Highly efficient gene editing via targeted Cas9 insertion into chicken housekeeping gene

Abstract

Achieving stable and efficient transgene expression is a key challenge in advancing avian genome engineering. Although viral vector-based and piggyBac-mediated transgenesis have been widely used in chickens, both approaches are prone to epigenetic silencing, leading to inconsistent, tissue-specific, and often diminished expression over time. This variability limits used of transgenes requiring robust and long-term expression across multiple tissues. In mammals, site-specific integration into genomic safe harbor loci, such as Rosa26, has enabled stable and predictable transgene expression without disrupting endogenous gene function; however, such strategy has not been established in birds. In this research, we hypothesized that integrating Cas9 into endogenous housekeeping genes (the ACTB and GAPDH) could achieve efficient gene editing in chickens through stable and ubiquitous transgene expression. Using two different approaches, 3′-targeted gene insertion and gene tagging, we inserted Cas9 and GFP cassettes into defined genomic loci in chicken DF-1 cells. Both approaches exhibited stable expression of transgenes in the cells, and functional assays confirmed that Cas9 showed highly efficient nuclease activity following guide RNA delivery. Additionally, we derived single-cell clones stably expressing Cas9, enabling uniform and reproducible genome editing in downstream applications. Targeted insertion of transgenes into active housekeeping genes as candidate safe harbor loci mitigates the limitations of random integration and promoter silencing, offering a robust platform for consistent transgene expression in poultry biotechnology and genome engineering.

Introduction

Advances in genome editing technologies, particularly the CRISPR/Cas9 system, have opened new possibilities in avian biotechnology, enabling precise genetic modifications in chickens for applications ranging from functional genomics to agricultural improvement and biopharmaceutical production. As genome editing tools become increasingly accessible and efficient, the development of reliable transgene expression systems has emerged as a critical need in the field of avian biotechnology (Jinek et al., 2012; Khwatenge and Nahashon, 2021; Kim, et al., 2023; Sid and Schusser, 2018).

One of the major challenges in avian transgenesis is achieving stable and consistent expression of transgenes across various tissues and throughout development. Traditional approaches, including viral vector-based delivery and transposon-mediated integration, have been widely used in chickens, particularly at early embryonic stages. However, both methods largely rely on random genomic integration, which makes them susceptible to epigenetic silencing, position effect variegation, and tissue-specific or mosaic expression patterns over time (McGrew, et al., 2004; Park and Han, 2012; Serralbo, et al., 2020). In many cases, the introduced transgene is silenced or downregulated due to local chromatin environments or the absence of endogenous regulatory elements. This unpredictability severely limits their utility in applications that demand robust and ubiquitous expression across multiple cell types or developmental stages (Cabrera, et al., 2022; Lee, et al., 2016; Mozdziak and Petitte, 2004).

In mammalian systems, the problem of variable transgene expression has been addressed through targeted knock-in at well-characterized “safe harbor” loci, such as Rosa26. These loci support long-term, ubiquitous expression of transgenes without disrupting endogenous gene function and have emerged as standard platforms in many mammalian models (Irion, et al., 2007; Soriano, 1999; Wu, et al., 2016). Recent work in chickens has shown that targeted knock-in of reporter cassettes into transcriptionally active endogenous loci can achieve stable and robust expression. For example, integration of green fluorescent protein (GFP) into the germ cell-specific Deleted In Azoospermia Like (DAZL) gene led to strong, heritable expression restricted to germ cells. This approach demonstrated that exploiting the transcriptional context of active endogenous genes can effectively circumvent silencing and positional effects (Choi, et al., 2022; Rengaraj, et al., 2022).

Inspired by these concepts, we sought to apply a similar intron-tagging approach to ubiquitously expressed endogenous genes, specifically, the housekeeping genes ACTB and GAPDH. To explore different expression strategies, we also designed constructs for targeted gene insertion into 3′ region of each housekeeping gene that could induce the transgene by an exogenous promoter at the same loci, allowing a comparative assessment of promoter-driven versus endogenous context-mediated transgene expression. Given that ACTB and GAPDH are stably expressed across nearly all cell types, they represent ideal candidates for supporting broad and persistent transgene expression in chickens. We aimed to determine whether targeted knock-in at these locations could serve as a platform for stable expression of functional genome-editing components. Furthermore, stable expression of CRISPR/Cas9 components is particularly valuable for establishing genome editing-ready cell lines, which serve as versatile platforms for functional studies, high-throughput screening, and the generation of transgenic animals. In mammalian models, constitutive Cas9 expression has enabled efficient, multiplexed genome modifications and has facilitated downstream applications, such as conditional knockouts (Ding, et al., 2020; Rieblinger, et al., 2021). Developing a stable Cas9-expressing chicken cell line would streamline genome engineering workflows and facilitate functional genomics applications in poultry research. Achieving this through targeted integration into ubiquitously active endogenous loci offers a rational and potentially robust solution.

Materials and methods

Construction of CRISPR/Cas9 vectors

Previously described all-in-one CRISPR/Cas9 plasmids were used to target the ACTB, GAPDH genes, and donor plasmids (Lee, et al., 2017). Guide RNAs (gRNAs) were designed using CRISPOR (Haeussler, et al., 2016), which evaluates both on-target activity and off-target risk based on sequence features and mismatch sensitivity near the PAM site. For insertion of gRNAs into the CRISPR/Cas9 plasmids (PX459), sense and antisense oligonucleotides were designed and synthesized by Integrated DNA Technologies, Inc. (IDT; Coralville, IA, USA). Annealing of sense and antisense oligonucleotides was carried out under the following thermocycling conditions: 95°C for 30 s, 72°C for 2 min, 37°C for 2 min, and 25°C for 2 min. The annealed oligonucleotides were ligated into the PX459 vector using the Golden Gate assembly method, and the constructed CRISPR/Cas9 vectors were validated by Sanger sequencing. Guide RNA-expressing plasmids were independently constructed by ligating annealed single-stranded oligonucleotides into a synthesized plasmid backbone containing only the U6 promoter and gRNA scaffold, without the Cas9 coding sequence targeting the intergenic region between DMRT1 and DMRT3. Sequence information for oligonucleotides is listed in Supplementary Table 1.

Construction of targeted knock-in vectors

For targeted gene insertion into the 3′ region of housekeeping genes, we utilized the CAG-Cas9-T2A-EGFP-ires-puro plasmid (Addgene #78311; RRID:Addgene_78311), a gift from Timo Otonkoski (Saarimaki-Vire, et al., 2017), in which Cas9 is derived from Streptococcus pyogenes and EGFP is derived from Aequorea victoria. For gene tagging, we had GENEWIZ (South Plainfield, NJ, USA) synthesize gene constructs containing the last intron and exon (with the stop codon removed) for each housekeeping gene. The donor constructs were designed with a single ∼200 bp homology arm on the left side of the intended insertion site, enabling targeted gene insertion via an end-joining–dependent mechanism. To clone these constructs, we first digested them with the restriction enzymes SalI and BglII. We then ligated the digested constructs into the CAG-Cas9-T2A-EGFP-ires-puro backbone, which replaced the existing CAG promoter.

Culture and transfection of DF-1 cells

Chicken DF-1 fibroblast cells (ATCC® CRL-12203, American Type Culture Collection, Manassas, VA, USA) were maintained in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; HyClone, Cytiva, Marlborough, MA, USA) and 1 × antibiotic-antimycotic solution (Gibco). Cultures were incubated at 37°C in a humidified atmosphere containing 5% CO₂, and passaging was routinely performed using TrypLE™ Express (Gibco).

To validate CRISPR/Cas9 constructs, DF-1 cells were transfected with plasmids targeting the ACTB and GAPDH loci. Cells were seeded in 12-well plates at a density of 3 × 10⁵ cells per well to reach ∼70% confluence prior to transfection. A total of 5 µg CRISPR/Cas9 plasmid DNA was diluted in Opti-MEM (Thermo Fisher Scientific) and mixed with Lipofectamine 2000 (Thermo Fisher Scientific) at a 1:1 (w/w) ratio. After a 5-minute incubation at room temperature, the transfection mixture was added dropwise to the culture wells. After 24 hours, the medium was replaced with fresh complete DF-1 medium.

For targeted knock-in, cells were co-transfected with donor plasmids harboring the gRNA target site and a puromycin resistance gene, together with CRISPR/Cas9 plasmids. For the 3′ targeted knock-in, donor plasmid targeting vector was additionally introduced. Transfection was carried out as above, using 5 µg of each plasmid (total 10 or 15 µg/well). Following a 24-hour incubation, cells were cultured in complete medium and then subjected to puromycin selection. To validate Cas9-mediated knock-in events in DF-1 cells, dual selection with puromycin (1 µg/mL) and neomycin (300 µg/mL; G418 sulfate, Thermo Fisher Scientific) was applied after gRNA plasmid targeting of the intergenic region between DMRT1 and DMRT3, to enrich for edited populations.

Culture and transfection of PGCs

Chicken PGCs were isolated from the circulating blood of Pure Columbian embryos at Hamburger and Hamilton (HH) stages 13–16 and maintained at 37°C with 5% CO₂ in Ca²⁺-free DMEM (Gibco) supplemented with 100 μM CaCl₂, 1 × B-27 (Gibco), 2 mM GlutaMAX (Gibco), 1 × non-essential amino acids (Gibco), 55 μM β-mercaptoethanol (Gibco), 1.2 mM sodium pyruvate (Gibco), 0.2% ovalbumin (Sigma-Aldrich, St. Louis, MO, USA), 0.2% sodium heparin (Sigma-Aldrich), 10 μg/mL ovotransferrin (Sigma-Aldrich), 25 ng/mL Activin A (R&D Systems, Minneapolis, MN, USA), and 4 ng/mL FGF2 (R&D Systems). For targeted knock-in, PGCs were co-transfected with donor plasmids and CRISPR/Cas9 plasmids. Transfection was performed as described above using 5 μg of each plasmid. Cells were allowed to recover in complete medium for 24 h, followed by puromycin selection (Thermo Fisher Scientific).

Clonal selection of DF-1 cells

For clonal isolation, puromycin-resistant knock-in single cells were seeded at one cell per well into 96-well plates. After approximately two weeks of culture, wells containing a single colony derived from a single cell were identified and selected for expansion. Genomic DNA and total protein were extracted from each resulting clone for downstream analysis. Passaging was performed every 3–4 days, and GFP expression was monitored over a period of approximately two months following clonal isolation.

T7E1 assay and genomic DNA sequencing

Genomic DNA was extracted from puromycin-selected cells to assess genome editing efficiency. Target regions were PCR-amplified using specific primers (Supplementary Table 1). Amplicons were denatured and reannealed to generate heteroduplex DNA, followed by treatment with 5 U of T7 endonuclease I (New England Biolabs, Ipswich, MA, USA) at 37°C for 20 minutes. Digestion products were resolved on a 1.5% agarose gel. For sequence verification, PCR products were cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA) and subjected to Sanger sequencing. Sequence data were analyzed using Geneious Prime software (Biomatters Ltd., Auckland, New Zealand).

Copy number validation

For copy number validation, quantitative PCR was performed on a QIAquant 96 5plex thermal cycler (Qiagen, Hilden, Germany) using SYBR Green Master Mix (Applied Biosystems™, Thermo Fisher Scientific) and genomic DNA as the template, with all reactions run in technical triplicates. The cycling conditions were 95°C for 15 s and 61°C for 60 s for 35 cycles. Cas9 copy number per genome was estimated by comparative Ct analysis using GAPDH as the internal reference gene, assuming two copies of GAPDH per diploid genome and comparable amplification efficiencies between GAPDH and Cas9.

Flow cytometry

Wild-type and clonal DF-1 cells were dissociated and resuspended in PBS. Samples were analyzed on a Cytek Aurora (Cytek Biosciences, Fremont, CA, USA), and data were processed using FlowJo software (Treestar, Ashland, OR, USA).

Western blot

Proteins from clonal DF-1 cells were extracted using RIPA buffer (Thermo Fisher Scientific) supplemented with a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific). Protein extracts from wild-type DF-1 cells served as negative controls. Approximately 7 µg of total protein was separated on 4–20% gradient SDS-PAGE gels (Bio‑Rad Laboratories, Hercules, CA, USA) and transferred onto iBlot 2 PVDF Regular Stacks (Thermo Fisher Scientific). Membranes were blocked with 5% bovine serum albumin (BSA) solution for 1 hour at room temperature and incubated overnight at 4°C with primary antibodies: anti-Cas9 (MA1-202; Thermo Fisher Scientific), anti-GAPDH (NB300-327; Novus Biologicals, Centennial, CO, USA), or anti–β-tubulin (A01410; Genscript, Piscataway, NJ, USA). After washing, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 hour at room temperature. Signals were visualized using an enhanced chemiluminescence (ECL) detection system (Thermo Fisher Scientific), imaged with the iBright 1500 Imaging System, and quantified using iBright Analysis Software (Thermo Fisher Scientific).

Statistical analysis

Statistical analysis was performed using Prism software (GraphPad, Boston, MA). One-way ANOVA was used to compare the means of each group with the control group (WT), and P < 0.05 was considered statistically significant.

Results

Targeting chicken ACTB and GAPDH in DF-1 cells by CRISPR/Cas9

To precisely target the 3′ region or last intron of the chicken ACTB and GAPDH genes without disrupting endogenous expression, we employed CRISPR/Cas9-mediated genome editing in chicken DF-1 cells. We first constructed CRISPR/Cas9 plasmids encoding Cas9 protein and gene-specific gRNAs to evaluate gRNA activity. For ACTB, two gRNAs were designed to target either the 3′ region or the last intron (Fig. 1A). DF-1 cells were transfected with the CRISPR/Cas9 plasmids, and T7E1 assays were performed to assess cleavage efficiency (Figs. 1B and 1D). Both gRNAs were functional, and we selected gRNA #2 for each target site for further validation by genomic DNA sequencing. A sequencing analysis revealed that ACTB 3′-targeting gRNA #2 induced nucleotide deletions with an editing efficiency of 14.3% (Fig. 1C), and the intron-targeting gRNA #2 achieved 100% editing efficiency (Fig. 1E) in TA clones.

Fig. 1.

CRISPR/Cas9-mediated targeting of ACTB and GAPDH genes in chicken DF-1 cells. (A, F) Schematic diagrams of the ACTB (A) and GAPDH (F) gene structures, showing CRISPR/Cas9 targeting sites. (B–E) Validation of ACTB targeting vectors. (B, D) T7E1 assays and (C, E) Sanger sequencing of DF-1 cells transfected with CRISPR/Cas9 constructs targeting the 3′ region (B, C) or intron (D, E). (G–J) Validation of GAPDH targeting vectors. (G, I) T7E1 assays and (H, J) Sanger sequencing of DF-1 cells transfected with constructs targeting the 3′ region (G, H) or intron (I, J). gRNA sequences are shown in red or blue, PAM sequences in yellow. Deleted bases are indicated by strikethrough lines, substitutions by italics, and insertions by lowercase letters.

For GAPDH, single gRNAs were designed to target the 3′ region and the last intron, respectively (Fig. 1F). DF-1 cells were transfected with the corresponding CRISPR/Cas9 plasmids, followed by T7E1 assay (Figs. 1G and 1I). Both gRNAs showed editing activity and were further validated by sequencing analysis in TA clones. The GAPDH 3′-targeting gRNA exhibited a 50% editing efficiency (Fig. 1H), and the GAPDH intron-targeting gRNA showed an editing efficiency of 28.6% (Fig. 1J). Based on these results, gRNA sequences with confirmed editing activity were selected and used to construct targeted knock-in vectors for Cas9 and GFP expression in DF-1 cells.

Targeting strategy for Cas9 and GFP expression

For the targeted gene insertion of Cas9 and GFP-expression cassette into 3′ region of each housekeeping gene, we inserted donor plasmids containing Cas9 and GFP-expression cassette driven by a cytomegalovirus enhancer/chicken beta-actin hybrid promoter (CBh) with puromycin selection marker. By co-delivery of CRISPR/Cas9 plasmid targeting 3′ region, CRISPR/Cas9 plasmid targeting donor plasmid (CRISPR/Cas9 donor plasmid targeting) and donor plasmid, we could induce CRISPR/Cas9-NHEJ-mediated gene insertion into the target region in chicken DF-1 cells (Supplementary Figures 1A and 1C).

For the targeted gene tagging of Cas9 and GFP-expression cassette into housekeeping genes, we constructed a donor plasmid comprising portions of the last intron and exon (without the stop codon) of ACTB or GAPDH and a cassette to express Cas9 and GFP fused with the 2A self-cleavage peptide. The donor plasmid also included a puromycin selection marker. By co-delivery of the promoter-less donor plasmid and CRISPR/Cas9 targeting the last intron region of each housekeeping gene, we intended to induce Cas9 and GFP expression driven by regulatory machinery of endogenous ACTB or GAPDH (Supplementary Figures 1B and 1D).

Targeted knock-in of Cas9 and GFP at ACTB and GAPDH loci in DF-1 cells

To achieve CRISPR/Cas9-NHEJ-mediated 3′ region knock-in or gene tagging of Cas9 and GFP (Fig. 2A), DF-1 cells were transfected with knock-in vector systems targeting either the ACTB or GAPDH locus, as described in the preceding section. Fluorescence microscopy revealed green fluorescence in the established cells, indicating successful expression of the Cas9–2A–GFP cassette. In contrast, non-transfected wild-type DF-1 cells showed no fluorescence under the same imaging conditions (Fig. 2B). Junction PCR was performed to confirm site-specific integration of the donor cassette. Amplification of locus-specific bands in transfected cells indicated successful targeted insertion at both the ACTB and GAPDH loci (Fig. 2C). Sanger sequencing of the amplicons from the junction PCR further validated precise integration at the intended genomic regions, with indels at the junctions consistent with NHEJ-mediated repair (Figs. 2D–G). These results indicate that both 3′ region knock-in and gene tagging strategies enabled stable and accurate incorporation of the Cas9–GFP construct into the DF-1 genome.

Fig. 2.

Validation of Cas9-GFP knock-in at the ACTB and GAPDH loci in DF-1 Cells. (A) Schematic illustration of the 3′ region targeted and tagging CRISPR/Cas9 approaches. (B) Detection of GFP in ACTB and GAPDH targeted chicken DF-1 cells. Non-transfected wild-type (WT) DF-1 cells are shown as a control, appearing without fluorescence under standard and fluorescence microscopy. Cells successfully transfected with the knock-in vector constructs targeting ACTB and GAPDH genes exhibit green fluorescence, indicating expression of the reporter gene. Scale bar, 100 µm. (C) Knock-in-specific junction PCR of targeted sites. (D, F) Sequencing analysis of the 3′ region targeted knock-in in chicken DF-1 cells. The schematic illustrates the gene locus following CRISPR/Cas9-mediated insertion of a donor cassette at the 3′ region targeting site via non-homologous end joining (NHEJ). Sanger sequencing of the junction PCR products confirmed integration of the donor sequence in the adjacent genomic regions with indel mutations. (E, G) This schematic depicts the post-integration structure of each gene following CRISPR/Cas9-NHEJ-mediated targeted gene tagging. The donor plasmid was designed with GFP flanked by genomic homology arms corresponding to sequences adjacent to the targeted intron. Sanger sequencing of the junction PCR products confirmed integration of the donor sequence in the adjacent genomic regions with indel mutation.

Cas9 activity validation in the genome edited DF-1 cell lines

To evaluate the genome-editing activity of integrated Cas9, four genome edited DF-1 cell lines (ACTB 3′ KI, ACTB tagging, GAPDH 3′ KI, and GAPDH tagging) were transfected with a plasmid expressing a gRNA targeting the intergenic region between DMRT1 and DMRT3 (Fig. 3A). The gRNA-expressing plasmid was introduced in each cell line, and targeted mutations were assessed using the T7E1 assay to detect indel formation at the target site. Following gRNA transfection, cleavage bands were detected in both GAPDH knock-in lines, indicating functional Cas9 activity capable of generating site-specific DNA breaks. In the ACTB knock-in lines, no cleavage products were observed under the tested conditions (Fig. 3B). To further confirm the targeted mutation in the GAPDH knock-in cell lines, Sanger sequencing was performed on PCR amplicons spanning the gRNA target site. Indel mutations were detected at the expected cleavage position in both the 3′ region (100% efficiency) and gene tagged lines (80%), consistent with NHEJ-mediated repair (Fig. 3C). These results demonstrate that Cas9 expressed from the GAPDH-targeted loci was robustly active in the genome edited DF-1 cells. Subsequent analyses focused on these lines for clonal isolation and further validation.

Fig. 3.

Validation of Cas9 activity in ACTB and GAPDH knock-in (KI) chicken DF-1 cells. (A) Gene structure of the intergenic region between DMRT1 and DMRT3 is depicted, showing exons as boxes and introns as lines, with the gRNA target site indicated. (B) T7E1 assay for KI DF-1 cells (ACTB 3′ KI, ACTB tagging, GAPDH 3′ KI, and GAPDH tagging) followed by transfection with gRNA expressing vector. (C) Sanger sequencing analysis of KI chicken DF-1 cells (GAPDH 3′ KI, and GAPDH tagging) transfected with DMRT gRNA are shown. gRNA sequences are shown in red, PAM sequences in yellow. The strikethrough lines indicate regions where base pairs have been deleted.

Clonal expansion and validation of Cas9 editing activity in single DF-1 clones

Clonal expansion was performed to generate uniform cell populations for precise characterization of transgene copy number, expression levels, and genome editing activity. The GAPDH tagging DF-1 line (GAPDH::Cas9) was used as a representative model for single-cell isolation and clonal analysis, primarily for practical reasons. Single cells were seeded into 96-well plates, and clonal expansion was monitored using both bright-field and fluorescence microscopy. The resulting clones exhibited diverse growth rates and GFP fluorescence intensities (Fig. 4A). A total of twelve GAPDH::Cas9 DF-1 clones were selected, based on overall cell health, morphology, and growth characteristics. To re-confirm the integration patterns at the single clone level, PCR targeting the 5′ junction was performed using genomic DNA from selected single clones. All twelve clones were positive for junction PCR (Fig. 4B) and sequencing analysis confirmed precise integration at the intended loci (Supplementary Figure 2). Junction sequencing revealed clone-specific indels near the gRNA cleavage site, but all analyzed clones showed a targeted integration of the Cas9 transgene at the same genomic locus. On the other hand, wild-type allele–specific PCR revealed that all GAPDH::Cas9 DF-1 clones retained the wild-type allele and were thus heterozygous (Fig. 4B). Sequencing the wild-type allele PCR products showed that, although targeted knock-in did not occur, small indels presumably resulting from NHEJ were present in the alleles of all tested clones (Supplementary Figure 3), indicating sequence alteration relative to the wild-type allele. This suggests that one allele was precisely modified by targeted gene insertion, whereas the other allele was repaired via NHEJ, generating disruptive indels. Quantitative PCR was performed to determine the relative Cas9 copy number per genome for each selected clone, and the analysis indicated that all clones harbored comparable copy numbers (Fig. 4C).

Fig. 4.

Generation and validation of single-cell clones with Cas9-GFP knock-in at the GAPDH locus in chicken DF-1 cells. (A) Bright-field (BF) and GFP fluorescence images obtained after subculture following single-cell seeding. Each panel represents a clonal population derived from a single genome-edited cell. A total of 16 single-cell-derived clones were identified from the 96-well plates, of which 12 maintained consistent growth after subculture. Clone numbers correspond to the original 16 identified clones, and images of the 12 viable clones are shown. Scale bar, 100 µm. (B) PCR analysis of 12 single-cell-derived clones following subculture. Intron-targeted knock-in alleles were confirmed by 5′ junction PCR using junction-specific primers. The presence of residual wild-type (WT) alleles in individual clones was assessed using WT allele–specific primers. GAPDH PCR served as a genomic DNA quality control. (C) Relative Cas9 copy number was estimated by quantitative PCR (qPCR) using genomic DNA from each clone, normalized to the endogenous GAPDH reference locus (two copies in diploid cells). Bars represent the mean ± SD of technical qPCR replicates (n = 3).

Comprehensive validation of selected Cas9 tagged clones

To enable further analysis, three heterozygous clones were randomly selected from twelve previously characterized GAPDH::Cas9 single-cell-derived clones. The use of heterozygous clones allowed precise assessment of knock-in copy numbers while preserving one wild-type allele, thereby providing flexibility for future sequential targeting or conditional genome editing. The selected clones were continuously cultured, reaching passage 12, to evaluate long-term stability of transgene expression. GFP fluorescence remained consistent over time (Supplementary Figure 4). Evaluation of GFP expression by fluorescence microscopy revealed variation in signal intensity among the clones, which was corroborated by flow cytometry (Fig. 5A). Although clones #13 and #16 showed bimodal GFP intensity profiles, all clones were derived from single cells, and the observed heterogeneity likely reflects variable transgene expression within clonal populations following serial passaging. Quantification of median fluorescence intensity (MFI) revealed that clone #13 exhibited the highest GFP expression, followed by clones #3 and #16 (Fig. 5B). Western blot analysis confirmed Cas9 protein expression in all three clones, with signal intensity corresponding to their respective GFP levels (Fig. 5C). To evaluate potential disruption of the endogenous GAPDH locus, GAPDH protein levels were also assessed. All clones retained near-normal GAPDH expression, and clone #13 showed reduced levels (Fig. 5C), potentially reflecting cell cycle–dependent variation (Mansur, et al., 1993). These results collectively demonstrate the successful generation of stable Cas9-expressing DF-1 clones with distinct integration and expression patterns. The observed variability among single-cell-derived clones underscores the importance of clonal isolation for accurately characterizing transgene behavior. This clone set provides a reliable platform for downstream functional validation and genome editing applications.

Fig. 5.

Characterization of single-cell-derived Cas9-expressing DF-1 clones. (A) Flow cytometry analysis of GFP expression levels in GAPDH tagging clones. (B) Median fluorescence intensity (MFI) of GFP in each clone. Data represents n = 3 biological replicates; bars show mean ± SD. ^⁎⁎⁎⁎P < 0.0001. (C) Western blot analysis of Cas9 and GAPDH protein expression in each clone. α-tubulin was used as a loading control. (D–E) Functional validation of genome editing capability in single-cell-derived Cas9-expressing DF-1 clones. A guide RNA (gRNA) expression vector targeting an internal region between DMRT1 and DMRT3 was transfected into each clone. As a control, wild-type (WT) DF-1 cells were co-transfected with the same gRNA vector and a transient Cas9 expression plasmid. (D) Genome editing activity was assessed by T7 endonuclease I (T7E1) assay. (E) Sanger sequencing of the target site confirmed indel formation at the expected genomic locus. gRNA sequences are shown in red, PAM sequences in yellow. Deleted bases are indicated by strikethrough lines, substitutions by italics, and insertions by lowercase letters.

To assess the genome-editing capabilities of the established Cas9 knock-in DF-1 clones, selected clones (GAPDH::Cas9 clones #3, #13, and #16) were transfected with a gRNA-expressing vector targeting the intergenic region between DMRT1 and DMRT3. As a control, wild-type DF-1 cells were co-transfected with plasmids encoding Cas9 along with a gRNA-expressing plasmid. Cas9 activity was evaluated using a T7E1 assay. Distinct cleavage bands were observed in all tested knock-in clones as well as in the control group, indicating successful induction of double-strand breaks at the target site in all conditions (Fig. 5D). To further validate editing outcomes, Sanger sequencing of PCR amplicons spanning the target region was performed. Indel mutations were detected at the expected Cas9 cleavage site in all tested samples, consistent with NHEJ-mediated repair (Fig. 5E). To benchmark editing performance, genome editing efficiencies were compared between the Cas9 knock-in clones and wild-type DF-1 cells transiently expressing Cas9. The editing efficiencies of the GAPDH::Cas9 clones were 77.8% (#3), 88.9% (#13), and 90% (#16). In comparison, wild-type DF-1 cells transfected with Cas9 exhibited a 33.3% editing efficiency (Fig. 5E). These results demonstrate that the integrated Cas9 in DF-1 cells is functionally active and capable of inducing efficient genome editing, with efficiencies exceeding those achieved by transient Cas9 expressions. The observed variation in editing efficiencies among individual clones facilitated the selection of lines with consistently high editing activity. Thus, GAPDH-targeted Cas9 knock-in DF-1 clones provide a stable and effective platform for CRISPR/Cas9-mediated genome engineering in chicken cells.

Discussion

In this study, we established a CRISPR/Cas9-mediated targeted knock-in strategy in chicken DF-1 cells by inserting Cas9 and GFP into either the 3′ region or the intronic regions of two well-characterized housekeeping genes, ACTB and GAPDH. Our results demonstrate that targeted integration into GAPDH enables robust and stable expression of Cas9, supporting efficient genome editing in single-cell-derived clonal lines. This targeted approach offers a powerful alternative to conventional methods that depend on random integration or episomal expression, which are often subject to epigenetic silencing and inconsistent transgene expression.

Both ACTB and GAPDH loci have been proposed as potential safe-harbor sites for stable transgene integration, and reporter gene expression from these loci has been demonstrated in several studies (Antonova, et al., 2018; Han, et al., 2019; Xiong, et al., 2020). Reporter-based knock-in at the ACTB locus has been experimentally validated in multiple systems, including insertion of a Venus reporter in mouse haploid embryonic stem cells and an mEGFP reporter in human embryonic stem cells (Bhat and Inamdar, 2025; Kimura, et al., 2015), indicating that this locus can accommodate foreign gene integration and expression. However, functional validation using a large effector protein, such as Cas9, has been rarely reported for either locus (Zhang, et al., 2025) and, to our knowledge, has not yet been demonstrated in avian cells. In this study, we demonstrated functional Cas9 activity from the GAPDH locus, whereas knock-in of Cas9 at the ACTB locus did not result in detectable genome-editing activity. This discrepancy is more plausibly explained by differences in effective Cas9 expression levels arising from the genomic insertion locus, rather than by silencing of the integrated transgene. Indeed, GFP reporter expression was detectable at the analyzed passage (Fig. 2B), indicating that complete transgene silencing is unlikely to be the primary cause of the lack of genome-editing activity. In this context, our results indicate that the GAPDH locus supports both stable transgene expression and efficient genome editing, highlighting its potential as a robust integration site in avian systems. By contrast, lower Cas9 activity observed in ACTB-targeted cells suggests that this locus may be less permissive for achieving functionally sufficient Cas9 expression, and further studies will be required to elucidate the underlying mechanisms.

Among the GAPDH-tagged clones, we observed considerable variability in Cas9 and GFP expression levels, despite comparable Cas9 knock-in copy numbers. Although the integration site was the same, this variability may reflect differences in chromatin accessibility or epigenetic regulation, as previous studies have shown that local chromatin context can influence Cas9 activity independently of its expression level, potentially accounting for the variability observed in this study (Pavani and Amendola, 2020; Williams, et al., 2008). Editing efficiency was not proportional to Cas9 expression levels in the clones, consistent with previous findings that factors beyond nuclease abundance, including cell cycle stage, end resection properties, local chromatin context, epigenetic modifications, and DNA repair pathway activity, can influence genome-editing outcomes (Javaid and Choi, 2021; Xue and Greene, 2021). Importantly, this clonal variability allowed the identification of high-performing lines with stable Cas9 expression and reliable editing competence. These findings highlight the complexity of transgene behavior at the clonal level and underscore the importance of thorough molecular and functional validation when establishing genome editing–competent cell lines.

The GAPDH-targeted knock-in DF-1 clones generated in this study offer a stable and reusable platform for diverse genome editing applications in avian systems. Constitutive Cas9 expression eliminates the need for repeated Cas9 delivery, thereby simplifying workflows for gRNA screening, high-throughput mutagenesis, and functional genomic studies. Moreover, the knock-in framework described here can be readily adapted for other genome engineering tools, such as base editors or prime editors, and for the integration of tissue-specific or inducible expression cassettes. While this study primarily focused on a fibroblast-derived cell line, we further confirmed that the same targeting strategy supports stable transgene expression in chicken primordial germ cells (PGCs) (Supplementary Figure 5). These findings indicate that this approach is applicable beyond fibroblasts and may provide a versatile platform for genome engineering across multiple chicken cell types, with potential utility for transgenerational genome modification. Recent studies have also identified the chicken ROSA-like (cROSA) locus, located upstream of the THUMPD3 gene, as a genome safe harbor in avian systems, analogous to the ROSA26 locus in mammals (Dehdilani, et al., 2023). Together with our findings at the GAPDH locus, these studies highlight that multiple permissive genomic loci may be exploited for stable and functional Cas9 integration in chickens.

Our study demonstrates the feasibility and utility of GAPDH-targeted knock-in for stable Cas9 expression in chicken DF-1 cells, but several limitations remain. First, all experiments were conducted in a single immortalized fibroblast cell line under in vitro conditions. It remains to be determined whether this approach can be extended to other cell types or applied in vivo. Second, while GAPDH showed favorable characteristics as a knock-in locus in this study, further comparative analysis of other housekeeping genes or candidate safe harbor sites would help validate the generalizability of this strategy. Finally, although we achieved efficient genome editing using integrated Cas9, the long-term stability of editing capacity, off-target effects and transgene expression under selective pressure or differentiation conditions warrants further investigation. Despite these limitations, our findings establish a proof-of-concept framework for safe and predictable transgene integration in avian systems and lay the groundwork for the development of genome editing–competent cell lines to advance poultry biotechnology.

CRediT authorship contribution statement

Kyung Min Jung: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Rachel Klein: Validation, Investigation. Sabrina I Mony: Validation. Paula R Chen: Writing – review & editing, Resources. Kiho Lee: Writing – review & editing, Resources. Hong Jo Lee: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization.

Disclosures

The authors have no conflicts of interest to declare.