Abstract
The CRISPR/Cas9 genome editing tool has the potential to improve the livestock breeding industry by allowing for the introduction of desirable traits. Although an efficient and targeted tool, the CRISPR/Cas9 system can have some drawbacks, including off-target mutations and mosaicism, particularly when used in developing embryos. Here, we introduced genome editing reagents into single-cell bovine embryos to compare the effect of Cas9 mRNA and protein on the mutation efficiency, level of mosaicism, and evaluate potential off-target mutations utilizing next generation sequencing. We designed guide-RNAs targeting three loci (POLLED, H11, and ZFX) in the bovine genome and saw a significantly higher rate of mutation in embryos injected with Cas9 protein (84.2%) vs. Cas9 mRNA (68.5%). In addition, the level of mosaicism was higher in embryos injected with Cas9 mRNA (100%) compared to those injected with Cas9 protein (94.2%), with little to no unintended off-target mutations detected. This study demonstrated that the use of gRNA/Cas9 ribonucleoprotein complex resulted in a high editing efficiency at three different loci in bovine embryos and decreased levels of mosaicism relative to Cas9 mRNA. Additional optimization will be required to further reduce mosaicism to levels that make single-step embryo editing in cattle commercially feasible.
Similar content being viewed by others
Introduction
CRISPR-mediated genome editing in livestock zygotes offers an attractive approach to introduce useful genetic variation into the next generation of cattle breeding programs. However, genetic mosaicism is particularly problematic for CRISPR-mediated genome editing in developing zygotes1,2. Genetic mosaicism complicates phenotypic analysis of F0 animals and may complicate screening multiple founders and breeding mosaic founders to produce an F1 generation. While this is routine in plant and mouse research, such approaches are time-consuming and essentially cost-prohibitive in uniparous large food animal species with long generation intervals like cattle.
A limited number of genome editing studies have been reported in bovine zygotes3, and indicate the frequent production of mosaic embryos. The frequency of mosaicism varies depending upon the type of site-directed nuclease used, the timing of editing relative to embryonic development, the form and efficiency of the targeting regents, the intrinsic properties of the target locus, and the method of delivery1.
Correspondingly, there are a number of experimental variables that need to be optimized to improve the efficiency of obtaining non-mosaic, homozygous genome edited founder cattle. In this study, we focused on the type of CRISPR/Cas9 system delivered (i.e. mRNA or protein) and report the impact on mutation efficiency, levels of mosaicism, and off-target mutations based on next generation sequencing when using CRISPR-mediated genome editing of bovine zygotes.
Results
Guide construction and testing
To determine the optimal parameters for CRISPR/Cas9-mediated genome editing in bovine zygotes, efficiency following microinjection was investigated for three gRNA per locus on three different chromosomes. Three gRNAs were designed targeting the POLLED locus on chromosome 1, a safe harbor locus (H11) on chromosome 17 and a locus (ZFX) on the X-chromosome downstream of the Zinc Finger, X-linked gene (Supplementary Table S1). Three gRNAs per locus were independently injected alongside Cas9 protein in groups of 30 zygotes, 18 h post insemination (hpi). Groups of 50 non-injected embryos were cultured as controls. The highest mutation rates were 76.9% for gRNA2 targeting the POLLED locus, 83.3% for gRNA1 targeting the H11 locus, and 77.8% for gRNA3 targeting the ZFX locus (Supplementary Table S2; χ2 test, P < 0.05). Overall, there was a decrease in the number of embryos that reached the blastocyst stage as the rate of mutation for a given gRNA increased. For each locus, the gRNA with the highest mutation rate was associated with the lowest developmental rate (Supplementary Table S2). gRNAs with the highest mutation rate were selected for further analysis.
Guides targeting the POLLED locus, the H11 locus and the ZFX locus were then injected in groups of 30 in vitro fertilized embryos 18hpi alongside either Cas9 mRNA or protein (Table 1). The blastocyst rate of uninjected controls (30.7%) was significantly higher than embryos that were microinjected with gRNA and Cas9 editing reagents (Fig. 1a; P < 0.001). The overall mutation rate did not differ among the three loci for a given form of Cas9 (Fig. 1b; P = 0.45;); however the probability of a mutation was higher (P = 0.002) when Cas9 protein was microinjected as compared to Cas9 mRNA (Fig. 1c).
Evaluation of mosaicism and off-target insertions and deletions
To evaluate the level of mosaicism, 69 blastocysts (19 gRNA2 targeting the POLLED locus (10 Cas9 mRNA, 9 Cas9 protein), 26 gRNA1 targeting the H11 locus (11 Cas9 mRNA, 15 Cas9 protein), and 24 targeting the ZFX locus (13 Cas9 mRNA, 11 Cas9 protein)) were collected, barcoded by PCR amplification and sequenced on a PacBio sequencer (Supplementary Table S3). Consensus sequences were called from raw reads using circular consensus sequencing (ccs) with a minimum of 3 passes, a minimum predicted accuracy of 99% and a maximum length of 700 bp (Supplementary Table S4). Unsorted ccs reads were aligned to each of the target sequences to analyze the types of insertions/deletions (indels) surrounding the predicted cut site with 26,460 reads aligned to the POLLED target site; 78,305 reads aligned to the H11 target site; and 66,780 reads aligned to ZFX target site (Supplementary Table S5). About half of the aligned sequences for the POLLED locus were wild type sequences (47.8%), while almost three quarters of the H11 and ZFX reads were wild type sequences (75.7% and 71.3%, respectively). The primary indels for reads aligned to the POLLED locus were 7 bp deletion (1672 reads), 11 bp deletion (1751), 4 bp deletion (6356 reads) and 1 bp insertion (2250 reads); aligned to the H11 locus were 11 bp deletion (3246 reads), 6 bp deletion (3813 reads), 3 bp deletion (4091 reads), and 1 bp deletion (7853 reads); and aligned to the ZFX locus were 14 bp deletion (4222 reads), 9 bp deletion (2998 reads), 3 bp deletion (3198 reads), 1 bp deletion (2194 reads) and 1 bp insertion (6532 reads) (Supplementary Table S5).
Ccs reads were then sorted by barcode and analyzed by individual embryos (Fig. 2). Seven samples were discarded from further analysis due to a lack of reads following the quality filtering step (Supplemental Table S3). A total of 10 samples contained only wild type sequence (7 Cas9 mRNA and 3 Cas9 protein), resulting in an overall mutation rate of ~ 84% (Table 2). Of the 62 samples injected 18hpi, four contained only mutated alleles, without evidence for any wild type sequence. All four samples were from embryos injected with Cas9 protein (Supplementary Table S6). Three of these samples contained only one allele and were presumably non-mosaic homozygous, although our analyses could not rule out an unmappable mutation (e.g. large insertion) at the second allele. Each of the mutated embryos containing more than a single allele had at least three individual alleles or a disproportion of reads for each allele, for example 75% wildtype and 25% mutant (Supplementary Fig. S1), suggesting these embryos were mosaic rather than heterozygous. This translates to 94.2% mosaicism when injecting Cas9 protein compared to 100% mosaicism when injecting Cas9 mRNA.
There was a decreased average number of alleles (3.0 ± 0.4) when targeting the POLLED locus using Cas9 protein (Fig. 1d; Table 2), as compared to Cas9 mRNA. There was no significant difference in the number of alleles for the other loci when comparing Cas9 mRNA or protein. However, there was a significant increase in the number of alleles when comparing pooled samples of embryos injected 18hpi with guides alongside Cas9 mRNA (5.23 ± 0.268), as compared to protein (4.23 ± 0.268) (ANOVA, P < 0.05). In addition, there was a significant increase in the percentage of wild type alleles present when injecting Cas9 mRNA compared to Cas9 protein for each of the three loci (42.5% vs. 9.1%, 70.9% vs. 33.7% and 79.7% vs. 43.5%, for POLLED , H11 and ZFX, respectively; P < 0.05).
A total of 24 potential off-target sites were predicted across 11 bovine chromosomes (1, 4, 7, 8, 10, 12, 14, 18, 21, 27 and X) (Supplementary Table S7) for the three loci. The 24 predicted off-target sites were PCR amplified, barcoded and sequenced using an Illumina MiSeq sequencer for each of the 69 samples (Supplementary Table S3). HTStream processed reads were aligned to the 24 predicted sites with 10,399,614 reads mapped with coverage ranging from 1X to 112X per sample per site (Supplementary Table S7). Genetic variation was found throughout the samples in each of the 24 predicted off-target sites with almost no indels present at the predicted off-target cut site with the exception of two targets. A 12 bp deletion 26 bp downstream from a predicted off-target cut site for the H11 gRNA targeting chr1: 7454978 was detected in 69,434 reads (6.8%) (Supplemental Table S7). Additionally, 2397 reads (0.51%) contained a 3 bp deletion 11 bp downstream from the predicted off-target cut site of the ZFX gRNA target chr21: 28506796 (Supplemental Table S7).
Discussion
The ability to efficiently generate non-mosaic, homozygous founder animals is important for the production of genome edited livestock. The use of the CRISPR/Cas9 system has been reported across many livestock species3, but few reports have characterized its use in bovine embryos. In this study, using the CRISPR/Cas9 system, we identified gRNAs that resulted in high rates of mutation at target locations in two autosomes and the X chromosome in bovine embryos with an overall high efficiency (81–90%; Table 1). Significant differences were observed in gRNA efficiency within a locus, but not between loci. It has been demonstrated that microinjection itself does not have a significant impact on the development of bovine embryos4, but we found that microinjection of editing reagents in zygotes reduced development to the blastocyst stage compared to non-injected controls as the mutation efficiency of a given gRNA increased (Supplementary Table S2). However, no difference was observed in the number of embryos that reached the blastocyst stage when comparing embryos injected with Cas9 mRNA or protein (16.2% vs. 16.4%; Fig. 1a). This finding was important because we observed a significantly higher mutation rate in blastocysts when injecting Cas9 protein compared to Cas9 mRNA (84.2% and 68.5%, respectively; Fig. 1b). This difference is likely due to the immediate availability of the gRNA/Cas9 ribonucleoprotein (RNP) complex to induce mutation in the embryo. When Cas9 mRNA is injected, there is a delay in genome editing as Cas9 mRNA must be translated into protein before it can combine with the gRNA to induce a DSB5.
Mosaicism, the presence of more than two alleles in an individual, is a common problem in livestock genome editing6, with a high rate of embryos resulting in multiple alleles (Table 3). Studies utilizing transcription activator-like effector nucleases (TALENs) have demonstrated lower mosaicism rates than we observed here; however, the proportion of edited embryos tends to be lower as well7,8. A study employing a zinc finger nuclease (ZFN) in bovine embryos demonstrated both high embryo editing efficiency and mosaicism rates as compared to those found in TALEN edited embryos9. However, the prevalence of mosaicism was reduced when injecting embryos at 8hpi compared to 18hpi, before S-phase had occurred9. While we were able to induce mutations in embryos at a high rate, we also observed a high level of mosaicism when injecting 18hpi. Many studies of editing in livestock zygotes similarly report high levels of mosaicism when utilizing CRISPR/Cas9 (Table 3). Many of these studies characterized mosaicism by sequencing the PCR amplicon of the genomic regions flanking the gRNA target sequence and then decomposing the resulting chromatogram data with the TIDE bioinformatics package10. Although this approach is cost-effective and rapid, next generation sequencing of the PCR products allows for a more accurate characterization of the different alleles that are present in a mosaic individual, and their relative abundance11. However, this approach does present some concern with PacBio sequencing being highly error prone in regards to indels and SNPs. However, given the short sequences of the target amplicons, we were able to generate circular consensus sequencing (CCS) reads, increasing the confidence in the accuracy of the alleles that were being called12.
In bovine embryos, DNA replication occurs approximately 12–14 h after fertilization13. When injecting at 18hpi, as is often done when using traditional in-vitro fertilization (IVF) protocols, most zygotes would be expected to have completed DNA replication14 and there would likely be more than two copies of each chromosome, thus more opportunities for multiple genomic edits to occur, resulting in mosaicism. Additionally, following cytoplasmic injection, the gRNA/Cas9 ribonucleoprotein (RNP) complex needs time to enter the nucleus, find its target and cleave the DNA. Furthermore, if injecting Cas9 mRNA, translation to Cas9 protein must also occur, further delaying the editing process, thus resulting in a higher rate of mosaicism. It has been suggested that injection of the CRISPR/Cas9 RNP prior to the S-phase of DNA replication could reduce mosaicism1.
One recent study with bovine embryos reported low rates (~ 30%) of mosaicism when introducing Cas9 RNA or protein into early stage zygotes (0 or 10hpi) prior to the S-phase of DNA replication14. In that study, the authors were targeting two genes simultaneously via microinjection of two gRNAs into either matured oocytes before IVF or into zygotes at various time points post IVF. Allele identification was first made by Sanger sequencing of an amplicon of the targeted region, and then by clonal sequencing of 10 colonies derived from the PCR product per embryo. PCR and cloning-based approaches can identify that a range of alleles exist but cannot accurately quantitate the abundance of each allelic species. The authors went on to employ next generation sequencing on 20 embryos per group to characterize the alleles in non-mosaic embryos. The authors considered embryos that contained biallelic mutations resulting in frame-disrupting alleles to be non-mosaic, regardless of the number of alleles.
In the current study, we employed next generation sequencing to quantitate the abundance of each allele. The fact that we observed multiple alleles occurring in only a small percentage of reads (< 25%) in many samples analyzed in this study (Fig. 2) suggests that editing continued in some subset of cells after the first cleavage division. Further, we considered an embryo containing more than one population of genetically distinct cells to be mosaic irrespective of whether the edit resulted in a missense or nonsense mutation. It is important to determine if founder animals are mosaic because mosaicism complicates the interpretation of the effect of a given genome alteration6, and subsequent breeding of mosaic founder animals to achieve non-mosaic animals can take years15. Additionally, mosaics do not fit easily into the proposed regulatory framework for genome edited food animals16.
Along with the level of mosaicism, one of the concerns raised with the generation of genome edited animals is the potential for off-target mutation events. Typically, online prediction tools are used to calculate the likelihood of off-target sites17,18,19. The top predicted sites can then be PCR amplified and the presence of a mutation determined by either next generation sequencing, TA cloning followed by Sanger sequencing, or mismatch cleavage assays followed by Sanger sequencing20. In this study, we used the targeted approach using online predictive tools to identify off-target sites rather than a genome-wide approach. Off-target cleavage can occur in the genome with three to five base pair mismatches in the PAM-distal sequence17,21,22,23. Cas9 specificity is determined by the seed region, or the 8 to 11-nt PAM-proximal sequence, making it the most vital part of the gRNA sequence21,24. In our gRNA design, we excluded all gRNAs with less than three mismatches across the off-target sequence. We determined this threshold based on previous studies showing reduced Cas9 activity in regions with at least three mismatches25.
In the 69 samples analyzed, there were two potential off-target mutations detected. One of these (H11) was in a region that had known annotated wild type 12 bp deletions (rs876383581 and rs521367917) around the potential cut-site. Additionally, 0.51% of total reads contained a 3 bp deletion 11 bp downstream from the predicted off-target cut site for the ZFX gRNA target chr21: 28,506,796 (Supplemental Table S7). This predicted site does not have any annotated variation. It is important to note that although this off-target location had three mismatches to the gRNA sequence, all three of the mismatches were located outside the seed region (8–11 bp upstream of the PAM sequence). This guide was designed using off-target prediction software and the Btau 4.6.1 bovine reference genome26, which was the only Bos taurus reference genome available with the online tool at the time. When the off-target prediction software was re-run for the off-target analysis, the most recent reference genome available was UMD 3.1.126. Using the new reference genome, this locus on chromosome 21 was identified as having the requisite three mismatches, but there were no mismatches in the seed region, as specified by our guide design criteria. More recently, an improved reference bovine genome ARS-UCD1.2 was published27. Using the online tool with the updated reference genome resulted in the same predicted off-target sites as UMD 3.1.1.
One of the stated concerns with off-target mutation events is that if they occur in functional regions, such as coding sequences or regulatory regions, they could potentially be detrimental to the health or development of the resulting animal. Neither of these two off-target deletions were in a region of annotated function. As there were approximately 20 individual blastocysts included in these analyses, these deletions may also have been the result of naturally occurring polymorphic variation. A detailed sequence analysis of 2703 individuals from different breeds of cattle revealed a high level of genetic diversity including 84 million single-nucleotide polymorphisms (SNPs) and 2.5 million small insertion deletions28. Data like these are essential to put naturally occurring variation, like that seen at the H11 locus, in context. Various studies in humans29,30, monkeys31, and rodents32,33 suggest that the off-target frequency of Cas9-mediated mutagenesis does not differ from the de novo mutation rate.
Overall, we demonstrated efficient CRISPR/Cas9 genome editing across three different loci on three different chromosomes. We found that injecting zygotes with Cas9 protein results in a significantly higher mutation rate compared to Cas9 mRNA (82.2% vs 65.4%). In addition, zygotes injected with Cas9 protein displayed a significantly lower number of alleles compared to those injected with Cas9 mRNA (4.2 vs 5.2). Although off-target events did not appear to be an issue, the rate of mosaicism was still high, and further optimization needs to be done before this technique is feasible in a livestock production setting.
Materials and methods
Guide construction
Guides sequences were designed using the online tools sgRNA Scorer 2.034,35 and Cas-OFFinder36 and targeting the POLLED locus on chromosome 1, a safe harbor locus (H11) on chromosome 17 and in the 3′ UTR of the Zinc-finger X-linked (ZFX) gene (ZFX) on the X-chromosome. Guides were selected with no less than three mismatches in the guide sequence for off-target sites using the UMD3.1.1 bovine reference genome26, and at least one mismatch in the seed region (8–11 bp upstream of the PAM sequence). Oligonucleotides were ordered from Eurofins USA (Louisville, KY) for the top four guides for construction of the gRNA and were used for in vitro transcription using the AmpliScribe T7-Flash Transcription kit (Lucigen, Palo Alto, CA) and purified using the MEGAclear Transcription Clean-Up kit (Thermo Fisher, Chicago, IL) as described by Vilarino et al11. Cleavage efficiency was tested using an in vitro cleavage assay by combining 60 ng of PCR amplified product, 100 ng of gRNA, 150 ng of Cas9 protein (PNA Bio, Inc., Newbury Park, CA), 1 μL of 10X BSA, 1 μL of NEB Buffer 3.1 and water bringing the total volume to 10 μL in a 0.2 μL tube and incubating at 37 °C for 1 h. The incubated product was then run on a 2% agarose gel with 5 μL of Sybr Gold at 100 V for 1 h and visualized using a ChemiDoc-ItTS2 Imager (UVP, LLC, Upland, CA).
Embryo production
Bovine ovaries were collected from a local processing plant and transported to the laboratory at 35–37 °C in sterile saline. Cumulus-oocyte complexes (COCs) were aspirated from follicles and groups of 50 COCs were transferred to 4-well dishes containing 400 μL of maturation media37. COCs were incubated for 21–24 hr at 38.5 °C in a humidified 5% CO2 incubator. Approximately 25 oocytes per drop were fertilized in 60 μL drops of SOF-IVF37 with 1 × 106 sperm per mL and incubated for 18 hr at 38.5 °C in a humidified 5% CO2 incubator. Presumptive zygotes were denuded by light vortex in SOF-HEPES medium37 for 5 min. 25 zygotes per drop were incubated in 50 μL drops of KSOM culture media (Zenith Biotech, Glendale, CA, USA) at 38.5 °C in a humidified atmosphere of 5% CO2, 5% O2, and 90% N2 for 7–8 days.
Guide testing
Mutation rate for each guide was determined by laser-assisted cytoplasmic injection4 of in vitro fertilized embryos with 6pL of a solution containing 67 ng/μL of in vitro transcribed gRNA alongside 133 ng/μL of Cas9 mRNA or 167 ng/μL of Cas9 protein (PNA Bio, Inc., Newbury Park, CA) incubated at room temperature for 30 min prior to injection. Injected embryos were incubated for 7–8 days. Embryos that reached blastocyst stage were lysed in 10 μL of Epicenter DNA extraction buffer (Lucigen, Palo Alto, CA) using a SimpliAmp Thermal Cycler (Applied Biosystems, Foster City, California) at 65 °C for 6 min, 98 °C for 2 min and held at 4 °C. The target region was amplified by two rounds of the polymerase chain reaction (PCR) using primers developed using Primer3 (Supplementary Table S1)38,39. The first round of PCR was performed on a SimpliAmp Thermal Cycler (Applied Biosystems, Foster City, California) with 10 μL GoTAQ Green Master Mix (Promega Biosciences LLC, San Luis Obispo, CA), 0.4 μL of each primer at 10 mM and 9.2 μL of DNA in lysis buffer for 5 min at 95 °C, 35 cycles of 30 s at 95 °C, 30 s at anneal temp (Supplementary Table S1), and 30 s at 72 °C, followed by 5 min at 72 °C. The second round of PCR was run with 10 μL GoTAQ Green Master Mix (Promega Biosciences LLC, San Luis Obispo, CA), 4.2 μL of water, 0.4 μL of each primer at 10 mM and 5 μL of first round PCR for 3 min at 95 °C, 35 cycles of 30 s at 95 °C, 30 s at anneal temp (Supplementary Table S1), and 30 s at 72 °C, followed by 5 min at 72 °C. Products were visualized on a 1% agarose gel using a ChemiDoc-ItTS2 Imager (UVP, LLC, Upland, CA), purified using the QIAquick Gel Extraction Kit (Qiagen, Inc., Valencia, CA) and Sanger sequenced (GeneWiz, South Plainfield, NJ).
Allelic variation and off-target analysis
Embryos that reached the blastocyst stage were lysed and underwent whole-genome amplification using the Repli-G Mini kit (Qiagen, Inc., Valencia, CA). To determine presumptive off-target sites, guide sequences were mapped against the bosTau8 bovine reference genome using the online tool Cas-OFFinder36. A total of 24 off-target sites were predicted using the online tool: eight off-target sites for the POLLED gRNA, eleven off-target sites for the H11 gRNA and five off-target sites for the ZFX gRNA (Supplementary Table S7). Whole-genome amplified samples were used for PCR amplification of cut-sites and presumptive off-target sites using a dual round PCR approach described above to barcode each sample with a reduction from 35 to 5 cycles in the first round of PCR. Primers were designed to amplify each region using Primer338,39 with a 15 bp adapter sequence attached to the forward (AGATCTCTCGAGGTT) and reverse (GTAGTCGAATTCGTT) (Supplementary Information S1). The second round of PCR amplified off the adapters adding an independent barcode for each sample to identify reads for pooled sequencing (Supplementary Table S1).
PCR samples targeting the gRNA cut site underwent SMRTbell library preparation and were sequenced on a PacBio Sequel II sequencer by GENEWIZ, LLC (South Plainfield, NJ, USA). Consensus sequences were called, reads sorted by barcode and BAM converted to individual FASTQ files using SMRT Link v8.0.0.80529 (https://www.pacb.com/support/software-downloads/). Reads were aligned to each target site using BWA v0.7.16a40. SAM files were converted to BAM files, sorted and indexed using SAMtools v1.941. Number and types of alleles were determined for each sample using CrispRVariants v1.12.042.
Off-target PCR samples underwent library preparation using the Illumina TruSeq library kit and were sequenced (300 bp paired-end) on an Illumina MiSeq Next Generation Sequencer by the DNA Technologies and Expression Analysis Cores at the UC Davis Genome Center. Paired-end reads were processed and overlapped to form high quality single-end reads using HTStream Overlapper v1.1.0 (https://github.com/ibest/HTStream). Processed reads were aligned to each target site using BWA v0.7.16a40. SAM files were converted to BAM files, sorted and indexed using SAMtools v1.941. Insertions and deletions were called using CrispRVariants v1.12.042.
Statistical analysis
Comparison between blastocyst development and mutation rates were evaluated using a logistic regression model with Cas9 form and gRNA modeled as fixed effects. To analyze the level of mosaicism, an ANOVA test was used to determine significance between number of alleles per sample and percent wild type when injecting alongside Cas9 mRNA or protein. Samples with only wild type alleles were removed from analysis. Differences were considered significant when P < 0.05.
Data availability
Raw sequence reads from PacBio Sequel II and Illumina MiSeq sequencing are available in the NCBI Sequence Read Archive as BioProject PRJNA623431 and SRA accession number SRR11850065. Individual results for the blastocyst development and mutation rate from each replicate (~ 30 embryos) of control and microinjected embryos are available in Supplementary Table S8.
References
Mehravar, M., Shirazi, A., Nazari, M. & Banan, M. Mosaicism in CRISPR/Cas9-mediated genome editing. Dev. Biol. 445, 156–162. https://doi.org/10.1016/j.ydbio.2018.10.008 (2019).
Vilarino, M. et al. Mosaicism diminishes the value of pre-implantation embryo biopsies for detecting CRISPR/Cas9 induced mutations in sheep. Transgenic Res. 27, 525–537. https://doi.org/10.1007/s11248-018-0094-x (2018).
Mclean, Z., Oback, B. & Laible, G. Embryo-mediated genome editing for accelerated genetic improvement of livestock. Front. Agric. Sci. Eng. https://doi.org/10.15302/J-FASE-2019305 (2020).
Bogliotti, Y. S., Vilarino, M. & Ross, P. J. Laser-assisted cytoplasmic microinjection in Livestock Zygotes. J. Vis. Exp. https://doi.org/10.3791/54465 (2016).
Glass, Z., Lee, M., Li, Y. & Xu, Q. Engineering the delivery system for CRISPR-based genome editing. Trends Biotechnol. 36, 173–185. https://doi.org/10.1016/j.tibtech.2017.11.006 (2018).
Teboul, L., Murray, S. A. & Nolan, P. M. Phenotyping first-generation genome editing mutants: A new standard?. Mamm. Genome 28, 377–382. https://doi.org/10.1007/s00335-017-9711-x (2017).
Carlson, D. F. et al. Efficient TALEN-mediated gene knockout in livestock. Proc. Natl. Acad. Sci. USA 109, 17382–17387. https://doi.org/10.1073/pnas.1211446109 (2012).
Proudfoot, C. et al. Genome edited sheep and cattle. Transgenic Res. 24, 147–153. https://doi.org/10.1007/s11248-014-9832-x (2015).
Wei, J. et al. Efficient introgression of allelic variants by embryo-mediated editing of the bovine genome. Sci. Rep. 5, 11735. https://doi.org/10.1038/srep11735 (2015).
Brinkman, E. K., Chen, T., Amendola, M. & van Steensel, B. Easy quantitative assessment of genome editing by sequence trace decomposition. Nucleic Acids Res. 42, e168–e168 (2014).
Vilarino, M. et al. CRISPR/Cas9 microinjection in oocytes disables pancreas development in sheep. Sci. Rep. 7, 17472. https://doi.org/10.1038/s41598-017-17805-0 (2017).
Wenger, A. M. et al. Accurate circular consensus long-read sequencing improves variant detection and assembly of a human genome. Nat. Biotechnol. 37, 1155–1162. https://doi.org/10.1038/s41587-019-0217-9 (2019).
Comizzoli, P., Marquant Le Guienne, B., Heyman, Y. & Renard, J. P. Onset of the first S-phase is determined by a paternal effect during the G1-phase in bovine zygotes. Biol. Reprod. 62, 1677–1684. https://doi.org/10.1095/biolreprod62.6.1677 (2000).
Lamas-Toranzo, I. et al. Strategies to reduce genetic mosaicism following CRISPR-mediated genome edition in bovine embryos. Sci. Rep. 9, 14900. https://doi.org/10.1038/s41598-019-51366-8
Bishop, T. F. & Van Eenennaam, A. L. Genome editing approaches to augment livestock breeding programs. J. Exp. Biol. https://doi.org/10.1242/jeb.207159 (2020).
Van Eenennaam, A. L., Wells, K. D., Murray, J. D. & Proposed, U. S. regulation of gene-edited food animals is not fit for purpose. NPJ Sci. Food 3, 3. https://doi.org/10.1038/s41538-019-0035-y (2019).
Hsu, P. D. et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nat. Biotechnol. 31, 827–832. https://doi.org/10.1038/nbt.2647 (2013).
Doench, J. G. et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nat. Biotechnol. 34, 184–191. https://doi.org/10.1038/nbt.3437 (2016).
Haeussler, M. et al. Evaluation of off-target and on-target scoring algorithms and integration into the guide RNA selection tool CRISPOR. Genome Biol. 17, 148. https://doi.org/10.1186/s13059-016-1012-2 (2016).
Jamal, M. et al. Keeping CRISPR/Cas on-Target. Curr. Issues Mol. Biol. 20, 1–12. https://doi.org/10.21775/cimb.020.001 (2016).
Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819–823. https://doi.org/10.1126/science.1231143 (2013).
Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–826. https://doi.org/10.1126/science.1232033 (2013).
Pattanayak, V. et al. High-throughput profiling of off-target DNA cleavage reveals RNA-programmed Cas9 nuclease specificity. Nat. Biotechnol. 31, 839–843. https://doi.org/10.1038/nbt.2673 (2013).
Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816–821 (2012).
Anderson, E. M. et al. Systematic analysis of CRISPR-Cas9 mismatch tolerance reveals low levels of off-target activity. J. Biotechnol. 211, 56–65. https://doi.org/10.1016/j.jbiotec.2015.06.427 (2015).
Elsik, C. G. et al. The genome sequence of taurine cattle: A window to ruminant biology and evolution. Science 324, 522–528. https://doi.org/10.1126/science.1169588 (2009).
Rosen, B. D. et al. De novo assembly of the cattle reference genome with single-molecule sequencing. Gigascience https://doi.org/10.1093/gigascience/giaa021 (2020).
Hayes, B. J. & Daetwyler, H. D. 1000 bull genomes project to map simple and complex genetic traits in cattle: Applications and outcomes. Annu. Rev. Anim. Biosci. 7, 89–102. https://doi.org/10.1146/annurev-animal-020518-115024 (2019).
Veres, A. et al. Low incidence of off-target mutations in individual CRISPR-Cas9 and TALEN targeted human stem cell clones detected by whole-genome sequencing. Cell Stem Cell 15, 27–30. https://doi.org/10.1016/j.stem.2014.04.020 (2014).
Yang, L. et al. Targeted and genome-wide sequencing reveal single nucleotide variations impacting specificity of Cas9 in human stem cells. Nat. Commun. 5, 5507. https://doi.org/10.1038/ncomms6507 (2014).
Luo, X. et al. Trio deep-sequencing does not reveal unexpected off-target and on-target mutations in Cas9-edited rhesus monkeys. Nat. Commun. 10, 5525. https://doi.org/10.1038/s41467-019-13481-y (2019).
Iyer, V. et al. Off-target mutations are rare in Cas9-modified mice. Nat. Methods 12, 479. https://doi.org/10.1038/nmeth.3408 (2015).
Willi, R., Smith, H. E., Wang, C., Liu, C. & Hennighausen, L. Mutation frequency is not increased in CRISPR–Cas9-edited mice. Nat. Methods 15, 756–758 (2018).
Aach, J., Mali, P. & Church, G. M. CasFinder: Flexible algorithm for identifying specific Cas9 targets in genomes. bioRxiv. https://doi.org/10.1101/005074 (2014).
Chari, R., Yeo, N. C., Chavez, A. & Church, G. M. sgRNA Scorer 2.0: A species-independent model to predict CRISPR/Cas9 activity. ACS Synth. Biol. 6, 902–904. https://doi.org/10.1021/acssynbio.6b00343 (2017).
Bae, S., Park, J. & Kim, J. S. Cas-OFFinder: A fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases. Bioinformatics 30, 1473–1475. https://doi.org/10.1093/bioinformatics/btu048 (2014).
Bakhtari, A. & Ross, P. J. DPPA3 prevents cytosine hydroxymethylation of the maternal pronucleus and is required for normal development in bovine embryos. Epigenetics 9, 1271–1279. https://doi.org/10.4161/epi.32087 (2014).
Koressaar, T. & Remm, M. Enhancements and modifications of primer design program Primer3. Bioinformatics 23, 1289–1291. https://doi.org/10.1093/bioinformatics/btm091 (2007).
Untergasser, A. et al. Primer3—new capabilities and interfaces. Nucleic Acids Res. 40, e115. https://doi.org/10.1093/nar/gks596 (2012).
Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv:1303.3997v (2013).
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079. https://doi.org/10.1093/bioinformatics/btp352 (2009).
Lindsay, H. et al. CrispRVariants charts the mutation spectrum of genome engineering experiments. Nat. Biotechnol. 34, 701–702. https://doi.org/10.1038/nbt.3628 (2016).
Petersen, B. et al. Efficient production of biallelic GGTA1 knockout pigs by cytoplasmic microinjection of CRISPR/Cas9 into zygotes. Xenotransplantation 23, 338–346. https://doi.org/10.1111/xen.12258 (2016).
Zhang, X. et al. Disruption of the sheep BMPR-IB gene by CRISPR/Cas9 in in vitro-produced embryos. Theriogenology 91, 163-172 e162. https://doi.org/10.1016/j.theriogenology.2016.10.025 (2017).
Crispo, M. et al. Efficient generation of Myostatin knock-out sheep using CRISPR/Cas9 Technology and microinjection into zygotes. PLoS ONE 10, e0136690. https://doi.org/10.1371/journal.pone.0136690 (2015).
Tao, L. et al. Efficient biallelic mutation in porcine parthenotes using a CRISPR-Cas9 system. Biochem. Biophys. Res. Commun. 476, 225–229 (2016).
Wang, Y. et al. Efficient generation of gene-modified pigs via injection of zygote with Cas9/sgRNA. Sci. Rep. 5, 8256. https://doi.org/10.1038/srep08256 (2015).
Daigneault, B. W., Rajput, S., Smith, G. W. & Ross, P. J. Embryonic POU5F1 is required for expanded bovine blastocyst formation. Sci. Rep. 8, 7753. https://doi.org/10.1038/s41598-018-25964-x (2018).
Namula, Z. et al. Genome mutation after the introduction of the gene editing by electroporation of Cas9 protein (GEEP) system into bovine putative zygotes. Vitro Cell Dev. Biol. Anim. 55, 598–603. https://doi.org/10.1007/s11626-019-00385-w (2019).
Tanihara, F. et al. Effects of concentration of CRISPR/Cas9 components on genetic mosaicism in cytoplasmic microinjected porcine embryos. J. Reprod. Dev. 65, 209–214 (2019).
Tanihara, F. et al. Generation of a TP53-modified porcine cancer model by CRISPR/Cas9-mediated gene modification in porcine zygotes via electroporation. PLoS ONE 13, e0206360. https://doi.org/10.1371/journal.pone.0206360 (2018).
Acknowledgements
The authors would sincerely like to thank Ashley Young and Rebecca Ozeran for diligent collection and transport of ovaries from the processing plant to UC Davis. The library preparation and sequencing for off-target analysis were carried out at the DNA Technologies and Expression Analysis Cores at the UC Davis Genome Center, supported by NIH Shared Instrumentation Grant 1S10OD010786-01. The authors would like to thank the University of California Davis Bioinformatics Core Facility for assistance with data analysis. This project was supported by Biotechnology Risk Assessment Grant Program competitive Grant No. 2015-33522-24106 from the U.S. Department of Agriculture, the Academic Federation Innovation Development Award at UC Davis, the Russell L. Rustici Rangeland and Cattle Research Endowment in the College of Agricultural and Environmental Science at UC Davis, the California Agricultural Experiment Station of the University of California, Davis, the Henry A. Jastro Research Fellowship in the College of Agricultural and Environmental Science at UC Davis and the National Institute for Food and Agriculture National Needs Graduate and Postgraduate Fellowship No. 2017-38420-26790 from the U.S. Department of Agriculture.
Author information
Authors and Affiliations
Contributions
S.L.H. and J.R.O. performed the experiments with additional input from P.J.R., A.L.V. and J.D.M.; S.L.H., J.R.O., J.C.L. and A.E.Y. participated in sample processing and data analysis. S.L.H., J.R.O., A.L.V. and J.D.M. wrote the manuscript with suggestions from all the co-authors. All authors read and approved the final version.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Hennig, S.L., Owen, J.R., Lin, J.C. et al. Evaluation of mutation rates, mosaicism and off target mutations when injecting Cas9 mRNA or protein for genome editing of bovine embryos. Sci Rep 10, 22309 (2020). https://doi.org/10.1038/s41598-020-78264-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-020-78264-8
- Springer Nature Limited
This article is cited by
-
Gene editing and therapy in acquired and inherited cardiovascular disorders
The Nucleus (2024)
-
Genome centric engineering using ZFNs, TALENs and CRISPR-Cas9 systems for trait improvement and disease control in Animals
Veterinary Research Communications (2023)
-
Technical considerations towards commercialization of porcine respiratory and reproductive syndrome (PRRS) virus resistant pigs
CABI Agriculture and Bioscience (2022)
-
Synergistic power of genomic selection, assisted reproductive technologies, and gene editing to drive genetic improvement of cattle
CABI Agriculture and Bioscience (2022)
-
A deletion at the polled PC locus alone is not sufficient to cause a polled phenotype in cattle
Scientific Reports (2022)