Single-hit genome editing in B cells to redirect their specificity toward tumor antigens

doi:10.21203/rs.3.rs-4317490/v1

Download PDF

Article

Single-hit genome editing in B cells to redirect their specificity toward tumor antigens

https://doi.org/10.21203/rs.3.rs-4317490/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 28 Sep, 2024

Read the published version in Scientific Reports →

You are reading this latest preprint version

T-cell-based adoptive immunotherapy is a new pillar of cancer care. Despite recent breakthroughs, B cells remain a challenging target in terms of survival after in vitro manipulation and functional expression of rewired immunoglobulin (Ig) genes. To address these limitations, we designed a single-chain Ig-encoding cassette (“scFull-Ig”) that, when inserted at a single genomic position, redirects antigen specificity but preserves all the Ig functional domains and the intrinsic regulatory mechanisms that modulate Ig expression in activated cells from the IgM B-cell receptor (BCR) expression to Ig secretion and/or class switching. Single-locus editing may then provide an efficient and safe strategy to both disrupt endogenous Ig expression and encode a new Ig paratope. As a proof of concept, the functionality of scFull BCR and/or secreted Ig was validated against two different classical tumor antigens, HER2 and hCD20. Once validated in cell lines, the strategy was extended to primary B cells, and successful engineering of BCR and Ig expression was confirmed, as the ability of scFull-Ig to undergo further class switching was confirmed. These results pave the way for future B-cell-based adoptive immunotherapy and strategies to express a therapeutic mAb with a variety of switched H-chains that provide complementary functions.

Biological sciences/Cancer

Biological sciences/Immunology

Effective monoclonal antibodies (mAbs) are available for the treatment of many chronic diseases, particularly cancer. The high specificity of such biotherapies hereby limits their side effects compared to those of conventional treatments such as cancer chemotherapy. However, mAbs can in some cases lead to poor responses or resistance, particularly in patients with the lowest serum trough concentrations ¹. Repeated injections of mAbs are then required to prevent relapses, which can be lengthy and costly, limiting the efficacy of these treatments. This is especially true for mAbs with rapid serum clearance, such as panitumumab or cetuximab, or for some bispecific antibodies ¹. Strategies that could lead to continuous infusion of such antibodies, for example, as endogenously produced by the patient’s own cells, would therefore represent a strong asset. The clinical use of some antibodies of high therapeutic interest but with chemical instability or manufacturing contingencies notably associated with some Ig classes could also be rescued by direct in vivo production in the patient’s body. This strategy could also be relevant for conferring long-term immunity to some immunodeficient patients with a preserved B-cell pool but poor response to classical vaccination.

B lineage cells are the most suitable cell type for the correct folding and secretion of immunoglobulins (Igs) because of their various membrane-bound or secreted formats ². At the plasma cell stage, these cells are capable of secreting large amounts of soluble antibodies, while memory B cells can also establish immune memory and respond to antigen (Ag) rechallenge, thereby contributing to protection against recurrent infections. B cells edited for Ig expression could therefore potentially endow patients with both aspects of a long-term immune response to a specific Ag. Several recent studies have shown that CRISPR-mediated editing of the B-cell genome is feasible in primary B cells and could reformat Ig production on purpose ^3–5. In the context of cancer immunotherapy and learning from the success of CAR-T cells, which have now achieved complete remission in many patients and have demonstrated the ability to safely integrate into patients’ immune system for years ⁶⁷⁸, further exploration and optimization of the conditions that might allow successful adoptive cell therapy based on B cells are warranted. In such a strategy, B cells taken from an individual are genetically modified to produce Ig with the desired specificity and then injected back into the original donor. Successful engraftment of engineered B cells would then be expected to result in long-term stable infusion of a therapeutic mAb into body fluids and/or artificially established immune memory against a tumor Ag.

For this purpose, we used CRISPR-mediated gene knock-in (KI), the best tool for precise gene editing ^9,10, to hijack the inherent ability of B cells to produce Ig molecules. The insertion site for the KI was chosen within the Ig heavy chain (IgH) locus to simultaneously disrupt endogenous IgH synthesis and integrate the KI cassette in a position that mimics all the normal features of Ig expression. These include B lineage-specific transcription, upregulated transcription in activated B cells and plasma cells, regulation of alternative splicing between membrane and secreted type transcripts, and the ability to undergo class switch recombination upon recombination between the endogenous Sµ region and a downstream S region of the IgH locus.

Since in vitro cell culture and maintenance of B cells, as well as their genetic engineering, remain challenging, we have optimized B-cell culture conditions and CRISPR-mediated gene editing and now report an efficient strategy for gene editing of human primary B cells to produce a complete Ig of a given specificity in a single editing step. The genetic engineering of Ig theoretically requires the modification of two loci, one for the heavy (H) and one for the light (L) chains. To express the desired Ig and to disrupt endogenous Ig production after a single gene modification, we developed a new single-chain format of a whole Ig, “scFull-Ig”. This format resolves any issue about the need for stoichiometric expression of both Ig chains and precludes any mispairing with an endogenous light chain. Moreover, this approach has the advantage of positioning both the V_HDJ_H and V_LJ_L regions immediately downstream of the V_H promoter in a position accessible to somatic hypermutation (SHM), with constant regions downstream. Using this strategy, modified primary B cells were shown to be able to successfully express BCR membrane molecules in the scFull-Ig-Upstream Switchable Module (USM) format, thereby acquiring the corresponding Ag-binding activity and secreted scFull antibodies.

Recombinant scFull-Ig production

To overcome the challenges associated with the expression of recombinant polymeric molecules, we designed a single-chain strategy to produce full-length Ig molecules by using two linker (Lnk) sequences to assemble the Ig domains in the following order: [V_H–Lnk-V_LC_L-Lnk-C_H]. We designed different expression vectors and gene cassettes first to validate the functionality of the expected soluble antibodies (anti-HER2 scFull-Ig and anti-hCD20 scFull-Ig) and second to edit the IgH locus in the B-cell genome to encode Ig in this scFull format (after integration of an “scFull-Ig USM” gene cassette).

All the vectors were assembled using the NEBuilder HiFi DNA Assembly strategy (New England Biolabs, Ipswich, MA) and combined with DNA fragments either PCR-amplified from human genomic DNA or custom synthesized on purpose (Genecust, Boynes, France).

The first aim was to validate the soluble form of sc-Full-Ig. Two vectors were assembled in the pcDNA3 expression vector backbone, which encoded the IgG1-class soluble forms of either the anti-HER2 scFull-Ig or the anti-hCD20 scFull-Ig, each following a [V_H–Lnk-V_LC_L-Lnk- g1(C_H1-C_H2-C_H3)] architecture. These vectors are driven by a CMV promoter and then comprise, in a 5’ to 3’ order, a V_H leader exon, a V_HDJ_H exon, a first linker [(G4S)3-SG3S], the V_LJ_L and C_L exons, a second linker [S-(G4S)3] and, ultimately, the g1 constant exons of the gene C_H1-C_H2-C_H3 (followed by a polyadenylation site). Both secreted antibodies were then produced by transfecting nonlymphoid cells, such as the XTEN-CHO cell line (ProteoGenix. Schiltigheim, France).

In a second aim, vectors were built to yield Ig with the same [V_H–Lnk-V_LC_L-Lnk] architecture after the IgH locus was modified. This was accomplished by precisely inserting into the IgH locus a cassette designed to be expressed in B cells and spliced to the endogenous IgH constant genes. This gene cassette started with a B-cell-specific pV_H promoter (a 180 bp fragment from the mouse VH4.1 gene; NCBI reference NC000078.7) and ended with a donor splice site sequence identical to that of a J_H element. This [pV_H-V_H–Lnk-V_LC_L-Lnk-USM] cassette was additionally flanked with 5’ and 3’ homology arms to promote homology-directed recombination (HDR) and precise insertion into the intron downstream of JH6. Once properly inserted, it could thus be expressed in association with the endogenous C_H genes from the targeted IgH locus.

Western blot

After reduction with NuPAGE Sample Reducing Agent (Invitrogen), scFull-Ig molecules were analyzed via Western blotting together with commercial rituximab (MabThera, Roche) and pertuzumab (Perjeta, Roche) as controls. After electrophoresis on 4–12% Bis-Tris 1.5 mm gels (Invitrogen), the proteins were transferred to nitrocellulose membranes using iBlot2 NC Mini Stacks and an iBlot2 Transfer Device (Invitrogen). The membranes were stained with 400 µg/ml goat anti-human κ chain antibody (Southern Biotech), 0.2 µg/ml donkey anti-goat IgG H + L HRP conjugated antibody (Bethyl Laboratories), or 1/10,000 diluted goat anti-human IgG HRP conjugated antibody (Sigma). Staining was detected using Pierce ECL Plus Western Blotting Substrate (Thermo Scientific).

Antibody-dependent and complement-dependent cytotoxicity assays

ADCC assays were performed using CD16-expressing reporter cells from the Jurkat-Lucia NFAT-CD16 cell line (InvivoGen) according to the manufacturer’s protocol. Briefly, target cells were incubated with antibodies for one hour, after which Jurkat-Lucia NFAT-CD16 cells were added. After incubating for 6 hours, the supernatant was transferred to a white microplate, and QUANTI-Luc solution (InvivoGen) was added. The light signal indicating Jurkat cell activation was quantified using a Synergy H1 luminometer (Agilent).

SUDHL4 target cells were incubated with antibodies for one hour, followed by the addition of rabbit complement (Cedarlane). After incubating for 45 min, the cells were washed, and viability was assessed via DAPI staining.

BCR signaling assays

The ability of the engineered BCR to trigger B-cell activation was demonstrated using phospho-flow assays, which monitor the intracellular phosphorylation of elements of the BCR signaling cascade (SYK, BLNK PLCγ2, pERK1 and BTK).

DG75 cells clonally expressing a scFull-Ig BCR were starved for 1 h with RPMI 1640 GlutaMAX medium (Gibco) supplemented with 1% fetal bovine serum (FBS). A stimulation step was then performed by adding the F(ab’)2 fragment goat anti-human IgA + IgG + IgM (H + L) (Jackson ImmunoResearch) at 20 µg/ml. The cells were then permeabilized and stained with appropriate antibodies specific for the phosphorylated forms of SYK, BLNK, PLCg2, BTK, and pERK1 (BD PhoSFL) according to the supplier's protocol. The stained cells were analyzed on a CytoFlex flow cytometer (Beckman Coulter).

Cell culture

DG75 cells were grown in RPMI 1640 GlutaMAX medium (Gibco) supplemented with 10% FBS, 1 mM sodium pyruvate (Gibco) and penicillin‒streptomycin-glutamine (PS-G; 1 unit/ml penicillin, 1 µg/ml streptomycin, and 2.92 µg/ml L-glutamine; Gibco).

MCF7 cells were grown in DMEM (Gibco) supplemented with 10% FBS (Gibco) and PS-G.

SUDHL4 and hCD20^+/− EL4 cells were grown in RPMI 1640 GlutaMAX medium supplemented with 10% FBS (Gibco), 1 mM sodium pyruvate and PS-G. BT474 cells were grown in RPMI 1640 GlutaMAX medium supplemented with 10% FBS, NEAA (Gibco), 10 mM HEPES (Gibco), 0.45% D-glucose (Gibco), 1 mM sodium pyruvate and 10 µg/ml bovine insulin (Sigma). SKBR3 cells were grown in RPMI 1640 GlutaMAX medium supplemented with 10% FBS, NEAA and 10 mM HEPES.

HEK293T cells were grown in DMEM supplemented with 10% FBS and 200 µg/ml G418 (Gibco). Transfections of HER293T cells were performed with MacsFectin reagent according to the manufacturer’s protocol.

Primary B-cell isolation and culture

Buffy coats from healthy volunteers were obtained from the Etablissement Français du Sang (Rennes, France) (after providing informed consent and under agreement #AC-2019-3853). B cells were isolated using the StraightFrom™ Buffy Coat CD19 Microbead Kit (Miltenyi Biotec) according to the manufacturer’s protocol. B cells were then cultured in RPMI 1640 GlutaMAX medium (Gibco) supplemented with 10% FBS (Gibco) and 1 mM sodium pyruvate (Gibco). B cells (0.75 10⁶ cells/ml) were stimulated for 4 days with the “1st step stimulation cocktail”, which included 1 µg/ml CpG oligodeoxyribonucleotide (CpG 2006; Miltenyi Biotec), 2.4 µg/ml F(ab’)₂ fragment goat anti-human IgA + IgG + IgM (H + L) (Jackson ImmunoResearch), 50 U/ml recombinant IL-2 (R&D Systems) and 100 ng/ml recombinant human soluble CD40L (Immunex). On day 2, 5 ng/ml IL-10 (R&D Systems) was added. On day 4, the activated B cells were washed and transferred at 0.5 10⁶ cells/ml into a “2nd step plasmablast differentiation cocktail”¹¹ containing 50 U/ml IL-2 (R&D Systems), 5 ng/ml IL-4 (R&D Systems) and 12 ng/ml IL-10 (R&D Systems). To obtain plasma cells, the cells were washed on day 7 and cultured at 0.5 10⁶ cells/ml with 50 U/ml IL-2 (R&D Systems), 250 U/ml IFN-α (R&D Systems), 20 ng/ml IL-6 (PeproTech) or 12 ng/ml IL-10 (R&D Systems).

Donor DNA preparation and gene editing

DNA templates for CRISPR-mediated KI were constructed from PCR-amplified fragments using the NEBuilder HIFI assembly kit (New England Biolabs). Donor DNA for CRISPR-mediated KI was prepared from plasmids by PCR amplification. PCR products were purified using a NucleoSpin Gel and PCR Clean-up Kit (Macherey-Nagel). The purified DNA was then ethanol precipitated and resuspended in H2O to a final target concentration of 2000–5000 ng/µl, as quantified by Qubit. For CRISPR/Cas9, the IgH locus between JH6 and the Eµ enhancer was targeted at a location chosen more than 0.4 kb downstream of JH6 to ensure sufficient 5’ homology for introducing a transgene, irrespective of the preexisting endogenous VDJ rearrangement. After preliminary trials with less efficient targeted sites (not shown), we focused our experiments on the best target in our hands, with the sequence GGAAAGAGAACTGTCGGAGT (human genome GRCh38, CHr14, 105862762–105862781.

For the DG75 cells, one day prior to the CRISPR experiments, the cells were seeded at 0.2 10⁶ cells/ml to be in the exponential growth phase. The cells were electroporated using a Nucleofector Kit V (Lonza) as follows: 1.10⁶–3.10⁶ cells were centrifuged and resuspended in 100 µl of supplemented Nucleofector solution. The RNP complex was formed by incubating 100 pmol of Alt-RTM S.p. HiFi Cas9 Nuclease V3 (IDT) with 500 pmol of sgRNA (Synthego) for 15 min at 37°C. RNP was then added to the cell suspension, and the cells were transfected using the Nucleofector I program “X-01”. Immediately after transfection, 900 µl of culture medium was added, and the cells were allowed to rest for 5–10 min before being transferred to 6-well plates in 5 ml of medium.

Primary B cells were electroporated using an ATx device (MaxCyte). A total of 3.10⁶ cells were centrifuged, washed and resuspended in 25 µl of electroporation buffer (MaxCyte). The RNP complex was formed by incubating 25 pmol of Cas9 with 125 pmol of sgRNA for 15 min at 37°C. RNP was subsequently added to the cell suspension, after which the cells were transfected with the OC-25 assembly using the program “B-cell 2”. Briefly, 0.5 U of Pulmozyme (Roche, 2500 U/2.5 ml) was added immediately after transfection, the cells were allowed to rest for 30 min at 37°C before being transferred to 24-well plates in 1.5 ml of medium, and Alt-R HDR Enhancer V2 (IDT) was added to the medium. After 24 hours, the HDR enhancer was removed by medium renewal, and the cells were incubated at 32°C for 24 hours.

For KI experiments, 2 µg or 6 µg of donor DNA was added to the RNP complex before addition to cell lines or primary B cells, respectively.

DNA and RNA analysis

For KI analysis, transfected cells were lysed using SDS and proteinase K at 56°C O/N. DNA was then precipitated with ethanol and resuspended in H₂O. KI analyses were performed via multiplex PCR using 3 primers that could competitively amplify either a WT band (897 bp) or a KI-specific band (772 bp).

For RNA analysis, cells were lysed using TRIzol reagent (Thermo), and RNA was extracted by phenol–chloroform extraction, washed in ethanol and resuspended in H₂O. Reverse transcription was performed using RT Superscript II (Thermo) with random primers (Invitrogen). scFull-Ig-specific transcripts were subsequently amplified via PCR using reverse primers specific for constant domains of different isotypes (IgM, IgG, IgA, and IgE) and a forward primer specifically hybridizing to the scFull-Ig cassette.

The primers used in the study are listed in Supplementary Table 1.

Flow cytometry

Cell surface staining for the anti-HER2 BCR was performed using biotinylated human HER2/ErbB2 protein (Sino Biological) and PECF594 streptavidin (BD Horizon). For staining, 18 µl of 20 ng of HER2-Biot and 10 ng of streptavidin were premixed and incubated for 30 min at room temperature before being added to 0.5 10⁵ cells, after which the mixture was incubated for 30 min on ice in a final volume of 50 µl. DAPI was used to label dead cells.

For staining with scFull-Ig supernatant, target cells were first incubated with 50 µl of supernatant for 1 hour on ice, washed and then labeled with 50 µl of PE-conjugated goat anti-human IgG Fc secondary antibody (Invitrogen) for 1 hour on ice. The stained cells were analyzed on a CytoFlex flow cytometer (Beckman Coulter).

ELISA

For detection and quantification of secreted anti-HER2 antibodies, MaxiSorp Clear Flat-Bottom 96-well plates (Thermo Scientific) were coated with 10 µg/ml streptavidin (Sigma) overnight at 4°C and blocked with PBS containing 5% BSA (Sigma) for one hour at room temperature. Then, 0.5 µg/ml biotinylated human HER2/ErbB2 protein (Sino Biological) was added, and the mixture was incubated for one hour at 37°C. The supernatants were subsequently added and incubated for 1–2 hours at 37°C. Finally, alkaline phosphatase-coupled goat anti-human kappa antibody (Southern Biotech) was added and incubated at 37°C for 1 hour, and binding was visualized using SIGMAFAST p-nitrophenyl phosphate tablets (Sigma) dissolved in H₂O. The reaction was stopped with 3 N NaOH, and the absorbance was read at 405 nm.

Statistical analysis

All the statistical analyses were performed using Mann‒Whitney tests with GraphPad Prism 8 software. The data are shown as the mean ± SD, with 95% confidence intervals (CIs); * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

The scFull-Ig mAb format can be functional

We investigated a new single-chain format of Ig, the scFull-Ig (Fig. 1. A), in which a first linker connects the heavy chain variable domain to the κ light chain variable and constant domains, followed by a second linker and the full constant region of the IgG heavy chain. To evaluate this format in the context of different associated B regions, we produced two different scFull-Igs, one against HER2 and one against hCD20, based on V region sequences from pertuzumab and rituximab, respectively. Single-chain versions of both antibodies were first produced in serum-free media in XTEN-CHO cells transfected with expression vectors encoding, in each case, the linked V_HDJ_H region + complete κ light chain and second linker ([VDJ-Lnk-VκJκ-Cκ-Lnk] cassette), followed by the constant domains of a human secreted γ1 heavy chain. The structure of the secreted antibodies was investigated by western blotting using anti-hIgG and anti-κ antibodies after gel electrophoresis. Under reducing conditions, the scFull-Ig complex consisted of a single 75 kDa protein, as expected, that combines the Ig heavy and light chains (Fig. 1.B).

We then validated the binding of these soluble scFull-Igs to their respective targets, as specifically expressed by the RAMOS, RAJI and hCD20⁺ EL4 cell lines for the anti-hCD20 scFull-Ig and MCF7, SKBR3 and MCF7 cells for the anti-HER2 scFull-Ig (Fig. 1. C/Sup Fig. 1). For the anti-HER2 mAbs, the staining results were quite similar for the classical format and the scFull-Ig format. For the anti-CD20 mAbs, specific staining was obtained, but the mean fluorescence intensity of the cells stained with the scFull-Ig was somewhat weaker than that of the cells stained with the commercial rituximab. This finding suggested preserved specificity but some loss of affinity after the single-chain transformation of this antibody, as has been observed for other scFv derivatives of rituximab ¹².

The functionality of the scFull-Ig molecules was also validated by assessing their ability to mediate cytotoxicity toward appropriate target cells (Fig. 1). Similarly to rituximab, anti-hCD20 scFull-Ig was found to induce complement-dependent cytotoxicity (CDC) in a dose-dependent manner, but with a shifted curve requiring approximately twice the concentration of the antibody to achieve the same cytotoxicity (Fig. 1D). Like with rituximab and pertuzumab, anti-hCD20 scFull-IgG and anti-HER2 scFull-IgG specifically induced ADCC in cells harboring the target antigen (i.e., RAJI cells and MCF7 cells, respectively), even when used at low doses in the range of 3.1 ng/ml (Fig. 1.E). Interestingly, we observed better induction of ADCC with the anti-HER2 agent scFull-Ig than with the commercial agent pertuzumab at an equivalent dose.

LEGEND TO FIGURE 1: Design and validation of scFull-Ig format

A. General architecture of the scFull-Ig design. B. Detection of scFull-IgG expression by western blot under reducing conditions. The membranes were stained with anti-IgG and anti-Igκ antibodies. C. Validation of target binding by the soluble forms of anti-HER2 and anti-CD20 scFull-IgG. The controls used were commercial pertuzumab and rituximab (n = 5). D. Validation of the capacity of scFull-Ig to induce complement-dependent cytotoxicity. CD20⁺ SUDHL4 target cells and rabbit complement were used. E. Validation of the capacity of scFull-Ig to activate the ADCC-related reporter cell line Jurkat-Lucia NFAT-CD16 in the presence of cognate HER2⁺ MCF7 target cells or hCD20⁺ RAJI target cells (n = 9).

Taken together, these data confirm that the scFull-Ig format is suitable for the production of Ag-specific antibodies. Given these results and the good efficacy of the anti-HER2 scFull-Ig format, this model was selected for subsequent gene editing experiments.

Knock-in of a spliceable single scFull-Ig-Upstream Switchable Module (scFull-Ig-USM) cassette allows functional BCR expression in a B lymphocyte cell line

In addition to the advantages of a single chain to ensure stoichiometry and physical association of the paired H/L structure, the scFull format was designed to allow recombinant Ig production through a single CRISPR-mediated genomic edit at the IgH locus. Accordingly, an editing site was chosen downstream of the J_H region and upstream of the Eµ enhancer (Fig. 2. A), as this position can allow the expression of a knock-in gene while concomitantly disrupting the expression of the endogenous IgH VDJ gene, which is thereby dissociated from Eµ. We preliminarily validated the high accessibility of this genomic site for CRISPR editing in the DG75 lymphoma cell line using an HDR template containing a pVH-promoter-tdTomato reporter cassette (Sup Fig. 2A). This experiment also validated the efficacy of 500 bp homology arms, which were ultimately designed with terminal truncated Cas9 target sequences (tCTS) at both ends of the HDR template to optimize HDR ¹³. Overall, this resulted in regular expression, as indicated by more than 30% fluorescent cells (Sup Fig. 2B).

Based on this result, the anti-HER2 scFull-Ig-USM cassette, preceded by a VH promoter, was flanked by the above validated HDR flanking arms to be inserted at the same genomic location downstream of JH. For proper splicing to endogenous IgH constant genes, the scFull-Ig-USM gene cassette was then terminated with a J_H splice donor site (J_H3’splice) downstream of the second linker. With such a complete ([pVH-VDJ-Lnk-VκJκ-Cκ-Lnk J_H3’splice]) configuration, this cassette can be expressed and spliced to the immediately downstream constant gene. It then replaces the endogenous V_HDJ_H exon in IgH transcripts (Fig. 2A). A KI at this position respecting the IgH C gene architecture is also expected to preserve the process of alternate splicing of membrane-type vs secreted-type IgH C transcripts. It does not compromise any of the elements controlling the process of class switch recombination (CSR) at the locus. In cells undergoing CSR and/or plasma cell differentiation, such an edited IgH locus is therefore expected to behave as a normal endogenous IgH locus and to potentially be expressed under several parallel or successive clonally related BCR and secreted Ig forms from the various Ig classes (but sharing the same clonotypic Ag specificity).

To confirm these expectations, we first inserted the anti-HER2 scFull-Ig-USM cassette into DG75 lymphoma cells with a mature B lymphocyte phenotype and IgM class BCR expression. This experiment assessed the ability of the membrane-bound scFull-IgM-USM to bind soluble HER2 after CRISPR-mediated insertion of the [VDJ_pertu-L-VκJκ_pertu-Cκ-L-J_H3’splice] gene cassette (Fig. 2.B). More efficient insertions were subsequently achieved by terminating HDR arms with tCTS sequences, which bind a noncleaving complex of specific gRNA and Cas9. Altogether, in DG75, approximately 37% of the cells were successfully edited and acquired the ability to bind soluble HER2, similar to the results obtained for the insertion of the tdTomato reporter cassette.

Validation of the on-target genomic integration of scFull-Ig-USM into the IgH locus was performed using a multiplex PCR assay (competitively involving 2 primers that flank the insertion site in the IgH locus and 1 primer that hybridized to the inserted scFull-Ig-USM cassette) (Fig. 2.C).

In addition to validation at the protein and DNA levels, in-frame splicing of the inserted scFull-Ig-USM cassette to the endogenous IgM constant regions was checked by reverse transcribing and sequencing IgH µ mRNA prepared from gene-edited cells, confirming the correct junction of the edited V-region cassette with the endogenous IgH constant region (Fig. 2.D).

To evaluate the functionality of the scFull-Ig-USM BCR, edited DG75 B cells were stimulated in vitro with BCR ligands, permeabilized and monitored by cell cytometry for the phosphorylation of several relevant members of the BCR signaling cascade (Syk, BLNK, PLCγ, Erk and Btk) (Fig. 2. E/Sup Fig. 3). Interestingly, BCR cross-linking by the multivalent form of HER2 Ag (biotinylated HER2 premixed with streptavidin) also triggered B-cell activation (Fig. 2.E).

Thus, functional BCR expression through the KI of the scFull-Ig-USM cassette is efficiently driven by CRISPR-mediated insertion into a B-cell line, prompting the extension of this strategy to primary B cells.

LEGEND TO Fig. 2. Knock-in in the DG75 cell line

A. Locus. The VDJ-L-Vκ-Cκ-L cassette is introduced between JH and the Eµ enhancer via CRISPR-mediated cleavage. “Created with BioRender.com.” B. Expression of membrane-bound scFull-Ig USM. Cell surface staining for anti-HER2 BCR was performed on gene-edited DG75 cells with biotinylated HER2 protein and PECF594-labeled streptavidin. tCTS, truncated Cas9 target sequence. C. Validation of the KI at the genomic level by multiplex competitive PCR. D. Validation of correct splicing of the knock-in exons on the endogenous IgH constant exons. RNA was extracted from the edited cells, and the IgH transcript was subsequently sequenced via reverse transcription and PCR. Correct splicing of the inserted cassette in the endogenous constant region was confirmed by amplification of the specific band via PCR and sequencing. E. Detection by flow cytometry of the intracellular phosphorylation of several members of the BCR signaling cascade (PLCg2 and pERK1) after cross-linking of the membrane-anchored anti-HER2 scFull BCR (n = 7) either with anti-BCR antibodies (anti-IgG/IgM/IgA) or with precomplexed antibodies [biotinylated HER2 + streptavidin-PECF594].

Edition for Ig expression in primary B cells

Primary B cells were magnetically sorted from buffy coats and cultured for two days in combination with our “1st step stimulation cocktail” (including CpG, anti-BCR, CD40L and IL-2). On day 2, CRISPR-mediated gene editing was performed under the same conditions as those used for DG75 cells but with the addition of the HDR enhancer (IDT) according to the supplier’s instructions. The expression of the anti-HER2 scFull-Ig-USM antibody at the cell membrane was assessed on day 7. Approximately 6% of the cells were successfully edited to bind soluble HER2 when the most efficient donor DNA template was used (flanked with tCTS sequences) after gene editing in the presence of the HDR Enhancer (Fig. 3A). Like in DG75 cells, a multiplex competitive PCR assay was performed on DNA from edited primary B cells to validate the on-target integration of scFull-Ig-USM (Fig. 3. B)

Correct splicing of the HDR template to endogenous constant regions was also confirmed by analyzing mature Ig transcripts, which subsequently evaluated the association of the scFull-Ig USM cassette with various IgH classes. While edited µ transcripts predominated on day 4, i.e., 2 days after transfection, their class-switched counterparts progressively increased during in vitro stimulation culture. With a significant increase from day 4 to day 10, in-frame splicing of the scFull-Ig-USM KI cassette onto all endogenous constant Ig gene classes was clearly detectable, indicating the expression of not only IgM but also class-switched IgA, IgG and IgE (Fig. 3.C).

Accordingly, the secretion of HER-binding Abs in the supernatant of edited and in vitro-stimulated B cells was detected in parallel in day 7 and day 10 samples (Fig. 3D).

Overall, these data confirm that the KI of the [VDJ_pertu-L-VκJκ_pertu-Cκ-L-J_H3’splice] cassette in the IgH locus upstream of the Eµ enhancer allows the expression of scFull-Ig-USM in primary B cells both in association with Cµ and with downstream class-switched isotypes and both as membrane-associated or secreted Ig.

LEGEND to Fig. 3. Knock-in in primary B cells

A. Expression of the membrane-bound scFull-Ig USM. Cell surface staining for the anti-HER2 BCR was performed as described in Fig. 2 (n = 9). B. DNA was extracted from edited primary B cells, and PCR was performed with 3 primers to confirm the correct genomic integration of the template (n = 4). C. RNA was extracted from edited primary B cells, and splicing was confirmed by performing RT‒PCR on IgH transcripts from the unswitched Cµ gene or Cγ, Cα and Cε genes after class switching (n = 6). D. Detection of secreted anti-HER2 scFull-Ig-USM by ELISA. Supernatants of gene-edited B cells were collected at different time points, and HER2-specific antibodies were quantified (n = 8).

Adoptive cell therapy has developed remarkably over the last decade. While T cells are currently the mainstay of such strategies, B cells have recently attracted some interest. B-lineage cells can abundantly secrete proteins, and initial attempts to make them produce therapeutic proteins have relied on random lentiviral integration, particularly for the secretion of Ig or coagulation factors^14,15. However, for Ig, such strategies run the risk of generating cells with dual specificity and/or with a mixture of desired and undesired H-L associations. Precise CRISPR editing of Ig loci is thus attractive because it can simultaneously terminate the expression of endogenous Ig and replace it with a therapeutic mAb^3–5. In this context, the reformatting of the IgH locus is of particular interest due to its remarkable ability to yield both membrane-bound and secreted Ig and its ability to regulate several other physiologic variations in expression. With such intrinsic properties of the IgH locus, edited human B cells could fulfill two distinct and complementary therapeutic goals: either to secrete a mAb of interest on a long-term basis when differentiated into long-lived CD138⁺ CD20⁻ plasma cells or to patrol the patient’s lymphoid tissues as Ag-specific CD20⁺ CD27⁺ memory B cells to secrete the engineered Ig only upon encounter with Ag.

We herein explored and validated a strategy for the modification of both Ig chain V domains through single editing of the IgH locus, which retains its physiological organization, remains compatible with class switching and can then provide edited IgM molecules together with a variety of secondary class-switched variants carrying diversified constant regions, i.e., susceptible to exerting a diversified spectrum of effector functions. In addition, the proposed strategy joins the two edited V_H and V_L regions within a unique IgH locus domain accessible to SHM (i.e., within 1.5 kb downstream of the functional IgH transcription initiation site)¹⁶. The edited Ig could theoretically follow the path of affinity maturation and either increase its initial affinity for the target Ag or adapt to antigenic variations. The latter property would be of particular interest for the Ag variations found in chronic viral infections and in some cancer cells. Previous studies have proposed KI strategies for the IgH locus to express both the Ig H and L chains by covalently linking both chains through a long linker peptide and then incorporating Ig constant sequences within the 1.5 kb domain affected by SHM downstream of the gene promoter^5,17. In contrast, our strategy is based on the expression of so-called scFull-Ig sequences starting from tandemly associated V_H and V_L sequences and can thereby hijack Ig production in B cells through a single edit of the IgH locus. If edited B cells are activated, the scFull-Ig-USM design should therefore spare the C exons from SHM while also exposing both the V_H and the V_L domains to affinity maturation. The scFull-Ig-USM design proposed in this study thus combines several advantages. This approach simplifies gene editing, guarantees the assembly of given IgH and IgL chains into a unique full Ig molecule, and complies with the goal of including both the V_H and V_L regions but not the C region in an upstream transcribed IgH domain exposed to high SHM in activated B lymphocytes. Another important expected outcome of such an editing of the IgH locus is the expression of both membrane-associated BCR and secreted antibodies with intact endogenous Fc domains, thereby capturing all Ig effector functions. The objectives of this study were to validate the functionality of the scFull-Ig-USM format in terms of its efficient secretion, Ag specificity, and ability of the scFv domain to functionally associate with endogenous Fc domains in chimeric molecules after an appropriate IgH KI.

Linkers incorporated into Fv domains may eventually affect the conformation of Ig, and preservation of Ag specificity and/or affinity is a preliminary requirement for a mAb expressed in a single-chain format, as is notably known for its use of rituximab V domains ¹². Accordingly, we verified the retained Ag binding of two classical therapeutic mAbs, anti-CD20 rituximab and anti-HER2 pertuzumab, when modified according to the scFull design. In both cases, antigen specificity was clearly retained, although somewhat lower binding was observed with rituximab in the scFull format, in agreement with the findings of a previous study with other single-chain derivatives of rituximab ¹². In addition to binding, we also verified that secreted scFull-IgG, with a fully preserved Fc region, also retains the ability to efficiently induce ADCC and CDC against target cells.

Using CRISPR-based editing of the IgH locus and KI of scFull-Ig-USM cassettes, we confirmed in the HER2 model that both a human B-cell line and human primary B cells can be efficiently edited for Ig expression: the upstream part of the cassette, including both V regions and the C_L exon, was correctly spliced onto the endogenous IgH C regions, and edited B lymphocytes hereby expressed a membrane BCR with the expected Ag specificity. Like its wild-type counterpart, this edited BCR was connected to the intracellular signaling cascade and induced intracellular phosphorylation upon BCR cross-linking by specific antibodies or by a specific Ag. Transcript sequencing also revealed that scFull-Ig-USM expressed in primary B cells was initially associated predominantly with IgM but later also associated with class-switched IgG, IgA and IgE. The observation that the number of class-switched scFull transcripts increased with time in in vitro-stimulated B cells showed that they mostly originated from ongoing CSR events occurring in culture after IgH locus CRISPR editing. This is notably striking for the ε transcripts since IgE^± cells are short-lived in vivo and absent from normal PBMCs ¹⁸.

Many adjustments need to be made before edited human B cells become good candidates for cell therapy, but improving the efficacy of KI in these cells is an important challenge. First, this efficiency relies on the method of delivery of the DNA template to B cells, which must combine efficiency with low toxicity. Although viral vectors such as adeno-associated virus (AAV) are efficient delivery vehicles, we decided to focus our efforts on a virus-free protocol for which safe and well-designed production and purification methods are easily transferable under good manufacturing practice (GMP) conditions, as required for future preclinical development. In terms of safety, naked DNA-based strategies can compete with other virus-free methods based on nanoparticles and mRNAs ^19–21 but can additionally provide stable expression.

One limitation of B-cell editing is the difficulty of performing HDR with an exogenous DNA template at the optimal insertion site of the IgH locus. We expressed gene cassettes after their integration at the IgH locus using an efficient protocol for HDR-mediated precise genomic insertion in viable B cells, with an optimized Cas9/sgRNA ratio, amount of DNA template, addition of HDR-enhancing molecules and optimized primary B-cell culture conditions prior to and after the KI experiment. The rate and homogeneity of the KI results were subsequently improved by using the tCTS-modified dsDNA template, as described by Nguyen et al.¹³. Overall, although the protocol proposed in this study is quite efficient, further optimization will remain useful. In the future, strategies that block the nonhomologous end-joining (NHEJ) pathway ²² may also be useful for targeting B cells to promote HDR. As B cells express high levels of intracellular Toll-like receptors, the HDR rate and cell survival could be further improved by using ssDNA templates ²³ or by inhibiting DNA sensors prior to electroporation to reduce the toxicity of transfected DNA, as reported for T-cell editing ²⁴.

In conclusion, we report an efficient strategy for editing human B cells by a single precise genetic modification of the IgH locus. This disrupts the expression of the endogenous V_H domain and replaces it with a cassette providing most of the [Fv/C_L] region of a mAb of interest, albeit in a single-chain format. This strategy rebuilds a complete Ig by reassociating the KI [Fv/C_L] region with the endogenous [CH1 + Fc] region encoded by the IgH locus, thereby guaranteeing the expression of the IgH chain under all possible normal variations, either expressed at the cell membrane as a BCR in B lymphocytes or eventually switched or mutated once these cells have been activated or finally secreted by differentiated plasma cells. Edited B cells will then be ready for transfer into recipient mouse models to assess antibody secretion and potential adoptive humoral immunity.

Data availability statement: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declaration of interests: The authors declare no competing interest.

Author Contributions: Conceptualization, M.Co.; Methodology, M.Co., N.A, M.Ca, S.D., J.L.., L.D & Y.D.; Formal Analysis, M.Co., N.A, M.Ca , C.S., J.M., A.G. & Y.D; Writing – Original Draft, M.Co., N.A, M.Ca & Y.D.; Writing – Review & Editing, M.Co; Funding Acquisition, M.Co & A.G.; Supervision, M.Co. & Y.D.

Acknowledgments: This work was supported by Fondation ARC (Grant PGA1 RF20180207070 to MC) and Agence Nationale de la Recherche (Grant SYN-B-18-CE18-022à03)

Paci, A. et al. Pharmacokinetic/pharmacodynamic relationship of therapeutic monoclonal antibodies used in oncology: Part 1, monoclonal antibodies, antibody-drug conjugates and bispecific T-cell engagers. Eur J Cancer 128, 107–118 (2020).
Slifka, M. K., Antia, R., Whitmire, J. K. & Ahmed, R. Humoral immunity due to long-lived plasma cells. Immunity 8, 363–372 (1998).
B, L. et al. Engineering of α-PD-1 antibody-expressing long-lived plasma cells by CRISPR/Cas9-mediated targeted gene integration. Cell death & disease 11, (2020).
Hartweger, H. et al. HIV-specific humoral immune responses by CRISPR/Cas9-edited B cells. J. Exp. Med. 216, 1301–1310 (2019).
Moffett, H. F. et al. B cells engineered to express pathogen-specific antibodies protect against infection. Sci Immunol 4, (2019).
Melenhorst, J. J. et al. Decade-long leukaemia remissions with persistence of CD4 + CAR T cells. Nature 602, 503–509 (2022).
Jackson, H. J., Rafiq, S. & Brentjens, R. J. Driving CAR T-cells forward. Nat Rev Clin Oncol 13, 370–383 (2016).
Grupp, S. A. et al. Chimeric antigen receptor-modified T cells for acute lymphoid leukemia. N. Engl. J. Med. 368, 1509–1518 (2013).
Johnson, M. J., Laoharawee, K., Lahr, W. S., Webber, B. R. & Moriarity, B. S. Engineering of Primary Human B cells with CRISPR/Cas9 Targeted Nuclease. Sci Rep 8, 12144 (2018).
Barrangou, R. & Doudna, J. A. Applications of CRISPR technologies in research and beyond. Nat. Biotechnol. 34, 933–941 (2016).
Le Gallou, S. et al. IL-2 requirement for human plasma cell generation: coupling differentiation and proliferation by enhancing MAPK-ERK signaling. J Immunol 189, 161–173 (2012).
Olafsen, T. et al. Recombinant anti-CD20 antibody fragments for small-animal PET imaging of B-cell lymphomas. J Nucl Med 50, 1500–1508 (2009).
Nguyen, D. N. et al. Polymer-stabilized Cas9 nanoparticles and modified repair templates increase genome editing efficiency. Nat. Biotechnol. 38, 44–49 (2020).
Levy, C. et al. Baboon envelope pseudotyped lentiviral vectors efficiently transduce human B cells and allow active factor IX B cell secretion in vivo in NOD/SCIDγc ^-/- mice. J Thromb Haemost 14, 2478–2492 (2016).
Fusil, F. et al. A Lentiviral Vector Allowing Physiologically Regulated Membrane-anchored and Secreted Antibody Expression Depending on B-cell Maturation Status. Molecular Therapy 23, 1734–1747 (2015).
Peters, A. & Storb, U. Somatic hypermutation of immunoglobulin genes is linked to transcription initiation. Immunity 4, 57–65 (1996).
Hartweger, H. et al. HIV-specific humoral immune responses by CRISPR/Cas9-edited B cells. J Exp Med 216, 1301–1310 (2019).
Laffleur, B. et al. Self-Restrained B Cells Arise following Membrane IgE Expression. Cell Rep 10, 900–909 (2015).
Pardi, N. et al. Administration of nucleoside-modified mRNA encoding broadly neutralizing antibody protects humanized mice from HIV-1 challenge. Nat Commun 8, 14630 (2017).
Nawaz, W. et al. Nanotechnology and immunoengineering: How nanotechnology can boost CAR-T therapy. Acta Biomater (2020) doi:10.1016/j.actbio.2020.04.015.
Smith, T. T. et al. In situ programming of leukaemia-specific T cells using synthetic DNA nanocarriers. Nat Nanotechnol 12, 813–820 (2017).
Fu, Y.-W. et al. Dynamics and competition of CRISPR-Cas9 ribonucleoproteins and AAV donor-mediated NHEJ, MMEJ and HDR editing. Nucleic Acids Res 49, 969–985 (2021).
Shy, B. R. et al. High-yield genome engineering in primary cells using a hybrid ssDNA repair template and small-molecule cocktails. Nat Biotechnol (2022) doi:10.1038/s41587-022-01418-8.
Kath, J. et al. Pharmacological interventions enhance virus-free generation of TRAC-replaced CAR T cells. Mol Ther Methods Clin Dev 25, 311–330 (2022).

No competing interests reported.

Download PDF

Journal Publication

published 28 Sep, 2024

Read the published version in Scientific Reports →

Editorial decision: Revision requested
09 Jul, 2024
Reviews received at journal
27 Jun, 2024
Reviews received at journal
11 Jun, 2024
Reviewers agreed at journal
05 Jun, 2024
Reviews received at journal
17 May, 2024
Reviewers agreed at journal
17 May, 2024
Reviewers agreed at journal
04 May, 2024
Reviewers invited by journal
04 May, 2024
Editor assigned by journal
30 Apr, 2024
Editor invited by journal
30 Apr, 2024
Submission checks completed at journal
29 Apr, 2024
First submitted to journal
24 Apr, 2024

You are reading this latest preprint version

Single-hit genome editing in B cells to redirect their specificity toward tumor antigens

Status:

Journal Publication

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

Recombinant scFull-Ig production

Western blot

Antibody-dependent and complement-dependent cytotoxicity assays

BCR signaling assays

Cell culture

Primary B-cell isolation and culture

Donor DNA preparation and gene editing

DNA and RNA analysis

Flow cytometry

ELISA

Statistical analysis

RESULTS

The scFull-Ig mAb format can be functional

Edition for Ig expression in primary B cells

DISCUSSION

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1