MAMMALIAN CELL LINE FOR PROTEIN PRODUCTION AND LIBRARY GENERATION

Abstract

According to a first aspect of the invention, a method for the generation of a cell line is provided, comprising the steps of (a) providing a plurality of mammalian B cells, wherein each of the plurality of B cells comprises a transgenic genomic DNA sequence encoding a marker protein inserted into an endogenous immunoglobulin locus comprised in said B cell, and wherein the transgenic genomic DNA sequence is amenable to cleavage by a site directed nuclease, particularly Cas9: (b) replacing the transgenic genomic DNA sequence encoding a marker protein with a second transgenic DNA sequence encoding a protein of interest; (c) sorting B cells based on the presence or absence of the marker protein; and (d) collecting B cells in which the marker protein is absent.

Claims

1.-20. (canceled)

21. A method for the generation of a modified B cell line, comprising the steps of: (a) providing a recombinant mammalian B cell comprising: (i) a first transgenic genomic DNA sequence inserted into an endogenous immunoglobulin gene locus, (ii) a nucleic acid encoding a CRISPR-associated endonuclease under the control of a non-immunoglobulin promoter, or a constitutive promoter, and (iii) a safe harbor locus; wherein the first transgenic genomic DNA sequence comprises: (a) a nucleic acid that encodes a marker protein that is not endogenous to a mammalian B cell and the marker protein is under the control of an endogenous immunoglobulin promoter; and (b) two homology arms corresponding to the endogenous immunoglobulin locus; wherein the recombinant mammalian B cell transiently or constitutively expresses the CRISPR-associated endonuclease; and wherein the nucleic acid encoding the protein marker replaces the endogenous immunoglobulin gene; (b) replacing the first transgenic DNA sequence encoding the marker protein with a second transgenic DNA sequence comprising a protein of interest that is not endogenous to the mammalian B cell, and two homology arms corresponding to the endogenous immunoglobulin gene locus using a CRISPR-mediated site-directed DNA cleavage, and subsequent inactivation of the first transgenic DNA sequence by non-homologous end joining (NHEJ), and integration of the second transgenic DNA sequence by homology directed repair (HDR) to generate a plurality of modified mammalian B cells; (c) sorting the plurality of modified mammalian B cells based on the presence or absence of the marker protein; and (d) selecting and collecting modified mammalian B cells in which the marker protein is absent.

22. The method of claim 21, wherein: (a) the marker protein is a fluorescent protein; (b) the marker protein is a fluorescent protein comprises a guide RNA target site that is amenable to cleavage by the CRISPR-associated endonuclease; or (c) the CRISPR-associated endonuclease is a CRISPR-Cas endonuclease, or a CRISPR-Cas9 endonuclease.

23. The method of claim 21, wherein: (a) the recombinant mammalian B cell constitutively expresses the CRISPR-associated endonuclease; or (b) the CRISPR-associated endonuclease is inserted into the safe harbor locus.

24. The method of claim 21, wherein the recombinant mammalian B cell further comprises a nucleic acid encoding an exogenous activation-induced cytidine deaminase (AID) under the control of an inducible promoter or titratable promoter.

25. The method of claim 24, wherein the nucleic acid encoding the exogenous AID is integrated into a safe harbor locus or into a native AID locus.

26. The method of claim 24, wherein the nucleic acid encoding the exogenous AID is integrated into the safe harbor locus.

27. The method of claim 21, wherein the safe harbor locus is selected from a murine Rosa26 locus or an AAVS1 locus.

28. The method of claim 24, wherein the exogenous AID is expressed under the inducible promoter.

29. The method of claim 28, wherein the inducible promoter is a TRE3GS promoter.

30. The method of claim 21, wherein the recombinant mammalian B cell further comprises an expression cassette comprising a TRE3GS promoter, a DNA sequence encoding an activation-induced cytidine deaminase (AID), a human phosphoglycerate kinase 1 promoter (hPGK), a Tet-On 3G transactivator protein, and a SV40 poly-A signal.

31. The method of claim 24, wherein the inducible expression of the AID generates multiple genomic mutations within the protein of interest by inducible synthetic somatic hypermutation (iSSHM), thereby generating a library of protein of interest variants.

32. The method of claim 21, wherein the recombinant mammalian B cell further comprises one or more randomized nucleic acid sequences that are homologous to one or more regions of the protein of interest.

33. The method of claim 32, wherein the one or more randomized nucleic acid sequences comprise a donor dsDNA, a donor ssDNA, degenerate nucleotides, or trinucleotide codons.

34. The method of claim 32, further comprising generating genomic mutations within the protein of interest by site-directed mutagenesis with the one or more randomized nucleic acid sequences, thereby generating a library of protein of interest variants.

35. The method of claim 21, wherein: (a) in the recombinant mammalian B cell, the endogenous VH gene and the endogenous VL gene are disrupted; (b) the endogenous immunoglobulin promoter is a VH promoter; (c) the recombinant B cell is a human cell; or (d) the recombinant B cell is selected from the group consisting of a primary B cell, an immortalized B cell, a hybridoma cell, a myeloma cell, a plasmacytoma cell, and a lymphoma cell.

36. The method of claim 21, wherein the protein of interest is selected from the group consisting of: (a) a full-length antibody, a synthetic antigen binding fragment, a humanized camelide antibody, an immunoglobulin antigen-binding fragment, or a full-length antibody comprising a synthetic antigen binding fragment (sFAb); (b) a designed ankyrin repeat protein; and (c) a polypeptide comprising an armadillo repeat, a leucine-rich repeat, a tetratricopeptide repeat, a protein A domain, a fibronectin domain FN3, a consensus fibronectin domain, a lipocalin domain, a Zinc finger domain, a Src homology domain 2 (SH2), a Src homology domain 3 (SH3), a PDZ domain, a gamma-crystallin domain, a ubiquitin domain, a cysteine knot domain, or a knottin domain.

37. A method for the generation of a library of protein variants, the method comprising: (a) providing a recombinant mammalian B cell comprising: (i) a first transgenic genomic DNA sequence inserted into an endogenous immunoglobulin gene locus, (ii) a nucleic acid encoding a CRISPR-associated endonuclease under the control of a non-immunoglobulin promoter, or a constitutive promoter, and (iii) a safe harbor locus; wherein the first transgenic genomic DNA sequence comprises: (a) a nucleic acid that encodes a marker protein that is not endogenous to a mammalian B cell and the marker protein is under the control of an endogenous immunoglobulin promoter; (b) two homology arms corresponding to the endogenous immunoglobulin locus; and (c) a guide RNA target site; wherein the recombinant mammalian B cell transiently or constitutively expresses the CRISPR-associated endonuclease; (b) replacing the first transgenic DNA sequence encoding the marker protein with an expression cassette comprising a DNA sequence encoding a protein of interest that is not endogenous to a mammalian B cell, a DNA sequence encoding an exogenous activation-induced cytidine deaminase (AID) under the control of an inducible promoter, two homology arms corresponding to the safe harbor locus, and a gRNA targeting the safe harbor locus, using a CRISPR-mediated site-directed DNA cleavage, and subsequent inactivation of the first transgenic DNA sequence by non-homologous end joining (NHEJ), and integration of the expression cassette by homology directed repair (HDR) to generate a plurality of modified mammalian B cells; (c) inducing the expression of the exogenous AID to generate multiple genomic mutations within the protein of interest by inducible synthetic somatic hypermutation (iSSHM), thereby generating a library of protein of interest variants; (d) generating a plurality of modified recombinant mammalian B cells, wherein each modified recombinant mammalian B cell comprises a transgenic genomic DNA sequence encoding a variant of the protein of interest; (e) sorting the plurality of modified recombinant mammalian B cells based on the presence or absence of the marker protein; and (f) selecting and collecting modified mammalian B cells in which the marker protein is absent.

38. The method of claim 37, wherein each variant of the protein of interest expressed by a member of the plurality of modified recombinant mammalian B cells is different from any other variant of the protein of interest expressed by another member of the plurality of modified recombinant mammalian B cells.

39. The method of claim 37, wherein each variant is different from another variant in one to five positions of its amino acid sequence.

40. The method of claim 37, wherein each variant is at least 80%, 85%, 90%, 95%, 97%, 98% or 99% identical to any another variant encoded by a member of the library of protein of interest variants.

Description

DESCRIPTION OF THE FIGURES

[0136] FIG. 1 shows the generation of PnP-mRuby cells. (A) Schematic shows wildtype (WT) hybridoma cells expressing antibody will be converted into PnP-mRuby. (B) Shown is the targeting of WT IgH genomic locus with the following annotations: leader sequence (L), mRNA splice sites (SS), V.sub.H, and IgG constant heavy region (CH1). The CRISPR-Cas9 gRNA target site (black) is in the intron between V.sub.H and CH1. The donor construct consists of mRuby gene with a stop codon flanked by two homology arms of 732 and 711 bp. The PnP-mRuby IgH locus is generated by transfection of WT cells with CRISPR-Cas9 plasmid (pX458) and donor construct, which will result in HDR-based exchange of the V.sub.H region with mRuby. (C) Flow cytometry dot plot shows WT cells are exclusively IgH-positive and mRuby-negative, where PnP-mRuby cells are exclusively mRuby-positive and IgH-negative. (D) PCR was performed on genomic DNA from WT and PnP-mRuby cells using a forward primer in 5 HA and reverse primer that is external of the 3 HA. Agarose gel shows the expected size of bands. The band from PnP-mRuby cells was extracted and Sanger sequencing confirmed mRuby exchange of the V.sub.H region. (E) Shown is the targeting of WT hybridoma IgK locus with the following annotations: V.sub.L, and IgK constant light region (CK), other annotations same as in shown in (A). Two gRNA target sites are utilized in order to delete the V.sub.L region. (F) Flow cytometry dot plot shows WT cells are strongly IgH- and IgK-positive, where PnP-mRuby cells are exclusively IgH- and IgK-negative. (E) PCR was performed on genomic DNA from WT and PnP-mRuby cells using a forward primer 5 of the gRNA-F site and reverse primer 3 of gRNA-H site. Agarose gel shows the expected size of band for WT cells and nearly no amplification product for the PnP-mRuby cell line. Throughout this figure, WT cells correspond to clone WEN1.3 and PnP-mRuby cells correspond to clone 1E9.C3 (see table 1).

[0137] FIG. 2 shows the generation of PnP-mRuby hybridomas reprogrammed to surface express and secrete a new antibody. (A) Schematic shows PnP-mRuby cells expressing mRuby will be converted back into hybridomas expressing a new antibody via sFAb. (B) Shown is simplified design of the sFAb donor construct (for complete design details, see FIG. S8). (C) Shown is the PnP-mRuby IgH locus where gRNA-J target site is in the mRuby gene. PnP-mRuby cells transfected with pX458.2 and sFAb donor will result in HDR-driven genomic replacement of mRuby. The new antibody will then be expressed on a single mRNA transcript. (D) Flow cytometry dot plot shows the different populations that emerge following transfection of PnP-mRuby with pX458 and sFAb donor. Cells which were positive for IgH expression were sorted. (E) Flow cytometry dot plot shows initial population of PnP-mRuby cells and resulting cells (PnP-HEL23) from sorted IgH-positive population in (D), which are now strongly positive for IgH and IgK expression. (F) Graph shows sandwich ELISA results (capture anti-IgK, primary detection anti-IgH) on hybridoma culture supernatant, PnP-HEL23 show IgG secretion levels similar to WT. (G) PCR was performed on the genomic DNA of WT, PnP-mRuby, PnP-HEL23 cells using primers shown in (C). Agarose gel from genomic PCR shows the predicted band size in PnP-HEL23. (H) RT-PCR from mRNA results in a visible band present of the correct size present only in PnP-HEL23. The bands from PnP-HEL23 were extracted and Sanger sequencing confirmed correct integration of the PnP-sFAb construct. Throughout this figure, WT cells correspond to clone WEN1.3, PnP-mRuby cells correspond to clone 1E9.C3, and PnP-HEL23 correspond to clone Y (see table 1).

[0138] FIG. 3 shows PnP-HEL23 cells that surface express and secrete an antigen-specific antibody. (A) Flow cytometry histogram shows PnP-HEL23 cells surface express antibody specific for cognate antigen HEL. (B) ELISA data shows that PnP-HEL23 cells secrete antibody specific for HEL. WT cells correspond to clone WEN1.3, PnP-mRuby cells correspond to clone 1E9.C3, PnP-IgG cells correspond to clone Y (see table 1).

[0139] FIG. 4 shows the rapid and reproducible generation of PnP-antibody producing cells. (A) Flow cytometry dot plot shows PnP-mRuby following transfection with pX458.2 and different PnP-sFAb donor constructs. Cells were sorted for IgH expression. (B) Flow cytometry dot plots show that all three PnP cell lines express IgH and IgK following sorting in (A). (C) As in FIGS. 2G and 2H, PCR and RT-PCR was performed on genomic DNA and mRNA, respectively. (D, E) Similar to (A) and (B) are flow cytometry dot plots for PnP-HEL23.2 cells, with the exception that Cas9 sorting step was omitted. PnP-HyHEL10 corresponds to clone U, PnP-EBV-2G4 corresponds to clone AA, and PnP-EBV-4G7 corresponds to clone AB, PnP-HEL23 corresponds to clone AC (see table 1).

[0140] FIG. 5 shows the generation of PnP hybridoma cell that express Cas9 and AID. (a) Shown is the CRISPR-Cas9 targeting of PnP-mRuby Rosa 26 locus. A donor construct will provide Cas9 gene with the 2A peptide and puromycin resistance gene. Integration into this locus will provide constitutive expression of Cas9 in all PnP-mRuby cells. (b) and (c) Shown is the integration of an inducible AID locus into PnP-mRuby-Cas9 cells, either by integration into the Rosa26 locus or into the native AID locus. (d) Schematic shows that large libraries will be generated by inducible synthetic somatic hypermutation via expression of AID. These libraries can then be used for directed evolution and high-throughput screening by flow cytometry.

[0141] FIG. 6 shows the validation of CRISPR-Cas9 targeting of immunoglobulin loci of hybridoma cells. (A) Flow cytometry dotplot shows expression of Cas9-2A-GFP in WEN1.3 cells following transfection with with pX458. (B) Surveyor results validate CRISPR-Cas9 targeting in IgH locus (agarose gels of all gRNA sites tested). (C) Surveyor results validate CRISPR-Cas9 targeting in IgK locus (agarose gels of all gRNA sites tested).

[0142] FIG. 7 shows targeting of the ROSA26 locus by CRISPR-Cas9. (A) ROSA26 locus (mouse chromosome 6) with CRISPR targets identified; displayed are also the primers used for fragment amplification for cleavage analysis. (B) DNA gels from Surveyor assay showing that all 4 tested guides induce successful CRISPR cleavage. The fragments match the expected size, shown in (A).

[0143] FIG. 8 shows design and validation of PnP-Cas9 cell lines. (A) The ROSA26 locus is targeted for CRISPR-Cas9 induced HDR integration of the constitutive Cas9 cassette. Contained in the cassette are two separate genes. The SpCas9-2A-Puromycin gene with bovine growth hormone (bCGh), the GFP gene is under transcriptional control of the murine pPGK promoter. The 5 and 3 homology arms are also present in the construct. 1: SV40 pA; 2: AID; 3: F2A; 4: GFP; 5: TRE3Gs promoter; 6: hPGK promoter; 7: Tet-On 3G; 8: SV40 pA; 9: homology arm; 10: pCAG promoter; 11: SpCas9; 12: T2A; 13: Puromycin; 14: bGH pA: 15: mPGK promoter; 16: GFP; 17: homology arm. (B) A close up of the Cas9 donor construct, shown are guide RNA target sites within GFP or pPGK, which are used to subsequently inactivate GFP by Cas9-induced NHEJ. (C) Sanger sequencing results before or after introduction of guide RNA in PnPCas9. (D) The T7E1 assay confirms that PnP-Cas9 cells in the presence of guide RNA lead to Cas9-induced NHEJ of GFP cells. (E) Flow cytometry plots show that in PnP-mRuby-Cas9 cells, the addition of gRNA targeting mRuby leads to knockout of mRuby expression in nearly all cells.

[0144] FIG. 9 shows generation and selection of PnP-iAID-mRuby cell lines. (A) Shown is the integration of the ISSHM donor cassette (GFP-2A-AID construct under a Doxinducible promoter system) is integrated into the Rosa26 locus of hybridoma cells by Cas9-induced HDR. Hybridoma cells express mRuby in their reprogrammed IgH locus. (B) Cell expressing Cas9 (2A-BFP) are sorted, then in the presence of Dox (or absence for negative controls), GFP-positive cells are single cell sorted and expanded. (C) Characterization of GFP expression with or without Dox in the single-cell sorted colonies from B.

[0145] FIG. 10 shows generation and selection of PnP-iAID-IgG cell lines. (A) Shown is the integration of the ISSHM donor cassette (GFP-2A-AID construct under a Doxinducible promoter system) is integrated into the Rosa26 locus of hybridoma cells by Cas9-induced HDR. Hybridoma cells express IgG through sFAb in their reprogrammed IgH locus. (B) Cell expressing Cas9 (2A-BFP) are sorted, then in the presence of Dox (or absence for negative controls), GFP-positive cells are single cell sorted and expanded. (C) Characterization of GFP expression with or without Dox in the single-cell sorted colonies from B.

[0146] FIG. 11 shows the general workflow for optimizing ssODN induced HDR and constructing a synthetic antibody (sAb) library with CDRH3 genetic diversity. (A) A.I: PnP-HEL23 sAb construct in the heavy chain gene locus with a CDRH3 sgRNA site to direct cleavage by Cas9 following transfection with the Cas9 vector, pX458. The Cas9 induced double stranded break introduces insertions/deletions (InDels) near the cut site through NHEJ causing a frameshift mutation and dysfunctional protein expression. The sequence shown in exploded view is SEQ ID No 17. A.IIa: Antibody expression can then be restored through HDR promoted by donor ssODNs with codon rearrangements for the CDRH3. A.IIb: Additional genetic diversity into the CDRH3 of the sAb cassette through HDR incorporation of degenerate ssODNs (NNK randomization). Sorting results: upper left: A1 upper panel; upper right: A1 lower panel (A A.IIa/b upper panel); lower left: A.IIa lower panel; A.IIb lower panel. (B) HDR percentages are estimated by flow cytometric analysis though labeling with a fluorescently tagged antigen. Percentages displayed are cells that had regained antigen specific antibody expression towards HEL. Data presented is representative of n=2 experiments after screening for transfection positive (GFP+) cells. 1: Cas9 plasmid; 2: Cas9 RNP; 3: Cas9 RNP+Modified ssODN; 4: Cas9 Cell+Modified ssODN.

[0147] FIG. 12: Evaluation of inducible-AID's mulation activity. (a) Experiment outline. PnP-mRuby cells selected for integrated TetOne-AID (via GFP-2A) were induced by Dox (1 g/ml, induction renewed daily) for 72 hours and FACS sorted for high (unchanged) or low (decreased) mRuby fluorescence. Genomic DNA (gDNA) was extracted from the two populations and mRuby gene was clone and Sanger sequenced. (b) FACS plots displaying, from left to right: cells at sorting (the displayed gates and percentages are not the original ones, but re-created post-analysis for illustrative purpose); the two sorted populations after recovery: PnP-mRuby-AID Red-high, sorted for high mRuby expression, and PnP-mRuby-AID Red-low, sorted for decreased mRuby expression; PnP-mRuby cells used as positive control. Genomic DNA was isolated for sequencing analysis. (c) Sequences were mapped to mRuby and investigated for the presence of mutations. This graph shows the percentage of mutated clones for each sample. By big deletions it is meant anything bigger than 1 nucleotide. The only sample yielding clones with big deletions was PnP-mRuby-AID Red-low. Values on top of the bar report the actual frequency in the cohort (before % conversion) (d) Mutations per kb in the analysed clones. For each of the four samples/cohorts, the sequenced nucleotides (mRuby ORF only. 711 bp) for all analysed clones were summed, and the frequency of mutated nucleotides per kb was calculated consequentially. Notably, in case of deletions, each missing nucleotide was calculated as a mutated one. Note: the data reported in (c) and (d) does not account for the coding or non-coding (silent) outcome of single nt substitutions.

TABLE-US-00001 TABLE 1 Summary of hybridoma clones Cell Type Type Description WT WEN1.3 Wen 1.3 cells are derived from a mouse infected with LCMV. They express lgG2c and are specific for LCMV GP-1 antigen. PnP-mRuby 1E9.C3 WEN 1.3 cells were transfected with pX458 with gRNA-E and mRuby donor construct and sorted for Cas9 positive expression (2A-GFP). This was followed by a first round of sorting for mRuby-positive cells, followed by a second single cell sort for mRuby. A single cell clone was selected and then transfected with pX458 with gRNA-F and H and sorted for Cas9 positive expression (2A-GFP). Cells were then sorted for IgK negative expression. A second round of single cell sorting was performed followed by genomic PCR to identify a clone with VL deletion. This final clone represents 1E9.C3 PnP-mRuby-pA D2 PnP-mRuby-pA cells include a polyadenylation signal following the mRuby gene's stop codon to increase cell fluorescence. PnP-mRuby-pA cells were generated in an identical manner to PnP-mRuby cells, but with a donor construct including the polyadenylation signal. PnP-mRuby-Cas9 1AD Clone D2 was transfected with pX458 with gRNA-P and (winner Cas9-2A-Puro-GFP HDR donor linearized. Cells were selection sorted for Cas9 positive expression (2A-BFP) and in expanded. Cells were sorted for GFP positive expression progress) and expanded. Cells were then cultured in growth medium supplemented with 2.5 ug/ml puromycin for up to one week. Single cells were sorted for GFP positive expression and expanded. A suitable clone was selected based on genotypic and phenotypic characterization. PnP-HEL23 Y Clone 1E9.03 was transfected with pX458 with gRNA-J and HEL23-2A HDR donor linearized. Cells were sorted for Cas9 positive expression (2A-BFP) and expanded. Cells were then sorted for surface IgH expression and expanded, and finally characterized for IgH and IgK expression. PnP-HEL23-IgH.sup. IgH.sup. Clone Y was transfected with pX458 with gRNA-Q. Cells were sorted for Cas9 positive expression (2A-GFP) and expanded. A single cell sort for cells lacking surface IgH expression was performed followed by genomic PCR and Sanger sequencing to genotypically characterize the individual clones. A suitable clone was selected based on genotypic characterization. PnP-HyHEL10 U Clone 1E9.C3 was transfected with pX458 with gRNA-J and HyHEL10-2A HDR donor linearized. Cells were sorted for Cas9 positive expression (2ABFP) and expanded. Cells were then sorted for surface IgH expression and expanded, and then they underwent a second IgH sort. They were finally characterized for IgH and IgK expression. PnP-EBV-2G4 AA Clone 1E9.03 was transfected with pX458 with gRNA-J and 2G4-2A HDR donor linearized. Cells were sorted for Cas9 positive expression (2A-BFP) and expanded. Cells were then sorted for surface IgH expression and expanded, and then they underwent a third sort for IgH and IgK expression. They were finally characterized for IgH and IgK expression. PnP-EBV-4G7 AB Clone 1E9.C3 was transfected with pX458 with gRNA-J and 4G7-2A HDR donor linearized. Cells were sorted for Cas9 positive expression (2A-BFP) and expanded. Cells were then sorted for surface IgH expression and expanded, and then they underwent a third sort for IgH and IgK expression. They were finally characterized for IgH and IgK expression. PnP-HEL23-2.0 AC Clone 1E9.C3 was transfected with pX458 with gRNA-J and HEL23-2A HDR donor linearized. Cells were NOT sorted for Cas9 positive expression. They were sorted for surface IgH expression and expanded. They were finally characterized for IgH and IgK expression. [In order to achieve a higher purity, cells were eventually sorted a second time for IgH and IgK expression] PnP-mRuby-PA- (winner PnP-mRuby-pA cells were transfected with px458-BFP AID selection with gRNA-O and sorted for Cas9 expression (2A-BFP). in Cells were then induced by 1 g/ml Doxycycline, single- progress) cell sorted for GFP expression, and characterized by further induction cycles (GFP+, mRuby knock-out activity), genotyping and transcript analysis. PnP-mRuby-pA- (winner PnP-HEL23-IgH.sup. cells were transfected with px458-BFP IgH.sup. selection with gRNA-O and sorted for Cas9 expression (2A-BFP). in Cells were then induced by 1 g/ml Doxycycline, single- progress) cell sorted for GFP expression, and characterized by further induction cycles, genotyping and transcript analysis.

TABLE-US-00002 TABLE2 ListofgRNAs Targetingsequence Resident Targetregion (5-3Sequence+PAM) plasmid gRNA-A Wen1.3leader-VH SEQIDNO01: pX458 intron GCTGTCGGGAGAAAGAAATTGTGG gRNA-B Wen1.3leader-VH SEQIDNO02: pX458 intron GCCCTATCTCCTCTTCAGATTGG gRNA-C Wen1.3leader-VH SEQIDNO03 pX458 intron GTTCCAATCTGAAGAGGAGATAGG gRNA-D Wen1.3JHdownstream SEQIDNO04 pX458 intron GGAGCATGACGGACTAATCTTGG gRNA-E Wen1.3JHdownstream SEQIDNO05 pX458 intron GTTGGTTTTAGCGGAGTCCCTGG gRNA-F Wen1.3VKleader SEQIDNO06 pX458 GGAGAAGCAGGACCCATAGCAGG gRNA-G Wen1.3VKleader SEQIDNO07 pX458 GGCTATGGGTCCTGCTTCTCTGG gRNA-H Wen1.3JH SEQIDNO08 pX458 downstreamintron GGGATCTTCTATTGATGCACAGG gRNA-I Wen1.3JH SEQIDNO09 pX458 downstreamintron GTGGCTAAATGAGCCATTCCTGG gRNA-J mRuby2 SEQIDNO10 pX458.2 GTCATGGAAGGTTCGGTCAACGG (BFP) gRNA-K mRuby2 SEQIDNO11 pX458.2 GCATGCCGTTGATCACCGCCTGG (BFP) gRNA-L ROSA26 SEQIDNO12 pX458 GAGACCTCCATCGCGCACTCCGGG gRNA-M ROSA26 SEQIDNO13 pX458 GCAGACCTCCATCGCGCACTCCGG gRNA-N ROSA26 SEQIDNO14 pX458 GCCTCGATGGAAAATACTCCGAGG gRNA-O ROSA26 SEQIDNO15 pX458 GCGATGGAAAATACTCCGAGGCGG (BFP) gRNA-P ROSA26 SEQIDNO16 pX458 AAGCATGTATTGCTTTACGTGGG (BFP) gRNA-Q HEL23-2ACDR3 SEQIDNO17 pX458 TGCGCGCGTGATAGCAGCGGCGG gRNA-R HEL23-2A-IgHCDR3 SEQIDNO18 pX458 ATTGCGCGCGTGATAGCAGGCGG

[0148] The sequences listed in this table (SEQ ID NO 01-SEQ ID NO 18) refer to the DNA sequences encoding the targeting sequences of the respective gRNAs.

Methods

Hybridoma Cell Culture Conditions

[0149] The WT hybridoma cell line (Wen1.3) was obtained as a gift from Prof. Annette Oxenius (ETH Zrich). All hybridoma cell lines were cultivated in high-glucose Dulbecco's Modified Eagle Medium [(DMEM), Thermo Fisher Scientific (Thermo), 11960-044] supplemented with 10% heat inactivated fetal bovine serum [(FBS), Thermo, 10082-147)], 100 U/ml Penicillin/Streptomycin (Thermo, 15140-122), 2 mM Glutamine (Sigma-Aldrich, G7513), 10 mM HEPES buffer (Thermo, 15630-056) and 50 M 2-mercaptoethanol (Sigma-Aldrich, M3148). All hybridoma cells were maintained in incubators at a temperature of 37 C. and 5% CO.sub.2. Hybridomas were typically maintained in 10 ml of culture in T-25 flasks (Thermo, NC-156367), and split every 48/72 hours.

Cloning and Assembly of CRISPR-Cas9 Targeting Constructs

[0150] Unless otherwise noted, cloning of CRISPR-Cas9 plasmids and HDR donor constructs was done by Gibson assembly and cloning with the Gibson Assembly Master Mix (NEB, E2611S) (Gibson et al., Nat Methods 2009, 6:343-345). When necessary, fragments for the Gibson assembly cloning were amplified with the KAPA HiFi HotStart Ready Mix [KAPA Biosystems (KAPA), KK2602]. All gRNAs were obtained from Integrated DNA Technologies (IDT) as single-stranded 5-phosphorylated oligonucleotides purified by standard desalting. The basis for CRISPR-Cas9 experiments relied on the plasmid pSpCas9 (BB)-2A-GFP (pX458), obtained as a gift from Feng Zhang (Addgene plasmid #48138) (Ran et al., Nat Protoc 2013, 8:2281-2308). An alternate version of pX458 was generated by replacing the GFP (eGFP variant) with BPF (TagBFP variant) (pX458.2 or pSpCas9 (BB)-2A-BFP). For cloning gRNAs, both versions of pX458 were digested with Bbsl [New England BioLabs (NEB), R0539S], gRNA oligonucleotides were ligated into plasmids with DNA T4 ligase (NEB, M0202S). The gene for mRuby (mRuby2 variant) was derived from the plasmid pcDNA3-mRuby2, a gift from Michael Lin (Addgene plasmid #40260) (Lam et al., Nat Methods 2012, 9:1005-1012; Jinek et al., eLife 2013, 2: e00471-e00471). The HDR donors (mRuby and the antibody constructs) were cloned in the pUC57 (Kan)-HDR plasmid, obtained from Genewiz. The vector was designed with homology arms according to the annotated mouse genomic sequence (GRCm38). The 2A antibody constructs were obtained as synthetic gene fragments (gBlocks, IDT). The HDR donor vectors were linearized by PCR with the KAPA HiFi HotStart ReadyMix (KAPA Biosystems, KK2602). All plasmid and linear versions of HDR donors, as well as pX458 and pX458-BFP, were ethanol precipitated as a final purification step.

Hybridoma Transfection with CRISPR-Cas9 Constructs

[0151] Hybridoma cells were transfected with the 4D-NucleofectorSystem (Lonza) using the SF Cell Line 4D-Nucleofector X Kit L (Lonza, V4XC-2024) with the program CQ-104. Cells were prepared as follows: 105 cells were isolated and centrifuged at 90G for 5 minutes, washed with 1 ml of Opti-MEM | Reduced Serum Medium (Thermo, 31985-062), and centrifuged again with the same parameters. The cells were finally re-suspended in 100 l of total volume of nucleofection mix, containing the vector(s) diluted in SF buffer (per kit manufacturer guidelines). For the exchange of V.sub.H locus, 5 g of pX458 (or pX458-BFP) with gRNA-E (targeting V.sub.H) or gRNA-J (targeting mRuby), and 5 g of the circular or linearized HDR donor constructs were nucleofected into cells. For V.sub.L deletion, 5 g each of pX458 with gRNA-F and gRNA-H were co-transfected into cells. Following transfection, the cells were typically cultured in 1 ml of growth media in 24-well plates (Thermo, NC-142475). When a significant cell expansion was observed, cells were supplemented 24 hours later with 0.5-1.0 ml of fresh growth media. After sorting, typically 48 hours after transfection, cells were recovered in 24-well plates, and progressively moved into 6-well plates (Thermo, NC-140675) and T-25 flasks, following expansion. After replacing the V.sub.H with mRuby, cells were single-cell sorted in U-bottom 96-well plates (Sigma-Aldrich, M0812) in a recovery volume of 100 pl. The clones were eventually expanded in 24-well plates, 6-well plates and T-25 flasks.

Genomic and Transcript Analysis of CRISPR-Cas9 Targeting

[0152] Genomic DNA of hybridoma cell lines were recovered from typically 106 cells, which were washed with PBS by centrifugation (250G, 5 minutes) and re-suspended in QuickExtract DNA Extraction Solution (Epicentre, QE09050). Cells were then incubated at 68 C for 15 minutes and 95 C for 8 minutes. For transcript analysis, total RNA was isolated from 106-5106 cells. The cells were lysed with TRIzol reagent (Thermo, 15596-026) and total RNA was extracted with the Direct-zol RNA MiniPrep kit (Zymo Research, R2052). Maxima Reverse Transcriptase (Thermo, EP0742) was used for cDNA synthesis from total RNA (Taq DNA Polymerase with ThermoPol Buffer, NEB, M0267S). Both genomic DNA and cDNA were used as templates for downstream PCR reactions.

[0153] The gRNAs targeting WT IgH and IgK loci and mRuby were initially tested for their activity by 30 induction of NHEJ. The targeted fragment was amplified by PCR with KAPA2G Fast ReadyMix (KAPA, KK5121) and the PCR product digested with the Surveyor nuclease for the detection of mismatches (Surveyor Mutation Detection Kit, IDT, 706020). For HDR evaluation, PCR was performed on genomic and cDNA using primers binding inside and outside homology arms, followed by fragment size analysis on DNA agarose gels. Selected PCR products were subjected to Sanger sequencing.

Flow Cytometry Analysis and Sorting of Hybridomas

[0154] Flow cytometry-based analysis and cell isolation were performed using the BD LSR Fortessa and BD FACS Aria III (BD Biosciences), respectively. At 24 hours post-transfection, approximately 100 l of cells were harvested, centrifuged at 250G for 5 minutes, resuspended in PBS and analyzed for expression of Cas9 (via 2A-GFP/-BFP). 48 hours post-transfection, all transfected cells were harvested and resuspended in Sorting Buffer (SB): PBS supplemented with 2 mM EDTA and 0.1% BSA). When labeling was required, cells were washed with PBS, incubated with the labeling antibody or antigen for 30 minutes on ice, protected from light, washed again with PBS and analyzed or sorted. The labeling reagents and working concentrations are described in table 3 below. For cell numbers different from 10.sup.6, the antibody/antigen and incubation volume were adjusted proportionally.

TABLE-US-00003 TABLE 3 Flow cytometry labeling reagents with their working concentrations Target Working Dilution Incubatn. antigen conc. from stock volume Fluorophore Product ID IgG2C 3.3 g/ml 1:150 100 l Allophycocyanin 115-135-208 (APC) (Jackson ImmunoResearch) IgG2C 1.6 mg/ml 1:100 100 l AlexaFluor 115-547-188 488 (Jackson ImmunoResearch) IgK 2.5 g/ml 1:80 100 l Brilliant Violet 409511 421 (BioLegend) Hen egg 0.1 g/ml 1:62.5 100 l AlexaFluor 62971-10G-F lysozyme 647 (Sigma- Aldrich)

Measurement of Antibody Secretion by ELISA

[0155] Sandwich ELISAs were used to measure the secretion of IgG from hybridoma cell lines. Plates were coated with capture polyclonal antibodies specific for Vk light chains (goat anti-mouse, Jackson ImmunoResearch, 115-005-174) concentrated at 4 g/ml in PBS (Thermo, 10010-015). Plates were then blocked with PBS supplemented with 2% w/v milk (AppliChem, A0830) and 0.05% v/v Tween-20 (AppliChem, A1389) (PBSMT). Supernatants from cell culture (10.sup.6 cells/sample, volume normalized to least concentrated samples) were then serially diluted (at 1:3 ratio) in PBS supplemented with 2% w/v milk (PBSM). As a positive control, a purified mouse IgG2b, K isotype control (BioLegend, 401202) was used at a starting concentration of 5 ng/l (diluted in hybridoma growth media) and serially diluted as the supernatants. After blocking, supernatants and positive controls were incubated for 1 hour at RT or O/N at 4 C., followed by 3 washing steps with PBS supplemented with Tween-20 0.05% v/v (PBST). A secondary HRP-conjugated antibody specific for mouse Fc region was used (goat anti-mouse, Sigma-Aldrich, A2554), concentrated at 1.7 g/ml in PBSM, followed by 3 wash steps with PBST. ELISA detection was performed using a 1-Step Ultra TMB-ELISA Substrate Solution (Thermo, 34028) as the HRP substrate. Absorbance at 450 nm was read with Infinite 200 PRO NanoQuant (Tecan). For antigen specificity measurements, plates were coated with purified hen egg lysozyme (Sigma-Aldrich, 62971-10G-F) concentrated at 4 g/ml in PBS. Blocking, washing, and supernatant incubation steps were made analogously to the previously described procedure, with the exception of serial dilutions of supernatants at 1:5 ratios. A secondary HRP-conjugated antibody was used specific for Vk light chain (rat anti-mouse, Abcam, AB99617) concentrated at 0.7 g/ml. ELISA detection by HRP substrate and absorbance reading was performed as previously stated.

Targeting of the ROSA26 Locus

[0156] The mouse safe harbor locus ROSA26 was amplified and Sanger sequenced from Wen1.3 cells, and the sequence obtained was used to design DNA cassette homology arms. Guide RNA target sequences (gRNA-L to gRNA-P) were individually validated by Surveryor Assay (FIG. 7). This locus was targeted for the creation of the PnP-mRuby-AID, PnP-IgG-AID and the PnP-mRuby-Cas9 cell lines.

[0157] For the targeting of the ROSA26 in the creation of the PnP-mRuby-AID and the PnP-mRuby-Cas9 cell lines, gRNA-O and gRNA-P were selected due to their high cleavage efficiency.Generation of the AID cell lines and induction by Doxycycline

[0158] Cloning of the donor cassette for the inducible AID (ISSHM) system was performed in three steps. (1) The Tet-One Inducible Expression System was purchased from Takara Clontech (634301); GFP-2A-AID was obtained as synthetic gene fragment (gBlocks, IDT) and cloned into the pTetOne vector by Gibson assembly cloning. (2) Homology arms (829 and 821 bp) for the hybridoma's ROSA26 locus were obtained by genomic DNA PCR and cloned in a pUC57 (Kan) plasmid. (3) Finally, the previously cloned-see point (1)-Tet-One-GFP-2A-AID construct (containing, in the forward orientation: the human phosphoglycerate kinase 1 promoter (hPGK), the Tet-On 3G transactivator protein, and the SV40 poly-A signal; in the reverse orientation: the P.sub.TRE3GS Inducible promoter, the GFP-2A-AID construct and the SV40 poly-A signal) was inserted between the homology arms through Gibson assembly cloning. The HDR donor was linearized by PCR with restriction digestion with the enzyme Ajul (Thermo, ER1951). gRNA-O was obtained and cloned as previously described in pX458-BFP. As an alternative NHEJ insertion design, the TetOne-GFP-2A-AID construct is linearized without homology arms. The cell lines engineered for introduction of the TetOne-iSSHM system are: PnP-mRuby-pA (PnP-mRuby with a bGH poly-A tail); PnP-HEL23-IgH (PnP-HEL23 with a frameshift insertion in the HCDR3 knocking out antibody expression-see next sections). The workflow for these cells are shown in FIGS. 9 and 10.

a. Cell Lines Generation

[0159] From the transfection stage, the cells were kept in Tet-free GM: regular growth media supplemented with Tet System Approved FBS, US-sourced (Takara Clontech, 631105). PnP-HEL23-IgH.sup. and PnP-mRuby-pA cells were transfected with 2.5 g px458-BFP with gRNA-O and 2.5 g linearized pTetOne-HDR donor (see previous section). 48 hours after transfection the cells were sorted for BFP and grown for recovery. Once recovered, induction experiments were performed to verify the system's functionality: Doxycycline (Takara Clontech, 631311) was dissolved in nuclease-free H.sub.2O at 1 mg/ml, sterile filtered and diluted in Tet-free GM at need directly before use. Concentrations in the range between 1 ng/ml and 2 g/ml were tested, with 1 g/ml proving to be the most efficient. Cells were induced by incubation at 37 C for 24 or 48 hours; however, 24 hour incubation gave the best results and was selected as main condition to check for positive integration and induction efficiency.

[0160] The cells were sorted after 24 hours of induction: from the GFP positive population, single-cell clones were isolated, grown and characterized. An initial screening was performed to select the most positive clones after Dox induction: each sample was seeded at 1 g Dox/10.sup.5 cells/1 ml culture and screened for GFP 24 hours later.

[0161] The best performing clones from the initial screening steps (9 for PnP-HEL23-IgH.sup., 4 for PnP-mRuby-pA) were used for genomic DNA extraction, locus-specific amplification and Sanger sequencing. According to the genomic sequence, one final clone was selected for each cell line.

b. Induction Optimization

[0162] After selection of the best clone for each transfected cell line, a second and tighter titration was performed with concentrations in the range between 500 ng/ml and 1.5 g/ml, and induction measured at different time points (ideally: 24 hrs; 48 hrs; 72 hrs; 96 hrs. Due to Doxycycline having a half-life of 24 hours, it was replaced in culture every 48 hours, as recommended by the manufacturer (Clontech).

[0163] For each time point, induction was assessed by: [0164] FACS analysis (GFP) [0165] RT-PCR (mRNA/cDNA) [0166] Western Blot

[0167] Amplification of AID from cDNA was performed with KAPA HiFi HotStart Ready Mix. For Western Blot, M-PER Mammalian Protein Extraction Reagent (Thermo, 78501), supplemented with Halt Protease Inhibitor Cocktail (Thermo, 78430) was used to obtain lysates from cultured hybridomas, typically from 106 cells. Anti-Human/Mouse Activation-Induced Cytidine Deaminase (AID) Purified (eBioscience, 14-5959-80) was used as primary antibody for AID detection via WB.

c. iSSHM (AID Activity) Assessment

[0168] For the PnP-mRuby-pA-AID cell line, hypermutation activity was first evaluated by FACS analysis and detection of decreasing mRuby fluorescence. For a more thorough evaluation, the mRuby gene was amplified from cDNA and analyzed by sanger or next-generation sequencing (NGS) using the method of molecular amplification fingerprinting (FIG. 12).

[0169] For PnP-HEL23-IgH-AID cell line, restored antibody expression and/or antigen-specificity was evaluated by flow cytometry after labelling cells (see previous sections) using anti-lgG2C and anti-IgK (any positivity arisen through random mutations) and HEL-647 (re-gained HEL positivity).

[0170] For more definitive assessment and optimization of the system, the iSSHM workflow was repeated for a cell line (obtained by either of the two starting platforms) expressing a functional antibody. To obtain such a situation, PnP-mRuby-pA cells were transfected to exchange mRuby with a sAb donor; PnP-HEL23-IgH-AID cells were transfected to exchange the knocked-out HEL23 sAb with a functional one with the same or another specificity; in an alternative setting, the HEL23 frame was restored by HDR with a 120 ssODN containing a codon-mutated version of the original HCDR3. In case of no previously tested binders, antibodies with a known and testable antigen were typically chosen to evaluate affinity maturation.

[0171] Once obtained a PnP-sAb-AID cell line, AID was induced like previously described (ON). After a 48-96 hours induction, Doxycycline was retracted from the system (OFF). To assess affinity maturation, FACS labelling was performed as previously described, but with decreasing antigen concentrations (typically 1-0.001 g/ml, decreasing ten-fold for each round of analysis/affinity maturation cycle). Positive cells were sorted and underwent a further iSSHM cycle; the cycle was repeated as needed. For each stage of affinity maturation, after Doxycycline retraction from the system, the effective switch OFF was evaluated by monitoring GFP fluorescence by FACS. Once in the OFF state, V.sub.L and V.sub.H regions were amplified from cDNA and iSSHM was evaluated by NGS using the method of molecular amplification fingerprinting.

Generation of the PnP-mRuby2-Cas9 Cell Line

[0172] Cloning of the donor cassette for constitutive expression of the Cas9 protein was performed in three steps. (1) The pSpCas9 (BB)-2A-Puro vector (pX459) and MDH1-PGK-GFP_2.0 vector were obtained from Addgene (plasmid #48139, #11375, respectively). The Cas9-2A-Puro and GFP gene fragments were obtained from their respective vectors through PCR (KAPA HiFi HotStart ReadyMix) and assembled together with Gibson assembly cloning. (2) Homology arms (1,000 and 976 bp) for the hybridoma's ROSA26 locus were obtained by genomic DNA PCR and assembled with the pUC57 (Kan) plasmid backbone through Gibson assembly cloning. (3) Finally, the previously assembled fragments-see points (1 and 2)-were assembled through Gibson assembly cloning. The HDR donor was linearized by restriction digestion with the Xhol and Mlul restriction endonucleases followed by gel electrophoresis purification. gRNA-P was cloned as previously described in pX458 (BFP). For the alternative NHEJ insertion design, the Cas9-2A-Puro-GFP construct was linearized without homology arms.

[0173] Following transfection with the Cas9-2A-Puro-GFP donor, GFP.sup.+ cells were sorted and expanded. Cells were then selected for stable integration of the donor construct by culturing in regular growth media supplemented with 2.5 g/ml of Puromycin (Thermo, A1113802) for up to one week before single-cell isolation, growth and PCR characterization. After identification of a single clone with correct integration of the Cas9-2A-Puro-GFP cassette, Cas9 activity within the cell was validated through transfection of a guide RNA targeting GFP. Cas9 cleavage activity was validated by T7E1 assay and Sanger sequencing of PCR amplicons (FIG. 8). GFP knock out effectiveness was confirmed by flow cytometry.

[0174] The cell lines engineered for constitutive Cas9 expression are: PnP-mRuby-pA; PnP-HEL23-IgH.sup.. A more comprehensive cell line was designed to incorporate both the constitutive Cas9 and the inducible AID; due to the constructs bearing homology with two different regions of the ROSA26 locus, it was possible to incorporate them in tandem like shown in FIG. 8A. This allowed us to merge the high HDR efficiency achieved by constitutively expressing Cas9 with the ISSHM workflow.

Cas9 Cells-In Vitro Transcription of Guide RNA or Synthetic Oligonucleotides (IDT)

[0175] The PnP-mRuby-Cas9 cell line, constitutively expressing Cas9, was transfected with the 10 appropriate HDR donor and already transcribed guide RNAs. The latter were obtained as oligonucleotides from IDT or in vitro transcribed. In the case of in vitro transcription, the previously described guide-DNA oligodeoxynucleotides (see Cloning and assembly CRISPR-Cas9 targeting constructs section) served as templates for the MEGAscript T7 transcription kit (Thermo, AM1334). An adapted protocol was as described previously (https://www.protocols.io/view/In-vitro-transcription-of-guide-RNAs-d4w8xd?step=3 accessed 10 Feb. 21, 2017).