Read Time:23 Minute, 14 Second

General methods

All molecular biology methods and cloning steps were performed as previously described13, including the utilization of USER enzyme (New England Biolabs, NEB, M5505L), Phusion U DNA Polymerase Green Multiplex PCR Master Mix (Thermo Fisher Scientific, F564L), Q5 Hot Start High-Fidelity 2X Master Mix (NEB, M0494L), Mach T1 competent cells (Thermo Fisher Scientific, C8681201) and ZymoPURE II Plasmid Midiprep kits (Zymo Research Corporation, D4201) in accordance with manufacturers’ protocols. Amino acid sequences for base editors highlighted in this study can be found in Supplementary Sequences 331. Sequences of sgRNAs used to target genomic sites can be found in Supplementary Table 3. Representative CABE-Ts and CBE-Ts used in this study have been deposited on Addgene.

Generation of TadA* and TADAC libraries for directed evolution

Synthetic libraries for directed evolution rounds one and two were obtained from Ranomics with the following specifications: evolution round one TadA*8.20 library—each amino acid position of the TadA*8.20 (from ABE8.20) sequence to be represented by all 20 amino acid substitutions at a frequency of 1–3 substitutions per library member (~10 million members). This library excluded all stop sequences and used only one codon per amino acid. This synthetic library was combined with a randomized library generated with error-prone PCR using TadA*8.19 (ref. 13) as a template as previously reported in ref. 12. Evolution round two synthetic library—each amino acid position of the TADAC1.02 sequence to be represented by all 20 amino acid substitutions via at a frequency of 2–3 substitutions per library member (~10 million members). These libraries were cloned into a bacterial expression plasmid containing dead Cas9 (dCas9 D10A and H840A) along with gRNAs targeting the chloramphenicol resistance gene through USER cloning.

Bacterial evolution of TadA variants

Directed evolution of TadA8.19 and TadA8.20 library (directed evolution round one) and TADAC1.02 library (directed evolution round 2) was conducted as previously described in ref. 13 with the following changes: libraries of various TadA* deaminase variants that are included in a bacterial plasmid containing TadA*-dCas9-UGI editor architecture were challenged to revert edits in the chloramphenicol resistance gene to survive treatment with lethal doses of antibiotic drug. In the first round of directed evolution, the evolution library was a combination of an error-prone ABE8.19m TadA* library and a synthetic ABE8.20m TadA* library where each amino acid position is represented by all 20 substitutions at a frequency of 1–3 substitutions per library member. To overcome the antibiotic challenge, 2 C-to-T reversions (proline reversion and active site His reversion) were needed. In the second round of evolution, a synthetic library of CABE-T1.2 was used, which was generated with the specifications as the ABE8.20 TadA* library but with 2–3 substitutions per library member. To overcome the antibiotic challenge, the same 2 C-to-T reversions plus 2 A-to-G STOP codon reversions were needed.

General HEK293T mammalian cell culture conditions

HEK293T cells (ATCC, CRL-3216) were cultured in DMEM + GlutaMAX (Gibco, 10569) supplemented with 10% (vol/vol) fetal bovine serum (Gibco, 10437) at 37 °C and 5% CO2 in accordance with standard protocols from ATCC and as previously described.13

General HEK293T transfection conditions

For all transfections, HEK293T cells were seeded at a density of 3.0 × 105 cells per well in BioCoat poly-d-lysine coated 48-well plates (Corning, 356509) 16–22 h before transfection. Plasmid transfections were performed using Lipofectamine 2000 (Invitrogen, 11668-019) as previously described.13 Transfections with mRNA were performed using Lipofectamine MessengerMAX in accordance with manufacturer protocols, with the following specifics: 500 ng (for saturating conditions) or 62.5 ng (subsaturating conditions) of mRNA encoding for editor or control and 100 ng of synthetic gRNA were combined in 12.5 μl total volume of OptiMEM serum reduced medium (Gibco, 31985). A 12.5 μl 1:12.5 (Lipo:OptiMEM) MessengerMAX mixture was then added to the mRNA/gRNA solution, and the entire contents were left to rest at ambient temperature for 15 min. For mRNA transfections at subsaturating conditions, 437.5 ng of carrier mRNA was also added to maintain equivalent amounts of transfected material. The entire 25 μl mixture was then used to treat the preseeded HEK293T cells. The sequences of sgRNAs used in this study are specified in Supplementary Table 3. Synthetic gRNAs for mRNA transfections have 5′/3′ end-modifications as previously described.13

Targeted amplicon next-generation sequencing of DNA samples

After 4 d of incubation, gDNA from HEK293T cells was harvested from the cells using 100 μl of Quick Extract DNA Extraction Buffer (Lucigen, QE09050) in accordance with manufacturer protocols. For allogeneic T cells, 50 μl of Quick Extract DNA Extraction Buffer was used on 1 × 105 cells at 5–6 d post-transfections. Genomic DNA samples from mammalian cell samples were amplified with primers for site-specific genomic DNA amplification containing adapter sequences compatible with Illumina’s TruSeq HT system (Adapter Read 1 sequence, AGATCGGAAGAGCACACGTCTGAACTCCAGTCA; Adapter Read 2, sequence AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT). The sequences of these primers are listed in Supplementary Table 4. Specifically, 2 μl of gDNA was added to a PCR reaction mixture containing Phusion U Green Multiplex Master Mix (Thermo Fisher Scientific, F564L) and 0.5 μM of each forward and reverse primer. These amplicons were then barcoded using Q5 Hot Start High-Fidelity 2X Master Mix, where 2 μl of amplicon from the first round of PCR was added to the master mix containing 0.5 μM of each unique combination of forward and reverse barcode primer. Thermocycling conditions are as follows: 95 °C × 2 min of initial denaturation; 95 °C × 15 s of cycle denaturation; 62 °C × 20 s of annealing; 72 °C × 20 s of extension, with cycle repeats of 30 for the initial amplicon generation and 10 for barcoding. Barcoded amplicons were purified, size selected via gel electrophoresis and gel extracted using the Qiaquick Gel Extraction Kit (Qiagen, 28706×4), and the resultant DNA concentrations were evaluated with a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific).

Data analysis of targeted amplicon next-generation sequencing

All targeted amplicon NGS data were analyzed using methods previously described, including the use of the following tools/software: trimmomatic (v0.39), bowtie2 (v2.35), samtools (v1.9) and bam-readcounts (v0.8).13

Data analysis of WGS data for guide-independent deamination

FASTQ files were aligned to the human genome (Gencode GRCh38v31 primary assembly) using BWA mem2 (bwa-mem2-2.2.1). Alignments were sorted by coordinates, merged if necessary, and duplicates were marked using Picard (v2.21.7) on default settings. Base-quality score recalibration was then performed using GATK (v4.1.4.1) to create a BAM file for input into LoFreq (v2.1.5) for variant calling. Bulk sample 1 was used as the normal sample and each clonally expanded cell was run as a separate tumor sample to identify somatic mutations specific to each cell. LoFreq was run with the ‘–min-cov 10’ flag to require a minimum of ten times coverage at the variant site and somatic variants were analyzed from the somatic_final_minus-dbsnp.snvs output file, to remove common variants that were likely false positives.

For the odds ratio plots, a single representative cell from the untreated clonally expanded cells is required as a reference point to compare with both the treated and untreated cells for both C-to-T and A-to-G deaminations. This cell was selected by ordering the untreated cells by proportion of A-to-G mutations and proportion of C-to-T mutations and selecting the one cell closest to the median for both metrics. N1 was in position 5/8 for C-to-T mutations and position 3/8 for A-to-G mutations, making it the best candidate for the reference cell across both CABE-T and CBE-T treatments.

Protein expression and purification

TadA*8.20 protein was cloned into a pET51b+ vector with His and SUMO tags at the N-terminus and expressed in E. coli BL21 Star (DE3) cells (NEB, C2527I) in LB media. Cell cultures were grown at 37 °C with shaking at 240 rpm, and protein expression was induced by 0.5 mM IPTG when OD600 reached 0.6. Cell culture was incubated with shaking at 18 °C overnight. Harvested cells were lysed by a high-pressure homogenizer in lysis buffer (25 mM Bis-Tris, 500 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol, pH 6.0 and 1 mM PMSF), and the cell lysate was clarified by ultracentrifugation. Clarified lysates were loaded onto Ni-NTA agarose resin by batch binding for 1 h at 4 °C. The resin was washed with lysis buffer with 20 mM imidazole on a gravity flow column followed by elution with the lysis buffer supplemented with 50/100/250 mM imidazole. The eluted sample was incubated with Ulp1 while dialyzed in 25 mM Bis-Tris, 300 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol and pH 6.0 overnight. The dialyzed sample was loaded onto Ni-NTA resin to remove uncleaved protein and Ulp1. The flowthrough from reverse Ni-NTA was loaded on a 5 ml Heparin HP column (Cytiva) and eluted using a 0–2 M NaCl gradient. Fractions containing TadA*8.20 protein were further purified by size exclusion chromatography on Superdex75 10/300 in 25 mM Bis-Tris, 300 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol, pH 7.0. TADAC-1.14 protein was expressed with N-terminal His-tag in pET51b+ vector and purified as described above, except that Ulp1 tag cleavage and reverse Ni-NTA steps were omitted. TADAC-1.17 and TADAC-1.19 were cloned in pD881 vector (ATUM) with N-terminal His-tag and SUMO tag and expressed in E. coli BL21 cells (NEB). Protein expression was induced by 0.2% (wt/vol) rhamnose at OD600 of 0.6, followed by incubation at 37 °C for 4 h. Purification was performed as described above. These deaminase variants were used for X-ray crystallography studies. All CBE-T base editor proteins used for biochemical studies were expressed and purified as described above with slight modifications.

Crystallization of TadA*8.20 with ssDNA

The crystallization condition of TadA*8.20 with ssDNA containing the adenine analog 2-deoxy-8-azanebularine (d8Az), 5′-G(1)C(2)T(3)C(4)G(5)G(6)C(7)T(8)d8Az(9)C(10)G(11) G(12)A(13)-3′, was identified and optimized using a Mosquito robot (SPT LabTech) at 20 °C. Drops were prepared by mixing 1 μl of protein plus ssDNA solution (0.15 mM TadA*8.20 in 25 mM Bis-Tris, 300 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol, pH 7 and 0.22 mM ssDNA with d8Az) and 1 μl of reservoir solution (27–29% (vol/vol) PEG 3,350, 0.22–0.26 M ammonium acetate, 0.1 M Tris pH 8.5), and equilibrated against 70 μl of reservoir solution. The crystals were transferred to a cryoprotectant solution (15% (vol/vol) glycerol, 29% (vol/vol) PEG 3,350, 0.26 M ammonium acetate, 0.1 M Tris pH 8.5) and flash-cooled in liquid nitrogen.

Crystallization of TADAC-1.17 with ssDNA

The crystallization condition of TADAC-1.17 with ssDNA containing the adenine analog 2-deoxy-8-azanebularine (d8Az), 5′-G(1)C(2)T(3)C(4)G(5)G(6)C(7)T(8)d8Az(9)C(10)G(11) G(12)A(13)-3′, was identified and optimized using a Mosquito robot (SPT LabTech) at 20 °C. Drops were prepared by mixing 1 μl of protein plus ssDNA solution (0.15 mM TADAC-1.17 in 25 mM Bis-Tris, 300 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol, pH 7 and 0.22 mM ssDNA with d8Az) and 1 μl of reservoir solution (4–8% (vol/vol) PEG 3,350, 8–10% Tacsimate pH 6) and equilibrated against 200 μl of reservoir solution. The crystals were transferred to a cryoprotectant solution (12% (vol/vol) PEG 3,350, 10% (vol/vol) Tacsimate pH 6, 25% (vol/vol) glycerol) and flash-cooled in liquid nitrogen.

Crystallization of TADAC-1.14 without ssDNA

The crystallization condition of TADAC-1.14 without ssDNA (TADAC-1.14-holo) was identified and optimized using a Mosquito robot (SPT LabTech) at 20 °C. Drops were prepared by mixing 1 μl of protein solution (0.18 mM TADAC-1.14 in 25 mM Bis-Tris, 450 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol, pH 7) and 1 μl of reservoir solution (1.8–2.0 M ammonium sulfate, 0.1 M HEPES pH 7.5) and equilibrated against 200 μl of reservoir solution. The crystals were transferred to a cryoprotectant solution (1.8 M ammonium sulfate, 0.1 M HEPES pH 7.5, 20% (vol/vol) glycerol) and flash-cooled in liquid nitrogen.

Crystallization of TADAC-1.19 without ssDNA

The crystallization condition of TADAC-1.19 without ssDNA (TADAC-1.19-holo) was identified and optimized using a Mosquito robot (SPT LabTech) at 20 °C. Drops were prepared by mixing 1 μl of protein solution (0.3 mM TADAC-1.19 in 25 mM Bis-Tris, 300 mM NaCl, 1 mM TCEP, 10% (vol/vol) glycerol, pH 7) and 1 μl of reservoir solution (6–12% (vol/vol) PEG 3,350, 0.3–0.5 M ammonium citrate tribasic pH 7.0) and equilibrated against 200 μl of reservoir solution. The crystals were transferred to a cryoprotectant solution (16% (vol/vol) PEG 3,350, 0.6 M ammonium citrate tribasic pH 7.0, 20% (vol/vol) glycerol) and flash-cooled in liquid nitrogen.

Data collection and structure determination of TadA*8.20 and TADAC-1 variants

Data collections were performed at the Frontier Microfocusing Macromolecular Crystallography (FMX) beamline of the National Synchrotron Light Source II or the ID30B beamline of the European Synchrotron Radiation Facility, or the BL13-XALOC beamline of the ALBA Synchrotron or the P13 beamline of the EMBL Hamburg at the PETRA III storage ring (DESY). Diffraction data were processed using XDS29 and scaled using AIMLESS30. The crystal structures of TadA*8.20, TADAC-1.17, TADAC-1.14 and TADAC-1.19 without or with ssDNA were determined by molecular replacement techniques implemented in Phaser31. For the TadA*8.20 structure, the coordinates of the E. coli TadA structure (Protein Data Bank (PDB) code: 1Z3A)32 were used to obtain the initial phases. For TADAC-1.17, TADAC-1.14 and TADAC-1.19 structures, the coordinates of the TadA*8.20 (this study) were used to obtain the initial phases. Following molecular replacement, simulated annealing was performed in phenix.refine33 to remove model bias. The models were refined by iterative rounds of model building and the addition of water molecules using Coot34. Refinement of the structures in phenix.refine used noncrystallographic symmetry restraints, positional and B-factor refinement, and TLS (translation, libration and screw) (except for TADAC-1.17 and TADAC-1.14). The crystals of TadA*8.20 and TADAC-1.17 are merohedrally twinned with twin fractions of 0.375 and 0.246 by Britton analyses (phenix.xtriage), respectively, and the twin law -h,-k,l was used in refinement. The data collection and refinement statistics are summarized in Supplementary Table 2. The residues and nucleotides visualized in the structures, of 167 residues and 13 nucleotides, are listed in Supplementary Table 5. Figures were created with PyMol Software (Schrodinger, 2010. The PyMOL Molecular Graphics System, Version 2.4.1.).

Biochemical characterization of deamination by ABEs, CABEs and CBEs

An sgRNA (mG*mA*mA*CACAAAGCAUAGACUGCGUUUUAGAGCUAGAAAUAGC

AAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCU*mU*mU*mU; modifications: m, 2′-O-methyl, and *, phosphorothioate linkage) was synthesized at Agilent Technologies and Integrated DNA Technologies (IDT). Substrate DNA was synthesized at IDT: DNA strand undergoing deamination (TTCGGTGGCTCCGTCCGTGAACACAAAGCATAGACTGCCGGCGTTTTGGTTGCTCTTCG) was labeled with 5′ ATTO-647 fluorophore and a complementary DNA strand undergoing nicking by D10A-Nickase (CGAAGAGCAACCAAAACGCCGGCAGTCTATGCTTTGTGTTCACGGACGGAGCCACCGAA) was labeled with 5′ 6-FAM fluorophore. For guide RNA-independent deamination, the ATTO-647 labeled single-strand DNA was used as is. For guide RNA-dependent deamination, dsDNA substrate was prepared by annealing the two strands, with twofold excess of the strand undergoing nicking (1:2 nmol). The duplexed DNA was purified by 7.5% Native-PAGE (29:1, acrylamide:bisacrylamide; Sigma). The acrylamide band containing the dsDNA was excised, crushed and rotated overnight in crush-and-soak buffer (400 mM NaCl and 25 mM EDTA) to elute the dsDNA. The eluted dsDNA was precipitated at –20 °C for 2 h after adding 1 volume of 100% 2-propanol, followed by centrifugation at 20,000g for 30 min at 4 °C. The DNA pellet was washed with 1 volume of 70% vol/vol ethanol and centrifuged at 20,000g for 30 min at 4 °C. The pellet was air-dried at room temperature for 30 min and resuspended in water.

RNP complexes were formed by mixing the sgRNA and the appropriate base editor protein in a 1.5:1 molar ratio in ‘RNP assembly and reaction buffer’ (20 mM HEPES-KOH pH 7.4, 100 mM KCl, 5 mM MgCl2, 5% vol/vol glycerol, 2 mM TCEP) and incubating at room temperature for 20 min.

For single turnover kinetics of guide RNA-dependent dsDNA deamination in vitro, to 1-µM final concentration of RNP, a final concentration of 10 nM dsDNA substrate (prepared at 100 nM in RNP assembly and reaction buffer) was added to initiate deamination. The reaction was incubated at 37 °C and aliquots of 5 µl were withdrawn at the indicated time intervals. The reactions were quenched in 50-µl quenching buffer (50 mM Tris–Cl, pH 8.5, 400 mM NaCl, 25 mM EDTA, 0.1% SDS, 1 µl thermolabile proteinase K (New England Biolabs, NEB P8111S) and 1 µl 15 mg ml−1 coprecipitant Glycoblue (Thermo Fisher Scientific, A9515)) for 15 min at 37 °C. The thermolabile proteinase K was inactivated at 75 °C for 15 min.

The quenched reaction time points were then precipitated with 2-propanol as described above. For detecting deaminated adenine (inosine) catalyzed by ABEs, the precipitated time points were treated with Endonuclease V as described previously in refs. 20,35. For detecting deaminated cytidine (deoxy-uridine) catalyzed by CBEs, the precipitated time points were treated with USER II (NEB, M5508L) according to manufacturer guidelines. The samples were mixed with equal volume of formamide gel loading buffer (95% formamide, 25 mM EDTA, 0.025% SDS and 0.025% bromophenol blue), heated to 98 °C for 5 min and resolved on denaturing 7.5% Urea-PAGE (19:1, acrylamide:bisacrylamide; National Diagnostics). The reaction was monitored by scanning the gel sequentially with FAM followed by Alexa-647 settings using ChemiDoc Imaging System (Bio-Rad). The intensities of the un-cleaved and cleaved DNA were quantified using ImageJ 1.53 K. Data were fit to a single exponential decay in Prism 9 (GraphPad Prism, v9.4.0) to calculate apparent deamination rates (kapp). Nicking of the substrate DNA by D10A-Nickase of base editor, constant across all base editors assayed, was detected with the 6-FAM fluorophore and used as control to ensure uniformly active recombinant proteins.

For single turnover kinetics of guide RNA-independent ssDNA deamination in vitro, the reaction was set up as described above but with the following modifications: the base editor was not programmed with sgRNA and was incubated with the ATTO-647 labeled ssDNA strand.

For in vitro end-point deamination assay to compare deamination by ABE 8.20, BE4, CABE-T2.17, CABE-T3.155, CBE-T1.14 and CBE-T1.52, the deamination reaction was set up with 1-µM BE RNP and 10-nM dsDNA substrate as described above. Instead of time points, the whole reaction was quenched after 24 h and precipitated as described above. The precipitated reaction was resuspended in water and split into four equal parts: untreated, treated with Endonuclease V as described, treated with USER II as described and treated with human Alkyl Adenine Glycosylase (hAAG; NEB 0313S) followed by AP Endonuclease 1 (APE1; NEB M0282L) according to manufacturer’s instructions. The combination of hAAG and APE1 was used because of our experimental observation that G:U (product of cytosine deamination) is a substrate for EndoV, which was confirmed by NEB (https://www.neb.com/tools-and-resources/selection-charts/dna-repair-enzymes-on-damaged-and-non-standard-bases). EndoV, therefore, could not be used when comparing ABEs, CABEs and CBEs for relative A-to-I and C-to-U deamination activities. hAAG is more specific, and only produced detectable cleavage product for A-to-I but not for C-to-U deamination under the same experimental conditions and thus was used for such comparisons. Following these treatments, the samples were resolved on Urea-PAGE and data were quantified as described above.

mRNA production of CABE-T, CBE-T and controls used in HEK293T, T cells and primary human hepatocytes

The mRNAs used in this study were produced through in vitro transcription of expression plasmids encoding our editors and controls, in accordance with protocols previously described in ref. 13.

Isolation of single cells by FACS and whole-genome sequencing

HEK293T cells were transfected via Lipofectamine MessengerMAX (Thermo Fisher Scientific, LMRNA001) with control (Cas9, SPACE, etc.) or editor-encoding mRNA along with synthetic gRNA (special order from Axolabs) targeting a region in β-2-microglobulin (B2M). The sequence of this synthetic guide is as follows (Axolabs-specific syntax): ascsusCACGCUGGAUAGCCUCCGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGGUGCusususU

The disruption of B2M upon successful targeting by ABE, CBE or Cas9 at this site has been internally validated. Three days after transfection, cells were dissociated with TrypLE Express, washed with cell staining buffer (Biolegend, 420201) via centrifugation, and resuspended in cell staining buffer containing 1:100 of PE-conjugated antihuman B2M antibody (Biolegend, 316306). After 30 min of incubation on ice in the dark, cells were washed three times with cell staining buffer via centrifugation and strained into standard 5-ml FACS tubes.

Single cells gated as PE-negative were sorted into 96-well plates containing DMEM + 20% FBS + 100 units per ml penicillin/streptomycin (Thermo Fisher Scientific, 15140122). For untreated control, single cells were sorted by live only. Representative gating strategies are provided in Supplementary Fig. 30. After 12 d of culture, gDNA was harvested from cells using the Agencourt DNAdvanced kit (Beckman-Coulter, A48705) in accordance with manufacturer protocols. Confirmation of successful editing of each clone was achieved through targeted amplicon sequencing of the B2M amplicon encompassing the target site. Sequence confirmed gDNA was then submitted to Novogene for library preparation and WGS.

Isolation and culture of allogeneic human T cells

Human T cells were isolated from leukapheresis products (Leukopaks, HemaCare) by positive selection using CD4 and CD8 MicroBeads (Miltenyi, 130045101 and 130045201). T cells were frozen at 25–50 × 106 cells per ml of Cryostor CS10 (Stemcell Technologies, 1001061). For editing experiments, T cells were thawed in a water bath at 37 °C and then allowed to rest overnight in ImmunoCult-XF T Cell Expansion Medium containing (Stemcell Technologies, 10981) 5% CTS Immune Cell SR, Glutamax, 10 mM HEPES, 1% Penicillin/Streptomycin (Thermo Fisher Scientific, 15140122). The next day, T cells were activated using 25 μl of ImmunoCult Human CD3/CD28/CD2 T Cell Activator (Stemcell Technologies, 10970) per ml of cells at 1 × 106 cells per ml plus 300 IU ml−1 of IL-2 (CellGenix, 1420050). Fresh IL-2 was added to T cells every 2–3 d. T cells were cultured at 37 °C and 5% CO2.

Electroporation of human T cells

T cells were transfected 72 h after activation. Cells were resuspended in P3 Primary Cell Nucleofector Solution containing Supplement 1 (Lonza, V4SP-3960). 1 × 106 T cells were edited with 1 μg of synthetic sgRNA (IDT) and 2 μg of editor mRNA in a total volume of 20 µl using P3 96-well Nucleocuvette kit (Lonza, V4SP-3960). The three sgRNAs used are as follows: B2M Exon 2 (B2M Ex.2), pmSTOP C6, CD247 pomSTOP C7 and PD-1 Ex.1 SA C7 are specified in Supplementary Table 3. T cells were electroporated with the 4D-Nucleofector system (Lonza, AAF-1003B and AAF-1003S) using program DH-102. All experiments were performed with two independent T cell donors. For NGS analysis, 1 × 105 T cells per condition at each timepoint were pelleted, supernatant was removed and pellets were resuspended in 50 ml of QuickExtract DNA Extraction buffer (Lucigen, QE09050) and transferred to a PCR plate for targeted amplicon sequencing.

Flow cytometry of human T cells

Protein knockout was evaluated by flow cytometry 5–6 d post-editing. T cells were stained with fluorophore-conjugated antibodies for TCRα/β (Biolegend, 306718), β2M (Biolegend, 316304) and PD-1 (Biolegend, 367422) via 1:33 dilution in standard PBS. For PD-1 analysis by flow cytometry, T cells were treated with Cell Activation Cocktail (without brefeldin A) (Biolegend) overnight before staining. Events were collected using a MACSQuant Analyzer 16 (Miltenyi). Data were analyzed using the FlowJo software (v10.8.1)

Generation and maintenance of primary human hepatocytes

Cryogenically frozen primary human hepatocytes (BioIVT) were thawed and plated at a density of 3.5 × 105 cells per well on BioCoat Collagen I 24-well plates (Corning, 354408) and maintained in CP Media supplemented with Torpedo Antibiotic Mix (BioIVT) in accordance with protocols provided by BioIVT. Once PHH monocultures were established overnight, generation of long-lived PHH cultures involved the additional coculturing of 3T3-J2 murine fibroblasts (Kerafast, EF3003) at 2.0 × 104 cells per well to the established PHH monocultures. PHH cocultures were maintained with media changes every 48 h throughout the duration of the study.

Transfection of primary human hepatocytes

PHH cocultures were transfected 48 h after coculture generation with 3T3-J2 murine fibroblasts. Transfections with mRNA were performed using Lipofectamine MessengerMax (Thermo Fisher Scientific, LMRNA003) in accordance with manufacturers’ protocols, with the following optimized specifics: 1 µg (for saturating conditions) of mRNA encoding for editor and 333 ng of synthetic gRNA (Synthego) were combined in 30 µl of OptiMEM serum reduced medium (Gibco, 31985). A 30 µl 1:15 (Lipofectamine:OptiMEM) mixture was added to the mRNA/gRNA solution with the resulting final mixture left to rest at ambient temperature for 15 min. The entire 60-µl solution was used to treat a well of cocultured primary human hepatocytes. Each study condition was run in triplicate and transfection amounts used were scaled up accordingly. At 9 d post-transfection, the PHH cocultures were lysed with a solution of 10 mM Tris–HCl pH8.0 (Thermo Fisher Scientific, 15568025), 0.05% SDS (Thermo Fisher Scientific, 15553027) and 500 µg proteinase K (Thermo Fisher Scientific, EO0491) at a total of 200 µl per well. Once lysed, lysate was treated at 85 °C for 15 min to inactivate proteinase K. The sequences of sgRNAs used in this study are specified in Supplementary Table 3.

Protein assays of transfected primary human hepatocytes

PCSK9 protein knockdown quantification was assessed using a Human PCSK9 SimpleStep ELISA kit (Abcam, ab209884) by measuring secreted PCSK9 concentration in supernatant collected every 48 h. Supernatant was ten times diluted using assay buffer, and the assay protocol was run in accordance with the manufacturer’s protocol. LDL-R quantification was assessed using a Human LDL-R SimpleStep ELISA kit (Abcam, ab209884) by measuring secreted LDL-R protein in supernatant collected every 48 h. Both SimpleStep ELISA kits employ an affinity tag labeled capture antibody and a reporter conjugated detector antibody. The capture antibody and detector antibody bind to sample analytes, which are then immobilized to an anti-tag antibody coating the assay well. Both colorimetric ELISA assays are read at an absorbance of 450 nm.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.


Source link

Happy
Happy
0 %
Sad
Sad
0 %
Excited
Excited
0 %
Sleepy
Sleepy
0 %
Angry
Angry
0 %
Surprise
Surprise
0 %
Previous post The economic downturn is hurting the creator economy, one of the most-hyped sectors of the past decade, in particular since its middle class hasn't yet emerged (Alex Kantrowitz/Big Technology) – The Knowledge Pal
Next post How Telegram, TikTok, Instagram, Facebook, and Twitter were used to boost election fraud claims in Brazil before riots hit congress and supreme court buildings (Elizabeth Dwoskin/Washington Post) – The Knowledge Pal