Introduction

Gene editing technology based on clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated protein (CRISPR/Cas) has rapidly developed in the field of livestock genome editing. The CRISPR/Cas system induces double-strand breaks, which can be repaired via non-homologous end joining (NHEJ) or the homology-directed repair (HDR) pathway1,2. NHEJ is predominantly used to induce highly efficient imprecise insertions and/or deletions, whereas HDR produces precise mutation by specific repair outcomes in the presence of DNA donors3. So far, two types of HDR exogenous DNA repair templates have been used: circular or linear double-stranded DNA (dsDNA) and single-strand oligodeoxynucleotides (ssODNs)1. Many single nucleotide polymorphisms (SNPs) have been reported to be associated with important economic traits in livestock. Therefore, introducing precise SNP mutation via HDR is critical for molecular breeding design and for a comprehensive understanding of the genetic mechanisms and functions of SNPs.

Litter size in sheep, a low-heritability quantitative trait controlled by polygenes, has attracted substantial attention in sheep breeding. The bone morphogenetic protein receptor type IB (BMPRIB) gene is the first major gene discovered to be associated with high fecundity in sheep. The effect of the BMPRIB gene on the litter size of sheep has long been a major focus of research4,5. This mutation, located on chromosome 6, is a non-synonymous substitution (c.746 A > G, p.Q249R) in the BMPRIB coding sequence. It was discovered to be an autosomal dominant gene with an additive effect on ovulation rate, designed as BMPRIB. Hu sheep and Small-tail Han sheep carrying the BMPRIB gene exhibit high litter size. Homozygous BMPRIB ewes have a reproduction rate of 220–240%, while heterozygote ewes have a reproduction rate of approximately 180%. Because reproduction rate is a low-heritability trait in singleton sheep, it is difficult to achieve viable genetic progress in a short period through traditional inbreeding selection approaches. Crossbreeding is a common method to introduce the FecB allele from high-fecundity breeds into singleton sheep. However, this method often results in undesired alterations, particularly affecting the fine wool quality of Merino sheep, thus limiting its practical use in breeding programs.

Recently, a number of gene-targeted sheep have been successfully generated via CRISPR/Cas9 using NHEJ or HDR, demonstrating considerable promise for further genetic editing6,7. Although the BMPRIB mutation was successfully introduced into the genome of Tan sheep using CRISPR/Cas9 without SCR78, the desirable high fecundity effect has not yet been fully documented. In this study, the BMPRIB gene-edited fine wool sheep were generated by microinjecting of a mixture of sgRNA, Cas9 mRNA, ssODN donor and SCR7 into sheep zygotes. The gene-edited founders were subsequently propagated using traditional breeding approaches, and the litter size of both founders and progeny was analyzed. Our research provides a new perspective by combining CRISPR/Cas9 gene-editing technology with traditional breeding schemes to produce gene-targeted founders and expand the gene-modified population with desired traits.

Materials and methods

Ethics statement and consent to participate

This study was conducted in accordance with the ethical guidelines of the Institutional Animal Care and Use Committee of the Xinjiang Academy of Animal Science (Approval ID: 2016ZX08010-004-009). The Chinese Merino fine wool sheep used in this research were maintained in optimal conditions at the Research Base of Sheep Breeding of the Xinjiang Academy of Animal Science. Surgical procedures were performed under strict anesthetic protocols to minimize animal suffering, with the specific anesthetic agents used being xylazole. Animal experimentation was approved by the research ethics committee and humane animal handling during experimentation and sample collection was applied. Furthermore, this study was carried out in compliance with the ARRIVE guidelines (https://arriveguidelines.org).

Animals

Chinese merino fine wool sheep used in this study are raised in The Research Base of Sheep Breeding & Reproduction of Xinjiang Academy of Animal Science. Animal suffering from surgery was minimized under stringent anesthetic guidelines. All methods for handling animals were approved by the Institutional Animal Care of Xinjiang Academy of Animal Science.

sgRNA and ssODN design

The sgRNA targeting sheep BMPRIB gene was designed by searching for a sequence consisting of 5ʹ -N (20) NGG- 3ʹ near the c.746 A > G base mutation target site (Fig. 1). The ssODN used for HDR template was synthesized with a single-base mutation of A > G at 746 locus, which has a total length of 120nt and homology arms of 67 to 52nt on both sides. Both sgRNA and ssODN were synthesized and purified through PAGE by Sangon Biotech (Shanghai) Corporation. The detailed sequences are provided in Supplementary Table S1.

Fig. 1
figure 1

Schematic diagram of precise editing BMPRIB by CRISPR/Cas9 -mediated HDR. SgRNA-targeting sequence is onlined and the PAM sequence is indicated in green. Exon 8 is indicated by closed box. The red nucleotide ‘A’ and ‘G’ are the target site of the intended mutation (c.746 A > G; p.Q249R). The amino acids are indicated by capital letters, with the amino acid variation indicated by the blue frame. ssODN, single-stranded oligo deoxynucleotide.

Full size image

Preparation of sgRNA and Cas9 mRNA

Two oligonucleotides (BMPRIB sgRNA, Table S1) were synthesized and annealed to form double-stranded oligos and subcloned into linearized pX330 vector (#42230, Addgene, Cambridge, MA, USA). The recombinant vector which served as templates for constructing sgRNAs vectors. SgRNA was transcribed in vitro and subsequently purified. The amplifed T7-Cas9 PCR product was gel purifed and used as the template for in vitro transcription as described previously8. The purified sgRNA and Cas9 mRNA were dissolved in nuclease-free water (Ambion, USA). Purity and mRNA concentrations were measured using a Nanodrop spectrophotometer. After being evaluated by gel electrophoresis, the mRNA was stored at -80 °C for further use.

Microinjection and screening of CRISPR/Cas9 targeted mutation of in vitro embryos

Sheep oocytes were obtained from ovaries and matured, cultured in vitro following the procedures outlined by Zhang and Crispo et al.8,9. Oocytes with homogenous cytoplasm and intact surrounding cumulus cells were chosen for maturation. The IVM (in vitro maturation) oocytes were then placed in an IVF (in vitro fertilization) medium and fertilized in vitro approximately 24 h following maturation. To avoid cleavage of zygote before microinjection, zygotes were collected 14 h subsequent to in vitro fertilization. One-cell zygotes were then transferred immediately into TCM199 medium (Sigma-Aldrich, USA). The fertilized oocytes were separated and the mixture of Cas9 mRNA (100ng/µL), sgRNA (50ng/µL) ssODN (50ng/µL) and SCR7 (1µmol/L, Thermo Fisher Scientific, USA) was subjected to cytoplasmic microinjection using a Nikon microinjection system. Injected zygotes were cultured in an in vitro culture medium (synthetic oviductal fluid supplemented with 3 mg/mL bovine serum albumin) for 7 days in a humidified atmosphere of 5% CO2 and 95% air at 38.6℃. The cleavage rate of the embryos was assessed at 48 h post fertilization, and the embryos were collected 48 h after culture.

To screen targeted gene editing events, each embryo was washed three times in PBS (Thermo Fisher Scientific, USA) and lysed as previously described8. The lysate was used directly as a template for PCR amplification of the target sequence for the 2-step PCR. The first round of PCR was performed in a total volume of 20 µl, containing 0.5 µl of each forward and reverse primer(BMPRIB-F2/R2, Table S1), 10 µl of 2×TransFast® Taq PCR SuperMix (TransGen Biotech, Beijing, China) and 5 µl lysate DNA, with ddH2O added to bring the final volume to 20 µl. The reaction conditions were as follows: 95 °C for 5 min, followed by 14 cycles of (95 °C for 30 s; 65–53 °C, decreasing by 2 °C per cycle, for 30 s; 72 °C for 45 s) and then 25 cycles of (95 °C for 30 s; 51 °C for 30 s; 72 °C for 45 s). The reaction was completed with a final extension at 72 °C for 7 min. After the first-round PCR products were diluted 1:25, 5 µl of each reaction was used as a template for the second round of PCR. The second-round PCR reaction mix contained 0.5 µl of each forward and reverse primer (BMPRIB-F1/R1, Table S1), 25 µl of 2×TransFast® Taq PCR SuperMix, and 5 µl of the diluted first-round PCR product. The final volume was brought to 50 µl with ddH₂O. The PCR conditions were as follows: 95 °C for 5 min, followed by 35 cycles of (95 °C for 30 s; 58 °C for 30 s; 72 °C for 45 s), with a final extension at 72 °C for 7 min. Subsequently, PCR products were purified and directly sequenced to confirm the presence of the mutations. Sequences with consecutive overlapping peaks were TA-cloned into a pMD19-T vector (TakaRa, Japan) and sequenced. Ten colonies were chosen at random from each sample and sequenced with the M13 primers. The Clustalx 1.83 and Chromas software were used to align the resulting sequences with the wild-type sequence (NM_001009431).

Generation of gene-modified sheep

Sheep zygotes were obtained from 15 donors via surgical oviduct flushing, estrus synchronization, and superovulation treatment, as previously described6. Gene-modified sheep were generated via microinjection of a mixture of 100 ng/µL Cas9 mRNA, 50 ng/µL sgRNA, 50 ng/µL ssODN and 1µmol/L SCR7 into the cytoplasm of the zygotes.

In brief, donors were treated with Controlled Internal Drug Release(CIDR, progesterone 300 mg, SANSHENG, Ningbo, China) for 12 days and superovulation was performed 36 h before the CIDR Device was removed. Sheep zygotes at the one-cell stage (approximately 20 h post-fertilization) were surgically collected and immediately transferred into Synthetic Oviduct Fluid (SOF) medium containing 3 mg/mL Bovine Serum Albumin (BSA, Sigma-Aldrich, USA). The zygotes were then subjected microinjection into cytoplasmic. The injected zygotes were cultured in in vitro culture medium (SOF supplemented with 3 mg/mL BSA) for 24 h at 38.6 °C and 5% CO2 until dividing into 2 ~ 4 cells.

Fifty-seven ewes between the age of 2 and 4 years old with regular estrus cycles were chosen as recipients. The same treatment was used to synchronize recipients and donor ewes for embryo transfer. The cleaved embryos were transferred into the ampullary-isthmic junction of the recipients’ oviducts. An ultrasound scan 60 days after transplantation confirmed pregnancy.

PCR-based assay

Tissue from a lamb’s tail was sampled for screening of editing event. Genomic DNA was extracted using the TIANamp Genomic DNA Kit (TIANGEN, Beijing, China). PCR reaction mix contained 50 ng template DNA, 0.5 µl of each forward and reverse primer (BMPRIB-F1/R1, Table S1), 25 µl of 2×TransFast® Taq PCR SuperMix,

The final volume was brought to 50 µl with ddH₂O. The PCR conditions were as follows: 95 ºC for 5 min followed by 95 ºC for 30s, 58 ºC for 30s, 72 ºC 45s, and 72 ºC for 7 min for 35 cycles.

DNA sequencing of the target gene

To verify the CRISPR/Cas9-mediated point mutations in the BMPRIB gene, PCR products were sequenced directly (primers used are listed in Table S1). Those with consecutive overlapping peaks were sub-cloned into the pMD-19T vector (TakaRa, Japan) to determine the sequences of each allele. Alignment of the sequenced alleles to the wild-type allele identified mutations.

Propagation of gene-edited sheep

The mating strategy was as follows: founders of BMPRIB gene-edited rams(F0 male) with a BB genotype were mated with wild-type ewes (++) or F0 gene-edited ewes though artificial insemination or natural mating to create an F1 progeny population. When the F1 ewes reached 1.5 years of age, homozygous F0 rams were mated with F1 ewes to produce F2 progeny. Gene-edited sheep were identified using the methods described above.

Results

BMPRIB gene editing in in-vitro zygotes

The genomic DNA of injected embryos were amplified and sequenced to identify editing events. Total of 52 embryos were subjected to microinjection and 37 embryos cleaved after 48 h were sampled for gene target screening. The cleavage rate of the embryo was 71.2% (37/52) (Table 1).

Table 1 Efficiency of HDR and indel mutation in in vitro sheep embryos.
Full size table

The genotypes of the sequencing diagrams were classified into four categories: wild type (c.746 A), homozygous mutation (c.746 A > G), heterozygous mutation (c.746 A > G and c.746 A) and Indel mutation. Among 25 targeting embryos, one was a homozygous mutation (biallelic substitution), 14 were heterozygous mutations (monoallelic substitution), and the other 10 embryos were Indel mutation. The efficiency of the precise base substitution was 42.9% (15/35) (Table 1), which indicated that the sgRNA and Cas9 system was effective on targeting precise mutation in sheep embryos.

Generation of BMPRIB gene modified sheep

Gene edited sheep with specific c.746 A > G mutation of BMPRIB gene were generated based on the editing system of in vitro embryo. A total of 119 fertilized zygotes were collected from 15 donor ewes, of which 101 qualified zygotes were subjected to cytoplasm microinjection of Cas9 mRNA, sgRNA, ssODN and SCR7. Out of 68 cleaved embryos, 57 were transferred to surrogate ewes. Approximately 150 days of gestation, 19 lambs were delivered (Fig. 2A; Table 2).

Fig. 2
figure 2

Analysis of exact modifications by sanger sequencing of gene-edited fine wool sheep. (A) Photos of BMPRIB gene-edited ewes and lambs were taken at one month old. (B) Chromatogram of sanger sequencing of each edited lamb. The black rectangle ‘G’ nucleotide is the introduced mutation. (C) Each type of modifications was illustrated in the sequence of each lamb. The sgRNA sequence was marked in red and PAM was in green; single nucleotide substitute, blue character; “-” represented deletion; “+” or “˄” represented insertion. The modifications were displayed in the right. At least 10 TA cloning colonies derived from the PCR products of each targeted lamb were sequenced. N/N indicated the number of sequences with respective modification to the total sequencing colonies.

Full size image
Table 2 Summary on generation of CRISPR/Cas9 targeted sheep by injection of BMPRIB-sgRNA/ssODN/Cas9mRNA into cytoplasm of one-cell stage zygotes.
Full size table

Genomic DNA from the tail skin tissue of 19 lambs was extracted, amplified and directly sequenced. The results showed that nine lambs (five females and four males) had genomic modifications: five with the homozygous mutation (BB genotype, biallelic substitution), one with a heterozygous mutation (B + genotype, monoallelic substitution), and three with indel mutations (Fig. 2B; Table 2).

Verification of precise mutation by clone sequencing

TA sequencing was performed on three lambs (#7, #8 and #9) with indel mutations to further characterize their mutations. The results showed that, in addition to the targeted c.746 A > G mutation, the three lambs exhibited different indels, including a 1 bp insertion in lamb #7, a 4 bp insertion in lamb #8, and both 1 bp and 4 bp insertion in lamb #9. The genotype of #7 is a precise c.746 A > G substitution, together with a insertion of single G nucleotide at position 747–748. The indel in #8 lamb occurs at position 744, where a single base substitution from T to A occurs, while a single base C is missing at position 745(Fig. 2B). The efficiency of genomic modifications was as high as 47.4% (9/19), and the efficiency of precise point substitution was as high as 31.6% (6/19; Table 2), indicating that Cas9-directed HDR repair is efficient in production of precise gene edited sheep.

Propagation of gene edited sheep and evaluation of reproductive performance

A total of 101 adult wild type ewes ( + + genotype) were raised and crossed with homozygous BMPRIB gene edited founder rams (F0, BB genotype). Eighty-three ewes gave birth to 90 lambs. Using PCR and sequencing screening, 90 lambs (41 males and 49 females) with the B + genotype were obtained. F0 gene edited ewes produced eight lambs with BB genotype. Lambing rate for F0 generation of gene-edited ewes was 160% (8/5), which indicates that the reproductive efficiency of these primiparous ewes was suboptimal, with all yielding singletons. These 98 lambs were F1 progeny of gene edited sheep. F1 gene edited ewes with B + and BB genotypes were mated with F0 BMPRIB gene edited rams with the BB genotype to produce129 lambs. Of these, 104 lambs are not yet of breeding age at present. Therefore, the availability of F2 generation for breeding is constrained. Notably, ewes of F1 generation exhibit the highest lambing rate, reaching up to 170%(129/76).

Litter size of ewes with different genotypes between parents and offspring in gene-edited sheep

Based on records of lambing numbers from different population of sheep, statistical analysis was conducted on the litter size of different genotypes of ewes. The litter size for the three different genotypes (++, B + and BB) of ewes were 1.08, 1.70, and 1.50, respectively. The litter size of BMPRIB gene-edited ewes with BB and B + genotypes was significantly higher (p < 0.01) than that of wild-type ewes with the + + genotype (Fig. 3; Table 3). Among the 147 offspring of BMPRIB gene-edited ewes, 75 had the B + genotype, and 72 had the BB genotype, following Mendel’s law of inheritance (Table S3).

Fig. 3
figure 3

Association between different genotypes of ewes and litter size of fine wool sheep. Ewes (n = 79) with the B + genotype exhibited the highest average litter size, which was significantly greater than that of the + + genotype (n = 83); Ewes (n = 8) with the BB genotype showed an intermediate litter size. Statistical significance is indicated by double asterisks (p < 0.01).

Full size image
Table 3 Analysis of lambing litters of gene edited ewes and wild type ewes.
Full size table

Influence of parity on lambing performance

All F0 primiparous ewes delivered single lamb. However, as parity increased, the reproductive capacity of multiparous ewes(≥2) improved, leading to twin births, as shown in Fig. 4. In the F1 population of gene-edited ewes, the lambing rate increased to 170% (129/76). Of these F1 primiparous ewes, 46.9% (23/49) gave birth to multiple lambing (two or more), whereas 53.1% (26/49) delivered singletons. For F1 multiparous ewes (≥2), the percentage of singletons was 40.7% (11/27), while the percentage of multiple births was 59.3% (16/27). Overall, the percentage of singletons in primiparous ewes was higher compared to that in multiparous ewes (≥2). As parity increased, ewes exhibited a higher frequency of multiple births, indicating that multiparous ewes have a greater likelihood of multiple births than primiparous ewes.

Fig. 4
figure 4

Heatmap of single and multiple lambing rates across generations and parties of ewes. The color gradient represents the percentage rates, with darker blue indicating higher rates and lighter blue indicating lower rates. This figure highlights the variation in reproductive performance between generations and parity statuses of ewes.

Full size image

Discussion

CRISPR/Cas9 technology has shown tremendous potential for genetic modifications across a wide range of fields, including agricultural, biomedical research, and more10,11. Its application in livestock breeding has enhanced production performance12,13 and enabled the precise introduction of desirable traits, such as wool fiber traits7,14,15, coat color6, reproductive performance8,16,17, tail traits18, disease resistance, and the development of human disease models19,20,21.In this study, we successfully applied CRISPR/Cas9-mediated homologous-directed repair (HDR) to introduce the FecB mutation (c.746 A > G) in the BMPRIB gene of Chinese Merino fine wool sheep. The results validate the feasibility of this gene-editing approach in sheep and offer valuable insights into enhancing reproductive performance.

There are numerous methods for the site-specific integration of foreign genes, which are based on two repair mechanisms, HDR and NHEJ22,23. In the presence of DNA donors, the HDR pathway enables specific repair outcomes3, whereas the NHEJ tends to produce random insertions or deletions in the absence of a donor template. When ssODN is used as a donor template, it results in high precision and efficiency compared to dsDNA templates. However, it has been observed that HDR-dependent genomic engineering frequently results in indels, confirming the complex mechanism of ssODN-mediated HDR. In previous studies, Niu et al.17 introduced G-A point mutations in the GDF9 gene in goat, resulting in single amino acid substitutions followed by indels, where the efficiency of the precise point substitution was 23.5% (4/17). Zhou et al.16 demonstrated the successful introduction of the BMPRIB mutation in Tan sheep using the CRISPR/Cas9 system without SCR7, resulting in an editing efficiency of 23.8%. However, the fecundity traits in these gene-edited Tan sheep have yet to be thoroughly characterized. In our study, the results indicated that CRISPR/Cas9-mediated HDR repair can achieve high levels of precision when combined with ssODNs and SCR7 as repair enhancers. The overall gene-editing efficiency in this study was 47.4% (9/19) and 31.6% (6/19) are precise base substitutions. These results suggest that CRISPR/Cas9-mediated HDR repair, when combined with SCR7 as enhancers, can achieve a high level of precise targeting. The efficiency is comparable to those of previous studies using similar approaches in other species, but it represents a notable achievement in sheep. Furthermore, the Mendelian inheritance of the BMPRIB gene in subsequent generations (F1 and F2) confirms the stability of the introduced mutation, offering a reliable method for propagating the gene-edited offspring through traditional breeding practices.

In recent years, BMPRIB has been utilized more and more for molecular marker-assisted selection breeding. According to Yuqing Chong et al.24, BMPRIB has essentially the same effect on litter size in domestic sheep, for every extra copy of the B gene, the litter size increases by 0.4–0.5. Chu et al.25 found that BB genotyped ewes had 1.4 more litter sizes than + + genotyped ewes (p < 0.01), while B + genotyped ewes had 1.11 more litter sizes than + + genotyped ewes (p < 0.01) in the Small-tail Han sheep population. The ewes with B + and BB genotype in our work produced a litter size that was substantially higher than that of ewes with the + + genotypes. Though the limited number of BB genotype ewes in our work only provide a suggestive lambing performance, it is indicated the increase of litter size in the founders and offspring. Future expansion of the population achieved through propagation could provide more data to evaluate the reproductive performance of gene edited sheep.

The single and multiple lambing rates in primiparous and multiparous ewes were examined in this study. Though the results indicated no significant difference in the multiple lambing rate between primiparous and multiparous ewes (χ² = 0.622, degrees of freedom df = 1, p > 0.05), nonetheless, compared to primiparous ewes, multiparous ewes tended to have higher lambing rates. This could be explained that the multiparous ewes may have better physiological state or more developed reproductive system, which both provide benefits for producing numerous births. To confirm this point, more sample size is required in the future.

Although the CRISPR/Cas9 system is considered superior to previously gene editing techniques in livestock26, there are still numerous limitations in the establishment of large animal models, including inadequate gene targeting efficiency, mosaicism, and off-targeting events27. One notable problem was the occurrence of indel mutation, approximately accounting for around 16% (3/19). These drawbacks suggest that though the overall efficiency of HDR-mediated editing is much high for generation of targeted mutation, optimizing techniques are expected to further reduce the risk of off-target effects. Additionally, large deletions at the target site, which are a common outcome of Cas9-mediated gene editing, were not addressed in this study. Such deletions could go undetected by PCR if they exceed the size of the PCR amplicon, potentially impacting the genotyping results, particularly for BB genotyped animals. Future studies should incorporate methods such as long-range PCR or next-generation sequencing (NGS) to detect larger structural variations that might not be captured by standard PCR. Mosaicism, where different tissues or cells within an organism have varying genotypes, remains an unresolved issue. In this study, we analyzed only a single tissue sample from each animal, which limits our ability to assess the full extent of mosaicism. While amplicon cloning and sequencing of in vitro embryos were used, this approach provides only partial data on mosaicism. To address this limitation, future research should incorporate ICE, TIDE, or NGS sequencing to more thoroughly analyze mosaicism in targeted embryos or animals. Finally, off-target effects were not screened in this study, a notable limitation. Although our approach focused on the precise introduction of the c.746 A > G mutation, we did not perform systematic off-target analyses to evaluate unintended genetic modifications at non-target sites. Off-target mutations, if present, could undermine the specificity of CRISPR/Cas9 editing. Future studies should include off-target screening methods such as GUIDE-seq or CIRCLE-seq to assess the specificity of the CRISPR/Cas9 system and ensure no unintended mutations occur. Future research could benefit from newer gene-editing technologies, such as base editors28,29 and prime editors30,31. These technologies offer improved precision in making targeted genetic changes with reduced risks of indels and off-target effects, which could further enhance the efficiency and specificity of genome editing in livestock.

In conclusion, Chinese Merino fine wool sheep with specified point mutations in the BMPRIB gene were successfully generated through CRISPR/Cas9-mediated HDR using an ssODN template. The efficiency of introducing precise point substitutions was 31.6% (6/19), demonstrating a high level of efficiency. The F2 BMPRIB-edited sheep population, which carries specific mutations for multi-lambing traits, has been successfully propagated. These results demonstrate that compared to conventional sheep breeding strategies, genome editing technology offers a more precise, effective, and time-saving method for improving sheep breeds. By enabling the direct manipulation of specific traits, genome editing facilitates the swift enhancement of breed characteristics, significantly accelerating the breeding process.