Alphafold prediction

The protein structure prediction was performed in the alphafold2 software under the default arguments by inputting the full-length amino acid sequence of chicken ovalbumin-coupled EGFP proteins with different linkers.

Animal care

The use and care of animals complied with the guideline of the Biomedical Research Ethics Committee of the State Key Laboratory for Conservation and Utilization of Subtropical Agro-bioresources (in Guangxi University, Nanning, China).

Plasmid construction

The CRISPR/Cas9 plasmid targeting ovalbumin was constructed base on the plasmid PX330 (PX330 was a gift from Feng Zhang; Addgene plasmid #42230). Briefly, we fused an EGFP after the coding region of Cas9 protein to construct a U6-sgRNA-CAG-Cas9-EGFP plasmid, linear the plasmid by BbsI (New England Biolab, NEB) digestion and inserted the annealed sgRNA oligos by T4 ligation (NEB). For donor plasmid, 800 bp DNA sequences around sgRNA targeted sites were cloned from the genomic of three-yellow-chicken and fused to the sides by EGFP-3′-UTR-CAG-mCherry fragments by NEBuilder® HiFi DNA Assembly Master Mix (NEB). The piggyBac transposon plasmid containing CAG-EGFP elements and transposase plasmid was a gift from Kehuan Lu professor at Guangxi university. All these plasmids were extracted following the manuals of E.Z.N.A Endo-free plasmid mini kit.

PGCs deriving

For primary PGCs deriving, fertilized eggs were collected from Donglan chickens, a black feather chicken breed native to Guangxi Province, China. Embryonic gonads were isolated from 7-day-old chicken embryos (stage HH27–31), trypsinized with 0.05% trypsin-EDTA at 37 °C for 10 min, neutralized in DMEM/F12 supplemented with 10% fetal bovine serum FBS and plated in 24-wells plates with 24-well transparent membrane (PET) inserts (1.0 μm; Millipore, Stafford, VA, USA). After 4–5 h incubation in 5% CO2 at 37 °C, suspended cells were collected by gentle pipetting transferred into PGCs culture medium coupled with an insert (mKO-Insert), and kept passaged or changed medium every 2 days. PGCs culture medium was based on a Knockout DMEM (osmolality:250 mOsmol/kg, customized in Thermo-Fisher) supplied with 0.2% chicken serum, 1% fetal bovine serum (FBS; Hyclone), 1 X NEAA, 0.1 mM of b-mercaptoethanol, 4 ng/mL human recombinant FGF(R&D), 1.2 mM sodium pyruvate, 1 X GS nucleoside (Milipores), 100 μg/mL sodium heparin (Sigma), 25 ng/mL activin A (PeproTech), and 1 X B27 supplement. Without the specific annotation, the reagents mentioned above were purchased from Thermofisher.

Cells transfection, FACs, and transplant into recipient embryos

For PGCs transfection, we used Lipofectamine3000 Reagent according to the manufacturer’s instructions. A total of 2 μg of plasmids (Cas9: donor = 1: 1) were used for each well of a 24-wells plate, and the transfection solution was removed 12 hours later. mCherry positive (mCherry+) cells were isolated by FACS 3 days after transfection and purified again by second FACS 4–6 days later. mCherry+ cells were cultured in the mKO-insert system and proliferated for transplantation into recipient embryos. For recipient embryo injection, 5–10 × 103 PGCs were injected into the vascular system of 53-55 h recipient embryos.

Germline transmission and chicken lines keeping of OVAL-E3-EGFP chickens

After the recipient chicken hatched, they were raised into sex matured. The sperms from recipient cocks were collected for fluorescence observation to confirm the germline transmission of mCherry+ PGCs. Recipient chickens were made with wild-type three-yellow chicken, and F1 offspring were confirmed by fluorescence detection. For OVAL-E3-EGFP chicken lines keeping, the male chickens were kept for generating offspring and female chickens were used for foreign protein production.

Immunofluorescence

The immunofluorescence (IF) staining was performed based on the paraffin dissection staining method. The oviducts from different chickens were freshly collected and overnight fixed with 4% paraformaldehyde (PFA), washed with PBS and dehydrated in a gradient series of alcohol solutions, and embedded in paraffin. Tissue in paraffin was then sliced up to sections (5 μm), de-paraffinized, and rehydrated for IF analysis. For IF analysis, the monoclonal antibody against EGFP (Abcam, ab184601) and Ovalbumin (Abcam, ab306591) were used as 1st antibody. The second antibody with Alexa Flour 488 label or Alexa Flour 568 (Thermofisher) was used to visualize the expressed protein in sections. For cell nuclear staining, DAPI with Alexa Fluor 405 (Thermofisher) was used.

Eggs fluorescence detection and analysis

EGFP concentration in eggs was measured according to the standard curve established based on the EGFP protein Standard sample. For the EGFP concentration standard curve established, the EGFP protein standard (Beyotime, P7410) was purchased and diluted into different concentrations as standard samples. Two μL of EGFP samples in different concentrations were then moved to the surface of a slice, covered by cover slide, taken photos under the fluorescence microscope by the same exposing time (Fig. S7), and analyzed by the ImageJ software. For EGFP concentration evaluation, the eggs from wild-type chickens, CAG-EGFP chickens, and OVAL-E3-EGFP chickens were respectively collected for fluorescence detection and analysis. The thick albumen and thin albumen were both extracted from the eggs and mixed well by carefully pipetting, and 1 mL of egg white mixture was 1:30 diluted into PBS as the sample for further analysis. The same density analysis protocol as EGFP standards was used to get the fluorescence density of egg white samples, and concentrations were calculated by using the formula of the standard curve.

Western blot analysis

Egg white from different chickens was collected and 1:1000 diluted into PBS and their concentration was detected by bicinchoninic acid (BCA) assay reagent (Beyotime Biotechnology). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer was then mixed with samples and boiled for 5 min and equal amounts of proteins (20-40 μg) were electrophoresed on SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad). For first antibodies incubation, the monoclonal antibody against EGFP (Abcam, ab184601) and Ovalbumin (Abcam, ab306591) was used, and PVDF membranes were overnight incubated in 4 °C with 1st antibodies. 2nd antibodies labeled with alkaline ahosphatase (Beyotime) and BCIP/NBT Alkaline Phosphatase Color Development Kit (Beyotime) were used for visualization of EGFP protein or ovalbumin protein bands. The band grey density assays were performed by using ImageJ.

Oviduct EGFP+/− cells WGS and RNA-seq analysis

For cell isolation, fresh oviducts from OVAL-E3-EGFP chickens were dissected and washed three times in PBS supplied with 5x Anti-Anti (Thermofihser). The tissue on the surface of the oviduct inner wall was collected and trypsinized by 0.25% enzyme for 10 minutes at 37 °C, the suspension was then put through a 70 μm cell strainers, and flowed cells were collected and re-suspended by DMEM medium supplied with 2% of FBS. Freshly isolated cells were separated into EGFP+ and EGFP- groups by FACS.

For WGS analysis, the genomic DNA of EGFP+/EGFP- cells was extracted by using the DNeasy blood and tissue kit (Qiagen) according to the manufacturer’s instructions. WGS was performed at mean coverages of 55x by BGI DNBSEQ-T7. The qualified sequencing reads in WGS data were mapped to the reference genome (GRCg7b) by using BWA (v0.7.17), and Picard tools (v2.25.7) were then used to sort and mark duplicates of the mapped BAM files. To save computing time and resources, Strelka (v2.9.10) was first run for the detection of whole genome de novo Indels and SNVs. Then 200 bp upstream and downstream of the mutation location were selected as candidate regions. To call variants with high confidence, Mutect2 (v4.2) and Lofreq (v2.1.5) were run for the detection of SNVs with candidate regions. In parallel, Mutect2, and Scalpel (v0.5.4) were run for the detection of Indels with candidate regions. The variants that overlap in the three algorithms would be considered true SNVs or Indels. In the variants calling, we used EGFP+ data as a control to identify the mutations that appeared in EGFP- data for each pair of samples. Furthermore, all identified and overlapped SNVs and Indels were confirmed by manual realignment, and variants locus in repeat sequence array would be removed.

For RNA-seq analysis, the total RNA from EGFP+/EGFP- cells was extracted by using the RNAeasy kit (Qaigen). RNA-seq was then performed in illumine platform, and raw data was trimmed and mapped by Hisat2 (v2.1.0), sorted by samtools (v1.10), and visualized by IGViewer (v2.12.3).

Statistical analysis

R version 4.2.1 (https://www.r-project.org) was used to conduct all the statistical analyses in this study. All tests conducted were two-sided, and the significant difference was considered at P < 0.05.