BIIB015, MLN2704, AMG172, AMG595, LOP628, vadastuximab butarilin / SGN-CD33A, SGN-CD70A, SGN-CD19B, SGN-CD123A, SGN-CD352A, rovalpituzumab tesirine / SC16LD6.5, SC-002, SC-003, ADCT-301 / HuMax-TAC-PBD, ADCT-402, ME DI3726 / ADC-401, IMGN779, IMGN632, gemtuzumab ozogamicin, inotuzumab ozogamicin / CMC-544, PF-06647263, CMD-193, CMB-401, trastuzumab duocarmazine / SYD985, BMS-936561 / MDX-1203, sacituzumab govitecan / IMMU-132, labetuzumab govitecan / IMMU-130, DS-8201a, U3-1402, milatuzumab doxorubicin Antibody-drug conjugates include IMMU-110 / hLL1-DOX, BMS-986148, RC48-ADC / hertuzumab-vc-MMAE, PF-06647020, PF-06650808, PF-06664178 / RN927C, rupartumab amadotin / BAY1129980, apurutumab ixadotin / BAY1187982, ARX788, AGS62P1, XMT-1522, AbGn-107, MEDI4276, and DSTA4637S / RG7861. In one example, the antibody-drug conjugate is gemtuzumab ozogamicin.

[0128] In some embodiments, the binding of the antibody-drug conjugate to an epitope of a cell surface protein can induce internalization of the antibody-drug conjugate and release the drug (or toxin) intracellularly. In some embodiments, the binding of the antibody-drug conjugate to an epitope of a cell surface protein can induce internalization of the toxin or drug, which can kill the cell expressing the cell surface protein (target cell). In some embodiments, the binding of the antibody-drug conjugate to an epitope of a cell surface protein can induce internalization of the toxin or drug, which can modulate the activity of the cell expressing the cell surface protein (target cell). The type of toxin or drug used in the antibody-drug conjugates described herein is not limited to any particular type.

[0129] In some embodiments, two or more (e.g., two, three, four, five or more) epitopes of a cell surface antigen are modified, allowing two or more (e.g., two, three, four, five or more) different cytotoxic agents (e.g., two ADCs) to be targeted against two or more epitopes. In some embodiments, the toxins carried by the ADCs may act synergistically to enhance efficacy (e.g., death of target cells). In some embodiments, two or more (e.g., two, three, four, five or more) epitopes of a cell surface protein are modified, allowing two or more (e.g., two, three, four, five or more) different cytotoxic agents (e.g., two ADCs) to be targeted against two or more epitopes of a cell surface antigen. In some embodiments, one or more (e.g., 1, 2, 3, 4, 5 or more) epitopes of a cell surface antigen are modified, one or more (e.g., 1, 2, 3, 4, 5 or more) epitopes of an additional cell surface protein are modified, allowing two or more (e.g., 2, 3, 4, 5 or more) different cytotoxic agents (e.g., two ADCs) to be targeted to an epitope of a cell surface antigen, and an epitope of an additional cell surface antigen. In some embodiments, targeting multiple cell surface antigens, or a cell surface antigen and one or more additional cell surface proteins / antigens, can reduce recurrence of hematopoietic malignancies.

[0130] In some embodiments, the methods described herein include administering an ADC that targets an epitope of a cell surface antigen that is mutated in a population of genetically engineered hematopoietic cells. In some embodiments, the methods described herein involve administering an ADC that targets an epitope of a cell surface antigen that is mutated in a population of genetically engineered hematopoietic cells and one or more additional cytotoxic agents that may target one or more additional cell surface proteins. In some embodiments, the agents may act synergistically to enhance efficacy by targeting multiple cell surface proteins.

[0131] The ADCs described herein may be used as a follow-on treatment for subjects who have received the combination therapy described herein.

[0132] In some embodiments, the methods described herein include administering to a subject a population of engineered cells lacking a non-essential epitope on a cell surface antigen and one or more immunotherapeutic agents (e.g., ADCs) that target cells expressing the cell surface antigen. In some embodiments, the methods described herein include administering to a subject a population of engineered cells lacking a non-essential epitope on a type 1 cell surface antigen and one or more immunotherapeutic agents (e.g., ADCs) that target cells expressing the cell surface antigen. In some embodiments, the methods described herein include administering to a subject a population of engineered cells lacking a non-essential epitope on a type 2 cell surface antigen and one or more immunotherapeutic agents (e.g., ADCs) that target cells expressing the cell surface antigen. In any of the embodiments described herein, for example, if the hematopoietic malignancy recurs, one or more additional immunotherapeutic agents can be further administered to the subject (e.g., targeting one or more additional epitopes and / or antigens).

[0133] ii. Immune cells expressing chimeric antigen receptors In some embodiments, the cytotoxic agents described herein that target specific cell surface antigen epitopes are immune cells that express a chimeric receptor, the chimeric receptor comprising an epitope-binding fragment that can bind to a cell surface protein epitope (e.g., FLT3, CD123, or Kit). Recognition of a target cell (e.g., a cancer cell) that has a specific protein epitope on its cell surface by the epitope-binding fragment of the chimeric receptor can transmit an activation signal to the signaling domain(s) (e.g., a costimulatory signaling domain and / or a cytoplasmic signaling domain of the chimeric receptor) that can activate effector functions in the immune cell that expresses the chimeric receptor.

[0134] In some embodiments, the immune cells express multiple chimeric receptors (e.g., 2, 3, 4, 5 or more), referred to as bispecific or multispecific immune cells. In some embodiments, the immune cells express multiple chimeric receptors, at least one of which targets an epitope of a cell surface antigen. In some embodiments, the immune cells express multiple chimeric receptors, each of which targets an epitope of a cell surface antigen. In some embodiments, the immune cells express multiple chimeric receptors, at least one of which targets an epitope of a cell surface antigen, at least one of which targets an epitope of an additional cell surface antigen. In some embodiments, targeting multiple cell surface proteins, or a cell surface protein and one or more additional cell surface proteins, can reduce recurrence of hematopoietic malignancies. In some embodiments, the immune cells express chimeric receptors that target multiple epitopes (e.g., multiple epitopes of one antigen, or epitopes of multiple antigens), referred to as bispecific chimeric receptors.

[0135] In some embodiments, epitopes of two or more lineage-specific cell surface proteins are targeted by a cytotoxic agent. In some embodiments, two or more chimeric receptors, e.g., bispecific chimeric receptors, are expressed on the same immune cell. Such cells may be used in any of the methods described herein. In some embodiments, cells expressing chimeric receptors are "pooled", i.e., two or more cell populations express two or more different chimeric receptors. In some embodiments, two or more cells expressing different chimeric antigen receptors are administered simultaneously. In some embodiments, two or more cells expressing different chimeric antigen receptors are administered sequentially. In some embodiments, epitopes of FLT3, CD123, and / or KIT are targeted by a cytotoxic agent. In some embodiments, chimeric receptors targeting FLT3, CD123, and / or KIT are expressed on the same immune cell (i.e., bispecific immune cell). Such cells may be used in any of the methods described herein. In some embodiments, cells expressing chimeric receptors targeting FLT3, CD123, and / or KIT are "pooled", i.e., two or more populations of cells express two or more different chimeric receptors. In some embodiments, two or more populations of cells expressing chimeric receptors targeting FLT3, CD123, and / or KIT are administered simultaneously. In some embodiments, two or more populations of cells expressing chimeric receptors targeting FLT3, CD123, and / or KIT are administered sequentially.

[0136] As used herein, chimeric receptor refers to a molecule of non-natural origin that can be expressed on the surface of a host cell and monitored by a binding domain that provides the specificity of the chimeric receptor (e.g., a cell surface lineage-specific protein). Generally, a chimeric receptor comprises at least two domains derived from different molecules. In addition to the epitope-binding fragments described herein, a chimeric receptor may further comprise one or more of the following: a hinge domain (e.g., CD28 hinge, IgG4 hinge, or CD8α hinge), a transmembrane domain (e.g., CD28 TM, CD8α TM, 4-1BB TM), a costimulatory domain (e.g., CD28z, 4-1BB, ICOS, OX40), a cytoplasmic signaling domain (e.g., CD3z), and combinations thereof.

[0137] In some embodiments, the chimeric receptors described herein include one or more hinge domain(s). In some embodiments, the hinge domain may be located between the epitope-binding fragment and the transmembrane domain. A hinge domain is generally an amino acid segment found between two domains of a protein and may allow flexibility of the protein and movement of one or both of the domains relative to each other. Any amino acid sequence that provides such flexibility and movement of the epitope-binding fragment relative to another domain of the chimeric receptor may be used.

[0138] The hinge domain may contain any one of about 10 to 200 amino acids, e.g., about 15 to 150 amino acids, 20 to 100 amino acids, or 30 to 60 amino acids. In some embodiments, the hinge domain may be at least any one of about 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, or 200 amino acids in length.

[0139] In some embodiments, the hinge domain is a naturally occurring protein hinge domain. Any protein hinge domain known in the art that contains a hinge domain is suitable for use in the chimeric receptors described herein. In some embodiments, the hinge domain is at least a portion of a naturally occurring protein hinge domain, providing flexibility to the chimeric receptor. In some embodiments, the transmembrane domain is derived from CD8α or CD28. In some embodiments, the hinge domain is a portion of the CD8α hinge domain, e.g., a fragment of the CD8α or CD28 hinge domain that contains at least 15 (e.g., 20, 25, 30, 35, or 40) consecutive amino acids.

[0140] Hinge domains of antibodies, such as IgG, IgA, IgM, IgE, or IgD antibodies, are also compatible for use in the chimeric receptors described herein. In some embodiments, the hinge domain is a hinge domain that joins constant domains CH1 and CH2 of an antibody. In some embodiments, the hinge domain is of an antibody and comprises an antibody hinge domain and one or more constant regions of the antibody. In some embodiments, the hinge domain comprises an antibody hinge domain and an antibody CH3 constant region. In some embodiments, the hinge domain comprises an antibody hinge domain and an antibody CH2 and CH3 constant region. In some embodiments, the antibody is an IgG, IgA, IgM, IgE, or IgD antibody. In some embodiments, the antibody is an IgG antibody. In some embodiments, the antibody is an IgG1, IgG2, IgG3, or IgG4 antibody. In some embodiments, the hinge region comprises an IgG1 antibody hinge region and a CH2 and CH3 constant region. In some embodiments, the hinge region comprises the hinge region and CH3 constant region of an IgG1 antibody.

[0141] In some embodiments, the chimeric receptors described herein comprise one or more transmembrane domain(s). The transmembrane domain for use in chimeric receptors can be in any form known in the art. As used herein, "transmembrane domain" refers to any protein structure that is thermodynamically stable in a cell membrane, preferably a eukaryotic cell membrane. A transmembrane domain that is compatible for use in the chimeric receptors described herein can be obtained from a naturally occurring protein. Alternatively, it can be a synthetic, non-naturally occurring protein segment, such as a hydrophobic protein segment that is thermodynamically stable in a cell membrane.

[0142] Further, transmembrane domains may additionally or alternatively be classified based on transmembrane domain topology, including the number of times the transmembrane domain passes through the membrane and the orientation of the protein. For example, a single-pass membrane protein crosses the cell membrane once, and a multi-pass membrane protein crosses the cell membrane at least twice (e.g., 2, 3, 4, 5, 6, 7, or more times). In some embodiments, the transmembrane domain is a single-pass transmembrane domain. In some embodiments, the transmembrane domain is a single-pass transmembrane domain that orients the N-terminus of the chimeric receptor to the extracellular side of the cell and the C-terminus of the chimeric receptor to the intracellular side of the cell. In some embodiments, the transmembrane domain is derived from a single-pass membrane protein. In some embodiments, the transmembrane domain is derived from CD28 or 4-1BB or CD8α.

[0143] In some embodiments, the chimeric receptors described herein comprise one or more costimulatory signaling domains. The term "costimulatory signaling domain" as used herein refers to at least a portion of a protein that mediates intracellular signal transduction to induce an immune response, such as an effector function. The costimulatory signaling domain of the chimeric receptors described herein can be a cytoplasmic signaling domain from a costimulatory protein that transmits signals and regulates responses mediated by immune cells, such as T cells, NK cells, macrophages, neutrophils, or eosinophils.

[0144] In some embodiments, the chimeric receptor comprises multiple (at least two, three, four or more) costimulatory signaling domains. In some embodiments, the chimeric receptor comprises multiple costimulatory signaling domains obtained from different costimulatory proteins. In some embodiments, the chimeric receptor does not comprise a costimulatory signaling domain.

[0145] Many immune effector cells require costimulation in addition to stimulating antigen-specific signals to promote cell proliferation, differentiation, and survival and activate the effector functions of the cells. Activation of costimulatory signaling domains in host cells (e.g., immune cells) can induce the cells to increase or decrease cytokine production and secretion, phagocytosis, proliferation, differentiation, survival, and / or cytotoxicity. The costimulatory signaling domain of any costimulatory molecule can be compatible for use in the chimeric receptors described herein. The type(s) of costimulatory signaling domain are selected based on factors such as the type of immune cell in which the chimeric receptor will be expressed (e.g., primary T cells, T cell lines, NK cell lines), and the desired immune effector function (e.g., cytotoxicity). Examples of costimulatory signaling domains for use in chimeric receptors can be cytoplasmic signaling domains of costimulatory proteins, including, but not limited to, CD27, CD28zeta (CD28z), 4-1BB, OX40, CD30, ICOS, CD2, CD7, LIGHT, NKG2C, B7-H3.

[0146] In some embodiments, the chimeric receptors described herein comprise one or more cytoplasmic signaling domain(s). Any cytoplasmic signaling domain may be used in the chimeric receptors described herein. In general, the cytoplasmic signaling domain relays a signal, such as the interaction of an extracellular ligand-binding domain with its ligand, to stimulate a cellular response, such as inducing a cellular effector function (e.g., cytotoxicity). In some embodiments, the cytoplasmic signaling domain is derived from CD3zeta (CD3z).

[0147] In some embodiments, provided herein are chimeric receptor constructs that target FLT3, CD123, FLT3+CD123, Kit, FLT3+Kit, or Kit+CD123. The constructs further comprise at least a hinge domain (e.g., from CD28, CD8α, or an antibody), a transmembrane domain (e.g., from CD28), one or more costimulatory domains (e.g., from one or more of CD28z), a cytoplasmic signaling domain (e.g., from CD3z), or a combination thereof. In some examples, the methods described herein include administering to a subject a population of genetically engineered hematopoietic cells (engineered to have mutant FLT3, CD123, FLT3+CD123, KIT, FLT3+KIT, or CD123+KIT) and a population of immune cells expressing chimeric receptors targeting FLT3, CD123, FLT3+CD123, KIT, FLT3+KIT, or CD123+KIT, respectively, which further comprise at least a hinge domain (e.g., from CD28, CD8α, or an antibody), a transmembrane domain (e.g., from CD28), one or more costimulatory domains (from one or more of CD28z), and a cytoplasmic signaling domain (e.g., from CD3z), or a combination thereof. In some embodiments, the administered immunotherapy product is a combination of immune cells expressing individual chimeric receptors targeting FLT3, CD123, and / or KIT.

[0148] Any of the chimeric receptors described herein can be prepared by routine methods, such as recombinant techniques. The method of preparing a chimeric receptor herein includes generating a nucleic acid encoding a polypeptide comprising each of the domains of the chimeric receptor, including an epitope-binding fragment and optionally a hinge domain, a transmembrane domain, at least one co-stimulatory signaling domain, and a cytoplasmic signaling domain. In some embodiments, the nucleic acids encoding the components of the chimeric receptor are linked using recombinant techniques.

[0149] Furthermore, any of the chimeric receptors can be expressed in immune cells and administered to a subject (e.g., a human subject) in a conventional manner. For example, T cells can be either derived from the T cells in the subject's own blood (autologous) or derived from the T cells of another healthy donor (allogeneic). Once isolated from the subject, these T cells are genetically engineered to express a specific CAR and programmed to target antigens present on the surface of tumors. The CAR-T cells are then routinely infused into the subject.

[0150] In some embodiments, the CAR is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of any one of SEQ ID NOs: 69, 71, 73, 75, 77, 79, 86, or 87, and the CAR retains the ability to bind to its respective cell surface lineage-specific protein (e.g., KIT, CD123, FLT3, or a combination thereof).

[0151] In some embodiments, the cell surface lineage-specific protein is KIT and the epitope-binding fragment comprises the following CDR sequences: GFNISVYMMH (SEQ ID NO: 88), SIYPYSGYTYYADSVKG (SEQ ID NO: 89), ARYVYHALDY (SEQ ID NO: 90), RASQRGLRNVAVA (SEQ ID NO: 91), SASSLYS (SEQ ID NO: 92), and QQWAVHSLIT (SEQ ID NO: 93). In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to GFNISVYMMHWVRQAPGKGLEWVASIYPYSGYTYYADSVKGRFTISADTSKNTAYLQMNSLRAEDTAVYYCARYVYHALDY (SEQ ID NO: 94), and the epitope-binding fragment retains the ability to bind to its respective KIT epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to RASQRGLRNVAVAWYQQKPGKAPKLLIYSASSLYSGVPSRFSGSRSGTDFTLTISSLQPEDFATYYCQQWAVHSLIT (SEQ ID NO: 95), and the epitope-binding fragment retains the ability to bind to its respective KIT epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 94, and a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 95. In some embodiments, the epitope-binding fragment comprises both SEQ ID NOs: 94 and 95.

[0152] In some embodiments, the cell surface lineage-specific protein is FLT3 and the epitope-binding fragment comprises the following CDR sequences: GYTFTSYWMH (SEQ ID NO: 96), EIDPSDSYKDYNQKFK (SEQ ID NO: 97), RAITTTPFDF (SEQ ID NO: 98), RASQSISNNLH (SEQ ID NO: 99), YASQSIS (SEQ ID NO: 100), and QQSNTWPYT (SEQ ID NO: 101). In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to GYTFTSYWMHWVRQRPGHGLEWIGEIDPSDSYKDYNQKFKDKATLTVDRSSNTAYMHLSSLTSDDSAVYYCARAITTTPFDF (SEQ ID NO: 102), and the epitope-binding fragment retains the ability to bind to its respective FLT3 epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to RASQSISNNLHWYQQKSHESPRLLIKYASQSISGIPSRFSGSGSGTDFTLSINSVETEDFGVYFCQQSNTWPYT (SEQ ID NO: 103), and the epitope-binding fragment retains the ability to bind to its respective FLT3 epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 102, and a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 103. In some embodiments, the epitope-binding fragment comprises both SEQ ID NOs: 102 and 103.

[0153] In some embodiments, the cell surface lineage specific protein is CD123 and the epitope-binding fragment comprises the following CDR sequences: GYSFTDYYMK (SEQ ID NO: 104), DIIPSNGATFYNQKFKG (SEQ ID NO: 105), ARSHLLRASWFAY (SEQ ID NO: 106), SQSLLNSGNQKNYLT (SEQ ID NO: 107), WASTRES (SEQ ID NO: 108), and QNDYSYPYT (SEQ ID NO: 109). In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to GYSFTDYYMKWARQMPGKGLEWMGDIIPSNGATFYNQKFKGQVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAY (SEQ ID NO: 110), and the epitope-binding fragment retains the ability to bind to its respective CD123 epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to ESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWASTRESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQNDYSYPYT (SEQ ID NO: 111), and the epitope-binding fragment retains the ability to bind to its respective CD123 epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 110, and a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 111. In some embodiments, the epitope-binding fragment comprises both SEQ ID NOs: 110 and 111.

[0154] In some embodiments, the cell surface lineage specific proteins are FLT3 and CD123 and the epitope-binding fragments comprise the following CDR sequences: GYTFTSYWMH (SEQ ID NO: 96), EIDPSDSYKDYNQKFK (SEQ ID NO: 97), RAITTTPFDF (SEQ ID NO: 98), RASQSISNNLH (SEQ ID NO: 99), YASQSIS (SEQ ID NO: 100), QQSNTWPYT (SEQ ID NO: 101), GYSFTDYYMK (SEQ ID NO: 104), DIIPSNGATFYNQKFKG (SEQ ID NO: 105), ARSHLLRASWFAY (SEQ ID NO: 106), SQSLLNSGNQKNYLT (SEQ ID NO: 107), WASTRES (SEQ ID NO: 108), and QNDYSYPYT (SEQ ID NO: 109). In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to GYTFTSYWMHWVRQRPGHGLEWIGEIDPSDSYKDYNQKFKDKATLTVDRSSNTAYMHLSSLTSDDSAVYYCARAITTTPFDF (SEQ ID NO: 102), and the epitope-binding fragment retains the ability to bind to its respective FLT3 epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to RASQSISNNLHWYQQKSHESPRLLIKYASQSISGIPSRFSGSGSGTDFTLSINSVETEDFGVYFCQQSNTWPYT (SEQ ID NO: 103), and the epitope-binding fragment retains the ability to bind to its respective FLT3 epitope. In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to GYSFTDYYMKWARQMPGKGLEWMGDIIPSNGATFYNQKFKGQVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAY (SEQ ID NO: 110), and the epitope-binding fragment retains the ability to bind to its respective CD123 epitope.In some cases, the epitope-binding fragment comprises a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to ESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWASTRESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQNDYSYPYT (SEQ ID NO: 111), and the epitope-binding fragment retains the ability to bind to its respective CD123 epitope. In some cases, the epitope-binding fragment is a sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 102, at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 103. In some embodiments, the epitope-binding fragment comprises all four of SEQ ID NOs: 102, 103, 110, and 111.

[0155] In some embodiments, the cell surface lineage specific protein is CD123 and the epitope-binding fragment comprises the following CDR sequences: DIIPSNGATFYNQKFKG (SEQ ID NO: 105), SQSLLNSGNQKNYLT (SEQ ID NO: 107), WASTRES (SEQ ID NO: 108), and QNDYSYPYT (SEQ ID NO: 109).

[0156] Exemplary CAR sequences are provided below: SEQ ID NO:69 - Fab79D-CAR (CD28 hinge, CD28 TM, CD28z, CD3z) amino acid sequence targeting KIT domain 4, variant I (epitope binding regions in italics, CDRs in bold) [ka]

[0157] SEQ ID NO:70 - Fab79D-CAR (CD28 hinge, CD28 TM, CD28z, CD3z) nucleotide sequence targeting KIT domain 4, variant I [ka]

[0158] SEQ ID NO:71 - Fab79D-CAR (CD28 hinge, CD28 TM, CD28z, CD3z) amino acid sequence targeting KIT domain 4, variant II (epitope binding regions in italics, CDRs in bold) [ka]

[0159] SEQ ID NO:72 - Fab79D-CAR (CD28 hinge, CD28 TM, CD28z, CD3z) nucleotide sequence targeting KIT domain 4, variant II [ka] [ka]

[0160] SEQ ID NO:73 - 4G8-CAR targeting FLT3 domain 4 (IgG4 hinge, CD28 TM, CD28z, CD3z) amino acid sequence (epitope binding regions in italics, CDRs in bold). [ka]

[0161] SEQ ID NO:74 - 4G8-CAR (IgG4 hinge, CD28 TM, CD28z, CD3z) nucleotide sequence targeting FLT3 domain 4 [ka]

[0162] SEQ ID NO:75 - CSL362-CAR (CD8a hinge, CD28 TM, CD28z, CD3z) amino acid sequence targeting the N-terminal domain of CD123 (epitope binding regions in italics, CDRs in bold) [ka]

[0163] SEQ ID NO:76 - CSL362-CAR (CD8a hinge, CD28 TM, CD28z, CD3z) nucleotide sequence targeting the N-terminal domain of CD123 [ka] [ka]

[0164] SEQ ID NO:77 - 4G8-CSL362-bispecific CAR (CD8a hinge, CD28 TM, CD28z, CD3z) amino acid sequence targeting both FLT3 and CD123, variant I (epitope binding regions in italics, CDRs in bold) [ka]

[0165] SEQ ID NO:78 - 4G8-CSL362-bispecific CAR targeting both FLT3 and CD123 (CD8a hinge, CD28 TM, CD28z, CD3z) nucleotide sequence, variant I [ka] [ka]

[0166] SEQ ID NO:79 - 4G8-CSL362-bispecific CAR (CD8a hinge, CD28 TM, CD28z, CD3z) amino acid sequence targeting both FLT3 and CD123, variant II (epitope binding regions in italics, CDRs in bold) [ka]

[0167] SEQ ID NO:80 - 4G8-CSL362-bispecific CAR targeting both FLT3 and CD123 (CD8a hinge, CD28 TM, CD28z, CD3z) nucleotide sequence, variant II [ka] [ka]

[0168] SEQ ID NO:86 - CSL362-CAR 2 targeting the N-terminal domain of CD123 nd Variant (IgG4 hinge, CD28 TM, CD28z, CD3z) amino acid sequences (CDRs in bold) [ka]

[0169] SEQ ID NO:87 - CSL362-CAR 3 targeting the N-terminal domain of CD123 rd Variant (IgG4 hinge, CD28 TM, CD28z, CD3z) amino acid sequences (CDRs in bold) [ka]

[0170] III. Methods of Treating Subjects The genetically engineered hematopoietic cells, such as HSCs, can be administered to a subject in need of treatment, alone or in combination with one or more cytotoxic agents that target one or more cell surface antigens described herein. Because the hematopoietic cells have been genetically edited in the genes of one or more cell surface antigens, the hematopoietic cells and / or their progeny will express one or more cell surface antigens in a mutant (e.g., functional) form. They can avoid targeting by cytotoxic agents, such as, for example, CAR-T cells.

[0171] Thus, the present disclosure provides a method for treating hematopoietic malignancies, comprising administering to a subject in need thereof (i) a population of genetically engineered hematopoietic cells as described herein, and optionally (ii) a cytotoxic agent, such as a CAR-T cell that targets a cell surface antigen, where the gene of the CAR-T cell that targets the cell surface antigen has been gene-edited in the hematopoietic cell such that the cytotoxic agent does not target the hematopoietic cell or its progeny. The administration of (i) and (ii) may be simultaneous or in any order. In some embodiments, the cytotoxic agent and / or the hematopoietic cells may be mixed with a pharmaceutically acceptable carrier to form a pharmaceutical composition, which is also within the scope of the present disclosure.

[0172] "Subject," "individual," or "patient" are used interchangeably herein and refer to a vertebrate, preferably a mammal, more preferably a human. Mammals include, but are not limited to, human primates, non-human primates, or murine, bovine, equine, canine, or feline species. In some embodiments, the subject is a human patient with a hematopoietic malignancy.

[0173] To carry out the methods described herein, an effective amount of genetically engineered hematopoietic cells can be administered to a subject in need of treatment. Optionally, the hematopoietic cells can be used in conjunction with a cytotoxic agent as described herein. As used herein, the term "effective amount" can be used interchangeably with the term "therapeutically effective amount" and refers to an amount of a cytotoxic agent, hematopoietic cell population, or pharmaceutical composition (e.g., a composition comprising a cytotoxic agent and / or hematopoietic cells). The term "effective amount" refers to an amount of a compound, cell population, or pharmaceutical composition sufficient to delay the onset of, halt the progression of, or alleviate or relieve at least one symptom of a disorder treated by the methods of the present disclosure. It is noted that when a combination of active ingredients is administered, the effective amount of the combination may or may not include the amount of each ingredient that was effective when administered individually.

[0174] The effective amount will be recognized by those skilled in the art to depend on the particular condition being treated, the severity of the condition, age, health, size, sex and weight, duration of treatment, nature of concomitant therapy (if any), the particular route of administration, and similar factors within the knowledge and expertise of the medical practitioner. In some embodiments, the effective amount results in palliating, alleviating, ameliorating, improving, reducing symptoms, or slowing the progression of any disease or disorder in the subject. In some embodiments, the subject is a human. In some embodiments, the subject is a human patient with a hematopoietic malignancy.

[0175] As described herein, the hematopoietic and / or immune cells expressing the chimeric receptor can be autologous to the subject. That is, the cells are obtained from a subject in need of treatment, engineered so that the cells do not bind to cytotoxic agents, and then administered to the same subject. The administration of autologous cells to a subject can result in reduced host cell rejection compared to the administration of non-autologous cells. For example, HSPCs are obtained from a biological sample from a subject, the HSPCs are genetically engineered, and the genetically engineered HSPCs are administered to the same subject. In some cases, the HSPCs are obtained from a biological sample, where the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood-derived CD34+ hematopoietic stem and progenitor cells.

[0176] Alternatively, the host cells are allogeneic cells. That is, the cells are obtained from a first subject, genetically engineered, and then administered to a second subject that is different from the first subject but of the same species. For example, allogeneic immune cells can be derived from a human donor and administered to a human recipient that is different from the donor. In some embodiments, the hematopoietic cells are further genetically engineered to reduce the host-versus-graft effect. For example, in some embodiments, the immune cells and / or hematopoietic cells can be subjected to gene editing or silencing methods to reduce or eliminate the expression of one or more proteins involved in the induction of the host immune response.

[0177] A typical amount of cells (i.e., immune or hematopoietic cells) administered to a mammal (e.g., human) is, for example, about 10 6 ~10 11 In some embodiments, the range may be 10 6 It may be desirable to administer fewer than 10 cells to a subject. In some embodiments, 11 It may be desirable to administer more than about 10 cells to a subject. In some embodiments, one or more doses of cells comprise about 10 6 ~ approx. 10 cells 11 cells, approximately 10 7 ~ approx. 10 cells 10 cells, approximately 10 8~ approx. 10 cells 9 cells, approximately 10 6 ~ approx. 10 cells 8 cells, approximately 10 7 ~ approx. 10 cells 9 cells, approximately 10 7 ~ approx. 10 cells 10 cells, approximately 10 7 ~ approx. 10 cells 11 cells, approximately 10 8 ~ approx. 10 cells 10 cells, approximately 10 8 ~ approx. 10 cells 11 cells, approximately 10 9 ~ approx. 10 cells 10 cells, approximately 10 9 ~ approx. 10 cells 11 cells, or approximately 10 10 ~ approx. 10 cells 11 Contains cells.

[0178] In some embodiments, the methods described herein include administering to a subject a population of genetically engineered hematopoietic cells and administering one or more immunotherapeutic agents (e.g., cytotoxic agents). As will be appreciated by one of skill in the art, the immunotherapeutic agents can be of the same type or different types (e.g., a therapeutic antibody, a population of immune cells expressing a chimeric antigen receptor(s), and / or an antibody-drug conjugate).

[0179] In some embodiments, the cytotoxic agent comprising an epitope-binding fragment that binds to an epitope of a cell surface protein (e.g., an immune cell expressing a chimeric receptor described herein) is administered prior to administration of the hematopoietic cells. In some embodiments, the agent comprising an epitope-binding fragment that binds to an epitope of a cell surface protein (e.g., an immune cell expressing a chimeric receptor described herein) is administered at least about 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 1 week, 2 weeks, 3 weeks, 4 weeks, 5 weeks, 6 weeks, 7 weeks, 8 weeks, 9 weeks, 10 weeks, 11 weeks, 12 weeks, 3 months, 4 months, 5 months, 6 months, or more prior to administration of the hematopoietic cells.

[0180] Alternatively, in some embodiments, the hematopoietic cells are administered prior to the cytotoxic agent comprising an epitope-binding fragment that binds to an epitope of a cell surface protein (e.g., immune cells expressing a chimeric receptor described herein). In some embodiments, the population of hematopoietic cells is administered at least about 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 1 week, 2 weeks, 3 weeks, 4 weeks, 5 weeks, 6 weeks, 7 weeks, 8 weeks, 9 weeks, 10 weeks, 11 weeks, 12 weeks, 3 months, 4 months, 5 months, 6 months, or more prior to administration of the cytotoxic agent comprising an epitope-binding fragment that binds to an epitope of a cell surface protein.

[0181] In some embodiments, the cytotoxic agent that targets a cell surface protein and the population of hematopoietic cells are administered substantially simultaneously. In some embodiments, the cytotoxic agent that targets a cell surface protein is administered, the patient is evaluated for a period of time, and then the population of hematopoietic cells is administered. In some embodiments, the population of hematopoietic cells is administered, the patient is evaluated for a period of time, and then the cytotoxic agent that targets a cell surface protein is administered.

[0182] Also within the scope of the present disclosure are multiple administrations (e.g., doses) of the cytotoxic agent and / or the population of hematopoietic cells. In some embodiments, the cytotoxic agent and / or the population of hematopoietic cells is administered to the subject once. In some embodiments, the cytotoxic agent and / or the population of hematopoietic cells is administered to the subject multiple times (e.g., at least 2, 3, 4, 5, or more times). In some embodiments, the cytotoxic agent and / or the population of hematopoietic cells is administered to the subject at regular intervals, for example, every 6 months.

[0183] Examples of routes of administration include intravenous, infusion, intradermal, subcutaneous, oral (eg, inhalation), transdermal (topical), transmucosal, and rectal administration.

[0184] Any of the methods described herein may be for the treatment of hematological malignancies in a subject. As used herein, the term "treat" or "treatment" or "treating" or "treating" refers to a therapeutic measure aimed at alleviating, slowing down the progression of, reducing symptoms of, and / or halting the progression of a pathological condition or disorder. Thus, a person in need of treatment includes a person who already has the disorder. For example, in some cases, treating cancer means stabilizing the progression of cancer. In some cases, treating cancer means slowing the progression of cancer. In some cases, treating cancer means stopping the progression of cancer. In some cases, treating cancer means reducing the size of cancer. In some cases, treating cancer means extending the overall survival of a subject diagnosed with cancer. Methods for assessing the progression of cancer are known in the art and include, for example, evaluation of target lesions using imaging (e.g., X-rays, computed tomography scans, magnetic resonance imaging, caliper measurements, or positron emission tomography scans). Cytology or histology, or expression of tumor marker(s) (see, e.g., Eisenhauer et al., 2009, European Journal of Cancer 45:228-247 and Schwartz et al., 2016, European Journal of Cancer 62:132-137, each of which is incorporated herein by reference in its entirety).

[0185] In some embodiments, the subject is a human subject with a hematopoietic malignancy. As used herein, hematopoietic malignancy refers to a malignant abnormality involving hematopoietic cells (e.g., blood cells, including progenitor cells and stem cells). Examples of hematopoietic malignancies include, but are not limited to, Hodgkin's lymphoma, non-Hodgkin's lymphoma, leukemia, or multiple myeloma. Exemplary leukemias include, but are not limited to, acute myeloid leukemia, acute lymphocytic leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia or chronic lymphoblastic leukemia, and chronic lymphocytic leukemia.

[0186] In some embodiments, the cells involved in the hematopoietic malignancy are resistant to conventional or standard therapeutic agents used to treat the malignancy. For example, the cells (e.g., cancer cells) can be resistant to chemotherapeutic agents and / or CAR T cells used to treat the malignancy.

[0187] In some cases, hematopoietic malignancies include: B-lymphoblastic leukemia (B-ALL), acute myeloid leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL), or blastic plasmacytoid dendritic cell leukemia (BPCDN).

[0188] IV. (Kits and Compositions) Any of the immune cells expressing the chimeric receptors described herein can be administered as a pharmaceutical composition in a pharma- ceutically acceptable carrier or excipient.

[0189] The phrase "pharmacologically acceptable" when used in connection with compositions and / or cells of the present disclosure refers to molecular entities and other components of such compositions that are physiologically tolerated and do not normally produce untoward reactions when administered to a mammal (e.g., human). Preferably, as used herein, the term "pharmacologically acceptable" means approved by a federal or state regulatory agency or listed in the United States Pharmacopeia or other generally recognized pharmacopoeias for use in mammals, more specifically humans. "Acceptable" means that the carrier is compatible with the active components of the composition (e.g., nucleic acid, vector, cell, or therapeutic antibody) and does not adversely affect the subject to which the composition(s) are administered. Any of the pharmaceutical compositions and / or cells used in the present methods may include pharmaceutically acceptable carriers, excipients, or stabilizers, in the form of a lyophilized formulation or aqueous solution.

[0190] Pharmaceutically acceptable carriers, including buffers, are well known in the art and may include phosphates, citrates, and other organic acids. Antioxidants, including ascorbic acid and methionine, preservatives, low molecular weight polypeptides, proteins such as serum albumin, gelatin, or immunoglobulins; amino acids, hydrophobic polymers, monosaccharides, disaccharides, and other carbohydrates, metal complexes, and / or non-ionic surfactants. See, for example, Remington: The Science and Practice of Pharmacy 20th Ed. (2000) Lippincott Williams and Wilkins, Ed. KE Hoover.

[0191] Also within the scope of the present disclosure is a kit for use in treating hematopoietic malignancies. Such a kit may include genetically engineered hematopoietic cells, such as HSPCs, and optionally one or more cytotoxic agents that target cell surface antigens, the genes of which have been edited in the hematopoietic cells. Such a kit may include a container that includes a first pharmaceutical composition that includes any of the genetically engineered hematopoietic cells described herein, and optionally one or more additional containers that include one or more cytotoxic agents that target cell surface antigens also described herein (e.g., immune cells expressing a chimeric receptor described herein).

[0192] In some embodiments, the kit may include instructions for use in any of the methods described herein. The included instructions may include instructions for administering the engineered hematopoietic cells and, optionally, instructions for administering one or more cytotoxic agents to the subject to achieve an intended activity in the subject. The kit may further include instructions for selecting a suitable subject for treatment based on identifying whether the subject is in need of treatment. In some embodiments, the instructions include instructions for administering the engineered hematopoietic cells and, optionally, one or more cytotoxic agents to a subject in need of treatment.

[0193] The instructions associated with the use of the genetically engineered hematopoietic cells and optionally cytotoxic agents described herein will generally include information regarding the dose, dosing regimen, and route of administration for the treatment of interest. The containers may be unit doses, bulk packages (e.g., multi-dose packages), or subunit doses. The instructions accompanying the kits of the present disclosure are typically instructions on a label or package insert. The label or package insert indicates that the pharmaceutical composition is used for treating, delaying the onset of, and / or ameliorating a disease or disorder in a subject.

[0194] The kit provided herein is suitable packaging. Suitable packaging includes, but is not limited to, vials, bottles, jars, flexible packaging (e.g., sealed Mylar or plastic bags), etc. Packages for use in combination with certain devices, such as inhalers, nasal administration devices, or infusion devices, are also contemplated. The container holds or contains the formulation and can have a sterile access port (e.g., the container can be an intravenous solution bag or a vial with a stopper that can be punctured by a hypodermic needle). The container can also have a sterile access port. At least one active agent included in the pharmaceutical composition is the chimeric receptor variant described herein.

[0195] The kit may optionally provide additional components such as buffers and interpretive information. Typically, the kit comprises a container and a label or package insert(s) on or associated with the container. In some embodiments, the disclosure provides an article of manufacture comprising the contents of the kit described above. EXAMPLES

[0196] The following specific examples are for the purpose of illustrating the invention and should not be construed as limiting the scope of the claims. To the extent that specific materials are mentioned, they are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art will be able to readily develop equivalent means or reactants that do not depart from the scope of the invention by using ordinary inventive skills.

[0197] Example 1: General protocol used in the following examples Plasmid cloning WT-Cas9 and NG-Cas9 base editor plasmids were obtained from Addgene (plasmid numbers 138495, 138491). SpRY-ABE8e-V106W 3xNLS and other base editor variants (see full list below) were cloned using NEB HiFi Assembly Master Mix and synthesized dsDNA inserts (IDT gBlocks). Single amino acid changes (i.e., K918N) or deletions (Blackjack variants) were introduced by standard site-directed mutagenesis techniques. Where appropriate, sgRNAs were cloned into a pLentiguide-Puro backbone (Addgene) or pLKO-mTagBFP2 backbone (Cloned) using BsmBI restriction enzyme and annealed and phosphorylated DNA oligos containing the desired spacer sequences. Plasmid maxipreps were purified using a Mackarey Nagel NucleoBond Xtra Maxi kit. [Table 3-1] [Table 3-2]

[0198] Flow cytometry ligand affinity assay A fluorescent ligand binding assay was developed to assess the binding affinity of mutant receptors to their ligands. Human SCF and FLT3L (Peprotech) were conjugated with Alexa Fluor 488 antibody labeling kit (Invitrogen Cat A20181) according to the manufacturer's recommendations. Cells expressing either FLT3 or KIT variants were incubated with FcR blocking reagent (Miltenyi 130-059-901), LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967), control antibodies (KIT 104D2 PE-Cy7, Biolegend, or FLT3 BV10A4 PE-Cy7, Biolegend) and the respective AF488-conjugated ligands for 15 min at room temperature. Samples were washed with PBS+2% FBS and analyzed on a BD Fortessa flow cytometer.

[0199] Reporter cell lines for FLT3 and CD123 editing Because K562 cells do not constitutively express FLT3 or CD123, these cells were engineered to overexpress FLT3, CD123, KIT, or any combination of these three genes from their endogenous genomic loci to serve as models for gene editing approaches to these genes. To obtain constitutive and robust expression of these genes, a fully human EF1 alpha promoter was integrated upstream of the transcription start site of the FLT3, CD123, or KIT gene via CRISPR-Cas9 (FLT3, KIT) or CRISPR-AsCas12a (CD123) gene editing strategies (Figure 8). A dsDNA linear donor with 50 bp long homology arms of the target region was prepared by PCR on a Sleeping Beauty plasmid encoding the EF1 alpha promoter. Wild-type K562 cells were electroporated using the Lonza 4D-Nucleofector system with Cas9 (FLT3, KIT) or Cas12a (CD123) RNP with gRNA targeting the promoter region of the respective gene. 5-10ug of dsDNA linear donor with matched homology arms was included in the electroporation reaction to serve as a template for integration by homology-directed repair. Cells were stained with FLT3 BV10A4 PE-Cy7 (Biolegend), KIT104D2 PE-Cy (Biolegend) or CD123 9F5 BV421 and CD123 7G3 BV711 (BD) and assessed by flow cytometry. FLT3, CD123, or KIT-high populations were sorted by FACS and single clones were isolated by limiting dilution (Figure 8). Clones with the highest MFI by flow cytometry for FLT3 (clone AH11) and CD123 (clone 3D5) were selected and expanded.

[0200] Western blot K562 cells overexpressing FLT3 / CD123 variants or NIH-3T3 or HEK-293T cells overexpressing KIT were cultured overnight in medium without FBS (serum starvation), after which one million cells from each condition were stimulated with or without 100 ng / mL FLT3L, IL3, or SCF (depending on the receptor evaluated) for 5 min at 37 °C. Cells were then washed three times with ice-cold PBS, lysed, and proteins were extracted for Western blot (Cell Extraction Buffer, ThermoFisher FNN0011 + 1 mM PMSF). After quantification with the BCA assay, protein extracts were mixed with Laemmli loading dye (Biorad 161-0747) and run on Novex Tris-Glycine Gels (Invitrogen). After transfer, membranes were blocked with 5% w / v BSA in TBST and incubated with primary antibodies recognizing pKIT (Y719, Cell Signaling 3391T) or pFLT3 (Tyr589 / 591 clone 30D4, Cell Signaling 3464S) overnight at 4°C. After washing, membranes were incubated with anti-rabbit IgG, HRP-conjugated antibody (Cell Signaling 7074) for 1 h at room temperature, followed by incubation with SuperSignal West Femto chemiluminescent HRP substrate (Thermo Scientific 34096) and analysis on an ImageQuant LAS4000. The same membranes were then incubated with Restore (ThermoScientific #21059) stripping buffer for 20 min at room temperature and secondary staining with anti-KIT (clone 1C5, Invitrogen MA5-15894) or anti-FLT3 (clone OTI7D6, Origene TA808157) primary antibodies. Secondary Ab staining was performed with anti-rabbit or anti-mouse IgG HRP-conjugated Abs according to the primary Abs, and membranes were developed and acquired as above. Anti-actin staining was used as an internal control for loading normalization.

[0201] Flow cytometry analysis (in vitro experiments) Edited cell lines were assessed by flow cytometry 72 hours after editing and stained with antibody clones binding either the therapeutic epitope or an irrelevant epitope to serve as control antibodies for surface expression of the edited protein. For FLT3 editing, cells were incubated with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, FLT3 BV10A4 PE-Cy7 2 / 100 (Biolegend313314) and FLT3 4G8 BV711 (BD 563908). For CD123 editing, cells were incubated with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, CD123 9F5 PE or BV421 (BD 555644) 1 / 100, CD123 7G3 BV711 or BV421 (BD 740722) 1 / 100. For KIT editing, cells were incubated with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, KIT 104D2 PE-Cy7 or BV711 2 / 100 (Biolegend 313212), KIT Fab79D AF488 or PE (Creative Biolabs) 2 / 100. Staining was performed at 4°C for 30 minutes in a volume of 100uL / sample. To assess the stem cell phenotype of cultured human CD34+ HSPCs, cells were harvested, resuspended in 100uL PBS, and stained with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, CD34 BV421 (Biolegend) 1.5 / 100, CD90 APC (BD) 3.5 / 100, CD45RA APC-Cy7 (Biolegend) 3.5 / 100, and CD133 / 2 PE (Miltenyi) 4 / 100. Staining was performed at 4°C for 30 minutes in a volume of 100uL / sample.Samples were analyzed on a four- or five-laser BD Fortessa flow cytometer.

[0202] Flow cytometry analysis (in vitro experiments) Peripheral blood from xenografted NSG mice was collected in 1.5 mL Eppendorf tubes, spiked with 10 uL of 0.5 M EDTA and stained for 15 min at room temperature with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100 uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, human CD45 BV786 (Biolegend) 1.5 / 100, mouse CD45 BV570 (Biolegend), CD34 BV421 (Biolegend) 1.5 / 100, CD19 BV650 (Biolegend) 3 / 100, CD3 BV711 (Biolegend) 2 / 100, and CD33 BB515 (BD) 2 / 100. Blood samples were then lysed with ACK Reagent (StemCell technologies) for 5 min at room temperature and washed twice with PBS + 2% FBS. Samples were analyzed on a 4-laser BD Fortessa flow cytometer.

[0203] Bone marrow from xenograft mice was obtained by crushing hind leg bones with a mortar, filtering through a 40 um cell strainer, and resuspending the cells in Miltenyi MACS running buffer. A fraction of cells was diluted with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, 7-AAD (BD Pharmigen) 3 / 100, human CD45 BV786 (Biolegend) 1.5 / 100, mouse CD45 BV570 (Biolegend), CD34 BV421 (Biolegend) 1.5 / 100, CD19 BV650 (Biolegend) or CD19 BV605 (Biolegend) 3 / 100, CD3 BV711 (Biolegend) or CD3 PE-Cy7 (Biolegend) 2 / 100, CD33 BB515 (BD), CD38 BV480 (BD) 1.5 / 100 or CD38 BUV396 (BD) 2 / 100, FLT3 Staining was performed with BV10A4 PE-Cy7 (Biolegend), CD123 9F5 PE (BD), CD90 APC (BD) 3.5 / 100, and CD45RA APC-Cy7 (Biolegend) 3 / 100 for 30 minutes at 4° C. Spleens were crushed on a 40 μm cell strainer and resuspended in Miltenyi MACS running buffer. Harvested cells were stained with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, human CD45 Pacific Blue (Biolegend) 1.5 / 100, mouse CD45 BV570 (Biolegend), CD3 BV711 (Biolegend) 2 / 100, EGFR AF488 (R&D) 1.5 / 100, CD62L PE (BD) 2 / 100, CD4 APC (BD) 2 / 100, CD8 BV750 (Biolegend) 2 / 100, CD45RA APC-Cy7 (Biolegend) 3 / 100, CD69 PerCP-Cy5.5 (Biolegend) 3 / 100 for 30 minutes at 4°C. Samples were analyzed on a four-laser BD Fortessa flow cytometer.In some experiments, BM cells were stained with either hCD45 BV786, mCD45 PerCP-Cy5.5, CD3 PE-Cy5, CD7 AF700, CD10 BUV737, CD11c BUV661, CD14 BV510, CD19 BV605, CD33 PE-Cy7, CD38 BUV396, CD45RA APC-Cy7, CD56 BUV496, CD90 APC, FLT3 PE or BV711, and KIT BV711 or CD123 PE antibodies, adding 50uL / sample of Brilliant Stain Buffer (BD Catalog No. 659611).

[0204] gDNA extraction, PCR amplification, and Sanger sequencing Genomic DNA was extracted from dried pellet samples using Qiagen DNeasy Blood&Tissue Kit or Lucigen QuickExtract reagent, quantified by Nanodrop 8000, and sequences of interest were amplified by PCR using Promega GoTaq G2 and respective primers. PCR products were purified using Promega SV Wizard Gel and PCR cleanup system and sent for Sanger sequencing via Genewiz. Base editing efficiency was calculated from Sanger traces by deconvolution with EditR R package using custom scripts for high-throughput sample analysis.

[0205] Colony formation assay Colony forming unit assays (CFU) were performed by plating 1000CD34+ cells / well for in vitro CD34+ HSPC experiments or 25000 total bone marrow cells / well for xenograft BM-derived assays unless otherwise stated. Cells were resuspended in Methocult H4034 medium (StemCell Cat. No. 04034) and plated in SmartDish meniscus-free 6-well plates. Wells were imaged after 2 weeks using the StemCell STEMvision system. For flow cytometry analysis, methylcellulose medium was softened with warm PBS, collected and washed twice before analysis.

[0206] Example 2: Design of receptor variants To design mutant receptor variants that are not recognized by the selected monoclonal antibody clones, epitopes targeted by different mAbs were identified according to available information in the literature or by screening individual mutant receptor variants or mutation libraries. The overall goal was the identification of minimally modified target variants that lack recognition by the selected therapeutic antibody clones but preserve the functions of surface expression, gene regulation, and signaling (Figure 1). Cells bearing such modified surface targets were conferred selective resistance when exposed to immunotherapies against the target molecules, including but not limited to naked monoclonal antibodies, toxin-conjugated antibodies, bispecific antibody constructs, and chimeric antigen receptor cells.

[0207] FLT3 epitope engineering: FLT3 is a type III tyrosine kinase receptor characterized by the presence of five immunoglobulin-like domains, constituted by i) an extracellular ligand-binding domain, ii) a single spanning transmembrane region, and iii) an intracellular portion containing a split tyrosine kinase domain. The first three extracellular domains are involved in binding with its dimeric ligand, FLT3 ligand (FLT3L), and this interaction induced dimerization of the receptor. After dimerization, activation of FLT3 is mediated by the positioning of the intracellular tyrosine kinase domains in close proximity to each other, which promotes their subsequent transphosphorylation.

[0208] Alignment shows that mouse and human FLT3 / FLT3L are 85.5% identical at the amino acid level, and that IgG-like domain 4 of mouse and human FLT3 is 82% identical (95.5% similar) at the amino acid level (Figure 3). Furthermore, mouse and human FLT3 / FLT3L pairs are cross-reactive with each other. Anti-human FLT3 therapeutic antibody clone 4G8 specifically recognizes human FLT3, with the epitope localized within extracellular domain 4. Clone 4G8 was confirmed to not recognize other orthologous variants (i.e., mouse Flt3) by staining K562 cells overexpressing wild-type human or mouse FLT3 via the Sleeping Beauty transposon system, in which the cDNA sequence of either human FLT3 or mouse Flt3 was cloned downstream of a constitutive EF1a promoter (Figure 2). Transduced cells were identified using the mCherry fluorescent reporter encoded by the Sleeping Beauty transposon. Based on this evidence, 16 residues located in extracellular domain 4 that are relatively unconserved across orthologous sequences (Figure 3) were replaced with the corresponding mouse amino acids, and the resulting variant (eFLT3-01, SEQ ID NO: 49) was not recognized by clone 4G8 (Figure 2). The resulting variant still maintained surface expression (confirmed using control anti-FLT3 clone BV10A4, which binds to extracellular domain 2), FLT3L binding (see Figures 4A and 4B), and intracellular signaling properties (see Figure 4C).

[0209] As a second step, the 16 mutations introduced to generate eFLT3-01 (SEQ ID NO: 49) were separated into two mutant pools by genomic localization within either FLT3 exon 9 or exon 10 (together encoding FLT3 extracellular domain 4). Overexpression of these two FLT3 variants showed that mutations restricted to FLT3 exon 9 (SEQ ID NO: 50) were sufficient to abolish the anti-FLT3 clone 4G8, similar to SEQ ID NO: 49. FLT3 variants with mutations restricted to exon 10 did not abolish the binding of clone 4G8.

[0210] To identify the key residues involved in the recognition of FLT3 extracellular domain 4 by the anti-FLT3 antibody FLT3 4G8 clone, a combinatorial library was designed containing all 16 residues previously identified, either wild type or mutant, and cloned into the Sleeping Beauty transposon transfer vector under the EF1 alpha promoter (GeneScript-Figure 5). An antisense cassette expressing mCherry resistance and puromycin resistance under the RPBSA promoter served as a transduction marker and selection method. K562 cells were electroporated using the Lonza 4D-Nucleofector system with 5, 10, or 100 ng of transfer vector and 500 ng of SB100x transposase expression plasmid. Nucleofected cells were selected with 1ug / mL puromycin for 7 days and then evaluated by flow cytometry. Cells positive for control antibody (FLT3 BV10A4 PE-Cy7) and negative for therapeutic antibody (FLT3 4G8 BV711) were FACS sorted on a BD Aria II sorter and expanded in vitro. Integrated FLT3 library regions were PCR amplified from gDNA samples of unsorted, single-positive sorted, and double-positive sorted cells, partial Illumina adapters were added to the amplicons, and subjected to NGS on the Illumina platform (GeneWiz). By comparing the three samples, only one codon (N399) showed differential mutation enrichment in single- and double-positive cells (Figure 5). Validation of the candidate mutation (N399D) in a Sleeping Beauty overexpression system confirmed that it was sufficient to abolish 4G8 clonal binding (Figure 5). Incubation of K562 cells overexpressing the FLT3 N399D variant with AF488-conjugated fluorescent FLT3L demonstrated ligand binding comparable to wild-type FLT3 (Figure 5).

[0211] KIT epitope engineering: Based on the available literature, the anti-KIT antibody clone Fab79D recognizes putative amino acid contact points in the extracellular domain 4 of KIT (Figure 3). Therefore, we generated a KIT variant (eKIT-01) that is not recognized by the anti-KIT antibody clone Fab79D by replacing 10 of the predicted contact amino acids with less conserved ones across the orthologous sequence (Figure 6A). This variant maintained human stem cell factor (SCF) affinity (Figure 6B) and phosphorylation of the intracellular domain upon SCF stimulation (Figure 6C). To identify the specific amino acids underlying clone Fab79D binding, each single amino acid change was individually cloned into a Sleeping Beauty transfer vector co-expressing mCherry and puromycin resistance. 293T cells were electroporated with 100 ng of the transfer vector and 500 ng of a plasmid expressing SB100x transposase using the Lonza 4D-Nucleofector system in SF solution (Figure 6D). Cells were selected with puromycin (2ug / mL) and stained with KIT Fab79D and KIT 104D2 control antibodies and analyzed by flow cytometry. The mutation with the lowest Fab79D MFI compared to KIT expression by clone 104D2 staining (KIT H378R) was selected as a candidate variant for developing base editing strategies (Figure 6D).

[0212] To expand on these findings, we designed a comprehensive library approach to further define alternative codons involved in Fab79D binding. A degenerate library, in which each codon in the KIT extracellular domain 4 was composed of a degenerate base (NNN), was cloned into a Sleeping Beauty transfer plasmid expressing human KIT cDNA, mTagBFP reporter, and puromycin resistance (Figure 7). HEK-293T cells were electroporated with the library plasmid and pSB100X transposase plasmid to allow stable integration of the transgene. After puromycin selection, cells were FACS sorted to obtain single- and double-positive populations. Library regions were PCR amplified and NGS sequenced. Comparison of the relative enrichment of amino acid variants in the single-positive population versus the double-positive population identified additional codons involved in Fab79D binding in addition to H378 (M318, I319, V323, D332, E360, Y362, E376).

[0213] CD123 epitope engineering: For CD123 (CD123), it has been reported in the literature that the anti-CD123 antibody clone 7G3 and its humanized version CSL362 recognize putative amino acid contact points within the N-terminal domain of CD123 (Figure 9). Similarly, other commercially available clones, 6H6 and S18016E, have been reported to bind to the N-terminal domain, but information on their contact points with CD123 is limited. IL3 is a ligand for CD123, and CD123 residues important for IL3 binding have been mapped by Ala scan or evolutionarily conserved amino acid changes (Figure 9).

[0214] To engineer target epitopes within the CD123 extracellular domain, we used a direct base editor screening approach by designing a set of sgRNAs targeting I50, E51, Y58, S59, R84, P88, or P89 in the N-terminal domain of CD123 (Figure 10). These gRNAs were tested in K562 cells overexpressing CD123 together with cytidine or adenine base editors (evo-APOBEC1-BE4, ABE8e-V106W). Briefly, 0.5 M cells were electroporated in SF solution using the Lonza 4D-Nucleofector system according to the manufacturer's instructions. The electroporation reaction contained 500 ng of base editor expression plasmid and 300–360 pmol of sgRNA (Integrated DNA Technologies). Cells were then cultured and samples for genomic DNA and flow cytometry analysis were harvested 72 h after editing. sgRNAs that generated CD123 variants that lacked recognition by several mAb clones (7G3 / CSL362, 6H6, S18016F) but retained surface expression by staining with the control antibody clone 9F5 were selected for further development (Figure 10, only the best performing candidate gRNAs are reported below: SEQ ID NO:24, SEQ ID NO:30, SEQ ID NO:32, SEQ ID NO:34).

[0215] A second round of sgRNA screening was performed, including additional sequences adjacent to the sgRNAs identified in the previous round, and several BEs, including variants with relaxed PAM specificity, were tested (evo-APOBEC1-BE4, NG-EA-BE4max, NG-A3A-BE5, SpRY-evo-APOBEC1-BE4-SEQ ID NO:8 and SEQ ID NO:11, NG-ABE8e, SpRY-ABE8e-V106E-SEQ ID NO:7 and SEQ ID NO:10, LbCas12a-ABE8e-Figure 11). Flow cytometric assessment of binding of clones 7G3, 6H6 and S18016E was performed as previously described, using clone 9F5 as a normalizer for surface expression. gDNA was extracted from conditions showing the desired loss of recognition and Sanger sequenced the CD123 exon 2 and 3 regions to obtain amino acid changes associated with the observed phenotype. For antibody clone 7G3 or its humanized counterparts, CSL362, CD123 S59P (introduced by gRNA-N, gRNA-R, and related variants with ABE) and S59F (introduced by variants with gRNA-L and CBE) were the best candidate variants. The Y58H mutation was also introduced by gRNA-N and gRNA-R by converting the bystander adenine from A to G. To abolish the binding of antibody clones 6H6 and S18016F, P88L / P89L (introduced by gRNA-F and related variants with CBE) was the best performing amino acid substitution. Additional gRNAs (gRNA-H and related variants) that introduced a pool of other amino acid substitutions (R84Q, V85M, V85I, A86T) by cytidine base editing were identified, resulting in reduced affinity for clones 7G3, 6H6, and S18016F (Figure 11).

[0216] Example 3: FLT3 variants are resistant to FLT3-targeted CAR-T cells To test whether FLT3 variants were selectively resistant to FLT3-targeting CAR-T cells, an in vitro killing assay was performed.

[0217] CAR-T cell production A generation III lentiviral construct expressing a second generation FLT3-specific 4G8 cloned chimeric antigen receptor with CD28 transmembrane region and CD28 costimulatory domain under the constitutive hPGK promoter was cloned using a synthetic dsDNA fragment (IDT gBlocks). An antisense cassette expressing a truncated variant of human EGFR cDNA under a minimal CMV promoter was included to serve as a marker for transduction and a safety switch for in vivo depletion using the anti-EGFR antibody cetuximab. VSV-G pseudotyped self-inactivating lentiviral particles were prepared by calcium-phosphate transient co-transfection of five plasmids (transfer vector, pMD2, pMDL-RRE, pREV and pAdvantage plasmid) in HEK-293T cells according to published methods. The viral particle-containing supernatant was concentrated 500-fold by ultracentrifugation (20000 rpm, 2 h at 20°C) and resuspended in PBS. The concentrated LVs were titrated by transducing 293T cells at different concentrations and calculating the transduction efficiency by flow cytometry or ddPCR.

[0218] Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood by ficoll gradient centrifugation. After estimation of T cell percentage by flow cytometry, either freshly isolated or thawed PBMCs were incubated with CD3-CD28 Dynabeads (Gibco 11131D) at a bead:T cell ratio of 3:1 for 45 min at room temperature with gentle agitation and then magnetically separated (DynaMag-5 magnetic, Invitrogen 12303D). Positively selected cells were cultured with 1 M / mL Dynabeads in IMDM supplemented with 10% FBS, 1% P / S, human IL-7 5 ng / mL (Peprotech), and human IL-15 5 ng / mL (Peprotech). 48 h after the start of Dynabeads stimulation, T cells were transduced with lentiviral particles encoding the CAR of choice at MOI 5 to MOI 10 depending on the experiment. On day 7 from the start of stimulation, Dynabeads were removed from the cultures by magnetic separation, and T cells were expanded for an additional 5–7 days in IMDM supplemented with 10% FBS, 1% P / S, human IL-7 5 ng / mL (Peprotech), and human IL-15 5 ng / mL (Peprotech). T cell phenotype and transduction efficiency (by EGFR surface staining) were assessed periodically by flow cytometry. Expanded CAR-T cells or untransduced T cells were used for either killing assays, in vivo dosing, or immediately frozen 12–14 days after the start of stimulation.

[0219] K562 cells (either unmodified, base edited, or overexpressing receptor variants after Sleeping Beauty transduction) were plated in 96-well plates (25000 target cells / well). Anti-FLT3 CAR-T cells were generated by transducing peripheral blood mononuclear cells (PBMCs) with a lentiviral vector encoding a second-generation 4G8-CAR construct bearing a CD28 costimulatory domain and co-expressing an EGFRt (truncated EGFR) safety switch (protocol above). FLT3-targeted 4G8 CAR-T cells or untransduced T cells were marked with CellTrace yellow (Invitrogen C34567) according to the manufacturer's recommendations. FLT3-targeted 4G8 CAR-T cells or untransduced T cells were then co-plated into the same wells at different effector:target ratios (E:T ratios), typically 10, 5, 2.5, 1.25, and 0.625, and incubated at 37°C in a humidified incubator with 5% CO2. After 4 hours, 50% of the culture volume was harvested for flow cytometry analysis by staining with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, FLT3 BV10A4 PE-Cy7 2 / 100 (Biolegend 313314), CD4 APC (BD) 2 / 100, CD8 BV750 (Biolegend) 2 / 100, CD107a BV711 (Biolegend) 2 / 100, CD69 PerCP-Cy5.5 (Biolegend) 2 / 100. The staining mix included flow counting beads (Biolegend Precision Count Beads) to normalize cell numbers. Cells were then washed and resuspended in Annexin V binding buffer (Biolegend) supplemented with Annexin V FITC (Biolegend) 3 / 100. Samples were analyzed on a 4- or 5-laser BD Fortessa flow cytometer.The remaining culture volume was assessed 48 hours after plating by flow cytometry stained with FcR blocking reagent (Miltenyi 130-059-901) 2 / 100uL, LIVE / DEAD fixable yellow dead cell stain (Invitrogen L34967) 1 / 1000, FLT3 BV10A4 PE-Cy7 2 / 100 (Biolegend 313314), CD4 APC (BD) 2 / 100, CD8 BV750 (Biolegend) 2 / 100, CCR7 BV421 (Biolegend) 2 / 100, CD45RA APC-Cy7 (Biolegend) 3 / 100, CD33 PerCP-Cy5.5 (Biolegend). Cells were then washed and resuspended in Annexin V binding buffer (Biolegend) supplemented with Annexin V FITC (Biolegend) 3 / 100.

[0220] Figure 12A shows the experimental design of the co-culture killing assay to evaluate the resistance of modified FLT3 variants to CAR-T cell-mediated killing. Figure 12B is a flow cytometry plot showing the co-culture composition of live cells after 4 and 48 hours of incubation. The top row shows FLT3 expression by flow cytometry in WT or engineered K562 cells. Cells expressing unmodified wild-type FLT3 are selectively killed, while engineered FLT3 (e9-FLT3) is spared. Figure 12C shows target cell viability by AnnexinV and LiveDead yellow staining after 4 hours of incubation. Figure 12D shows selective T cell degranulation by CD107a surface staining at 4 hours only for FLT3-CAR-T exposed to target cells expressing unmodified FLT3. Figure 12E shows selective T cell proliferation by dye dilution (CellTrace yellow) after 48 hours of co-culture of FLT3-CAR-T alone exposed to target cells expressing unmodified FLT3. Unmodified K562: wild type cells; WT FLT3 OE: K562 reporter cell line with FLT3 overexpression from the endogenous promoter (see Reporter Cell Line Generation); "WT FLT3 OE (Sleeping Beauty)" and "e9 FLT3 OE (Sleeping Beauty)": K562 cells transduced with Sleeping Beauty transposon driving overexpression of FLT3 variants. The e9-FLT3 variant had 12 amino acid changes compared to WT-FLT3, including N399D (SEQ ID NO: 50).

[0221] Example 4: Homology-directed repair to introduce N399D into FLT3 To test whether FLT3 N399D could be introduced by homology-directed repair, a CRISPR Cas homology-directed repair (HDR) strategy was designed. SpCas9 or AsCas12a nucleases combined with several gRNAs targeting the FLT3 exon 9 locus were tested in combination with 200 nt long single-stranded oligodeoxynucleotides (ssODNs) as donor templates for HDR. Each donor template contains silent mutations selected in bystander amino acids to reduce the risk of re-cutting by the CRISPR-Cas9 RNP nuclease complex after successful DNA repair. The reverse complement of each ssODN donor (referred to as A, C, H, F) was also tested. The sequences of the ssODN template donors are reported in SEQ ID NOs: 40-43 (see Table 2 above). K562 reporter cells overexpressing FLT3 by targeted integration of the EF1-a promoter upstream of the endogenous FLT3 locus were electroporated using the Lonza 4D-Nucleofector system in SF solution supplemented with 50 pmol of 3xNLS Cas9 nuclease (IDT) complexed with 62.5 pmol of annealed trRNA:gRNA or 50 pmol of AsCas12a Ultra (IDT) complexed with 62.5 pmol of sgRNA according to experimental conditions. Cas9 gRNA e9-4-NGG was tested in combination with ssODN-A and C and their reverse complements (5 uM final concentration). Cas12a gRNA e9-15-TTTV and e9-16-TTTV were tested in combination with ssODN-H and F and their reverse complements (5 uM final concentration). IDT HDR Enhancer 0.2uL / 20uL was included in the electroporation reaction (30uM). The results of the editing procedure were evaluated by flow cytometry 72 hours after electroporation. Figure 13A (top) shows exon 9 of FLT3 with the N399 residue highlighted and its relative position with respect to three gRNAs (one SpCas9 gRNA (e9-4-NGG) and two Cas12a gRNAs (e9-15-TTTV and e9-16-TTTV)). The FLT3 exon 9 targeting gRNAs are reported in SEQ ID NOs: 13-16 (see Table 1 above).Figure 13A (bottom) shows the single-stranded oligo-deoxynucleotide donor template utilized to insert the N399D (arrow) mutation by homology-directed repair. An additional silent single nucleotide change was included in the template design to reduce the rate of re-cutting by the Cas9-gRNA complex (dark square).

[0222] Figure 13B shows a FACS plot 72 hours after electroporation of K562 reporter cells stained with FLT3 clone 104D2 as a normalizer for surface expression and clone 4G8 to assess the efficiency of the N399D mutation. Successfully edited cells are highlighted with a black rectangle, indicating that the N399D mutation can be inserted into human cell lines via CRISPR-Cas homology-directed repair resulting in loss of recognition by mAb clone 4G8.

[0223] Example 5: FLT3 mutation at position N399 can be introduced by CRISPR adenine base editor CRISPR-Cas base editing was tested to introduce the desired single codon changes at the FLT3 locus with high efficiency and low toxicity without introducing double-stranded DNA breaks. A panel of sgRNAs predicted to introduce N399D or N399G mutations in combination with an adenine base editor was designed. The CRISPR-Cas9 base editor ABE8e (TadA-8e V106W) was selected for development of the editing strategy and further optimized by relaxing PAM specificity by mutating the Cas9 nickase protein to enable editing in the absence of the traditional NGG PAM. To this end, both NG-SpCas9 and SpRY-Cas9 variants of the base editor were cloned. To further increase efficiency, a third nuclear localization site (NLS) was fused to the C-terminal part of the protein. Unless otherwise stated, base editing experiments were performed by electroporation of reporter K562 overexpressing the FLT3 gene with 500 ng of base editor plasmid and either 300-360 pmol of sgRNA (Integrated DNA Technologies). Cells were then cultured and samples for genomic DNA and flow cytometry analysis were harvested 72 hours after editing.

[0224] Figure 14 shows a representative experiment illustrating the PAM requirement for N399 base editing. Two SpCas9 sgRNAs with NG PAM (SEQ ID NO: 17 and SEQ ID NO: 18, see Table 1 above) were tested with WT-SpCas9 nickase (PAM requirement: NGG), NG-SpCas9 nickase (PAM requirement: NG), and SpRY-Cas9 nickase (PAM requirement: NRN), ABE constructs. Figure 14A shows the relative position of the sgRNA and the N399 codon in the FLT3 gene. Figure 14B, gRNA protospacer sequence with the target adenine underlined. Figure 14C shows the design of three adenine base editor constructs (wild-type Cas9, NG-Cas9, or SpRY-Cas9 variants linked to the TadA deaminase domain with or without the V106W mutation). The V106W mutation is associated with reduced RNA editing, a known undesirable effect of adenine base editors. ABE8e-V106W and NG-ABE8e are available through Addgene (catalog product numbers 138495 and 138491, respectively). Cloned SpRY-ABE8e-V106W (SEQ ID NO: 7 and SEQ ID NO: 8) contained an additional C-terminal nucleoplasmin nuclear localization sequence (NLS). A FACS plot of the base editing results 72 hours after electroporation is presented in Figure 14D. Here, K562 cells were stained with anti-FLT3 clones BV10A4 (normalization) and 4G8 (therapeutic Ab). The percentage of edited cells is reported at the bottom right of the FACS plot for each condition. Only base editor variants with relaxed PAM specificity (NG-ABE8e and SpRY-ABE8e-V106W) showed efficient editing, with up to 41.1% of cells becoming negative for clone 4G8 staining.

[0225] To improve the efficiency of base editing of the N399 codon, the location of the desired editing window was achieved by screening additional sgRNAs in combination with the nearly PAM-free SpRY-Cas9 variant. Figure 15A shows the genomic context of FLT3 N399 and its location relative to the five sgRNAs. The sgRNAs include two sgRNAs from the previous example (SEQ ID NO: 17 and SEQ ID NO: 18), as well as three additional gRNAs with NRN PAM (SEQ ID NO: 19, SEQ ID NO: 20, and SEQ ID NO: 23). Figure 15B shows the sgRNA protospacer sequence with the target adenine underlined. Figure 15C is a FACS plot of a base editing experiment by electroporation of 500 ng of base editor expression plasmid and 360 pmol of sgRNA into K562 cells. The percentage of edited cells is reported at the bottom right of the FACS plot for each condition. SpRY-ABE8e-V106W in combination with sgRNA FLT3_e9_18 achieved the highest efficiency by flow cytometry, with up to 66.3% of edited cells not recognized by clone 4G8.

[0226] Example 6: FLT3 mutations introduced by ABE preserve protein expression and ligand binding Sanger sequencing of base-edited cells from Example 5 revealed A to G editing of both adenines in the N399 codon (sequence: AAC), potentially generating either N399D or N399G mutations. To further test whether FLT3 variants containing mutations at N399 still maintain physiological FLT3L binding, fluorescent ligand binding assays were performed. Figure 16A shows fluorescent ligand binding assays performed on K562 cells expressing various FLT3 variants: N339D (SEQ ID NO: 51), N399G (SEQ ID NO: 52), exon 9 mutation (SEQ ID NO: 50), or WT FLT3 (SEQ ID NO: 48). The fluorescence ratio between FLT3L AF488 and FLT3 BV10A4 PE-Cy7 is reported in each plot. Figure 16B shows the distribution (histogram) of the fluorescence ratio between FLT3L AF488 and FLT3 BV10A4 PE-Cy7 for each variant. The data show that the N399D and N399G variants caused by FLT3 adenine base editing maintain physiological FLT3L binding affinity.

[0227] Example 7: Use of SpRY-ABE8e-V106W mRNA results in highly efficient base editing in human leukemia cell lines and human CD34+ HSPCs To translate base editing procedures into primary cells, suitable delivery methods for base editors need to be developed, as bacterial plasmid transfection has been reported to be toxic to stem cells.

[0228] The base editing protocol was translated into human CD34+ HSPCs using base editor mRNA produced by in vitro transcription.Functional mRNA encoding the adenine base editor (SpRY-ABE8e-V106W 3xNLS or SpRY-K918N-ABE8e-V106W 3xNLS) was produced by in vitro transcription (IVT) using the MEGAscript T7 Transcription Kit (Invitrogen AM1333) or the NEB T7 HiScribe Kit (E2040S) and a custom plasmid (SEQ ID NO: 81, Figure 17A and see below) template (encoding the reading frame of the base editor downstream of the T7 promoter sequence, a minimal 5'UTR, and upstream of two copies of HBB (hemoglobin B) 3'UTR and a polyA sequence (60-120 bp long)). Figure 17B illustrates the in vitro transcription workflow utilized to generate co-transcriptionally 5'-capped and 3'-polyadenylated mRNA for primary human CD34+ base editing. Co-transcriptional capping was achieved by replacing 80% of the GTP with the 3'-O-Me-m7G(5')ppp(5')G RNA cap structure analog (NEB S1411). IVT reaction products were purified using either Qiagen RNAesy miniKIT or NEB Monarch mRNA CleanUp (T2050L), quantified by Nanodrop, and analyzed by Agilent Fragment Analyzer for quality control (a typical electropherogram is shown in Figure 17C).

[0229] pmRNA plasmid for in vitro transcription of SpRY_ABE8e_V106W adenine base editor, containing SEQ ID NO:81-5'UTR, HBB 3'UTRx2, and a 120bp long polyA tail [ka] [ka] [ka] [ka] [ka] [ka]

[0230] Figure 18 illustrates base editing experiments in K562 reporter cells using different doses of electroporated IVT SpRY-ABE8e-V106W 3xNLS base editor mRNA (0.5-5ug in 20uL electroporation volume) and either 180 or 360pmol of FLT3-e9-18 sgRNA. Edited cells were stained with 4G8 clone (targeting antibody) and BV10A4 clone for expression normalization (Figure 18A). The percentage of edited cells by flow cytometry is reported at the bottom right of each plot. Figure 18B shows the relationship between mRNA / gRNA dose and editing efficiency by flow cytometry. Figure 18C shows the relationship between mRNA / gRNA dose and cell viability by flow cytometry (LiveDead yellow staining). Figure 19 is a heatmap showing base editing efficiency by Sanger sequencing of PCR amplified gDNA from samples of the experiment in Figure 19A. Sanger traces were deconvoluted using the EditR package and A to G conversion efficiency was calculated. gRNA and mRNA dosages are reported on the left, and the numbering of targeted adenines within the gRNA sequence is reported as columns (PAM corresponds to positions 21-23). ​​Figure 19B reports the relationship between editing efficiency by flow cytometry (4G8-cells) and Sanger sequencing.

[0231] To confirm that FLT3 base editing of primary human CD34+ hematopoietic stem and progenitor cells is feasible, we performed in vitro base editing and expansion culture experiments. Mobilized peripheral blood-derived CD34+ HSPCs were thawed and cultured at 500,000–750,000 cells / mL in StemCell SFEMII medium supplemented with 1% penicillin / streptomycin, SCF 100ng / mL (Peprotech), FTL3L 100ng / mL (Peprotech), TPO 50ng / mL (Peprotech), Stemregenin-1 0.75uM (StemCell technologies), and UM171 35nM (Selleckhem). Forty-eight hours after thawing, 0.15–0.25 million (M) HSPCs were electroporated using the Lonza 4D-Nucleofector system in P3 electroporation solution (Lonza) supplemented with 2.5–7.5 ug of base editor mRNA (SpRY-ABE8e-V106W) and 250–450 pmol of sgRNA (Integrated DNA Technologies) per 20 uL reaction. Cells were cultured for 5–7 days in the aforementioned medium. To test for specific resistance to CAR-T cell killing, edited or unedited CD34+ cells were expanded in vitro for 3 days after editing and then co-cultured with either anti-FLT3 CAR-T cells or untransduced T cells.

[0232] FIG. 20A shows the experimental layout and timeline for in vitro expansion culture and FLT3 base editing of mobilized peripheral blood-derived CD34+ HSPCs. Cells were electroporated with either 1, 2.5, or 5ug of SpRY-ABE8e-V106W 3xNLS mRNA and 180 or 360pmol of FLT3-e9-18 sgRNA. FIG. 20B illustrates the gating strategy for flow cytometry immunophenotyping of CD34+ cells during in vitro expansion culture. FIG. 20C shows the in vitro fold expansion of CD34+ cells compared to different base editing conditions, indicating limited toxicity of the base editing procedure compared to untreated controls. FIG. 20D summarizes the flow cytometric composition of cultured CD34+ at days 3 and 6 post-editing, indicating no bias of stem and progenitor cell subsets upon electroporation and mRNA base editing.

[0233] Editing efficiency by Sanger sequencing 6 days after electroporation is reported in Figure 21A. gRNA and mRNA dosages are reported on the left, while the numbering of targeted adenines within the gRNA sequence is reported as columns (PAM corresponds to positions 21-23). ​​The relationship between base editing efficiency in human CD34+ HSPCs and different gRNA x mRNA dosages is reported as a scatter plot in Figure 21B.

[0234] Example 8: FLT3 base editing of human FLT3-expressing leukemia cell lines or human CD34+ HSPCs with SpRY-ABE8e-V106W mRNA confers resistance to 4G8 CAR-T cells To test whether the base-edited cells obtained from the experiment performed in Figure 18 are resistant to 4G8 CAR-T cells, co-culture experiments were performed as previously reported. Flow cytometry plots of viable cells after 4 hours of co-culture at different effector:target ratios are reported in Figure 22A. Edited cells (editing efficiency approx. 89%) are protected from CAR-T cell killing and survive at higher E:T ratios. Figure 22B reports the viability of target cells (unedited or FLT3 base-edited K562) by Annexin V and LiveDead yellow staining 4 hours after co-culture with either 4G8 anti-FLT3 CAR-T cells or untransduced T cells, highlighting the selective killing of unmodified cells. CAR-T / T cell degranulation by surface CD107a staining at 4 hours of co-culture with either unedited or FLT3 base-edited K562 cells is reported in Figure 22C. Only CAR-T cells exposed to unmodified cells show significant degranulation.

[0235] Similarly, to test whether base-edited human hematopoietic stem and progenitor cells are resistant to FLT3-targeted immunotherapy, CD34+ cells edited in the experiment performed in FIG. 20 were co-cultured with 4G8 CAR-T cells or untransduced T cells. FIG. 23 shows a co-culture experiment of CAR-T cells with unmodified or FLT3 base-edited mPB CD34+ cells (editing efficiency approx. 46%) from the third day of culture of the experiment in FIG. 20. Flow cytometry plots of viable cells at 48 hours of co-culture with either 4G8 anti-FLT3 CAR-T cells or untransduced T cells at different effector:target ratios are reported in FIG. 23A. Specific killing of CD34+ cells co-cultured with either 4G8 anti-FLT3 CAR-T cells or untransduced T cells at different effector:target ratios is reported in FIG. 23B. Specific killing of stem cell-enriched CD34+90+ subsets is reported in FIG. 23C. Although the differences in CAR-T-mediated killing are not as pronounced compared to experiments performed with cell lines, the modified HSPCs exhibit the resistance conferred by FLT3 editing due to the relatively low expression of FLT3 by cultured CD34+ HSPCs exposed to FLT3L, and therefore a greater difference among the more stem cell-enriched CD90+ cells.

[0236] Example 9: FLT3-edited HSPCs are protected by CAR-T cell-mediated killing in vivo To further confirm the protective role against on-target killing mediated by anti-FLT3 CAR-T cells, an in vivo xenotransplantation experiment in immunodeficient NSG mice was performed (Figure 24A shows the experimental design). Human total mobilized peripheral blood derived CD34+ HSPCs were edited as exemplified above and transplanted at 1 million per mouse by tail vein injection 24 hours after sublethal irradiation (2.5 Gy). Mice xenotransplanted with non-edited HSPCs served as the control group. Each group was further divided into two treatment subsets at 7 weeks post-transplantation, either vehicle (PBS) or 4G8-CAR T cells, 1.5 million cells per mouse. Mice were euthanized at week 8 (1 week after CAR-T cell treatment). Human bone marrow engraftment (% of human CD45+ cells, excluding CD3+ cells) and human absolute CD45+ cell counts by flow cytometry are reported in Figure 24B. CD3+ are excluded from human engraftment, since they originate from the injected CAR-T cells. A more severe depletion of human engraftment is observed in mice transplanted with unedited cells. The human engraftment composition on bone marrow (excluding CD3+ cells) is reported in Figure 24C. This shows the multilineage engraftment generated by edited HSPCs.

[0237] The frequency of hematopoietic stem and progenitor cells (CD34+CD38-) and absolute abundance in bone marrow at sacrifice are reported in Figure 25A. This shows the protective effect of FLT3 base editing against CAR-T-mediated killing. Exemplary flow cytometry plots obtained from pooled events from bone marrow analysis of each group of mice are reported in Figure 25D (gating strategy) and Figure 25E (fraction of lineage-CD34+38- persister cells). The relatively high activation of CAR-T cells in the spleens of treated mice by surface CD69 staining is shown in Figure 25B. CAR-T cells were identified by EGFR staining (co-expressed with CAR as a marker and safety switch). CAR-T cell phenotype by expression of CD62L and CD45RA in the spleens of treated mice is reported in Figure 25C. Naïve: CD45RA+CD62L+; Central Memory (CM): CD45RA-CD62L+; Effector Memory (EM): CD45RA-CD62L-; Terminally Differentiated Effector Memory (TEMRA): CD45RA+CD62L-. A relative bias towards effector subsets is evident in treated mice xenografted with unedited HSPCs.

[0238] To further evaluate the protection of CD34+ HSPCs by FLT3 base editing of N399 against CAR-T cell killing, a second experimental in vivo experiment was performed in NBSGW mice. This allows xenotransplantation and engraftment of human HSPCs without irradiation, allowing a higher level of human hematopoietic reconstitution (homozygous W41 allele) due to the hypomorphic Kit mutation in mouse HSPCs. Figure 32 reports the experimental setup and FLT3 base editing efficiency in engrafted human cells at 8 weeks of bleeding. The treatment group received 2.5 million 4G8 CAR-T cells at 11 weeks post-transplantation and was then euthanized at 13 weeks post-transplantation. This experimental setup allowed for improved evaluation of myeloid progenitor and lineage protection conferred by FLT3 base editing. Figure 33A shows the relative abundance of granulocytes (polymorphonuclear cells, PMN) by flow cytometry in bone marrow at the time of sacrifice with clear protection in mice transplanted with FLT3-edited HSPCs compared to AAVS1 control edited cells. This indicates some degree of protection at the progenitor level, since mature granulocytes do not express FLT3. Consistent with this hypothesis, granulomonocytic progenitors (GMPs), defined as lineage-CD34+CD38+FLT3+CD45RA+, are selectively depleted by FLT3 CAR-T cells only in mice xenografted with mock-edited HSPCs (Figure 33A). Figure 33B reports the FLT3 base editing efficiency at several time points in 4G8-CAR-treated or untreated groups. Upon 4G8-CAR administration, there is a progressive negative selection of unmodified cells, resulting in the selection of FLT3-edited cells (this is more evident in the progeny derived from colony-forming unit assays plated with bone marrow samples from treated mice). Representative FACS plots showing granulocyte, granulocyte-monocyte progenitor (GMP) and HSPC (lineage-CD34+CD38-CD90+CD45RA-) populations are reported in FIG.

[0239] Example 10: CD123 base editing of human CD34+ HSPCs with SpRY-ABE8e-V106W mRNA is feasible and can be multiplexed with FLT3 base editing. Similar to the FLT3 base editing experiments, we used SpRY-ABE8e-V106W mRNA and CD123 gRNA-N, alone or in combination with FLT3 N399 base editing, to transduce human CD34+ HSPCs. Mobilized peripheral blood-derived CD34+ HSPCs were thawed and cultured at 500,000–750,000 cells / mL in StemCell SFEMII medium supplemented with 1% penicillin / streptomycin, SCF 100ng / mL (Peprotech), FTL3L 100ng / mL (Peprotech), TPO 50ng / mL (Peprotech), Stemregenin-1 0.75uM (StemCell technologies), and UM171 35nM (Selleckhem). 48 hours after thawing, 0.15-0.25 million (M) HSPCs were electroporated using the Lonza 4D-Nucleofector system in P3 electroporation solution (Lonza) supplemented with 5-7.5ug of base editor mRNA (SpRY-ABE8e-V106W) and 420-450pmol of sgRNA (Integrated DNA Technologies) per 20uL reaction. Cells were cultured for 7 days in media as described above. Figure 26A illustrates the experimental layout and media composition. Figure 26B shows in vitro expansion of cultured CD34+ HSPCs. Figure 26C shows loss of binding of 7G3 clones to successfully base-edited CD34+ HSPCs. Flow cytometry plots show CD123 base editing of HSPCs by loss of 7G3 staining when gated on stem-enriched CD90+ subsets and normalized to clone 9F5 staining. The editing efficiencies of FLT3 single-edited, CD123 single-edited, and FLT3+CD123 double base-edited conditions are reported in Figure 27. Although it is feasible to introduce the S59P mutation by base editing, the efficiency obtained in our setting is relatively low (up to 29.3%) compared to FLT3 N399 (>64%).

[0240] Example 11: Optimization of sgRNA position and introduction of K918N substitution in the adenine base editor Cas9 protein can improve CD123 base editing efficiency To see whether the editing efficiency of CD123 S59 codon can be improved, Cas9 protein containing K918N mutation, which is associated with improved Cas9 catalytic activity, was tested in combination with our SpRY-ABE8e design.As shown in Figure 28A, the introduction of K918N mutation has sequence-specific effects on editing efficiency, improving CD123-gRNA-N but slightly reducing FLT3-gRNA-18 editing.Figure 28A shows the results of base editing by flow cytometry 3 days after electroporation of 500ng of base editor expression plasmid and 360pmol of sgRNA into K562 reporter cells.Edited cells were stained for CD123 with 7G3 (targeting antibody) plus 9F5 clone (normalization). gRNA CD123-N (CD123_gRNA_N, SEQ ID NO: 24) was tested in combination with SpRY-ABE8e-V106W-3xNLS BE as in the previous experiment, or in combination with SpRY-K3918N-ABE8e-V106W-3xNLS (SEQ ID NO: 9 and SEQ ID NO: 12, sequences above). The percentage of edited cells by flow cytometry is reported at the bottom right of each plot.

[0241] To further improve the editing efficiency of CD123, two additional sgRNAs (CD123-gRNA-R and its 21 bp long version, CD123-gRNA-R21) were cloned together with the reference CD123-gRNA-N and FLT3-gRNA-18. The pHKO-mTagBFP2 plasmid was generated under the human U6 promoter. Figure 28B reports the relative position and sequence of the gRNA targeting CD123 S59. Additionally, additional adenine base editor designs were cloned and tested along with SpRY-ABE8e-V106W (Figure 28C): SpRY-K918N-ABE8e-V106W, SpRY-HF1-ABE8e-V106W, SpRY-HF1-BlackJack-ABE8e-V106W, SpRY-BlackJack-ABE8e-V106W, SpRY-Sniper-ABE8e-V106W SpRY-Sniper-BlackJack-ABE8e-V106W. The HF1 variant contains substitutions N497A, R661A, Q695A, Q926A that are associated with reduced tolerance to gRNA spacer mismatched bases and reduced off-target editing (high fidelity). The Sniper variant contains substitutions F539S, M763I, and K890N, which produce another high-fidelity SpCas9 variant with maintained on-target efficiency. BlackJack is a Cas9 variant designed to tolerate sgRNAs longer than 21 bp, which is relatively important for high-fidelity variants. Co-electroporation of 500 ng of base editor plasmids and 500 ng of sgRNA expression plasmids in K562 reporter cells was performed for all combinations of sgRNAs (CD123-gRNA-N, CD123-gRNA-R, CD123-gRNA-R21, FLT3-gRNA-18) and results were evaluated 72 hours after editing by flow cytometry. Figure 28D summarizes the results of the screening procedure and highlights the improved combinations for CD123 S59 base editing.Notably, excluding the HF1 variant, CD123-gRNA-R is on average 1.42-fold more efficient than CD123-gRNA-N, whereas its 21-bp-long counterpart (CD123-gRNA-R21) is only 1.21-fold more efficient. Introduction of the K918N mutation improves CD123 editing by 1.08-fold for gRNA-R and 1.05-fold for gRNA-N. Combined with the Sniper mutation, K918N improves efficiency by 1.22-fold for gRNA-R and 1.16-fold for gRNA-N. In conclusion, switching to CD123-gRNA-R and introduction of the K918N mutation partially compensates for the lower efficiency observed for CD123 editing compared to FLT3, and the K918N substitution appears to be relatively less impactful.

[0242] Example 12: Improved duplex base editing efficiency by SpRY-K918N-ABE8e-V106W mRNA in CD34+ HSPCs Improved CD34+ HSPCs base editing efficiency of FLT3 and CD123 loci using SpRY-K918N-ABE8e-V106W in vitro transcribed mRNA was achieved by optimizing multiple parameters (mRNA preparation, K918N mutagenesis, sgRNA selection). An exemplary experiment is reported in Figure 29. Figure 29A summarizes the experimental design. Figure 29B reports the gating strategy for CD123 editing evaluation by flow cytometry. Meanwhile, Figure 29C shows similar proliferation of FLT3+CD123 base edited cells when compared to untreated (electroporation only) controls. Sanger sequencing revealed efficiencies of up to 85% for FLT3 N399 editing and up to 41% for CD123 S59 editing (Figure 30).

[0243] Example 13: The KIT H378R mutation can be introduced into CD34+ HSPCs via adenine base editing Using the same experimental setup as in Example 12, we tested whether the H378R mutation could be introduced via base editing in human CD34+ HSPCs. Editing efficiencies of up to 60% were achieved by electroporating SpRY-K918N-ABE8e-V106W in vitro transcribed mRNA (4ug / 20uL electroporation volume) and KIT-gRNA-Y (SEQ ID NO: 37).

[0244] Example 14: Generation of anti-FLT3 and CD123 bispecific chimeric antigen receptors Rationally designed anti-FLT3 and CD123 bispecific chimeric antigen receptors should provide potent anti-leukemia efficacy. Figure 29A is a schematic of a putative second generation bispecific chimeric antigen receptor targeting both FLT3 domain 4 and the N-terminal domain of CD123. Figure 29B is a design of a second generation monospecific CAR targeting CD123 (CSL362 clone, SEQ ID NOs: 75-76) and FLT3 (4G8 clone, SEQ ID NOs: 73-74), as well as a bispecific chimeric antigen receptor with different orientations of the scFv domains and / or different extracellular linkers, transmembrane domains, intracellular costimulatory domains (SEQ ID NOs: 77-80).

[0245] Example 15: Base editing generates stealth receptors The cytokine receptors FLT3, KIT, and CD123 (IL3RA) were chosen as targets for the epitope engineering strategy. Fms-like tyrosine kinase 3 (FLT3, CD135) and the proto-oncogene c-KIT (KIT, CD117) are class III receptor tyrosine kinases that are expressed in 93% and 85% of AML cases, respectively, in either wild-type (WT) or mutated forms. 32~37 CD123 is found on the surface of >75% of AML cases and is the alpha subunit of the IL-3 receptor (IL3RA), a type I cytokine receptor that is overexpressed on the surface of leukemic stem cells. 38~40 These genes are present at various stages of normal hematopoietic development, and their overexpression on AML cells is associated with higher relapse rates and poorer overall survival after HSCT in both adult and pediatric patients.33、36、41~43 To develop our approach, we investigated the efficacy and safety of a monoclonal antibody (mAb) currently being evaluated for the development of anti-AML immunotherapy, namely clone 4G8. 22、44 (FLT3), Fab-79D 21、45 (KIT), and 7G3 46~48 We selected FLT3 extracellular domain (CD123) as the target of our study. Previous studies have reported that 4G8 recognizes FLT3 extracellular domain (ECD) 4, whereas clone BV10A4 recognizes an unrelated epitope within ECD2, and therefore could serve as a control to assess FLT3 surface expression. 44 Because 4G8 was generated by immunizing BALB / c mice with a human FLT3-transfected cell line, we reasoned that 4G8 recognizes a human-specific epitope, despite the high degree of homology (85.8% identity and 91.5% similarity) and cross-reactivity of the FLT3 ligand (FLT3L) between human and mouse FLT3. 49 .

[0246] We first confirmed that a chimeric human FLT3 with ECD4 replaced by its mouse orthologue (16 codon changes) resulted in loss of 4G8 binding without affecting FLT3L binding and intracellular kinase phosphorylation (Figure 36A,B left,C top). To identify the minimal number of residues involved in 4G8 binding, we designed a Sleeping Beauty transposon-based combinatorial library using human or mouse codons at each of the 16 mismatched positions in ECD4 (Figure 36D, top left). Flow cytometry analysis of K562 cells transduced with this library showed a population of cells that were positive for a control antibody (BV10A4) and negative for 4G8 (Figure 36D, top right). Comparison of the relative abundance of human vs. mouse codons at each position, measured by targeted deep sequencing of sorted single-positive (BV10A4+4G8-) and double-positive (BV10A4+4G8+) cells, revealed enrichment for a single amino acid substitution (N399D; Figure 36D bottom, and Figure 37A). To validate this result, K562 cells were transduced with the FLT3 N399D variant and showed surface expression of FLT3 at levels comparable to wild type, but with loss of 4G8 binding (Figure 37B).

[0247] Next, we evaluated gene editing strategies to introduce the N399D substitution. To easily evaluate the results of our genome engineering procedure in cells that do not depend on FLT3 signaling, we generated K562 reporter cells that express FLT3 from the endogenous locus by targeted integration of the EF1α constitutive promoter upstream of the transcription start site (Figure 37C). We then confirmed that the N399D mutation can be inserted by homology-directed repair (HDR) using either SpCas9 or AsCas12a nuclease and a 200bp ssODN donor template, and that successfully edited cells show loss of 4G8 binding while maintaining FLT3 surface expression (Figure 37D). Nevertheless, the use of nucleases carries an inherent risk of genotoxicity associated with DNA double-strand breaks (DSBs) and gene knockout, which occurs in the majority of non-edited cells (Figure 37D).

[0248] Because epitope engineering can be achieved by the introduction of single point mutations, we reasoned that base editing (BE) may be a preferred and safer option for epitope editing by circumventing the need for DSBs. The asparagine at position 399 is encoded by an AAC codon, which can be converted to GAC (aspartic acid) or GGC (glycine) by adenine base editing (ABE). We tested this hypothesis by electroporating FLT3 reporter cells with several sgRNAs (with target adenines at positions 3-9 of the protospacer, offset by 1 bp) in combination with advanced generation TadA-8e deaminase coupled to either SpCas9 nickase (NGG PAM) or Cas9 variants with relaxed PAM specificity (NG-SpCas9n and SpRY-Cas9n, Figure 36F). Flow cytometric evaluation showed successful epitope editing with loss of 4G8 recognition, with the highest efficiency (66.3%, Figure 36F) achieved by SpRY-ABE8e in combination with FLT3-sgRNA-18. In contrast to HDR-based strategies, both base-edited and non-base-edited cells maintained normal surface FLT3 expression without significant gene knockout. As both N399D and N399G are potential consequences of our editing strategy, we included this mutation in all further validation analyses.

[0249] A similar strategy was applied to epitope mapping of Fab-79D, a KIT-targeting mAb that has been reported to bind to KIT ECD4 and block ligand-induced dimerization. 45First, we confirmed the loss of Fab-79D binding and verified the preservation of stem cell factor (SCF) binding and intracellular kinase phosphorylation by introducing 10 amino acid changes (from the KIT orthologue) at positions previously predicted as potential contact points with KIT ECD4 (F316S, M318V, I319K, V323I, I334V, E360K, P363V, E366D, E376Q, H378R) (Figure 36B right, C bottom). To comprehensively screen the interaction of ECD4 mutations with cloned Fab-79D, we performed a screen using a degenerate codon library in which each position of KIT ECD4 was replaced with a random amino acid (Figure 36E left). We transduced the library into HEK-293T cells and selected cells with reduced binding to Fab-79D but retained KIT expression and Fab-79D staining (Figure 36E right). NGS sequencing of library regions and comparison of enrichment of specific codons at each position in single positive cells revealed several mutations that could reduce the affinity of Fab-79D (Figure 36E, bottom). To validate these findings, we cloned 20 amino acid substitutions individually by selecting mutations that could be reproduced by adenine or cytidine base editing at 10 positions identified by the library (M318, I319, V323, D332, I334, D357, E360, E376, and H378), and selected H378R for further development because it showed the highest reduction in Fab-79D binding while maintaining binding to SCF when normalized to the KIT control Ab (clone 104D2) (Figure 37E). Furthermore, H378R can be inserted by ABE similar to FLT3 N399D, thus allowing potential combinatorial and dual epitope engineering. As with FLT3, three different sgRNAs were screened in K562 cells in combination with SpRY-ABE8e using editing windows aimed at H378 (target A at positions 5, 6, or 7) and KIT-Y was identified as the most efficient combination (Figure 37C,F).

[0250] For CD123, both epitopes and amino acid substitutions that affect binding of the therapeutic clone 7G3 (or its humanized counterpart CSL362-talatuzumab) have been previously reported. 48 To develop our epitope editing strategy, we designed a target BE screen on K562 reporter cells by testing sgRNAs targeting E51, Y58, S59, R84, P88, and P89 in the N-terminal domain of CD123 and rationally combining them with CBE (evo-APOBEC1-BE4, including NGG, NG, and SpRY Cas variants) and ABE (ABE8e, NG and SpRY variants, Figure 36G) (Figure 37E). Although several combinations introduced mutations that reduced the affinity of 7G3 to CD123, only gRNA CD123-N and CD123-R combined with SpRY-ABE8e and gRNA CD123-L combined with SpRY-evo-APOBEC1-BE4 were able to completely abolish 7G3 binding (Figure 36G and Figure 37H). Sequencing of edited cells revealed that ABE and CBE resulted in substitutions of S59P and S59F, respectively. A bystander mutation (Y58H) could also be introduced by adenine base editing with gRNA CD123-R with higher efficiency than CD123-N and was therefore included for further validation. S59P introduced by gRNA CD123-R and ABE was selected for further development due to its potential to edit multiple epitopes in combination with FLT3 and KIT ABE. Finally, by testing our base-edited CD123 reporter cells with two additional CD123-targeting mAbs, we discovered that cytidine base editing of P88 / P89 resulted in combinations of P88S / P89S or P88L / P89L, resulting in an almost complete loss of recognition by clones 6H6 and S18016E. Thus, the pool of potential mAbs that can be applied in combination with epitope-engineered cells is broadened (Figure 37I).

[0251] To precisely quantify the loss of affinity of our selected Abs to the epitope-engineered receptors, we transduced each receptor variant (FLT3 WT, N399D, N399G, KIT WT, H378R, CD123 WT, S59P and Y58H-S59P) into K562 cells by Sleeping Beauty transposase and observed that for the WT receptor, all tested variants almost completely failed to recognize it, even at saturating concentrations of mAb (>5000ng / mL, Figure 36H). Notably, FLT3 N399G (introduced by editing both adenines at the AAG codon at position 399) did not differ from N399D in its ability to reduce 4G8 binding. Similarly, the addition of the Y58H bystander mutation to CD123 S59P showed a similar reduction in 7G3 binding without affecting CD123 expression on the cell membrane.

[0252] Overall, we conclude that epitope engineering of FLT3, KIT, and CD123 is feasible and can be achieved with high efficiency without the need for DSBs by selecting the appropriate combination of gRNA and base editing enzymes.

[0253] Example 16: Epitope editing maintains receptor function As the selected targets are fundamental cytokine / growth factor receptors expressed in human HSPCs with relevant roles in stem cell maintenance and lineage differentiation, we rigorously evaluated whether our engineering procedure altered the function of the receptors. Using fluorescently conjugated FLT3L, SCF, and IL-3 ligands, we confirmed equivalent binding to their respective WT or epitope-engineered receptors across all concentrations tested (1 to >1000 ng / mL, Figure 36I). As K562 cells do not express CSF2RB by flow cytometry, CD123 variants were co-expressed with a common beta subunit (CSF2RB / CD131) to form a heterodimeric IL-3 receptor and allow signal transduction (Figure 37J). Activation of the intracellular signaling cascades of FLT3 and KIT was confirmed by Western blot, showing equivalent and dose-dependent phosphorylation of the kinases upon ligand binding (Figure 38A,B). For CD123, we confirmed ligand-mediated receptor activation by measuring phosphorylation of the downstream STAT5 signaling factor, which was equally activated by both WT and epitope-engineered variants at all IL-3 concentrations tested (Figure 38C). Finally, to confirm that the ligand-dependent proliferative response was intact, we performed kinase complementation assays on BaF3 cells, a murine IL-3-dependent lymphoblastoid cell line that requires signaling through the STAT5 pathway for proliferation and survival. 50 We observed comparable and dose-dependent rescue of cell proliferation by WT and epitope-engineered receptor variants following exposure to human FLT3L, SCF, and human IL-3 during mouse IL-3 starvation (Figure 38D).

[0254] Example 17: Stealth receptors are resistant to CAR-T cells In a recent study, the anti-FLT3 clone 4G8 22 , Anti-KIT Fab-79D 21 , or anti-CD123 CSL362 (a humanized variant of clone 7G3 48、51It was shown that CAR-T cells generated from the mAb had remarkable efficacy against human AML cells. To evaluate the resistance of epitope-engineered cells to targeted CAR-T therapy, single-chain variable fragments (scFvs) of 4G8, Fab79D, and CSL362 were cloned into second-generation CAR constructs containing a CD28 costimulatory domain and optimized for expression using lentiviral vectors containing bidirectional promoters. 52 A truncated EGFR selection / depletion marker (tEGFR, FIG. 39A) was co-expressed. For the production of CAR-T cells, a culture protocol was used that expands cells with a T stem cell memory phenotype based on CD3-CD28 bead stimulation in the presence of IL-7+IL-15. 53 (Figure 39A, Figure 40A,B). High CAR transduction efficiency (>85% by tEGFR staining) was obtained with >20-fold in vitro expansion compared to culture initiation (Figure 39B).

[0255] By performing in vitro killing assays on K562 reporter cells, we found that the majority of cells overexpressing WT FLT3, KIT, or CD123 were killed by their respective specific CAR-Ts (E:T=10, <2% surviving cells compared to E:T=0), whereas K562 cells expressing epitope-engineered variants were resistant to CAR-T cell killing (both in absolute numbers and cell viability) across different effector:target ratios (E:T 0.625-10) and survived until the end of the experiment (Figure 39C-E and Figure 40C,D). T cell activation (expression of CD69) and degranulation (surface expression of CD107a) were significantly higher in conditions cultured with cells expressing the WT genes, consistent with the lack of recognition of the epitope-edited variants by CAR-Ts (Figure 40C,D). Moreover, surviving K562 reporter cells still expressed the targeted receptor at levels comparable to untreated controls (Figure 39C,D,E right). Untransduced T cells showed neither target killing nor upregulation of CD69 when cultured in all conditions tested (Figure 40D). To further confirm strict epitope-specific killing by CAR-T cells, mixed populations of either unmodified or base-edited dual-targeted cells (K562 cells expressing FLT3 and CD123) were co-cultured and plated with the corresponding CAR-T cells, and selective resistance of the epitope-edited population was observed (Figure 40E).

[0256] Overall, these data provide rigorous validation that cells overexpressing epitope-engineered FLT3, KIT, and CD123 variants are resistant to CAR-T cell recognition and killing.

[0257] Example 18: Efficient epitope editing of human HSPCs To effectively introduce our nucleotide variants into endogenous genes in human primary HSPCs, we optimized a base editing protocol for mobilized peripheral blood (mPB)-derived CD34+ cells based on co-electroporation of chemically modified sgRNA and in vitro transcribed (IVT) SpRY-ABE8e mRNA (Figure 41A and Figure 42A-C). After optimizing in vitro mRNA transcription, culture, and electroporation conditions, as well as editing at time points after HSPC stimulation (Figure 42D,E,F), we achieved up to 86.6%, 78.6%, and 67.9% of targeted A>G conversions for FLT3, KIT, and CD123, respectively. We were able to efficiently edit targeted adenines within the window of FLT3-18, KIT-Y, and CD123-R sgRNAs (Figure 41B). Analysis of treated cells showed no bias in the composition of phenotypically identified progenitors (LMPP, CD90-45RA+, MPP, CD90-45RA-, HSC, CD90+45RA-) during in vitro culture in the presence of stem maintenance compounds (StemRegenin, SR-1 and UM171, Figures 41D and 42E). HDR-Mediated Editing 54 In contrast to what was previously observed, base editing efficiency was comparable in bulk cells, committed progenitor cells (CD90-), and the more primitive HSPC subset (CD90+45RA-, Figure 41C).

[0258] FLT3 N399 and CD123 S59To confirm the resistance of epitope-engineered HSPCs to immunotherapy, killing assays by plating were performed on HSPCs edited with 4G8 and CSL362 CAR-T cells at different E:T ratios 3–5 days after electroporation. Specific killing by FLT3- and CD123 CAR-T was most prominent within the CD45RA+ and CD90+ subsets, respectively. These were therefore used to assess the results of these experiments by absolute numbers. Cells edited at the control site (AAVS1 safe genome harbor) were eliminated by CAR-T cell coculture, whereas epitope-edited cells showed higher viability and absolute numbers, similar to those observed in K562 coculture (Figure 41E,F). As KIT has known extrahematopoietic expression in humans, 55 In this study, we focused on the use of mAbs instead of CAR-T cells, which may result in less severe on-target toxicity and can be used for non-genotoxic conditioning or in vivo selection of epitope-engineered cells. By plating edited HSPCs with increasing concentrations of Fab-79D mAb, which blocks dimerization, we were able to inhibit KIT in response to SCF. H378 Maintenance of proliferation kinetics of edited HSPCs was observed, whereas AAVS1 BE The cells were inhibited in a dose-dependent manner (FIG. 41G).

[0259] To confirm the preservation of functionality of receptor-engineered CD34+ HSPCs, treated cells were cultured for an additional 4 days with increasing concentrations of the respective ligands and without other cytokines. At the end of the culture, a dose-dependent expansion of HSPCs was observed for all three targets, as well as CD123 at IL-3 concentrations ranging from 1 to 10 ng / mL. S59With the exception of BE, there was no difference between receptor-edited and AAVS1-edited controls (Figure 42H). Nevertheless, no counterselection of BE HSPCs was observed, and a uniform proliferation of CD34+ cells was confirmed, independent of CD123 epitope editing (Figure 42I). Furthermore, to confirm the minimal impact on proliferation of receptor-edited HSPCs and to evaluate the differentiation capacity of receptor-edited HSPCs, cell culture and colony-forming unit (CFU) assays were performed on HSPCs bases edited for FLT3, CD123, KIT or AAVS1 genes to observe absolute cell numbers after in vitro expansion, and the number of myeloid and erythroid colonies was comparable to untreated controls (Figure 41H). Xenotransplantation of treated HSPCs into female NBSGW immunodeficient mice resulted in the differentiation of FLT3 and erythroid cells. N399 We demonstrated maintenance of HSPC engraftment, repopulation, and multilineage differentiation capacity (Figure 41I, J), which was comparable to AAVS1-edited controls. The percentage of FLT3 editing was comparable to that measured in input cells (BE efficiency ~35%) and remained stable up to 13 weeks post-transplant (Figure 41K). This demonstrates successful editing of the most primitive HSPC subset, with FLT3 N399 The absence of counterselection of FLT3 cells was confirmed. Transplantation of bone marrow (BM) cells into secondary recipients resulted in high human hematopoietic engraftment, with no difference in lineage distribution up to 17 weeks after secondary transplantation (Fig. 43A,B), and FLT3 editing levels remained comparable to those measured in primary recipients (Fig. 41K). Because mouse Flt3l is cross-reactive with human FLT3, these results are consistent with the results of epitope-engineered FLT3. N399 Further confirming the functionality of the variant. Similarly, KIT H378 In vivo repopulation capacity and multilineage differentiation of edited HSPCs are consistent with AAVS1 BE Comparable to controls (Figures 43C-E), there was no lineage differentiation bias and no counterselection of edited cells, further supporting that the base editing procedure does not affect the fitness and functionality of treated cells.

[0260] Overall, these data demonstrated that epitope-engineered HSPCs could be efficiently achieved by adenine base editors without affecting stem cell function and differentiation capacity.

[0261] Example 19: FLT3 BE HSPCs are resistant to 4G8 CAR-T in vivo treatment FLT3 CAR-T cells are N399 To evaluate whether FLT3+ cells could be effectively used to treat AML while sparing hematopoiesis, we transfected NBSGW mice with CD34+ HSPCs (FLT3+ cells). BE or AAVS1 BE (either AML or CAR-T cells), as well as a human patient-derived AML xenograft (PDX-1) characterized by mutations in MLL-AF9 and FLT3-ITD and pre-transduced with a reporter gene (mNeonGreen) to facilitate their detection within mixed hematopoiesis (Fig. 43F, G). Ten days after PDX challenge, mice were treated with 4G8 CAR-T cells and their hematopoietic composition was monitored by cytofluorometric analysis of serial blood samples and hematopoietic organs (BM, spleen, SP; Fig. 44A) at the end of the experiment. As observed in previous experiments, before administration of AML PDX or CAR-T cells, transplanted mice showed similar peripheral blood composition in both editing groups (Fig. 45A), and editing levels were comparable to input cells (~85%) and did not differ within myeloid and lymphoid lineages (FACS-sorted CD33+ and CD19+ cells, respectively, Fig. 44B). Mice treated with 4G8 CAR-T showed CAR-T cell engraftment and complete eradication of AML in both the BM and SP (Figure 44C-F), with FLT3 in the BM. N399 Multiparametric flow cytometry analysis of BM showed a small but significant increase in the percentage of AAVS1 cells treated with 4G8 CAR (88% vs. 90% in myeloid cells and 89% vs. 94% in lymphoid cells; Figure 44B). BE A relative depletion of CD19+ B cells (pre-B and pro-B cells) was evident only in the FLT3 group, whereas N399Mice engrafted with hPSCs were protected (Figure 4G-I). Within differentiated myeloid cells (CD33 / 66b+, excluding AML cells), immature granulocytes (CD14-10-11c-SSCs) were 高 ) 56 The proportion of FLT3 N399 Compared with mice, AAVS1 BE (Fig. 44J). N399 BE is a potent marker for the differentiation of lineage-negative progenitors (lin-CD34+, Fig. 45B-D), in particular granulocyte-monocyte progenitors (GMP, lin-CD34+38+45RA+FLT3+, Fig. 44K,L) and lymphoid-primed multipotent progenitors (LMPP, lin-CD34+38-45RA+90-10-, Fig. 44N,O) (which may instead be a signaling pathway for AAVS1 BE group) was almost completely eliminated in the AAVS1 BE Against FLT3 N399 (1.4% vs. 26.6% GMP in lin-CD34+38+ and 4.8% vs. 43.3% LMPP in lin-CD34+38-, respectively). To more precisely identify the hematopoietic subsets that are depleted by 4G8 CAR and selectively protected by epitope engineering, we used treated FLT3 N399 and AAVS1 BE The absolute numbers of different lineages were compared between groups. Common myeloid progenitor cells (CMP, lin-CD34+38+CD45RA-FLT3+), dendritic cells (CD33+14-11c+FLT3+SSC 低 ), and the absolute number of GMP is AAVS1 BE Reduced under the conditions (AAVS1 BE Against FLT3 N399 Fold reduction of CMP 0.48-fold, GMP 0.01-fold, and cDC 0.41-fold (Fig. 44M). Lymphoid lineage and progenitor cells LMPP, pre-B / NK (lin-CD34+38+10+), as well as downstream subsets (B-prolymphocytes, pro-B, and pre-B) were upregulated by FLT3. N399 vs AAVS1 BE Protected by group (AAVS1 BE Against FLT3 N399In contrast, fold changes in LMPP (0.02-fold), preB / NK (0.19-fold), proB (0.2-fold), and preB (0.18-fold, FIG. 44P). The increase in mature B cells (which are FLT3-) in the CAR-treated condition may reflect proliferation in response to CAR-mediated crosstalk. Furthermore, AAVS1 cells exposed to 4G8 CAR increased fold changes in LMPP (0.02-fold), preB / NK (0.19-fold), proB (0.2-fold), and preB (0.18-fold, FIG. 44P). BE Persistent pre-B / NK, pro-B lymphocytes, pro-B and pre-B cells, monocytes and myeloblasts (CD33 / 66b+14-11c-34-SSCs) in 低 ) FLT3 median fluorescence intensity (MFI) is FLT3 N399 This was lower than that measured in the same population of FLT3-edited genes (Figure 45E). N399 This provides further evidence that cells can maintain expression of FLT3 while evading CAR-mediated killing. Interestingly, CAR-T cells detected at the end of the experiment showed similar phenotypes (mainly effector and central memory) in all groups (Figure 45F), but FLT3 N399 Those exposed to hematopoiesis were AAVS1 BE Compared to controls, they showed lower proliferation and significantly reduced expression of PD-1 (Figure 45G). This suggests an overall reduction in activation and / or exhaustion caused by a lower antigen load to which CAR-T is exposed due to lack of recognition of the epitope-engineered hematopoiesis. Importantly, FLT3 N399 Epitope editing provided the same protection against 4G8 CAR killing regardless of the presence of human PDX engraftment, highlighting the possibility to selectively eliminate AML cells while maintaining hematopoietic reconstitution.

[0262] Overall, these data suggest that in the NBSGW model, FLT3 CAR-T cell immunotherapy preferentially depletes B cell and progenitor subsets (GMP, LMPP), while FLT3 N399 It was confirmed that epitope editing confers protection against these subpopulations.

[0263] Example 20: CD123 BE HSPCs are resistant to CSL362 CAR-T in vivo treatment Similar to FLT3 epitope editing, CD123 S59 Xenotransplantation of HSPCs into NBSGW mice was performed, and AAVS1 BE We confirmed engraftment and multilineage repopulation capabilities similar to HSPCs (Figure 46A and Figure 45H), and confirmed that the percentage of edited cells was high and stable (Figure 46B). The transplanted mice were then injected with PDX-1, which also expresses CD123+ (Figure 43G), and treated with CSL362 CAR-T cells 10 days later. Similar to 4G8 CAR-T therapy, CSL362 CAR-T cells almost completely eradicated AML cells (Figure 46C,D), and expressed CD123 S59 Compared to AAVS1 BE Mice engrafted with HSPCs show high proliferation (Figure 46E). Flow cytometric analysis of BM at endpoint revealed that CD123 S59 AAVS1 while protecting the progeny of HSPCs BE A significant reduction in the absolute numbers of human hematopoietic cells (CD45+, after excluding AML and CAR T cells, FIG. 46F) and a relative depletion of myeloid lineage cells (CD33 / 66b+), including mature and immature granulocytes, in mice engrafted with HSPCs were highlighted (FIGS. 46H-K). Unlike the killing pattern observed with the 4G8 CAR, within the lymphoid lineage, only a percentage of pro-B cells was upregulated by CSL362 CAR-T-treated AAVS1 BE The CD123 expression level in mice showed a tendency to decrease (Fig. 46G). 高 Dendritic cells (DCs), including plasmacytoid DCs, were depleted by treatment with CSL362 CAR-T, whereas CD123 S59 The BE group was maintained (Figure 46L-N). Similar to FLT3 epitope editing, lin-CD34+ progenitor cells expressed CD123 S59 The absolute numbers of myeloid populations, including CMPs, GMPs, myeloblasts, granulocytes, and DC subsets, were relatively maintained in the AAVS1 group (Figure 46O and Figure 45F, G). BE and CD123 S59 Protected by epitope editing. In lymphoid cells, AAVS1 BE and CD123 S59A partial depletion of B-prolymphocytes versus mature B cells was observed when comparing AAVS1 treated with CSL362 CAR to FLT3 (Figure 46P,Q). As observed with FLT3, CD123 median fluorescence intensity (MFI) of persistent GMPs, myeloblasts, monocytes, cDCs, and pDCs was significantly higher in AAVS1 treated with CSL362 CAR than in AAVS1 treated with CSL362 CAR. BE Compared with the condition, CD123 S59 Although the IL-16 expression was higher in the lymphoid subsets, no relevant differences were observed in the lymphoid subsets (Figure 47A).

[0264] Overall, these data suggest that CD123 S59 We show a reduction in on-target toxicity induced by epitope-edited CD123 CAR-T cells against hematopoiesis that would normally result in myeloid lineages, depletion of DCs, and an overall reduction in absolute numbers of hematopoietic cells.

[0265] Example 21: Multiple editing enables more effective AML therapy We reasoned that editing two or more epitopes may enable more effective immunotherapy by simultaneously attacking multiple AML targets without increasing hematopoietic toxicity. To evaluate the feasibility of multiple epitope editing in primary HSPCs, SpRY-ABE8e mRNA was co-electroporated with two gRNAs into CD34+ cells, and editing efficiencies comparable to those measured for the same targets after a single BE were observed, without any obvious increase in cytotoxicity (Figure 48A and Figure 42D, F). To obtain proof-of-concept of protection from multiple CAR-T therapy, we performed dual editing of FLT3 and CD123 in a dual reporter K562 cell line (achieving >55% double-edited cells, Figure 48B), and treated the sorted single-edited cells as well as the double-edited cells with bispecific CAR-T cells obtained by co-transducing both 4G8 CAR and CSL362 CAR. After 2 days of co-culture, only the dual epitope edited cells did not induce T cell activation (CD69), degranulation (CD107a), or proliferation (CellTrace dilution) and showed complete protection from killing while maintaining expression of FLT3 and CD123 (Figure 48B,C).

[0266] To model the increased efficacy conferred by dual targeted immunotherapy, we used human FLT3 alone or in combination with other immunotherapies. N399 We identified AML PDXs that were not effectively eradicated by 4G8 CAR-T cell treatment (PDX-2, characterized by MLL-AF10 and mutated TP53, Fig. 43F,G) when xenotransplanted in combination with epitope-edited hematopoiesis (Fig. 47B,C). N399 When administered to engrafted mice, it showed a trend towards higher anti-leukemic efficacy against PDX-2 (Figure 47C). This suggests a detrimental role of extratumoral CAR activation on successful treatment. We next tested whether any combination of CAR-T cells targeting two of the selected antigens could provide more effective elimination of PDX-2 AML cells, and the combination of 4G8 CAR-T with both Fab79-D and CSL362 CAR-T was the most effective (Figure 47D). This is consistent with improved killing when two essential AML antigens are targeted simultaneously. To obtain formal evidence that dual-edited HSPCs are resistant to combination CAR-T cell therapy, we used dual FLT3 N399 / CD123 S59 Epitope-edited HSPC or AAVS1 BE Controls were xenografted into NBSGW mice. After confirming multilineage engraftment (Fig. 47E) and injecting PDX-2 AML cells, we treated the mice with both 4G8 and CSL362 CAR-T cells (1:1 ratio co-injection, Fig. 48D). As suggested by our CAR combination experiments, dual CAR-T therapy was able to completely eradicate AML cells from the BM and SP (Fig. 48E,F). Both FLT3-edited and CD123-edited cells persisted to the endpoint and were similarly detected in lymphoid and myeloid cells (Fig. 48G). By analyzing non-malignant hematopoiesis, we found that AAVS1 cells treated with dual CAR-T cells were significantly increased in AML cells compared to control mice. BEIn mice, we observed widespread depletion of myeloid and lymphoid lineages, with the strongest depletion of GMPs, granulocytes, cDCs, LMPPs, pre-B / NK, pro-B and pre-B cells, with less pronounced but significant reductions across almost all other subpopulations, suggesting a mixed depletion pattern between that observed with 4G8 and CSL362 CARs alone (Figure 48H).

[0267] Overall, these data provide a proof of concept that multiplex epitope engineering can be efficiently obtained in HSPCs, enabling potent multitargeted immunotherapy that reduces the off-tumor hematopoietic toxicity of overlapping on-targets.

[0268] See Examples 15-21: 1. Brentjens, RJet al.CD19-targeted T cells rapidly induce molecular remissions in adults with chemotherapy-refractory acute lymphoblastic leukemia.Sci Transl Med 5, 177ra38(2013). 2.Turtle,CJet al.Durable molecular remissions in chronic lymphocytic leukemia treated with CD19-specific chimeric antigen receptor-modified T cells after failure of ibrutinib.J.Clin.Oncol.35, 3010-3020 (2017). 3.Fry,TJet al.CD22-targeted CAR T cells induce remission in B-ALL that is naive or resistant to CD19-targeted CAR immunotherapy. Nat.Med.24, 20 (2018). 4.Brentjens,R.J.et al.Safety and persistence of adoptively transferred autologous CD19-targeted T cells in patients with relapsed or chemotherapy refractory B-cell leukemias.Blood 118,4817(2011). 5.Spiegel,J.Y.et al.CAR T cells with dual targeting of CD19 and CD22 in adult patients with recurrent or refractory B cell malignancies:a phase 1 trial. Nat Med 27, 1419 (2021). 6.Kantarjian,H.et al.Acute myeloid leukemia:current progress and future directions. Blood Cancer J 11, 41 (2021). 7.Longo,D.L.,Dohner,H.,Weisdorf,D.J.&Bloomfield,C.D.Acute Myeloid Leukemia.n engl j med 12, 1136-52 (2015). 8.Dohner,H.et al.Diagnosis and management of AML in adults:2017 ELN recommendations from an international expert panel. Blood vol. 129 424-447 Preprint at https: / / doi.org / 10.1182 / blood-2016-08-733196(2017). 9.Isidori,A.et al.Immunotherapy in Acute Myeloid Leukemia:Where We Stand. Article 11,1(2021). 10.Pollyea,D.A.&Jordan,C.T.Therapeutic targeting of acute myeloid leukemia stem cells.Blood vol.129 1627-1635 Preprint at https: / / doi.org / 10.1182 / blood-2016-10-696039 (2017). 11.Hofmann,S.et al.Chimeric Antigen Receptor (CAR) T Cell Therapy in Acute Myeloid Leukemia (AML).J Clin Med 8, 200(2019). 12.Haubner,S.et al.Coexpression profile of leukemic stem cell markers for combinatorial targeted therapy in AML.Leukemia 2018 33:1 33, 64-74(2018). 13.Kugler,M.et al.A recombinant trispecific single-chain Fv derivative directed against CD123 and CD33 mediates effective elimination of acute myeloid leukaemia cells by dual targeting.Br Jematol 150, 574-86(2010). 14.Lambert,J.et al.Gemtuzumab ozogamicin for de novo acute myeloid leukemia:final efficacy and safety updates from the open-label,phase III ALFA-0701 trial. Haematologica 104, 113-119 (2019). 15.Chen,L.et al.Targeting FLT3 by chimeric antigen receptor T cells for the treatment of acute myeloid leukemia.Leukemia vol.31 1830-1834Preprint at https: / / doi.org / 10.1038 / leu.2017.147(2017). 16.el Khawanky,N.et al.Demethylating therapy increases anti-CD123 CAR T cell cytotoxicity against acute myeloid leukemia.Nature Communications 2021 12:1 12,1-20(2021). 17.Zhang,H.et al.Successful Anti-CLL1 CAR T-Cell Therapy in Secondary Acute Myeloid Leukemia.Front Oncol 10,685(2020). 18.Kenderian,S.S.et al.CD33-specific chimeric antigen receptor T cells exhibit potent preclinical activity against human acute myeloid leukemia.Leukemia 29,1637-47(2015). 19.Wang,J.et al.CAR-T cells targeting CLL-1 as an approach to treat acute myeloid leukemia.doi:10.1186 / s13045-017-0553-5. 20.Casucci,M.et al.CD44v6-targeted T cells mediate potent antitumor effects against acute myeloid leukemia and multiple myeloma. (2013) doi:10.1182 / blood-2013-04. 21.Myburgh,R.et al.Anti-human CD117 CAR T-cells efficiently eliminate healthy and malignant CD117-expressing hematopoietic cells.Leukemia 34,2688-2703(2020). 22.Jetani,H.et al.CAR T-cells targeting FLT3 have potent activity against FLT3-ITD + AML and act synergistically with the FLT3-inhibitor crenolanib.Leukemia 32,1168-1179(2018). 23.F,B.et al.Gene-edited stem cells enable CD33-directed immune therapy for myeloid malignancies.Proc Natl Acad Sci USA 116, 11978-87. 24.Kim,M.Y.et al.Genetic Inactivation of CD33 in Hematopoietic Stem Cells to Enable CAR T Cell Immunotherapy for Acute Myeloid Leukemia.Cell 173,1439-1453.e19(2018). 25.Ulyanova,T.,Blasioli,J.,Woodford-Thomas,T.A.&Thomas,M.L.The sialoadhesin CD33 is a myeloid-specific inhibitory receptor. Eur.J.Immunol. 29, 3440-3449 (1999). 26.Paul,S.P.,Taylor,L.S.,Stansbury,E.K.&McVicar,D.W.Myeloid specific human CD33 is an inhibitory receptor with differential ITIM function in recruiting the phosphatases SHP-1 and SHP-2.Blood 96, 483-490(2000). 27.Ishida,A.et al.Negative regulation of Toll-like receptor-4 signaling through the binding of glycosylphosphatidylinositol-anchored glycoprotein,CD14,with the sialic acid-binding lectin,CD33.J.Biol.Chem.289, 25341-25350 (2014). 28.Jacoby,E.et al.CD19 CAR immune pressure induces B-precursor acute lymphoblastic leukaemia lineage switch exposing inherent leukaemic plasticity.Nat Commun 7,(2016). 29.Ruella,M.&Maus,M.v.Catch me if you can:Leukemia Escape after CD19-Directed T Cell Immunotherapies. Comput Struct Biotechnol J 14, 357-362(2016). 30.Sotillo,E.et al.Convergence of Acquired Mutations and Alternative Splicing of CD19 Enables Resistance to CART-19 Immunotherapy.Cancer Discov 5,1282-1295(2015). 31.Vago,L.et al.Loss of mismatched HLA in leukemia after stem-cell transplantation.N Engl J Med 361, 478-488(2009). 32.Grafone,T.,Palmisano,M.,Nicci,C.&Storti,S.An overview on the role of FLT3-tyrosine kinase receptor in acute myeloid leukemia:Biology and treatment. Oncology Reviews vol.6 64-74 Preprint at https: / / doi.org / 10.4081 / oncol.2012.e82012)(. 33.Mekawy,M.A.,Eissa,D.S.,Hamza,M.T.,Hamed,G.M.&Youssef,M.K.FLT3 receptor / CD135 expression by flow cytometry in acute myeloid leukemia:Relation to FLT3 gene mutations and mRNA transcripts. Egyptian Journal of Medical Human Genetics 19, 345-351(2018). 34.F Kuchenbauer,W.K.C.S.A.K.W.H.T.H.Detailed analysis of FLT3 expression levels in acute myeloid leukemia.Haematologica 90,1617-25(2005). 35.Valverde,L.R.et al.C-kit receptor (CD117) expression in acute leukemia.Ann Hematol 72,11-15(1996). 36.Gao,X.et al.High Expression of c-kit mRNA Predicts Unfavorable Outcome in Adult Patients with t(8;21) Acute Myeloid Leukemia.PLoS One 10,e0124241(2015). 37.Schwartz,S.et al.Expression of the c-kit receptor (CD117) is a feature of almost all subtypes of de novo acute myeloblastic leukemia (AML),including cytogenetically good-risk AML,and lacks prognostic significance.Leuk Lymphoma 34,85-94(1999). 38.Bras,A.E.et al.CD123 expression levels in 846 acute leukemia patients based on standardized immunophenotyping.Cytometry B Clin Cytom 96, 134(2019). 39.Ehninger,A.et al.Distribution and levels of cell surface expression of CD33 and CD123 in acute myeloid leukemia.Blood Cancer J 4,(2014). 40.Haubner,S.et al.Coexpression profile of leukemic stem cell markers for combinatorial targeted therapy in AML.Leukemia 2018 33:1 33, 64-74(2018). 41.Perriello,V.M.et al.CD123 Is Consistently Expressed on NPM1-Mutated AML Cells.CancersBasel)13, 496 (2021)(. 42.Kandeel,E.Z.,Madney,Y.,Eldin,D.N.&Shafik,N.F.Overexpression of CD200 and CD123 is a major influential factor in the clinical course of pediatric acute myeloid leukemia.Exp Mol Pathol 118,(2021). 43.Cheng,J.,Qu,L.,Wang,J.,Cheng,L.&Wang,Y.High expression of FLT3 is a risk factor in leukemia.Mol Med Rep 17,2885(2018). 44.Hofmann,M.et al.Generation,selection and preclinical characterization of an Fc-optimized FLT3 antibody for the treatment of myeloid leukemia. (2012) doi:10.1038 / leu.2011.372. 45.Reshetnyak,A.v et al.Structural basis for KIT receptor tyrosine kinase inhibition by antibodies targeting the D4 membrane-proximal region.doi:10.1073 / pnas.1317118110. 46.Herzog,E.et al.Pharmacodynamic Activity and Preclinical Safety of CSL362,a Novel Humanised,Affinity Matured Monoclonal Antibody Against Human Interleukin 3 Receptor.Blood 120,1524-1524(2012). 47.Xie,L.H.et al.CD123 target validation and preclinical evaluation of ADCC activity of anti-CD123 antibody CSL362 in combination with NKs from AML patients in remission.Blood Cancer J 7,(2017). 48.Broughton,S.E.et al.Dual Mechanism of Interleukin-3 Receptor Blockade by an Anti-Cancer Antibody.CellReports 8, 410-419(2014). 49.Li,Y.et al.A novel Flt3-deficient HIS mouse model with selective enhancement of human DC development.Eur J Immunol 46, 1291-1299(2016). 50.Warmuth,M.,Kim,S.,Gu,X.J.,Xia,G.&Adrian,F.Ba / F3 cells and their use in kinase drug discovery.Curr Opin Oncol 19, 55-60(2007). 51.el Khawanky,N.et al.Demethylating therapy increases anti-CD123 CAR T cell cytotoxicity against acute myeloid leukemia.Nat Commun 12,(2021). 52.Vavassori,V.et al.Modeling,optimization,and comparable efficacy of T cell and hematopoietic stem cell gene editing for treating hyper-IgM syndrome.EMBO Mol Med 13,(2021). 53.Cieri,N.et al.IL-7 and IL-15 instruct the generation of human memory stem T cells from naive precursors.Blood 121, 573-584(2013). 54.Genovese,P.et al.Targeted genome editing in human repopulating haematopoietic stem cells.Nature 510, 235-240(2014). 55. Lennartsson,J.& Ronnstrand,L.Stem cell factor receptor / c-Kit:From basic Science to clinical implications. Physiol Rev 92,1619-1649(2012). 56.Basso-Ricci,L.et al.Multiparametric Whole Blood Dissection:A one-shot comprehensive picture of the human hematopoietic system. Cytometry 91, 952(2017). 57.Dolgin,E.First CD123-targeted drug approved after wowing in rare cancer.Nat Biotechnol 37, 202-203(2019). 58.Biasco,L.et al.In Vivo Tracking of Human Hematopoiesis Reveals Patterns of Clonal Dynamics during Early and Steady-State Reconstitution Phases.Cell Stem Cell 19, 107-119(2016). 59.Leibowitz,M.L.et al.Chromothripsis as an on-target consequence of CRISPR-Cas9 genome editing.Nat Genet 53, 895-905(2021). 60.Schiroli,G.et al.Precise Gene Editing Preserves Hematopoietic Stem Cell Function following Transient p53-Mediated DNA Damage Response.Cell Stem Cell24,551-565.e8(2019). 61.Ferrari,S.et al.Efficient gene editing of human long-term hematopoietic stem cells validated by clonal tracking.Nat Biotechnol 38,1298(2020). 62.Gill,S.et al.Preclinical targeting of human acute myeloid leukemia and myeloablation using chimeric antigen receptor-modified T cells.Blood 123,2343-2354(2014). 63.Russkamp,NF,Myburgh,R.,Kiefer,JD,Neri,D.&Manz,MGAnti-CD117 immunotherapy to eliminate hematopoietic and leukemia stem cells.Exp Hematol 95,31-45(2021). 64.Hand,THet al.Catalytically Enhanced Cas9 through Directed Protein Evolution.CRISPR Journal 4, 223-232(2021).

[0269] Other embodiments While the invention has been described in conjunction with the above embodiments, it should be understood that the foregoing description and examples are intended to be illustrative and not limiting of the scope of the invention. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims

Claim 1: A genetically engineered hematopoietic stem cell (HSPC) comprising an edited gene encoding a cell surface protein comprising one or more mutations within an epitope to which a cytotoxic therapeutic antibody binds, the one or more mutations reduce or eliminate binding of the cytotoxic therapeutic antibody to the cell surface protein compared to the corresponding wild-type cell surface protein; The genetically engineered hematopoietic stem cell (HSPC), wherein the mutant cell surface protein has substantially the same biological activity as the wild-type cell surface protein. (a) the edited gene encoding the cell surface protein is a genetically engineered FLT3 gene and the cytotoxic therapeutic antibody is a therapeutic anti-FLT3 antibody, wherein the genetically engineered FLT3 gene has been engineered to reduce binding of the encoded cell surface protein to the therapeutic anti-FLT3 antibody; or (b) the edited gene encoding the cell surface protein is a genetically engineered CD123 gene and the cytotoxic therapeutic antibody is a therapeutic anti-CD123 antibody, wherein the genetically engineered CD123 gene has been engineered to reduce binding of the encoded cell surface protein to the therapeutic anti-CD123 antibody; or (c) the edited gene encoding the cell surface protein is a genetically engineered KIT gene and the cytotoxic therapeutic antibody is a therapeutic anti-KIT antibody, wherein the genetically engineered KIT gene has been engineered to reduce binding of the encoded cell surface protein to the therapeutic anti-KIT antibody. (a) the engineered FLT3 gene comprises at least one mutation in exon 9 of the FLT3 gene, optionally wherein the at least one mutation in exon 9 of the engineered FLT3 gene results in a polypeptide having a mutation at position N399, optionally wherein the mutation at position N399 is N399D or N399G; or (b) the engineered CD123 gene comprises at least one mutation in exon 2 of the CD123 gene, optionally wherein the at least one mutation in exon 2 of the engineered CD123 gene results in a polypeptide having a mutation at position S59, optionally wherein the mutation at S59 is S59P or S59F; or (c) the engineered CD123 gene comprises at least one mutation in exon 3 of the CD123 gene, optionally wherein the at least one mutation in exon 3 of the engineered CD123 gene results in a polypeptide having a mutation at position P88, optionally wherein the mutation at P88 is P88L or P88S; or (d) the engineered KIT gene comprises at least one mutation in exon 7 of the KIT gene, optionally wherein the at least one mutation in exon 7 of the engineered KIT gene results in a polypeptide having a mutation at position H378, optionally wherein the mutation at position H378 is H378R. (a) the therapeutic anti-FLT3 antibody is the anti-FLT3 clone 4G8 antibody, or an antibody that has the same six CDRs as the 4G8 antibody or competes with it; or (b) the engineered CD123 gene comprises at least one mutation in exon 2 of the CD123 gene, and the therapeutic anti-CD123 antibody is the clone 7G3 antibody or its humanized counterpart CSL362; or (c) the engineered CD123 gene comprises at least one mutation in exon 3 of the CD123 gene, and the therapeutic anti-CD123 antibody is anti-CD123 clone 6H6 antibody or anti-CD123 clone S18016F antibody; or (d) the therapeutic anti-KIT antibody is an anti-KIT clone Fab79D antibody.

5. A population of genetically engineered hematopoietic stem cells (HSPCs) according to any one of claims 1 to 4 for use in therapy.

6. 1. A pharmaceutical composition for use in a method of treating a hematopoietic malignancy, comprising: The pharmaceutical composition comprises a population of genetically engineered hematopoietic stem cells (HSPCs), The method comprises administering to a human subject: (a) a population of genetically engineered hematopoietic stem cells comprising the genetically engineered FLT3 gene of claim 2; (b) a therapeutically effective amount of at least one agent comprising an anti-FLT3 antibody binding domain or an antibody or antibody fragment comprising an anti-FLT3 binding domain; The pharmaceutical composition, wherein the at least one agent reduces or eliminates binding to the engineered FLT3 protein encoded by the engineered FLT3 gene compared to wild-type FLT3 protein.

7. The at least one agent is (a) a chimeric antigen receptor (CAR)-expressing immune cell, optionally a CAR-T cell; or (b) The pharmaceutical composition of claim 6, which is an antibody-drug conjugate (ADC).

8. A pharmaceutical composition for use in a method for treating a hematopoietic malignancy, comprising: The pharmaceutical composition comprises a population of genetically engineered hematopoietic stem cells (HSPCs), The method comprises administering to a human subject: (a) a population of genetically engineered hematopoietic stem cells comprising the genetically engineered CD123 gene of claim 2; (b) a therapeutically effective amount of at least one agent comprising an anti-CD123 antibody binding domain or an antibody or antibody fragment comprising an anti-CD123 binding domain; The pharmaceutical composition, wherein the at least one agent reduces or eliminates binding to the engineered CD123 protein compared to wild-type CD123 protein.

9. The at least one drug (a) a chimeric antigen receptor (CAR)-expressing immune cell, optionally a CAR-T cell; or (b) The pharmaceutical composition of claim 8, which is an antibody-drug conjugate (ADC).

10. A pharmaceutical composition for use in a method for treating a hematopoietic malignancy, comprising: The pharmaceutical composition comprises a population of genetically engineered hematopoietic stem cells (HSPCs), The method comprises administering to a human subject: (a) a population of genetically engineered hematopoietic stem cells comprising the genetically engineered KIT gene of claim 2; (b) a therapeutically effective amount of at least one agent comprising an anti-KIT antibody binding domain or an antibody or antibody fragment comprising an anti-KIT binding domain; The pharmaceutical composition, wherein the at least one agent reduces or eliminates binding to the engineered KIT protein compared to wild-type KIT protein.

11. The at least one drug (a) a chimeric antigen receptor (CAR)-expressing immune cell, optionally a CAR-T cell; or (b) The pharmaceutical composition of claim 10, which is an antibody-drug conjugate (ADC).

12. The CAR, (a) a hinge domain that is a CD28 hinge, an IgG4 hinge, or a CD8α hinge; and / or (b) a transmembrane domain that is CD28 TM, CD8α TM, or 4-1BB TM; and / or (c) a costimulatory domain that is CD28z, 4-1BB, ICOS, or OX40; and / or (d) a cytoplasmic signaling domain that is CD3z 12. The pharmaceutical composition of any one of claims 7, 9 and 11, comprising:

13. The pharmaceutical composition according to any one of claims 6, 8 and 10, wherein the hematopoietic malignancy is B-lymphoblastic leukemia (BLL), acute myeloid leukemia (AML), T-cell acute lymphoblastic leukemia (T-ALL) or blastic plasmacytoid dendritic cell leukemia (BPCDN).

14. An ex vivo method for preparing the genetically engineered HSPCs of any one of claims 1 to 4, comprising: (a) obtaining HSPCs from a biological sample from a human subject; and (b) genetically engineering HSPCs from said biological sample by introducing one or more mutations into a gene encoding a cell surface protein comprising an epitope to which a cytotoxic therapeutic antibody binds, wherein said one or more mutations reduce or eliminate binding of said cytotoxic therapeutic antibody to said cell surface protein compared to the corresponding wild-type cell surface protein. The ex vivo method comprising:

15. 15. The method of claim 14, wherein the biological sample is bone marrow cells, blood, umbilical cord blood cells, or mobilized peripheral blood-derived CD34+ hematopoietic stem and progenitor cells.

16. A pharmaceutical composition comprising a population of genetically engineered hematopoietic stem cells (HSPCs) according to any one of claims 1 to 4.

Images