Genetic improvement of horticultural crops mediated by CRISPR/Cas: a new horizon of possibilities

The burden of the current global challenge involving food security lies in the need to improve crop production. In this regard, biotechnology stands out as an essential tool to generate plants able to cope with pests, diseases, and harsh climatic conditions, and more efficient in the use of natural resources. An advanced approach to create genetic variability in a precise and targeted way, the genome-editing technique CRISPR/Cas (clustered regularly interspaced short palindromic repeats/CRISPR associated proteins), has drawn the attention of breeders. The genome editing CRISPR/Cas system relies on a guiding RNA that directs a nuclease to generate a double-strand break (DSB) at a target DNA, activating the cell repair systems and eventually leading to deletions or insertions of nucleotides. Therefore, CRISPR/Cas is a toolbox to achieve many goals, from basic science investigations to the development of crops with improved agronomic traits, with potential to bring innovative solutions to food production. The CRISPR/Cas system has been applied in a large number of plants, including some horticultural species. In this review, we present details of the CRISPR/Cas natural and artificial systems, its possibilities as a biotechnological tool, advantages over other breeding techniques, regulatory issues, and its applicability in horticultural crops, as well as future challenges.

fruit quality traits linked explicitly to consumer preferences, such as flavor and nutraceutical compounds (Hansson et al. 2018;Neves et al. 2018).
Over the years, conventional plant breeding has been proved successful in addressing several of these critical characteristics.However, in some cases, conventional strategies have limited application due to non-availability in the need to improve horticultural crop production (Van den Broeck & Maertens, 2016).Thus, the development of cultivars with enhanced tolerance to biotic and abiotic stresses accounts for the primary targets of genetic breeding programs.Furthermore, the goals in plant breeding have been expanded to the increasing of shelf-life, to reduce postharvest losses, and the improvement of H orticultural crops, which comprise vegetables, fruits, flowers, as well as medicinal and aromatic plants, are an essential part of agriculture production.In particular, fruits and vegetables are crucial to a balanced diet, providing energy and essential vitamins and minerals that are beneficial to human health.The burden of the current global challenge involving food security lies Received on May 2, 2018; accepted on September 6, 2018 of target alleles in the germplasm of a particular species.In this case, transgenic technology represents a potential tool for genetic enhancement since it allows the introduction of specific genes which may come from a non-crossable-plant species or a nonplant-organism (Anami et al., 2013).A large number of transgenic horticultural crops have been developed in the last 30 years carrying relevant modifications (Parmar et al., 2017).Although the use of genes from different species is still an important tool for developing improved crops for a variety of traits, a more advanced and precise approach has drawn the attention of the breeders: the genome editing using engineered nucleases.The clustered regularly interspaced short palindromic repeats (CRISPR) associated proteins (CRISPR/ Cas) system is the most recent tool developed for this purpose, allowing researchers to delete, replace, or insert specific sequences in a targeted location of the genome to generate new valuable traits with potential to bring innovative solutions to agriculture (Cardi, 2016;Liu et al., 2016).

Genome editing
Over the years, artificial genetic manipulation has been used to unravel the function of genes and their regulation in metabolic pathways to create a phenotype of interest.Initially, genetic modifications were artificially performed mainly by mutagenesis with radiation or chemical agents, followed by the identification of individuals presenting the desired or the aberrant phenotypes.
Although this method has contributed to the study and understanding of several cellular and molecular processes (Chen et al., 2012;Haydon et al., 2013), it requires the screening of thousands of individuals carrying random mutations.Consequently, it is challenging to apply it in plant breeding programs, especially those involving quantitative agronomic traits, such as yield, nutritional quality, and resistance and/or tolerance to biotic and abiotic stresses.
Advances in science created the opportunity to drive mutations at specific sites in the genome, saving time and the laborious work of inducing and screening random mutations.This great challenge was surpassed in the 90s when the first tools emerged, relying on engineered nucleases linked to components capable of recognizing DNA sequences.The first technique successfully used for this purpose was the zinc finger nuclease (ZFN), which identify specific DNA sequences through protein-DNA interactions and guide an associated endonuclease (commonly Fok I), allowing the manipulation of particular targets in the genome (Durai et al., 2005).Later, a similar genome editing tool was developed, based on transcription effectors identified in the plant pathogen Xanthomonas spp., called transcription activatorlike effectors (TALEs) (Bogdanove et al., 2010).These effectors can be customized to recognize and bind to specific DNA guiding endonucleases to induce site-specific editions in the genome in a complex system called TAL effector nucleases (TALENs) (Bogdanove & Voytas, 2011).Although the ZFN and TALENs technologies have been used for genome editing in several organisms, the high cost and complexity of synthesizing DNA-binding proteins have limited their use for the study and genetic improvement of plants.
The most recent genome editing tool is the CRISPR/Cas system, identified as part of the immune mechanism against exogenous DNA in bacteria and Archaea.CRISPR/Cas uses small sequences of non-coding RNAs to guide nucleases to cleave a target DNA (Horvath & Barrangou, 2010) or a target RNA, as recently discovered (Wolter & Puchta, 2018).By using RNA as guiding molecules, this technique dispenses the laborious and expensive step of building and optimizing complex proteins (such as ZFN and TALEs) for DNA recognition, representing a more flexible and viable tool for genome manipulation (Song et al. 2016).Compared to other genome editing strategies, the CRISPR/CAS technique is more straight forward, more costeffective, precise and is highly efficient even at multiplex genome editing (Wang et al., 2018).Therefore, the innumerable advantages of this new technology have gradually made previous tools, such as ZFN and TALENs, become increasingly obsolete.
The CRISPR/Cas system T h e o p p o r t u n i t y t o e x p l o i t t h e C R I S P R / C a s s y s t e m a s a biotechnological tool came from a deeper understanding of the underlying molecular mechanisms of the natural process in prokaryotes, specifically of the CRISPR/Cas 9 system from Streptococcus pyogenes.This microbial adaptive immune system mediates defense against foreign genetic elements through three main steps: immunization, expression, and interference (Bhaya et al., 2011).
During the immunization step, the host incorporates small sequences of the invader DNA (ranging from 21-48 bp) to a specific region of its own genome, the CRISPR locus, in the form of spacers between short palindromic repeats.The recognition of the region to be incorporated as a spacer depends on the presence of small conserved nucleotide sequences (2-3 specific nucleotides), called adjacent protospacer motifs (PAMs), which represent the anchoring site of the nuclease and determine the cleavage point of the target DNA (Mojica et al., 2009).The expression of these regions results in small noncoding RNAs, called CRISPR-RNAs (crRNAs) and trans-activating crRNA (tracrRNA) (Makarova et al., 2011).The tracrRNA participates in the maturation process of the crRNAs, forming a two-RNA structure that guides Cas9 to promote double-strand breaks (DSBs) in the foreign DNA, inactivating the genetic material of the invader at the interference stage (Deltcheva et al., 2011).
Thus, in summary, the CRISPR/Cas system allows the host to construct a "library" containing an array of small fragments of foreign DNA that have previously invaded the cell, which prevents future infections by cleaving exogenous genetic materials that include library-like fragments mediated by the action of endonucleases.Briefly, it relies on two essential components: the nuclease Cas9 and a guide, which is composed of two RNA sequences.Among the adaptions of the natural process to form the biotechnological Genetic improvement of horticultural crops mediated by CRISPR/Cas: a new horizon of possibilitie t o o l i s t h e s i n g l e g u i d e R N A (sgRNA), a combination of the dual-tracrRNA:crRNA (type II system), which simplifies the system (Jinek et al., 2012).Moreover, the ability to drive the Cas9 nuclease to the target site comes from the specificity of a sequence of about 20 bp located in the 5' portion of the guide RNA, which has homology to the sequence of interest (Cong et al., 2013).By manipulating this 20 bp sequence, it is possible to create specific sgRNA molecules for different targets in the genome, which makes the CRISPR/Cas technique an extremely versatile tool, applicable to a wide range of species.
As mentioned above, an essential requirement for the functionality of the system is that the sgRNA must have homology to a fragment of DNA that is adjacent to the anchoring site of the endonuclease, a small conserved nucleotide sequence (2-3 nucleotides), the PAM (Mojica et al., 2009).Interestingly, the sequence that determines PAM varies for homologous Cas present in different organisms.The Cas9 of S. pyogenes (SpCas9), for example, recognizes the sequence PAM 5'-NGG-3' and performs a blunt DSB at approximately three nucleotides upstream of the PAM sequence (Jinek et al., 2012).Although Cas9 of S. pyogenes is the most widely used wild endonuclease, homologous Cas9, such as Staphylococcus aureus (SaCas9), are important since they represent an extension of the range of PAM sequences that can be used in the CRISPR/Cas system.(Kleinstiver et al., 2015).
Furthermore, although the wild Cas9 nuclease of S. pyogenes proved to be functional for editing the genome of plants and other eukaryotes, several research groups performed optimizations on the original Cas9 codons, aiming for greater efficiency in the species studied, for instance monocots or dicots (Wang et al., 2014, Michno et al., 2015).The addition of the nuclear localization signal (NLS) to the enzyme sequence in the transformation vector was another adaptation to the editing of eukaryotic genomes (Michno et al., 2015).These signals ensure that after translation in the cytoplasm of the cell, the nucleases are transported to their site of action, the nucleus.
Recently, the diversity of this system has been investigated.Researchers screened for other nuclease families with features distinct from Cas9 and found some efficient enzymes, such as Cpf1 and Cms1 (CRISPR from Microgenomates and Smithella).These nucleases utilize a T-rich and ATrich PAM, respectively, mediating robust DNA cleavage via a staggered (sticky-end) DNA DSB (Zetsche et al., 2015).Identifying these variant nucleases and distinct mechanisms of interference broaden our understanding of the CRISPR/Cas toolbox, increases the diversity of options available to researchers and advances genome editing applications.
In summary, billions of years of evolution have developed and improved a system that cuts prokaryotic DNA in a site-specific manner.Researchers investigated and adapted it to become a biotechnological tool for editing the genome of many species, including eukaryotes.After the DSB at the genome target region caused by CRISPR/Cas, natural repair mechanisms present in the cell are recruited to the cleavage region, the main ones being the nonhomologous end joining (NHEJ), which is prone to errors, and the better, but more complicated, homology-directed repair (HDR) (Dexheimer, 2013).
During the NHEJ repair, the ends of the break are ligated and small insertions and deletions (Indels) at this site may occur, leading to imperfect repair and various mutations.These modifications may result in changes in the reading frame during translation of the mRNA and in the onset of a premature stop codon, culminating in the knockout of the gene of interest or loss of functionality of the encoded protein.The NHEJ repair system is the primary mechanism used in genome editing strategies in plants.
The HDR mechanism maintains the genome integrity by using a model DNA to correct a DBS, a sister chromatid in dividing cells, for instance (Escribano-Díaz et al., 2013).In genetic manipulation, a donor DNA homologous to the site of cleavage is provided to the cell.However, it may contain a new sequence between the homology borders, which is going to be copied by the cell.This action represents a highly specific genome editing technique, wherein the new sequences can be stipulated, unlike the random mutations the NHEJ repair system causes.The sequences inserted can be fragments of a gene, whole genes, and site of transcription factors in the promoter region or even entire promoters.It has not been vastly used for editing plant genomes because of the low frequency of target mutation and recombination (Puchta, 2005).However, the discovery and application of new endonuclease families have increased the efficiency of the HDR pathway.
Inducing a DSB in the DNA of an organism to activate a repair pathway has a great potential application for plant breeding, such as silencing of undesired native genes; alteration of the nucleotide sequence encoding amino acids, aiming at improving a protein activity (the transfer of complete or partial gene sequences or entire allele swapping); and fine-tune expression of native genes of interest acting upon the promoter region (repressing or adding artificial transcription factors, for example).Therefore, not only can native genes be modified and manipulated but novel traits can be inserted as well.
A f t e r d e f i n i n g t h e t y p e o f endonuclease and the approach of the system (NHEJ or HDR), plant genomic editing via CRISPR/Cas can be divided into four main steps: 1) selection of the target regions of the sgRNA (that must be unique in the genome and adjacent to the PAM specific to the nuclease chosen); 2) design and construction of CRISPR/ Cas transformation vectors containing the sgRNAs, the Cas nuclease and the donor DNA (in the case of HDR); 3) transformation of explants and regeneration of transformed plants, and 4) screening of mutations in T0 plants (Liang et al., 2016).
Among the prerequisites to perform genomic editing, the target organism must have its genome sequenced, to identify specific sites in the sgRNA, thus avoiding modifications at undesired sites (off-target edition).There are some horticultural crops with sequenced genomes of reference available, such as tomato (Solanum lycopersicum) (The Tomato Genome Consortium, 2012), pepper (Capsicum annuum) (Kim et al., 2014), cucumber (Cucumis sativus) (Huang et al., 2009), sugar beet (Beta vulgaris) (Dohm et al., 2014), potato (Solanum tuberosum) (The Potato Genome Sequencing Consortium, 2011) among others.The screening of the plant genome avoids designing sgRNAs that bind to DNA sequences that present homology to other regions, which would lead to DSBs at sites other than the target, eventually leading to aberrant or unwanted phenotypes.In addition to specific sgRNAs, an efficient method for introducing DNA into plant cells, competence for regeneration (wellestablished protocol to generate whole plants from the genetically modified cells cultivated in vitro) and a suitable selection system are required.Several horticultural crops still lack these basic requirements, which represent the greatest challenge for the CRISPR/Cas tool to be applied.Further studies are necessary.
The CRISPR/Cas system has some advantages when compared to the genetic engineering tool commonly used, the transgenic.It is important to highlight that CRISPR/Cas enables direct DNA insertion at strategic locations of the genome, presenting rates of expression higher than that observed for transgenic, that induces random insertions.Besides, the CRISPR/Cas expression cassette may be introduced in one chromosome and perform the edition in another.Therefore, traits can segregate in cross-breeding, generating non-transgenic mutant plants (that no longer has the CRISPR/Cas locus) (Figure 1) (Chandrasekaran et al., 2016).Besides, a recent study has shown the possibility of inducing modifications in the genome using the CRISPR/ Cas9 technology transiently expressed in calluses, which were later used to regenerate whole plants, edited and nontransgenic.The CRISPR/Cas9 system is delivered as a protein-RNA complex degraded after edition and not as a DNA sequence permanently introduced in the plant genome (Zhang et al., 2016).As such, this is a tremendous breakthrough for perennial crops, wherein the time range of each generation is a drawback and does not allow several crosses.
To generate an edited non-transgenic plant becomes even more critical when it comes to species used directly as human food, like most vegetables and fruits.Although the world uses transgenic crops for more than 30 years, population acceptance of genetically modified organisms (GMOs) is still a substantial impediment to the commercialization of these products (Bradford et al., 2005).Furthermore, the development and release to the market of transgenic crops require very high investments of both time and resources, amounting to an estimate of $136 million, from the discovery of the trait to the authorization and commercial launch (McDougall, Step 1) Plant transformation, introduction of the expression cassette containing the sequences of a Cas nuclease and a sgRNA via Agrobacterium tumefaciens; Step 2) Selection of the transformed cells that received the system sequence at chromosome 1 (randomly); Step 3) Expression of the components Cas and SgRNA of the system in the cytoplasm; Step 4) Cas and the sgRNA form a complex and are directed to the nucleus where it performs a DBS at the DNA target inchromosome 2; Step 5) chromosome independent segregation in the crossing generations (2 nd Mendel's law) makes it possible to select plants with chromosome 1 without the CRISPR/Cas insertion and Chromosome 2 carrying the edition.Londrina, UEL, 2018.
2011).From this total, the regulatory testing and the registration processes demand about 26% ($35.1 million) and 5.5 years.The high cost of the whole process limits it to companies with the required resources, and the few ones capable of developing transgenic crops focus on a few economically valuable traits.
With minor exceptions, the GMOs presently in use are cultivars of crops used for feed, consumable oil, and fiber production (maize, soybean, canola, and cotton), developed by multinational companies for farmers in the United States and Canada, which were later adapted to some other countries (Brazil, Argentina, India).The vast majority presents the same two functional traits (herbicide and insect tolerance), both introduced in 1996 (Pingali & Feder, 2017).Therefore, the high cost of developing a crop with a biotechnological advantage narrows the possibility of more public and private companies to generate GMOs addressing the needs of developing country farmers or focused on other traits such as nutritional quality.
Although the use of genes from different species is still an essential tool in developing improved plants for a variety of traits, the high cost of regulation of a transgenic crop and population constraints hinders the generation and commercialization of these products.The possibility of using CRISPS/Cas strategies that generate non-transgenic mutants would save millions of dollars and years of work, circumventing the need to regulate a transgenic.Such situation would benefit researchers when addressing agricultural challenges and also the public sector, universities and other parties which would have more chances of developing biotechnological assets.At the same time, it would increase public acceptance, allowing advances in food production and contributing to food security.
These new breeding technologies (NBTs), which include the CRISPR/Cas system, have required special attention under the regulatory point of view and have been discussed and established in the adopting countries such as Brazil, Argentina, China, USA, and Canada.As mentioned above, some of the genetic manipulations may not be considered as transgenic, since the technique can induce new genetic variations without leaving traces of genes from another species in the final product, generating improved plants identical to the original plants, except for the targeted mutation.It is crucial to implement similar criteria in the analysis and deliberation of products to avoid commercialization impediments.
CRISPR/Cas in horticultural crops Genome editing using the CRISPR/ Cas system in plants was first reported in 2013, and since then it has received extensive attention (Li et al., 2013;Nekrasov et al., 2013;Shan et al., 2013).A complete search on Scopus and Web of Science databases, using the following query: "CRISPR" and "plants" showed several results (Figure 2).It can be observed that in a short time, numerous examples of studies involving the CRISPR system in plants were published, demonstrating its potential as a genomic tool for plant biology studies and crop improvement.
Among the studies, there are many reports of successful gene editing by CRISPR/Cas9 in horticultural crops (Table 1).Since it is still a very recent method, many authors have sought to introduce mutations in genes that would result in a distinctive and immediately recognizable phenotype, to test and optimize the efficacy of the CRISPR/ Cas9 technique.Klimek-Chodacka et al. (2018) directed mutations in the carrot F3H gene, essential for anthocyanin biosynthesis in purple-colored carrot, resulting in the regeneration of discolored carrot calli.Tian et al. (2017) selected the ClPDS gene required for chlorophyll biosynthesis, and watermelon mutant plants exhibited evident albino phenotype.The gene SLAGO7 regulates organ polarity in tomato, and loss-of-function mutant plants had abnormal leaves -leaflets without petioles and leaves lacking laminae -early in the tissue culture phase (Brooks et al., 2014).Lawrenson et al. (2015) generated Cas9-induced mutations in the BolC.GA4, a gene from Brassica oleracea, leading to plants with the expected dwarf phenotype associated with target gene knockout.
Genome editing has been widely used in plant research also to study gene functions and fundamental biological processes.The CRISPR/Cas9 system was applied to mutate two genes likely to be involved in auxin biosynthesis and signaling, FveTAA1 and FveARF8, in wild strawberry Fragaria vesca.The authors generated transgenic plants harboring knockout mutations in the target genes.The arf8 mutant seedlings showed faster growth than wild-type plants, suggesting that the FveARF8 gene plays a repressor role in auxin signaling (Zhou et al., 2018).CRISPR/ Cas9-engineered mutations in the tomato SlBOP gene caused severe inflorescence defects and allowed to prove its dominant role in flowering and inflorescence architecture (Xu et al., 2016).
S i m i l a r l y, t o i m p r o v e t h e understanding of the tomato photoperiod response, Soyk et al. (2018) generated a loss of day-length-sensitive flowering in tomato, with mutations in the SP5G gene.The knockout of the SlAGL6 gene underlined its role in facultative parthenocarpy, leading to the production of seedless tomato fruits with regular weight and shape, under heat stress (Klap et al., 2017).Site-directed mutagenesis in tomato has also been used to study the regulation of ripening, wherein a series of mutations that potentially eliminate the function of a RIN gene (Ito et al., ) and the lncRNA1459 gene (Li et al., 2018) resulted in mutants with deficient-ripening fruit production.
The ability to modify genomes in a site-directed manner has also been applied to develop cultivars with new traits.Parthenocarpic fruit is an attractive attribute since it allows the production of seedless fruits.Ueta et al. (2017) demonstrated a CRISPR/ Cas9 strategy to effectively introduce mutations into SlIAA9, a crucial gene controlling parthenocarpy in tomato.Transgenic tomato plants of Micro-Tom and commercial cultivar Ailsa Craig, carrying bi-allelic and homozygous mutations, exhibited the typical phenotypes of parthenocarpy, described as fruit development before pollination, leading to seedless fruits.A small number of fertilized fruits developed a few seeds, which produced tomato plants exhibiting the mutations and associated phenotypes.
The long shelf life is another important characteristic in fleshy fruit that influences fruit marketability and can reduce fruit loss.For exploiting this quality attribute in tomato, the CRISPR/ Cas9 system was applied for obtaining the tomato ALC gene replacement in the presence of the homologous repair template (replacement of thymine by adenine in position 317 of the coding sequence).The replacement efficiency was low in T0 transgenic plants, and only one individual with a heterozygous mutation was obtained, requiring further segregation to generate the homozygous mutation.In the T1 generation, it was possible to generate recessive homozygous alc mutants free of the CRISPR/Cas9 components which presented excellent storage performance (Yu et al., 2017).Nonaka et al. (2017) report another study involving fruit quality in tomato.The authors increased γ-aminobutyric acid (GABA) accumulation in tomato fruit, by removing the autoinhibitory domain of SlGAD2 and SlGAD3 genes, through the CRISPR/Cas9 system.The consumption of GABA-enriched foods in daily life can bring anti-hypotensive effects and would be an interesting path to prevent hypertension in humans.
Steroidal glycoalkaloids (SGAs), such as α-solanine and α-chaconine, are naturally occurring toxic compounds in potato tubers that can cause a bitter taste and undesirable effects in humans when present at high levels.Nakayasu et al. (2018) demonstrated a CRISPR/Cas9 strategy to reduce the SGA content in potato.The authors produced transgenic hairy roots carrying multiple mutations at different sites in St16DOX gene, encoding a steroid 16α-hydroxylase in SGA biosynthesis.These mutations lead to complete abolition of the SGA accumulation in potato hairy roots.
The CRISPR/Cas technology has also found application in developing disease-resistant plants.A tomato variety resistant to the powdery mildew fungal pathogen Oidium neolycopersici was developed by creating loss-of-function mutations in the mildew resistant locus O1 (Mlo1), which encodes a membrane-associated protein that confers susceptibility to the fungal pathogen.Lines from the T1 generation with mutations in both alleles were fully resistant to the pathogen compared to the wild-type (Nekrasov et al., 2017).Chandrasekaran et al. (2016) reported the generation of a cucumber with a broad virus resistance, by knocking out the cucumber eIF4E gene, a plant cellular translation factor essential for the Potyviridae life cycle.Homozygous T3 progeny exhibited immunity to the Cucumber vein yellowing virus (Ipomovirus) infection and resistance to the potyviruses Zucchini yellow mosaic virus and Papaya ringspot mosaic virus-W.The disease resistant tomato and cucumber plants were also transgene-free.In both instances, since the transgene Cas9/sg RNA and the gene mutation sites were at different genome locations and had independent segregation, it was possible to select non-transgenic mutants in later generations, i.e., plants containing the mutation without carrying any foreign DNA.The same strategy was used to generate longshelf life tomato lines (Yu et al., 2017).These examples emphasize the considerable potential of this tool, beyond that of transgenics, since it can offer an efficient method to avoid random gene insertions throughout the genome and, more important, it allows for the possibility of removing foreign DNA, a big concern related to transgenic plants.
The value of eliminating CRISPR/ Cas9 components inserted in the genome via selfing or backcrossing is more complicated in genetically complex and vegetative propagated species, such as potato.To overcome this difficulty, Andersson et al. (2017) demonstrated the transient transfection of the CRISPR/ Cas9 construct in protoplasts isolated from tetraploid potato and, subsequently, shoot regeneration containing the desired mutations.Using this approach, the authors were able to produce a few lines lacking any DNA integration and carrying mutations in at least one allele, multiple or four alleles of the gene StGBSS, encoding granule-bound starch synthase.Complete elimination of the StGBSS enzyme function in four-allele mutated lines led to the production of starch with altered amylose synthesis and a concomitant increase in the amylopectin/amylose ratio, a desirable trait in potato tubers for culinary and industrial processes.
Another alternative to generate genome-edited plants lacking any DNA integration is delivering CRISPR/Cas9 as ribonucleoproteins (RNPs) into cells.To test this technique in potato, Andersson et al. (2018) employed the same previous approach (knockout of the StGBSS gene function) but using ribonucleoproteins for protoplast transfections rather than plasmid DNA.Briefly, the authors obtained regenerated plants with the specific desired mutations, without DNA integration, but with the advantage of a higher frequency of transgene-free mutated lines, which can reduce the size of screening populations.The development of genome-edited plants that do not possess any foreign DNA (transgene) would be an essential step to increase consumers' acceptance, as well as to reduce the cost and time for releasing new cultivars since these plants may not be subject to the current regulatory approval process applied to genetically engineered plants.
A Koltun et al.

FINAL CONSIDERATIONS
The CRISPR/Cas technology is a powerful tool for creating genetic variability in a precise and targeted way, representing a new era in crop breeding.The system is versatile, fast and low-cost, allowing genome editing strategies to be more accessible and efficient than other technologies (mutagenesis, ZFNs, and TALENs).Its effectiveness has already been proven in several species, many of them of commercial interest, so the speed of diffusion of the technique in the scientific and commercial environment depends more on the regulatory aspects related to biosafety.The possibility of generating an edited non-transgenic plant may facilitate the development of biotechnological crops, saving time and money and thus broadening the range of public and private companies capable of working on different species and traits, as well as increasing the acceptance biotechnological crops by the population that still has some resistance to transgenics.
Several studies with horticultural species have already been carried out as proof of concept but also addressing economically important traits, showing the applicability of the technique.Some horticultural crops still lack basic requirements for the CRISPR/Cas tool to be applied: information about the genome and established protocols for genetic transformation and in vitro regeneration.Therefore, expanding knowledge on horticultural crops will allow the precise targeted engineering of endogenous genetic traits and the introduction of new traits, possibly reshaping global horticulture and assuming a leading role towards food security.

Figure 1 .
Figure 1.Schematic representation of the process to develop an edited non-transgenic plant.Step 1) Plant transformation, introduction of the expression cassette containing the sequences of a Cas nuclease and a sgRNA via Agrobacterium tumefaciens;Step 2) Selection of the transformed cells that received the system sequence at chromosome 1 (randomly); Step 3) Expression of the components Cas and SgRNA of the system in the cytoplasm; Step 4) Cas and the sgRNA form a complex and are directed to the nucleus where it performs a DBS at the DNA target inchromosome 2; Step 5) chromosome independent segregation in the crossing generations (2 nd Mendel's law) makes it possible to select plants with chromosome 1 without the CRISPR/Cas insertion and Chromosome 2 carrying the edition.Londrina, UEL, 2018.

Figure 2 .
Figure 2. Number of papers reporting the use of the CRISPR/Cas system in plants from 2013 to 2017.Londrina, UEL, 2018.