Fig 2: Combinations of mAbs and EGFR‐specific TKIs downregulate in vitro several receptors for survival factors and effectively block the ERK and AKT pathways NSCLC expressing single‐site mutants of EGFR, PC9 (3 × 106) or H3255 (8 × 106), were seeded in 10‐cm dishes. On the next day, complete media were replaced with media containing serum (1%) and the cells were treated for 24 h with different EGFR‐specific TKIs (erlotinib, 50 nM; osimertinib, 50 nM, or afatinib, 10 nM), either alone or in combination with 2XmAbs (cetuximab and trastuzumab, 5 μg/ml each). Thereafter, cells were washed with cold saline and extracted. Proteins were separated using gel electrophoresis and transferred onto nitrocellulose membranes. After blocking, membranes were incubated overnight with the indicated primary antibodies, followed by incubation with horseradish peroxidase‐conjugated secondary antibodies (60 min), and treatment with Clarity™ Western ECL Blotting Substrates (Bio‐Rad). ECL signals were detected using the ChemiDoc™ Imaging System (Bio‐Rad) and images were acquired using the ImageLab software. Signals (relative to Control) were quantified and normalized to the signals of GAPDH (numbers shown below each lane).PC9 cells (1 × 106) were seeded in 6‐well plates and treated as in (A). After washing with acidic buffer (glycine 100 mM, pH 3.0), cells were incubated with fluorescently labeled antibodies against EGFR, HER2, and HER3, and surface expression levels of each receptor were analyzed using flow cytometry. The normalized fluorescence intensity is shown as averages + SEM of four experiments. Significance was assessed using two‐way ANOVA followed by Dunnett’s multiple comparisons test. Note that non‐significant comparisons are not shown. See P‐values in Appendix Table S2.PC9 cells were seeded on coverslips and treated for 24 h as in (A). Cells were washed in acidic buffer, fixed in paraformaldehyde (4%), and incubated with specific primary antibodies, followed by an Alexa Fluor 555‐conjugated secondary antibody (pseudo colored in green). DAPI (blue) was used to stain nuclei. Images were captured using a confocal microscope (40× magnification). Scale bars, 40 μm. Source data are available online for this figure.

Fig 3: Characterization and in vitro activity of ErbB2CAR-T cells: (A) The ErbB2CAR construct is comprises the ScFv specifically directed against ErbB2 followed by the co-stimulation domain of CD28 and the activation domain from FcγR. (B,C): Retroviral supernatant collected from ErbB2-expressing packaging cells (PG13) was used to infect PBLs from a healthy donor. Infected and non-infected cell cultures were grown for an additional 3 days following transduction. On the day they were harvested, the cells were analyzed by flow cytometry and the CAR expression level was calculated. (B) A representative flow cytometry analysis of lymphocytes transduced with ErbB2CAR-GFP. (C) Average transduction expression of the ErbB2CAR-GFP. The dots indicate four independent experiments, and the lines show mean ± STDEV. (D) Expression of CD4/CD8 gated from CD3 positive cells out of ErbB2CAR-GFP transduced T cells. Transductions of four donors are shown; a line indicates the average. (E) Expression of ErbB2 antigen on the human OC cell lines NAR, OVCAR8, and SKOV3. Cells were stained with anti-human ErbB2-APC or mouse IgG1k-APC isotype control and analyzed by flow cytometry. (F) Killing assay using effector (E) and target cells (T). Cells were incubated for 16 h with ErbB2CAR or UNT T cells (as control) at different E:T ratios as indicated. The percentage of killing was measured by methylene blue colorimetric assay. An average of five ErbB2CAR versus UNT experiments are presented with ± SEM. * p < 0.05; ** p < 0.01. (G) T cells transduced with ErbB2CAR (black bars) or UNT cells (gray bars) were co-cultured with the indicated ovarian cell lines at an E:T ratio of 2:1. After 16 h, IFNγ levels in culture supernatants were measured by ELISA. The results shown are the average of three experiments ±STDEV. ErbB2CAR versus UNT * p < 0.05; ** p < 0.01. F/G statistical analysis was performed using Student’s t-test between ErbB2CAR versus UNT treatment.

Fig 4: Hit-and-run epigenetic editing by Ezh2-dCas9, but not KRAB-dCas9, with targeted DNA methylation establishes and maintains a local heterochromatic environment. a Schematic representation of epi-dCas9. dCas9 fusions to effector domains (ED) contained N-terminal and C-terminal nuclear localization domains (NLSs), as well as an N-terminal 3XFLAG epitope tag with a 15-aa linker [(GGS)5] separating dCas9 and the EDs. Effector domains KRAB, Ezh2 and DNMT3A with domains are shown next to epigenetic marks they induce. A detailed view around the human HER2 promoter region shows an annotated CpG island (green bar), gRNAs targeting HER2 promoter region (blue boxes) and the regions interrogated by bisulfite cloning (gray bar) and ChIP-qPCR (black bar). The transcription start site (TSS) is indicated. UCSC browser tracks of DNAse1 hypersensitivity and H3K27ac ChIP-seq enrichment at the HER2 locus (human genome assembly hg19) are also shown. b Diagram of experimental design to assay for long-term repression using flow cytometry to select for low HER2 expression (HER2−) 4 days after transient transfection assays. Relative epi-dCas9 expression was determined by RT-qPCR in HCT116 cells after co-transfection of plasmids expressing dCas9,KRAB-dCas9 and Ezh2-dCas9 with three gRNAs targeted to the HER2 promoter (Dunnett’s test P < 0.001; n = 3 independent experiments per epi-dCas9 fusion; mean ± SEM). Representative flow cytometry data for HER2 expression in HCT116 cells treated with plasmids expressing Ezh2 + D3A + D3L (Ezh2-dCas9, D3A-dCas9 and overexpressed D3L) or KRAB + D3A + D3L (KRAB-dCas9, D3A-dCas9 and overexpressed D3L) in the presence of three gRNAs targeting the HER2 promoter. HER2− cells collected are indicated. Untreated control to identify HER2+ population is shown on the left. c Long-term repression of HER2 mRNA levels in HER2− cells was monitored for 50 days using RT-qPCR after co-transfection of plasmids expressing the indicated epi-dCas9 fusions and three gRNAs targeted to the HER2 promoter (n = 3 independent experiments each; mean ± SEM). d Average methylation levels were determined by bisulfite cloning at the HER2 promoter 10 days after transfection with the indicated Ezh2-dCas9 and KRAB-dCas9 cocktails or dCas9 (no effector domain). Each line represents a single clone, filled and empty circles represent a single methylated or unmethylated CpG, respectively. Average % methylation is indicated for each treatment. e Methylation level of CpG probes along 7 kb of the HER2 locus 17 days after HCT116 cells were co-transfected with indicated epi-dCas9 and three gRNAs targeting the HER2 promoter. CpG Island is shown in green. f Transient (4 days) and persistent (24 days) histone methylation was determined by ChIP-qPCR at the HER2 promoter in HCT116 cells co-transfected with either dCas9 with no ED or the indicated epi-dCas9 treatment cocktail and three gRNAs targeted to the HER2 promoter (Student’s t test, *P < 0.05, ***P < 0.001; n = 2–4; mean ± SEM). Due to higher cell number required for ChIP-qPCR, cells were cross-linked immediately after puromycin selection (4 days); sort for HER2− cells by FACS was omitted. H3K27me3 and H3K9me3 ChIP enrichment was assayed for Ezh2 + D3A + D3L and KRAB + D3A + D3L, respectively and compared to dCas9 with no ED. Decrease of H3K27ac enrichment was measured by ChIP-qPCR at the HER2 promoter in HCT116 at 4 and 24 days after co-transfection with three gRNAs targeted to the HER2 promoter and dCas9 with no ED or the epi-dCas9 treatment cocktail Ezh2 + D3A + D3L (Student’s t test, **P < 0.01, *P < 0.05; n = 2–3; mean ± SEM)

Fig 5: Full-length DNMT3L is required and sufficient to establish long-term repression facilitated by KRAB-dCas9 and Ezh2-dCas9. Long-term repression was evaluated using three different approaches for targeted DNA methylation: a D3A-based recruitment (D3A-dCas9 + overexpressed D3L), b Co-targeting of D3A + D3L (D3A-dCas9 + dCas9-D3L) and c D3L-based recruitment (dCas9-D3L). a Comparison of the D3A-based recruitment approach with the targetable D3a3L fusion (dCas9-D3a3L), HER2 mRNA levels in HCT116 cells and Snurf mRNA levels in Neuro2A cells were determined by RT-qPCR 14 days after co-transfection of plasmids expressing the indicated epi-dCas9 fusions with the three gRNAs targeted to the HER2 or Snurf promoter (Dunnett’s test, *P < 0.05, **P < 0.01, ***P < 0.001; n = 3; mean ± SEM). b Long-term repression by co-targeting D3A and D3L fusions. Performance of D3L fusions to the N or C terminus of dCas9 was compared. HER2 mRNA levels in HCT116 cells and Snurf mRNA levels in Neuro2A cells were determined by RT-qPCR 14 days after co-transfection of plasmids expressing the indicated epi-dCas9 fusions with the three gRNAs targeted to the Snurf or HER2 promoter (Dunnett’s test, *P < 0.05, **P < 0.01, ***P < 0.001; n = 3; mean ± SEM). c Long-term repression was evaluated by RT-qPCR 14 days after co-transfection of dCas9-DNMT3L with KRAB-dCas9 or Ezh2-dCas9 with gRNAs targeted to the SNURF and HER2 promoter in LNCaP cells. Expression levels were compared to dCas9 with no ED (Dunnett’s test, *P < 0.05, **P < 0.01, ***P < 0.001; n = 3; mean ± SEM). d Schematic of overexpressed full-length DNMT3L and design of dCas9 fusions. DNMT3A or DNMT3L were fused to the N-terminus of dCas9 (D3A-dCas9 and D3L-dCas9, respectively) or the C terminus of dCas9 (dCas9-D3L). dCas9-D3a3L contains D3a and the C-terminal portion of D3L (D3a3L) fused to the N-terminus of dCas9. e Average methylation levels were determined by bisulfite cloning (left panel) and high-throughput bisulfite amplicon sequencing (right panel) at the HER2 promoter 10 days after transfection with the indicated Ezh2-dCas9 and KRAB-dCas9 cocktails or dCas9 (no effector domain). Methylation after treatment with only dCas9-D3L and three HER2 gRNAs was also evaluated. Bisulfite cloning results are displayed as lollipop plots. Each line represents a single clone, filled and empty circles represent a single methylated or unmethylated CpG, respectively. Analysis of bisulfite amplicons sequencing was performed from two biological replicates. Average % methylation is plotted for each indicated epi-dCas9 treatment. Methylation obtained with targetable dCas9-D3L (orange bars) was compared with methylation status after combinatorial KRAB-dCas9 (K) and Ezh2-dCas9 (E) with D3A-dCas9 (D3A) and overexpressed D3L (gray bars). All treatments resulted in significant HER2 target methylation when compared to cells treated with dCas9 or cells that were not transfetcted (NT) (Dunnett’s test, ***P < 0.001; n = 2; mean ± SEM)