Transgenic Research

, Volume 26, Issue 6, pp 715–726 | Cite as

Genome editing in livestock: Are we ready for a revolution in animal breeding industry?

Review

Abstract

Genome editing is a powerful technology that can efficiently alter the genome of organisms to achieve targeted modification of endogenous genes and targeted integration of exogenous genes. Current genome-editing tools mainly include ZFN, TALEN and CRISPR/Cas9, which have been successfully applied to all species tested including zebrafish, humans, mice, rats, monkeys, pigs, cattle, sheep, goats and others. The application of genome editing has quickly swept through the entire biomedical field, including livestock breeding. Traditional livestock breeding is associated with rate limiting issues such as long breeding cycle and limitations of genetic resources. Genome editing tools offer solutions to these problems at affordable costs. Generation of gene-edited livestock with improved traits has proven feasible and valuable. For example, the CD163 gene-edited pig is resistant to porcine reproductive and respiratory syndrome (PRRS, also referred to as “blue ear disease”), and a SP110 gene knock-in cow less susceptible to tuberculosis. Given the high efficiency and low cost of genome editing tools, particularly CRISPR/Cas9, it is foreseeable that a significant number of genome edited livestock animals will be produced in the near future; hence it is imperative to comprehensively evaluate the pros and cons they will bring to the livestock breeding industry. Only with these considerations in mind, we will be able to fully take the advantage of the genome editing era in livestock breeding.

Keywords

Genome editing Livestock Breeding Industry 

Genome editing is a revolutionary technology

The development of genome editing technology

Genome editing technology can precisely modify the genome of an organism. The current nuclease-based genome-editing tools include zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) and clustered regularly interspaced short palindromic repeats/associated nuclease Cas9 (CRISPR/Cas9) (Doyon et al. 2008; Christian et al. 2010; Ran et al. 2013). All of these nucleases work by creating site-specific DNA double-strand breaks (DSBs) at desired location in the genome, and then the cell’s repair system “patches” up the DSBs either through the non-homologous end joining (NHEJ) or homology directed repair (HDR) pathway. In cells, NHEJ is a major DNA repair pathway which ligates the break ends directly without the need for a template. During the NHEJ repair process nucleotide insertions or deletions happen thus causing indels and repair error related frameshift. As a result, gene knock-out can be efficiently made by nuclease-assisted NHEJ. In contrast, HDR happens at a lower frequency compared to NHEJ in cells, but repair is done through homologous recombination between a donor DNA template and the target genomic locus, resulting in a precise insertion of the donor DNA. Nuclease-assisted HDR is therefore used to generate targeted gene knock-in or allele replacement. These nuclease-based genome editing techniques, particularly CRISPR/Cas9 due to its ease of use, has sparked a revolution and wide application in the entire field of biology.

In 2007, Sangamo Biosciences Inc. made modifications to ZFNs which is consisted of zinc finger DNA-binding domain and FokI DNA cleavage domain, that reduced its off-target rate (Miller et al. 2007). Subsequently, Sangamo successfully performed ZFN-mediated gene knockout in zebrafish (Doyon et al. 2008). The editing efficiency reached 20%, which was far beyond (> 200 fold) traditional targeting efficiency (~ < 0.1%). Genome editing technology rapidly gained extensive attention and promptly spread into research in zebrafish, humans, mice, swine, cattle and other species (Doyon et al. 2008; Hockemeyer et al. 2009; Li et al. 2011; Yang et al. 2011; Liu et al. 2013; Whyte et al. 2011). However, this technology is very difficult to adopt in regular labs due to high costs, technical difficulties, and the long time required for ZFN preparation; and thus its application has been limited. In 2010, the Voytas lab used naturally existing transcription activator-like effector (TALE) repeat arrays fused with FokI (a nuclease isolated from the bacterium Flavobacterium okeanokoites) to construct the artificial TALENs and verified that TALENs performed targeted cleavage of target DNA (Christian et al. 2010). Compared to ZFNs, the preparation procedure for TALENs was simpler, and moreover, TALENs have more target selection options. TALENs enabled the use of genome editing technology in regular research labs, thereby promoting a new front of the genome editing revolution. However, only about 2 years later in February 2013, research teams at Massachusetts Institute of Technology and Harvard University used the CRISPR enzymes from Streptococcus pyogenes and Streptococcus thermophiles and synthesized RNA to perform genome editing in mouse and human cells (Cong et al. 2013; Mali et al. 2013). For the first time, these researchers demonstrated the application of CRISPR/Cas9 system in genome editing in mammalian cells. Compared to the ZFN and TALEN technologies, the construction and use of the CRISPR/Cas9 gene-editing system, like Cas9/gRNA Ribonucleoproteins (RNPs), was much simpler (Table 1). The entire system includes only the Cas9 protein and a single guide RNA (sgRNA). The vector construction can be completed in a week, and its gene editing efficiency is high (up to 100% in certain cases). Further, CRISPR/Cas9 can be multiplexed, i.e. multiple genetic loci can be targeted at the same time. Moreover, it has been proven to work in all species tested so far. The CRISPR/Cas9 system is robust, efficient and cost-effective, and it has pushed genome editing technology to a new level. Today, almost every lab can conduct genome editing experiments, covering any species of interest.
Table 1

Comparison of ZFN, TALEN and CRISPR/Cas9

 

ZFN

TALEN

CRISPR/Cas9

DNA binding mode

Protein-DNA

Protein-DNA

RNA-DNA

Structural components

ZFP-FokI

TALE-FokI

Cas9, sgRNA

Efficiency

Comparable

Comparable

Comparable

Off-target rates

Comparable

Comparable

Comparable

Multiple genetic loci edited

Difficult

Difficult

Yes

Cost

High

Middle

Low

Time for vector construction

Months

2–4 weeks

1–2 weeks

An overwhelming trend in genome editing in livestock

Genome editing technologies have inevitably transformed transgenic animal production in livestock industry. In the past, gene targeting, including gene knockout (KO) and knock-in have been extremely challenging, if not impossible, in livestock breeds, largely due to the lack of germline transmittable embryonic stem cells.

One important consideration in improving livestock production and/or quality through genetic manipulation is the requirement of precise genome editing at multiple loci. Growth and development of animals are influenced by multiple genes that function throughout the entire body, and different production traits are often controlled by different and/or multiple genes. Different mutation sites or mutation types in the same gene may also have a large impact on the production performance. Therefore, significant enhancement of multiple livestock production traits requires precise genome editing at multiple genomic sites. Improving the efficiency in multiplexed genome editing will promote the efficiency of livestock breeding and bring livestock breeding to a new height.

Genome-editing technologies have rapidly attracted the attention of animal experts and been applied in swine, cattle, sheep and goats.

Swine

In Carlson et al. (2012) pioneered the use of TALENs to edit the genes in swine and cattle, but they did not produce live animals. In Lillico et al. (2013) edited the RELA gene in pigs by injecting the TALENs to zygotes directly and obtain live-born pig successfully. In Lee et al. (2014) used TALEN technology to edit the porcine recombination-activating gene 2 (RAG2) and generated immunodeficient swine, which can be used in medical studies. Li et al. (2014) identified the Rosa26 locus in swine and used TALENs to successfully generate EGFP transgenic swine by Cre recombinase-mediated ROSA26-targeting knock-in. Using another pair of loxP sites and recombinase-mediated gene exchange, these authors successfully generated swine expressing the red fluorescent protein tdTomato by replacing the EGFP gene with the tdTomato gene. Using this model, researchers can insert any gene into the Rosa26 locus through recombinase-mediated gene exchange. Researchers at the Institute of Animal Sciences, Chinese Academy of Agricultural Sciences, employed TALEN technology to edit the porcine MSTN gene and introduce precise mutations that resulted in a genotype similar to the Belgian blue cow. Yao et al. (2014) used TALENs to edit the alpha-1,3-galactosyltransferase (GGTA1), Parkin and DJ-1 genes in porcine fibroblast genome without using any selection marker gene during the editing process. Subsequently, they used the fibroblasts as nucleus donors to generate gene-edited swine by somatic cell nuclear transfer (SCNT) technology and provided good animal models for biomedical studies. In Sato et al. (2014) used the CRISPR/Cas9 system to knockout the porcine GGTA1 gene. Hai (2014) and Whitworth et al. (2014) used the CRISPR/Cas9 system and direct embryo injection to successfully generate gene-edited swine. In Whitworth’s study, they created CD163 or CD1D knock-out pigs using somatic cell nuclear transfer (SCNT) and embryo injection method via CRISPR/Cas9 (Whitworth et al. 2014). In Feng et al. (2015) used the CRISPR/Cas9 system to edit the genome and generated induced pluripotent stem (iPS) cells. Our group developed a targeted integration system at the porcine H11 genomic safe harbor locus using CRISPR/Cas9 (Ruan et al. 2015). This system can be used to insert any gene into the H11 site with high efficiency with high-level gene expression. In Wang et al. (2015a) performed a precise repair of the porcine MSTN gene using the CRISPR/Cas9 technology. Qian et al. (2015) used ZFN technology to edit the second exon of the myostatin (MSTN) gene of a domestic swine breeds in China (Meishan swine) and generated gene-edited swine (Cai et al. 2017). Because the MSTN gene can suppress muscle growth, the muscle growth of this edited swine is expected to be dramatically elevated after the MSTN modification. Mu et al. used ZFN technology to edit the porcine Rosa26 locus. The exogenous Fat1 gene was successfully inserted into this locus to generate gene-edited swine with a higher unsaturated fatty acid content and healthier meat (Tao et al. 2016). In Lillico et al. (2016) produced the RELA gene edited (HDR) pigs using ZFNs successful. The ability to efficiently achieve interspecies allele introgression in one generation opens unprecedented opportunities for agriculture and basic research.

Cattle

Yu et al. performed targeted gene mutations in cattle using the ZFN technology and generated gene-edited cattle with a ZFN-induced β-lactoglobulin gene mutation (Yu et al. 2011). In Liu et al. (2013) inserted the human lysozyme (Hlyz) gene into the cattle β-casein locus using ZFN, so that the mammary gland of the transgenic cattle can secrete human lysozymes, and the milk of transgenic cattle has the ability to kill Staphylococcus aureus (Liu et al. 2014). In Wu et al. (2015) used TALEN technology to insert the SP110 gene into the cattle genome and generate transgenic cattle that are less susceptible to tuberculosis. Wei et al. edited the beta-lactoglobulin (LGB) gene in cattle embryos in high efficiency using the ZFNs and TALENs method (Wei et al. 2015). Proudfoot et al. (2015) obtained the MSTN gene knock out cattle successful using TALENs. In Carlson et al. (2016) used the TALENs to knock out the Holstein cattle POLLED allele, and obtain the hornless dairy cattle. This study has the potential to improve the welfare of millions of cattle annually.

Goats and sheep

In Ni et al. (2014) used the CRISPR/Cas9 system to knockout the MSTN gene in goats, thereby increasing the amount of muscle. In Wang et al. (2015b) co-injected the Cas9 mRNA and sgRNAs targeting two functional genes (MSTN and FGF5) into a single-cell stage embryo to successfully generate a transgenic goat with one or two modified genes. The targeting efficiency of MSTN and FGF5 was up to 60% in fibroblasts under in vitro culture conditions, whereas the modification efficiencies of MSTN and FGF5 were 15 and 21%, respectively, in the 98 tested goats. The efficiency of the dual-genic modification was 10%. Proudfoot et al. (2015) made the sheep MSTN gene knock out successful using TALEN method by direct embryo injection.

Current challenges in livestock breeding

Extinction of many existing breeds and loss of breeds diversity

According to a survey of 6500 livestock and poultry breeds in more than 170 countries conducted by the United Nations Food and Agriculture Organization, more than 1000 breeds have become extinct worldwide in the twentieth century, and this situation continues to elevate. Currently, 6379 breeds of domesticated animals are recorded in the global agricultural animal genetic resource database, and data can be found for 4183 breeds. Of these 4183 breeds, 740 breeds are extinct, and 1335 breeds have a “high risk” status. Therefore, a total of ~ 2000 livestock and poultry breeds will be lost if no actions are taken in the next 20 years. The major reason is that people eliminate or modify local breeds that do not have competitive advantages in production performance and economic profits. Currently, the proportion of fine breeds and improved breeds in China is over 95% for swine, 60% for cattle, 80% for sheep and goats, and 95% for poultry. At the same time, no adequate measures exist to appropriately protect fine regional breeds, causing a large number of local breeds to disappear in a short period of time. From a global perspective, over the last 100 years, 450 local breeds out of more than 1000 cattle breeds have disappeared worldwide due to the widespread introduction of breeds to hybridize with the local breeds. The reduction of genetic resources limits the breeding of good, novel breeds and threatens the sustainable development of agriculture. Local livestock and poultry resources are a requirement for the continuous development of agriculture and animal husbandry. The extinction of a breed does not merely mean the disappearance of the breed; the more important issue is the vanishing of the genes the breed carries if the lost genetic resource cannot be regenerated. The rapid decline in the number of local livestock and poultry breeds and the increasingly narrow genetic base has to a great extent limited the breeding of new superior varieties. Under continuous directional selection, the difference between varieties is becoming smaller, which may lead to the selection limit or even collapse breeds variety. Therefore, the threat of breeding on “shrinking” genetic resources cannot be ignored.

Animal breeding has occurred in four stages, resulting in a long breeding cycle and slow effects

The history of animal breeding stretches for several thousands of years. Archaeological excavations found that humans started to tame dogs, sheep and goats during the Paleolithic period, and the wild boar was domesticated in the early Neolithic period (approximately 8000–10,000 years ago). Swine bones excavated from Jarmo (an archaeological site in Iraqi Kurdistan) dated back approximately 8500 years and are the oldest record of domesticated swine worldwide. And modern livestock breeding essentially began with Robert Bakewell and his use of selective breeding since 1700s (Wood 1973). During this period, the genetic improvement of livestock and poultry was mostly a phenotypic aim to “keep every good outcome” using phenotypic selection techniques. Subsequently, after 1961, the emergence of computer techniques promoted the development of animal breeding. The popularization and application of computers focused animal breeding on major economic traits and changed from phenotypic selection to breeding value selection, which greatly improved the accuracy of selection and allowed animal breeding techniques to enter the second phase (Ollivier and Sellier 1982). Molecular breeding is the third phase of animal breeding. This practice relies on molecular genetics and molecular quantitative genetics theories and uses recombinant DNA technology to improve livestock and poultry varieties (Davoli and Braglia 2007; Chen et al. 2009). With the development of genomic sequencing technology, the fourth phase of breeding technology (genomic breeding) appeared on the scene. Genomic breeding uses DNA marker techniques to directly select or improve important production trait loci and can simultaneously take into account multiple production trait loci or even the entire genome of the individual animal (Jonas and de Koning 2015; Duijvesteijn et al. 2010; Yang et al. 2014; Meuwissen et al. 2001). Therefore, this practice can be called a genome scan for selection. Genome-wide association studies (GWASs) are used to detect genetic markers within the scope of the entire genome based on an association analysis. This strategy is an effective method to study complex disease and trait heritability. A large number of SNP (single nucleotide polymorphism) markers in livestock and poultry genomes have been discovered with the development of high-throughput sequencing technology and completion of the genomic sequencing of swine, cattle, sheep, goats and other major livestock and poultry. Illumina and other companies have launched high-density chips for cattle, swine, chickens and other livestock and poultry that have proved convenient for the genomic breeding of livestock and poultry. GWAS makes it possible to screen mutation sites associated with target traits in the genome. Animal breeders have identified a large number of candidate SNP loci and genes associated with economic livestock and poultry traits through GWAS studies. Although traditional breeding and genomic breeding have played important roles, the corresponding breeding processes are inefficient due to their long cycles.

Traditional breeding methods have encountered a bottleneck

Traditional breeding generally refers to selective breeding and cross-breeding. A series of superior livestock varieties (e.g., Large White swine, Landrace swine, Duroc swine, Holstein cattle, and Simmental cattle) have been obtained using this method. The various production performances have been dramatically improved compared with their original counterparts.

However, the improvement of each superior trait has to be obtained from a large group of foundation stock, and subsequently, many generations of breeding are needed to stabilize the trait. It takes a very long time to achieve a good trait, and the breeding cost is very high. Sometimes it can take more than a decade to slightly improve the percentage of lean meat. Some of today’s superior domesticated livestock varieties went through over one hundred years of breeding (e.g., the Large White swine were dated back to the eighteenth century in England). Although traditional breeding made some progress in the early days, it is difficult to make improvement in a short time when various production traits reach a certain degree and inappropriate selection and management in applying this method in a later stage can easily lead to degeneration of the breeds, causing production traits to deteriorate. Traditional breeding methods are obviously encountering a bottleneck.

Genomic breeding has limitations

Molecular breeding helps to shorten the breeding cycles associated with traditional breeding, but still faces many challenges.
  1. A.

    The breeding cycle is long

     
Molecular marker selection can help to speed up the breeding process by eliminating the need of effects from environmental conditions and developmental stages. However, the predicament of a long breeding cycle still exists. Many traits are controlled by multiple genes, and multiple molecular markers need to be considered for breeding selection. This process requires a large investment in time and money to obtain individuals with multiple homozygous genetic loci during breeding. Therefore, the molecular breeding improvement cycle is still very long.
  1. B.

    Limited by genetic resources

     
The principle behind genomic breeding is to select and improve genetic loci of important production traits using the DNA marker technique. This requires that the traits to be selected already exist in the breeds. It cannot create de novo traits that are not present in the starting breeds or genetic resource. For example, the MSTN gene has natural mutants in cattle, dogs, and humans, and individuals with mutations in this gene will produce a large amount of muscle (Schuelke et al. 2004; Nishi et al. 2002; McPherron and Lee 1997). In contrast, swine do not have individuals with mutant MSTN genes, therefore genomic breeding cannot be used to obtain swine with mutant MSTN genes. The CD163 molecule, which is a scavenger receptor protein that participates in a variety of immune activities, is closely associated with porcine reproductive and respiratory syndrome (PRRS). Mutation of the CD163 gene can elevate swine’s resistance to such diseases. However, although there are many natural SNPs/mutations in the CD163 gene, none of these lead to a PRRSV resistant phenotype.; therefore, genomic breeding method cannot generate swine harboring a mutant CD163 gene to be resistant to reproductive and respiratory syndrome.

Impact of genome editing in livestock on animal husbandry

Genome editing technology emerged when genetic breeding of animals was at a bottleneck stage. This technology can better solve the abovementioned problems encountered in breeding and compensate for the drawbacks in traditional breeding and genomic breeding.

Genome editing greatly reduces target trait breeding duration

Genome editing technology can rapidly edit key genes that affect target traits and obtain the required genotype through a single generation of editing, thereby greatly reducing breeding time. It would not be trivial to develop a viable line of livestock based on multiple genome edits. Many edited founders would need to be produced to maintain background genetic variation and avoid excessive inbreeding levels.

Creating genetic diversity

We can improve genetic resources with defects by precisely modifying key genes that cause defects while keeping other superior traits untouched. Using the information on key SNP loci obtained from molecular breeding screening, we can perform precise single base modifications in animal genomes using genome editing technology and directly edit these loci to obtain a genotype with desired traits. Thus, we can generate a novel germplasm when other options fail.

Improving traits that do not naturally exist in the species

Natural genetic resource types are limited, and not all favorable genetic traits are available. Genome editing technology allows the generation of virtually all mutation types. As a result, animals can gain traits that do not exist in natural genetic sources. Gene-edited and new varieties of livestock (e.g., low-phosphorus discharge and environmentally friendly gene-edited swine) will provide more and better new materials for livestock breeding.

Product quality will also be greatly improved. Using genetic modification, scientists have already obtained swine carrying synthesized fatty acid desaturase-1 (sfat-1), peroxisome proliferator-activated receptor γ (PPARγ), and PPARγ coactivator 1α (PGC1α). The ω-3 polyunsaturated fatty acid content in the muscle tissues of sfat-1 transgenic swine is much higher than the content in the control swine (Lai et al. 2006). Moreover, the meat quality trait of the PPARγ and PGC1α transgenic swine is improved, the data have not been published. By maintaining a high lean meat rate, the intramuscular fat content of PPARγ transgenic swine is increased, so is the tenderness. Similarly, in PGC1α transgenic swine, the muscle L value, drip loss and water loss rate decreased dramatically, whereas the intramuscular fat, muscle moisture, water holding capacity, tenderness and other indicators have all improved. The improved meat quality will potentially benefit human health.

Disease resistant varieties in livestock will emerge. With technology development, novel livestock varieties carrying disease-resistant genes have been developed with good outcomes. The University of Missouri used precise CRISPR/Cas9 technology to silence the production of the CD163 protein after piglets were born to prevent the virus from spreading inside their bodies (Whitworth et al. 2014, 2016). And scientists obtained another new bi-allelic mutation of CD163 using genome editing technology; CRISPR/Cas9 was applied to precisely modify the endogenous porcine CD163 gene to remove its capacity to mediate PRRSV invasion. Two CRISPR/Cas9 targeting vectors were designed and constructed to target the 7th exon of the porcine CD163 gene, leading to an expressed CD163 protein lacking domain 5 and complete resistance to PRRSV (Burkard et al. 2017). This study has a significant impact on anti-PRRS research and economic potential. In Lillico et al. (2016) introduced the alleles from the warthog which is associated with resilience to African Swine Fever into pigs using ZFNs. This transfer of warthog alleles into pigs is impossible by natural breeding. Using TALEN-mediated gene targeting technology, eighteen anti-tuberculosis cloned cattle were generated by inserting an expression vector containing the promoter of macrophage scavenger receptor 1 (MSR1) driving the intracellular pathogen resistance 1 (Ipr1) gene at a specific genomic locus (Wu et al. 2015). After infection with Mycobacterium bovis, the apoptotic rate, necrotic rate and bacterial load of mononuclear cells from the peripheral blood of the transgenic cloned cattle and non-transgenic cloned cattle were examined. The blood mononuclear cells from the Ipr1 gene-targeted cattle exhibited significant resistance to Mycobacterium bovis, with an increase in disease resistance of over 50%. Similar work was done in other species. Transgenic sheep containing the TLR4 gene may increase anti-brucellosis ability (Li et al. 2016). Lysozyme secretion by anti-mastitis TLR2 transgenic sheep can be increased, which effectively removed pathogens (Deng et al. 2012). With further development of genome-editing technology, more disease-resistant livestock will emerge in the future.

Potential risks of genome edited livestock products

Although genome editing technology can solve many problems encountered in conventional livestock breeding, cautions should be taken, as we should for any new technology, for their applications in the twentyfirst century livestock industry.

First, it is inevitable that hundreds, if not thousands, of gene edited livestock animals will be produced in the near future. Simply banning gene edited animals are not only short-sighted, but also hard to implement. The number alone adds burden to the regulatory agencies. Moreover, unlike conventional transgenic animals which normally contain “foreign” sequences as compared with the host genome. In gene edited animals, the changes are often footprint free. This will require substantial technology and resource input to ensure the proper registration and tracking of the creation, reproduction, and consumption of these animals and their products. Scientists, industry stakeholders, and government agencies are obligated to work together to educate the general public as well as people who are directly involved in the production and distribution of genome edited animals and their products, on both benefits and risks, to establish industry standards, and develop regulatory laws and policies.

Off-target editing is a concern. The current genome editing technologies (ZFNs, TALENs and CRISPR/Cas9) all have potential to induce off-target mutations in the genome. Although these mutations may not have any impact on the health of individual animals, they still carry a potential risk and can create obstacles for the future promotion of genome editing. In most cases, off-target mutations will either be selected against, if they are deleterious, or will likely disappear by drift if they are neutral.

Evaluation principles for genome-edited livestock

The policies towards genome editing products in different countries

USA

The US Department of Agriculture (USDA) informed that GM (genetically modified) products edited by genome editing tools with no exogenous DNA insertion would fall outside the regulations. In 2010, USDA informed that corn edited by ZFN technology from Dow Agro Sciences will not be regulated. In 2012, a new herbicide-resistant transgenic rape using genome editing technology was confirmed by USDA (Waltz 2012). In 2016, the USDA informed that it would not regulate the mushroom that has been genetically modified with the gene-editing tool CRISPR–Cas9 (Waltz 2016a). But before products come to consumer’s table, they should be confirmed by the US Environmental Protection Agency (EPA) and US Food and Drug Administration (FDA). GM livestock are regulated as drugs by the FDA and the FDA has stated that they will regulate genome edited livestock.

EU

The EU plays it cautiously to GM food. The attitude is mainly influenced by the social factors such as science and technology, politics, economy and religion ethics. In 2012, the European Food Safety Authority (EFSA) GMO Panel informed that although the third kind genome editing products is different from traditional transgenesis or cisgenesis for the DNA insertion into a predefined region of the genome, GM food should still be assessed under the European Community regulations (Araki et al. 2014). For genome edited products that with no or a few base pair insert the attitude of EU is still unclear.

Argentina, Australia and New Zealand

In 2011 Argentina issued that genome edited products with no base pair insert would fall outside the regulations of the Argentinian regulatory framework, but products with a large DNA fragment insertion should be regulated (Araki et al. 2014). Moreover, genome edited products with a few base pair insert would be regulated on a case-by-case basis if its use entails the introduction of coding sequences. And the attitude to GMOs in Australia and New Zealand are similar to that in Argentina (Araki et al. 2014).

China

According to ‘Regulations on Administration of Agricultural Genetically Modified Organisms Safety’ promulgated by China, all genome edited products should be considered as GMOs. Although the government of China regulates traditional GMO products strictly, the attitude to the genome edited products is still not clear (Table 2).
Table 2

Comparison of regulation on genome edited products in diffident countries

 

With no exogenous DNA insertion

With a few base pair DNA insertion

With large DNA fragment insertion

US

Regulated

Regulated

Strict

EU

Not clear

Not clear

Strict

Argentina

Unregulated

Case-by-case

Strict

Australia

Unregulated

Case-by-case

Strict

New Zealand

Unregulated

Case-by-case

Strict

China

Not clear

Not clear

Strict

Gene edited livestock versus GMO

Up to 2015, about 363 kinds of GM crops were confirmed. It mainly included soybean, corn, rape and cotton (Li et al. 2017). In 2014, the global GM crop planted area reached 181.5 million hectares, and the area is increasing every year (Aldemita et al. 2015). GMO livestock animals, in general, have more biosafety advantages than GMO plants because livestock are primarily raised in a pen, which allows easier management and reduces the escape of edited animals and avoid the occurrence of gene drifting (Tang et al. 2011, 2014). However, only one kind of GMO animal, salmon, was confirmed as food by FDA in 2015 (Waltz 2016b). One primary reason is the low public acceptance to GMO meat product at this moment.

The new genome editing tools offer a safer alternative to the traditional GMOs, and hopefully it will be easier for the public to accept. The main difference between gene edited livestock and traditional GMOs are: (1) no addition of foreign genes are needed for gene edited livestock, hence no risks from “unknown” ingredients. (2) Footprint free; precise editing of the genome without any exogenous traces. (3) Similar to natural mutation with no or only a few base pair change in gene edited livestock.

Based on these differences, gene edited livestock are different from conventional GMO animals, and they do not have the same risks as GMO animals. Gene edited livestock can be regarded as being produced by mutagenesis and that animals produced by chemical or radiation induced mutagenesis are not regulated (Table 3).
Table 3

Comparison of genome edited livestock and traditional GMO

 

Genome edited livestock

GMO

Foreign gene

No addition of foreign genes

Foreign gene insertion

Footprint

Footprint free

Have footprint

Compared to natural mutation

Similar

Different

Policy considerations for gene edited livestock animals

Distinguish gene edited livestock from conventional GMO livestock

Assessment of gene-edited livestock should be conducted using a case-by-case analysis. Safety assessment steps designed for traditional transgenic animals can be omitted during the safety evaluation process of gene-edited livestock due to the lack of integration of exogenous sequences. Mimicking natural mutants can also be exempted. For cases that few nucleotide deletions that generates truncated proteins should be investigated on a case-by-case review process. If novel traits are introgressed by genome editing, such genome edited animal should be considered as GMO.

Public education is key to public acceptance

Public education is very important. One main reason that prevents GMO being accepted by the public is that people are led by their tradition or sometimes misconception. For example, some Chinese believe that what natural offers is the best, and GMO is harmful to people. Outreach on educating the general public on the safety of GMO or gene edited livestock is necessary although it may take some time.

Comprehensive monitoring system to ensure biosafety

Comprehensive monitoring system is critical for biosafety. Gene edited livestock can be divided into three types based on the type of gene modification: (a) the first type introduces mutations, including deletions and insertions of one or a few base pairs using NHEJ (non-homologous end joining) mechanism; (b) the second type introduces DNA into the genome using HDR (homology-directed repair) with a short repair donor as a template; (c) the third type introduces a large piece of DNA at a specific genomic site (targeted gene addition or replacement). Safety monitoring system should be individualized for each type. For most animals containing large DNA fragment insertion (the 3rd type described above), it is reasonable to treat these livestock the same as traditional transgenic animals with the same policy attitude as conventional GMO livestock. But for the first and second type of gene edited livestock, there are no or a few base pair insert, which is more similar to natural mutation. The policy attitude can be more optimistic. In the case that there is no base pair insert, it is reasonable to categorize them to be outside of the GMO regulatory framework like what the US policy is. It is worth mention that some large DNA fragments can be safe although they fall under “the third type” edit. For example, the polled Holstein cattle produced by genome editing (Carlson et al 2016) contained a large region of DNA from a Hereford animal replacing the same region in the Holstein genome. It is simply swapping one allele with another.

Conclusion

Livestock breeding has made remarkable achievements after experiencing various developmental stages. However, it has encountered a bottleneck due to long breeding cycle, slow effect, and decrease in genetic resources. Fortunately, genome editing technology has emerged and can significantly shorten the breeding cycle, lower breeding cost, and rapidly increase genetic diversity. It can also replace molecular breeding and whole-genome breeding. Currently, a large number of genome-edited livestock has been generated, including an anti-PRRS gene-edited pig, an MSTN gene-edited pig, and a transgenic cow that is less susceptible to tuberculosis. These gene-edited livestock have demonstrated significant improvements in the lean meat rate, disease resistance and other favorable traits. Gene-edited livestock are easier to manage and contain than crops. Safety assessments can be performed using a case-by-case analysis. The assessment procedure can be simplified for gene-edited livestock that do not have exogenous genes introduced or mimic natural mutations.

Taking advantage of the genome editing revolution, livestock breeding will enter into a new development opportunity and provide humans with higher quality, healthier and lower cost products with richer varieties. Genome editing technology will ignite a revolution in livestock breeding.

Notes

Acknowledgements

This work was supported by The National Transgenic Project of China (2016ZX08006-001) and National Key Basic Research Program of China (2015CB943101) and Foshan University Initiative Scientific Research Program.

References

  1. Aldemita RR, Reano IM, Solis RO, Hautea RA (2015) Trends in global approvals of biotech crops (1992–2014). GM Crops Food 6:150–166CrossRefPubMedPubMedCentralGoogle Scholar
  2. Araki M, Nojima K, Ishii T (2014) Caution required for handling genome editing technology. Trends Biotechnol 32:234–237CrossRefPubMedGoogle Scholar
  3. Burkard C, Lillico SG, Reid E, Jackson B, Mileham AJ, Ait-Ali T, Whitelaw CB, Archibald AL (2017) Precision engineering for PRRSV resistance in pigs: macrophages from genome edited pigs lacking CD163 SRCR5 domain are fully resistant to both PRRSV genotypes while maintaining biological function. PLoS Pathog 13:e1006206CrossRefPubMedPubMedCentralGoogle Scholar
  4. Cai C, Qian L, Jiang S, Sun Y, Wang Q, Ma D, Xiao G, Li B, Xie S, Gao T, Chen Y, Liu J, An X, Cui W, Li K (2017) Loss-of-function myostatin mutation increases insulin sensitivity and browning of white fat in Meishan pigs. Oncotarget 8:34911–34922PubMedPubMedCentralGoogle Scholar
  5. Carlson DF, Tan W, Lillico SG, Stverakova D, Proudfoot C, Christian M, Voytas DF, Long CR, Whitelaw CB, Fahrenkrug SC (2012) Efficient TALEN-mediated gene knockout in livestock. Proc Natl Acad Sci USA 109:17382–17387CrossRefPubMedPubMedCentralGoogle Scholar
  6. Carlson DF, Lancto CA, Zang B, Kim ES, Walton M, Oldeschulte D, Seabury C, Sonstegard TS, Fahrenkrug SC (2016) Production of hornless dairy cattle from genome-edited cell lines. Nat Biotechnol 34:479–481CrossRefPubMedGoogle Scholar
  7. Chen H, Li C, Fang M, Zhu M, Li X, Zhou R, Li K, Zhao S (2009) Understanding Haemophilus parasuis infection in porcine spleen through a transcriptomics approach. BMC Genom 10:64CrossRefGoogle Scholar
  8. Christian M, Cermak T, Doyle EL, Schmidt C, Zhang F, Hummel A, Bogdanove AJ, Voytas DF (2010) Targeting DNA double-strand breaks with TAL effector nucleases. Genetics 186:757–761CrossRefPubMedPubMedCentralGoogle Scholar
  9. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA, Zhang F (2013) Multiplex genome engineering using CRISPR/Cas systems. Science 339:819–823CrossRefPubMedPubMedCentralGoogle Scholar
  10. Davoli R, Braglia S (2007) Molecular approaches in pig breeding to improve meat quality. Brief Funct Genomic Proteomic 6:313–321CrossRefPubMedGoogle Scholar
  11. Deng S, Yu K, Zhang B, Yao Y, Liu Y, He H, Zhang H, Cui M, Fu J, Lian Z, Li N (2012) Effects of over-expression of TLR2 in transgenic goats on pathogen clearance and role of up-regulation of lysozyme secretion and infiltration of inflammatory cells. BMC Vet Res 8:196CrossRefPubMedPubMedCentralGoogle Scholar
  12. Doyon Y, McCammon JM, Miller JC, Faraji F, Ngo C, Katibah GE, Amora R, Hocking TD, Zhang L, Rebar EJ, Gregory PD, Urnov FD, Amacher SL (2008) Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat Biotechnol 26:702–708CrossRefPubMedPubMedCentralGoogle Scholar
  13. Duijvesteijn N, Knol EF, Merks JW, Crooijmans RP, Groenen MA, Bovenhuis H, Harlizius B (2010) A genome-wide association study on androstenone levels in pigs reveals a cluster of candidate genes on chromosome 6. BMC Genet 11:42CrossRefPubMedPubMedCentralGoogle Scholar
  14. Feng W, Dai Y, Mou L, Cooper DK, Shi D, Cai Z (2015) The potential of the combination of CRISPR/Cas9 and pluripotent stem cells to provide human organs from chimaeric pigs. Int J Mol Sci 16:6545–6556CrossRefPubMedPubMedCentralGoogle Scholar
  15. Hai T, Teng F, Guo R, Li W, Zhou Q (2014) One-step generation of knockout pigs by zygote injection of CRISPR/Cas system. Cell Res 24:372–375CrossRefPubMedPubMedCentralGoogle Scholar
  16. Hockemeyer D, Soldner F, Beard C, Gao Q, Mitalipova M, DeKelver RC, Katibah GE, Amora R, Boydston EA, Zeitler B, Meng X, Miller JC, Zhang L, Rebar EJ, Gregory PD, Urnov FD, Jaenisch R (2009) Efficient targeting of expressed and silent genes in human ESCs and iPSCs using zinc-finger nucleases. Nat Biotechnol 27:851–857CrossRefPubMedPubMedCentralGoogle Scholar
  17. Jonas E, de Koning DJ (2015) Genomic selection needs to be carefully assessed to meet specific requirements in livestock breeding programs. Front Genet 6:49CrossRefPubMedPubMedCentralGoogle Scholar
  18. Lai L, Kang JX, Li R, Wang J, Witt WT, Yong HY, Hao Y, Wax DM, Murphy CN, Rieke A, Samuel M, Linville ML, Korte SW, Evans RW, Starzl TE, Prather RS, Dai Y (2006) Generation of cloned transgenic pigs rich in omega-3 fatty acids. Nat Biotechnol 24:435–436CrossRefPubMedPubMedCentralGoogle Scholar
  19. Lee K, Kwon DN, Ezashi T, Choi YJ, Park C, Ericsson AC, Brown AN, Samuel MS, Park KW, Walters EM, Kim DY, Kim JH, Franklin CL, Murphy CN, Roberts RM, Prather RS, Kim JH (2014) Engraftment of human iPS cells and allogeneic porcine cells into pigs with inactivated RAG2 and accompanying severe combined immunodeficiency. Proc Natl Acad Sci USA 111:7260–7265CrossRefPubMedPubMedCentralGoogle Scholar
  20. Li H, Haurigot V, Doyon Y, Li T, Wong SY, Bhagwat AS, Malani N, Anguela XM, Sharma R, Ivanciu L, Murphy SL, Finn JD, Khazi FR, Zhou S, Paschon DE, Rebar EJ, Bushman FD, Gregory PD, Holmes MC, High KA (2011) In vivo genome editing restores haemostasis in a mouse model of haemophilia. Nature 475:217–221CrossRefPubMedPubMedCentralGoogle Scholar
  21. Li X, Yang Y, Bu L, Guo X, Tang C, Song J, Fan N, Zhao B, Ouyang Z, Liu Z, Zhao Y, Yi X, Quan L, Liu S, Yang Z, Ouyang H, Chen YE, Wang Z, Lai L (2014) Rosa26-targeted swine models for stable gene over-expression and Cre-mediated lineage tracing. Cell Res 24:501–504CrossRefPubMedPubMedCentralGoogle Scholar
  22. Li Y, Lian D, Deng S, Zhang X, Zhang J, Li W, Bai H, Wang Z, Wu H, Fu J, Han H, Feng J, Liu G, Lian L, Lian Z (2016) Efficient production of pronuclear embryos in breeding and nonbreeding season for generating transgenic sheep overexpressing TLR4. J Anim Sci Biotechnol 7:38CrossRefPubMedPubMedCentralGoogle Scholar
  23. Li R, Quan S, Yan X, Biswas S, Zhang D, Shi J (2017) Molecular characterization of genetically-modified crops: challenges and strategies. Biotechnol Adv 35:302–309CrossRefPubMedGoogle Scholar
  24. Lillico SG, Proudfoot C, Carlson DF, Stverakova D, Neil C, Blain C, King TJ, Ritchie WA, Tan W, Mileham AJ, McLaren DG, Fahrenkrug SC, Whitelaw CB (2013) Live pigs produced from genome edited zygotes. Sci Rep 3:2847CrossRefPubMedGoogle Scholar
  25. Lillico SG, Proudfoot C, King TJ, Tan W, Zhang L, Mardjuki R, Paschon DE, Rebar EJ, Urnov FD, Mileham AJ, McLaren DG, Whitelaw CB (2016) Mammalian interspecies substitution of immune modulatory alleles by genome editing. Sci Rep 6:21645CrossRefPubMedPubMedCentralGoogle Scholar
  26. Liu X, Wang Y, Guo W, Chang B, Liu J, Guo Z, Quan F, Zhang Y (2013) Zinc-finger nickase-mediated insertion of the lysostaphin gene into the beta-casein locus in cloned cows. Nat Commun 4:2565PubMedPubMedCentralGoogle Scholar
  27. Liu X, Wang Y, Tian Y, Yu Y, Gao M, Hu G, Su F, Pan S, Luo Y, Guo Z, Quan F, Zhang Y (2014) Generation of mastitis resistance in cows by targeting human lysozyme gene to beta-casein locus using zinc-finger nucleases. Proc Biol Sci 281:20133368CrossRefPubMedPubMedCentralGoogle Scholar
  28. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM (2013) RNA-guided human genome engineering via Cas9. Science 339:823–826CrossRefPubMedPubMedCentralGoogle Scholar
  29. McPherron AC, Lee SJ (1997) Double muscling in cattle due to mutations in the myostatin gene. Proc Natl Acad Sci USA 94:12457–12461CrossRefPubMedPubMedCentralGoogle Scholar
  30. Meuwissen TH, Hayes BJ, Goddard ME (2001) Prediction of total genetic value using genome-wide dense marker maps. Genetics 157:1819–1829PubMedPubMedCentralGoogle Scholar
  31. Miller JC, Holmes MC, Wang J, Guschin DY, Lee YL, Rupniewski I, Beausejour CM, Waite AJ, Wang NS, Kim KA, Gregory PD, Pabo CO, Rebar EJ (2007) An improved zinc-finger nuclease architecture for highly specific genome editing. Nat Biotechnol 25:778–785CrossRefPubMedGoogle Scholar
  32. Ni W, Qiao J, Hu S, Zhao X, Regouski M, Yang M, Polejaeva IA, Chen C (2014) Efficient gene knockout in goats using CRISPR/Cas9 system. PLoS ONE 9:e106718CrossRefPubMedPubMedCentralGoogle Scholar
  33. Nishi M, Yasue A, Nishimatu S, Nohno T, Yamaoka T, Itakura M, Moriyama K, Ohuchi H, Noji S (2002) A missense mutant myostatin causes hyperplasia without hypertrophy in the mouse muscle. Biochem Biophys Res Commun 293:247–251CrossRefPubMedGoogle Scholar
  34. Ollivier L, Sellier P (1982) Pig genetics: a review. Ann Genet Sel Anim 14:481–544PubMedPubMedCentralGoogle Scholar
  35. Proudfoot C, Carlson DF, Huddart R, Long CR, Pryor JH, King TJ, Lillico SG, Mileham AJ, McLaren DG, Whitelaw CB, Fahrenkrug SC (2015) Genome edited sheep and cattle. Transgenic Res 24:147–153CrossRefPubMedGoogle Scholar
  36. Qian L, Tang M, Yang J, Wang Q, Cai C, Jiang S, Li H, Jiang K, Gao P, Ma D, Chen Y, An X, Li K, Cui W (2015) Targeted mutations in myostatin by zinc-finger nucleases result in double-muscled phenotype in Meishan pigs. Sci Rep 5:14435CrossRefPubMedPubMedCentralGoogle Scholar
  37. Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE, Scott DA, Inoue A, Matoba S, Zhang Y, Zhang F (2013) Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell 154:1380–1389CrossRefPubMedPubMedCentralGoogle Scholar
  38. Ruan J, Li H, Xu K, Wu T, Wei J, Zhou R, Liu Z, Mu Y, Yang S, Ouyang H, Chen-Tsai RY, Li K (2015) Highly efficient CRISPR/Cas9-mediated transgene knockin at the H11 locus in pigs. Sci Rep 5:14253CrossRefPubMedPubMedCentralGoogle Scholar
  39. Sato M, Miyoshi K, Nagao Y, Nishi Y, Ohtsuka M, Nakamura S, Sakurai T, Watanabe S (2014) The combinational use of CRISPR/Cas9-based gene editing and targeted toxin technology enables efficient biallelic knockout of the alpha-1,3-galactosyltransferase gene in porcine embryonic fibroblasts. Xenotransplantation 21:291–300CrossRefPubMedGoogle Scholar
  40. Schuelke M, Wagner KR, Stolz LE, Hubner C, Riebel T, Komen W, Braun T, Tobin JF, Lee SJ (2004) Myostatin mutation associated with gross muscle hypertrophy in a child. N Engl J Med 350:2682–2688CrossRefPubMedGoogle Scholar
  41. Tang M, Zheng X, Cheng W, Jin E, Chen H, Yang S, Cui W, Li K (2011) Safety assessment of sFat-1 transgenic pigs by detecting their co-habitant microbe in intestinal tract. Transgenic Res 20:749–758CrossRefPubMedGoogle Scholar
  42. Tang M, Qian L, Jiang S, Zhang J, Song P, Chen Y, Cui W, Li K (2014) Functional and safety evaluation of transgenic pork rich in omega-3 fatty acids. Transgenic Res 23:557–571CrossRefPubMedGoogle Scholar
  43. Tao C, Yang Y, Li X, Zheng X, Ren H, Li K, Zhou R (2016) Rapid and sensitive detection of sFAT-1 transgenic pigs by visual loop-mediated isothermal amplification. Appl Biochem Biotechnol 179:938–946CrossRefPubMedGoogle Scholar
  44. Waltz E (2012) Tiptoeing around transgenics. Nat Biotechnol 30:215–217CrossRefPubMedGoogle Scholar
  45. Waltz E (2016a) Gene-edited CRISPR mushroom escapes US regulation. Nature 532:293CrossRefPubMedGoogle Scholar
  46. Waltz E (2016b) GM salmon declared fit for dinner plates. Nat Biotechnol 34:7–9CrossRefPubMedGoogle Scholar
  47. Wang K, Ouyang H, Xie Z, Yao C, Guo N, Li M, Jiao H, Pang D (2015a) Efficient generation of myostatin mutations in pigs using the CRISPR/Cas9 system. Sci Rep 5:16623CrossRefPubMedPubMedCentralGoogle Scholar
  48. Wang X, Yu H, Lei A, Zhou J, Zeng W, Zhu H, Dong Z, Niu Y, Shi B, Cai B, Liu J, Huang S, Yan H, Zhao X, Zhou G, He X, Chen X, Yang Y, Jiang Y, Shi L, Tian X, Wang Y, Ma B, Huang X, Qu L, Chen Y (2015b) Generation of gene-modified goats targeting MSTN and FGF5 via zygote injection of CRISPR/Cas9 system. Sci Rep 5:13878CrossRefPubMedPubMedCentralGoogle Scholar
  49. Wei J, Wagner S, Lu D, Maclean P, Carlson DF, Fahrenkrug SC, Laible G (2015) Efficient introgression of allelic variants by embryo-mediated editing of the bovine genome. Sci Rep 5:11735CrossRefPubMedPubMedCentralGoogle Scholar
  50. Whitworth KM, Lee K, Benne JA, Beaton BP, Spate LD, Murphy SL, Samuel MS, Mao J, O’Gorman C, Walters EM, Murphy CN, Driver J, Mileham A, McLaren D, Wells KD, Prather RS (2014) Use of the CRISPR/Cas9 system to produce genetically engineered pigs from in vitro-derived oocytes and embryos. Biol Reprod 91:78CrossRefPubMedPubMedCentralGoogle Scholar
  51. Whitworth KM, Rowland RR, Ewen CL, Trible BR, Kerrigan MA, Cino-Ozuna AG, Samuel MS, Lightner JE, McLaren DG, Mileham AJ, Wells KD, Prather RS (2016) Gene-edited pigs are protected from porcine reproductive and respiratory syndrome virus. Nat Biotechnol 34:20–22CrossRefPubMedGoogle Scholar
  52. Whyte JJ, Zhao J, Wells KD, Samuel MS, Whitworth KM, Walters EM, Laughlin MH, Prather RS (2011) Gene targeting with zinc finger nucleases to produce cloned eGFP knockout pigs. Mol Reprod Dev 78:2CrossRefPubMedPubMedCentralGoogle Scholar
  53. Wood RJ (1973) Robert Bakewell (1725–1795), pioneer animal breeder, and his influence on Charles Darwin. Folia Mendeliana 58:231–242PubMedGoogle Scholar
  54. Wu H, Wang Y, Zhang Y, Yang M, Lv J, Liu J, Zhang Y (2015) TALE nickase-mediated SP110 knockin endows cattle with increased resistance to tuberculosis. Proc Natl Acad Sci USA 112:E1530–E1539CrossRefPubMedPubMedCentralGoogle Scholar
  55. Yang D, Yang H, Li W, Zhao B, Ouyang Z, Liu Z, Zhao Y, Fan N, Song J, Tian J, Li F, Zhang J, Chang L, Pei D, Chen YE, Lai L (2011) Generation of PPARgamma mono-allelic knockout pigs via zinc-finger nucleases and nuclear transfer cloning. Cell Res 21:979–982CrossRefPubMedPubMedCentralGoogle Scholar
  56. Yang S, Li X, Li K, Fan B, Tang Z (2014) A genome-wide scan for signatures of selection in Chinese indigenous and commercial pig breeds. BMC Genet 15:7CrossRefPubMedPubMedCentralGoogle Scholar
  57. Yao J, Huang J, Hai T, Wang X, Qin G, Zhang H, Wu R, Cao C, Xi JJ, Yuan Z, Zhao J (2014) Efficient bi-allelic gene knockout and site-specific knock-in mediated by TALENs in pigs. Sci Rep 4:6926CrossRefPubMedPubMedCentralGoogle Scholar
  58. Yu S, Luo J, Song Z, Ding F, Dai Y, Li N (2011) Highly efficient modification of beta-lactoglobulin (BLG) gene via zinc-finger nucleases in cattle. Cell Res 21:1638–1640CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2017

Authors and Affiliations

  1. 1.College of Life ScienceFoshan UniversityFoshanChina
  2. 2.Center for Advanced Models for Translational Sciences and TherapeuticsUniversity of Michigan Medical CenterAnn ArborUSA
  3. 3.Applied StemCell, IncMilpitasUSA
  4. 4.Institute of Animal ScienceChinese Academy of Agricultural SciencesBeijingChina

Personalised recommendations