Please wait...




Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters


Methods and Materials for High Throughput Testing of Mutagenized Allele Combinations


This application claims the benefit of United States Provisional Application No. 62/294,539, filed February 12, 2016, herein incorporated by reference in its entirety.


This document relates to methods and materials involved in improving traits in organisms. For example, this document provides plants and materials and methods for making plants and plant products, where crops of the cultivated plants achieve improved agronomic characteristics or plant material quality.


The sequence listing that is contained in the file named "CRES032WO_ST25.txt" which is 127 kilobytes as measured in Microsoft Windows operating system and was created on February 7, 2017, is filed electronically herewith and incorporated herein by reference.


Modern elite varieties of cultivated plant species are generally highly bred, meaning that they have undergone a large number of cycles of artificial selection for improved agronomic traits. This process has resulted in the accumulation of a large number of favorable alleles in elite genetic backgrounds. Hence, for further improvements to elite performers, much of an existing genetic material needs to be maintained. But, generation of new genetic diversity by means of mutagenesis or wide crosses often begins with massive changes of the genetic material.

A laborious isolation and evaluation process generally results in relatively few new loci that are sufficiently favorable to justify incorporation into improved elite varieties. Moreover, specific pairs of allelic variants may exhibit positive epistasis, but they may elude identification if their occurrence emerges only from random mutagenesis or recombination. A need exists for an improved system for generation of saturation or near- saturation mutagenesis of agronomic trait candidate loci and for testing the effect of combined mutations, preferably within an otherwise uniform elite genetic background.


This document provides methods and materials for improving one or more phenotypic features in an organism. For example, this document provides high throughput methods for identifying combinations of mutations that can be used to improve a phenotypic feature in an organism. As described herein, large populations of organisms (e.g., plants) containing different combinations of mutations can be grown side-by-side. These large populations of organisms are genetically nearly identical, except for the specific combinations of mutations in each plant. The particular combinations that result in desirable phenotypes can be identified based on improved quality or performance in the field or greenhouse or lab testing. Combinations of mutations can generate significant phenotypes as their effects may be additive or synergistic.

In one aspect, this document features a method for identifying combination of genetic mutations that improves a phenotype of a plant. The method includes selecting a plurality of genomic targets (e.g., 4, 5, 6 or more genomic targets); making a plant cell that has both gRNAs designed to mutate the selected genomic targets and a Cas polypeptide, so that a plant descended from the plant cell will have germline mutations; sexually crossing a first parental plant

comprising at least a subset of the germline mutations to a second parental plant to produce a progeny population; selecting at least one progeny plant the population as having an improved phenotype to obtain a selected progeny plant; and determining which mutations are present within the selected progeny plant, thereby identifying a combination of mutations that improves a phenotype of a plant. The method can include repeating all the steps by either selecting genomic targets determined to be mutated in the selected progeny, using in the sexual cross a parent related by lineage to the selected progeny, or both. The plants could be corn (Zea mays) and the genomic targets can comprise at least one of SEQ ID NOs: 26-39. The plants could also be sorghum (Sorghum bicolor), wheat (Triticum aestivum), or rice (Oryza sativa).

In some embodiments at least some of the different gRNAs are designed to mutate distinct residues of the same genomic target, n some embodiments at least some of the different gRNAs are designed to mutate residues within conserved sequences of paralogous genes. In some embodiments a plant cell is made by inserting gRNA-expressing transgenes. In some embodiments a plant cell is made by adding a Cas9 polypeptide-expressing transgene, which may be accomplished by crossing. In some embodiments a plant cell is made by contacting the cell with pre-assembled gRNA-Cas9 ribonucloeoproteins. In some embodiments the first parental plant is a progeny of selfing a plant with germline mutations. In some embodiments the first parental plant is a progeny of a cross of the plant having germline mutations to a wild type plant or to another plant, so that the germline mutations of the first parental plant are

heterozygous. In some embodiments the first parental plant does not have a Cas9 polypeptide-expressing transgene. In some embodiments in the second plant also has germline mutations. In some embodiments the first and second parental plants are isogenic and belong to

complementary heterotic groups. In some embodiments the first parental plant or the second parental plant is cytoplasmically male sterile.

Selecting at least one individual with an improved phenotype from the progeny population may be based at least in part on performance of the plant under field testing conditions. The selection may also be based at least in part on other criteria such as the selected plant's water use efficiency, nitrogen use efficiency, seed oil content, or plant density stress performance. In some embodiments, the progeny population from which an individual is selected may be itself be selected by genotyping, for example by seed chipping. A collection of seeds can be made with embryonic cells having a combination of genetic mutations identified according to the methods described.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used to practice the invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

The details of one or more embodiments of the invention are set forth in the

accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims. The word

"comprising" in the claims may be replaced by "consisting essentially of or with "consisting of," according to standard practice in patent law.


Figure 1: Progenitor plants have multiple gRNA-expressing transgenes or a Cas9-expressing transgene. When crossed, the progeny combining the expression of Cas9 and gRNAs produces germline mutations of the gRNA targets. A cross with a wild type plant of a similar genetic background allows for segregation of Cas9 and gRNA transgenes, as well as making the mutations heterozygous. Crossing plants heterozygous for the mutations, here shown on the female side, with plant of a complementary heterotic group, results in formation of hybrids that recombine the mutated alleles. These hybrids can be phenotyped for the traits of interest to select favorable combinations of mutations.


This document relates to methods and materials for identifying or optimizing

combinations of mutations that improve one or more phenotypic features of an organism. For example, the methods described herein can be used in plants to improve grain yield; tolerance to an abiotic stress such as drought stress, osmotic stress, or nitrogen deficiency; soil aluminum; cold stress; frost stress; density stress; heat stress; oxidative stress; low light tolerance; herbicide stress; as well as improved water use efficiency; nitrogen use efficiency; phosphate use efficiency; seed oil or protein content; lignin content; biotic or pest resistance; biomass;

heterosis; chemical composition such as higher percentage of sucrose; plant architecture such as increased tillering or branching, decreased or increased apical dominance, or increased root mass; flowering time; and/or biofuel conversion properties in a plant. "Water use efficiency,"

"nitrogen use efficiency," or "phosphate use efficiency" refers to increased yield under the same levels of input, i.e., same level of water, nitrogen, or phosphate.

In general, the methods described herein can include obtaining first and second parental organisms, wherein at least one of the parents includes a plurality of mutations introduced by genome editing, sexually crossing the parent organisms to produce a progeny population, and identifying the combinations of mutations that improve a phenotypic feature. In some cases, one of the parental organisms can be a wild type plant. In some cases, each parent can be mutated and can include one or more mutations. As described in more detail below, the first and/or second parental organisms can be heterozygous for the mutations, and the gametes of one or both parental organism can include independently segregating subgroups of the plurality of mutations.

Mutations are generated in selected genomic targets of an organism (e.g. a crop plant) by genome editing using a CRISPR/Cas system. Accordingly, a guide RNA (gRNA) that is directed to a residue of the genomic target and a Cas endonuclease must be simultaneously present in the same cell as the a genomic target to be mutated. The gRNA binds the endonuclease and guides it to the genomic target at the location where it is complementary to the engineered gRNA sequence. After the endonuclease cleaves the genomic target, many types of mutations (e.g. insertions, deletions, substitutions) around the nucleic acid residues of the target of the respective gRNA will be formed, often by the error-prone Non-Homologous End Joining (NHEJ) pathway. Consequently, the same gRNA/Cas can act to make multiple mutations, and different mutations can produce different phenotypes.

In some embodiments, as depicted in Figure 1, gRNAs and Cas endonucleases are produced by the expression of transgenes. Consequently, Cas expression can be maintained in a

separate plant from the one or many gRNA-expressing transgenes, as needed for carrying out the present methods. Then, Cas can be brought together with selected gRNAs by crossing two parents having these transgenes, as long as their co-expression will occur at least in germline cells. In this illustrated progeny, Cas and gRNA-expressing transgenes are hemizygous, and their simultaneous expression of both gRNAs and Cas molecules generates mutations. But individual progeny of this cross will not be uniformly mutated, since the same gRNA can generate different mutations. The progeny population may be useful at this point to identify combinations of mutants with improved phenotypes, especially if co-expression occurs early in development such as in the fertilized egg or the embryo. In some cases, however, it is desirable to segregate at least the Cas transgene so as to both avoid additional mutations and form different combinations of the mutations. This can be accomplished with a different cross, illustrated with a wild type (wt) parent in Figure 1. The progeny of this second cross will have the mutations in a heterozygous state. Consequently, a cross of this progeny with another plant will result in recombination of all the mutations, and the progeny will be a population having different segregating mutations. This population, noted as Fl in Figure 1, can be phenotyped for identifying outstanding pairs or combinations of mutations.

The methods described herein provide a number of advantages when compared to alternative solutions, although not all advantages may be present in a specific embodiment.

For most breeding objectives, commercial breeders work within germplasm that is often referred to as the cultivated type. This germplasm is easier to breed with because it generally performs well when evaluated for agronomic performance. The performance advantage the cultivated type provides is sometimes offset by a lack of allelic diversity. This is the tradeoff a breeder accepts when working with cultivated germplasm: better overall performance, but a lack of allelic diversity. Breeders generally accept this tradeoff because progress is faster when working with cultivated material than when breeding with genetically diverse sources.

In contrast, when a breeder makes either intra- specific crosses, or inter- specific crosses, a converse trade off occurs. In these examples, a breeder typically crosses cultivated germplasm with a non-cultivated type. In such crosses, the breeder can gain access to novel alleles from the non-cultivated type, but may have to overcome the genetic drag associated with the donor parent. Because of the difficulty with this breeding strategy, this approach often fails because of fertility and fecundity problems. The difficulty with this breeding approach extends to many crops, and is exemplified with an important disease resistant phenotype that was first described in tomato in 1944 (Smith, Proc. Am. Soc. Hort. Sci. 44:413-16). In this cross, a nematode disease resistance was transferred from L. peruvianum into a cultivated tomato. Despite intensive breeding, it was not until the mid-1970's before breeders could overcome the genetic drag and release successful lines carrying this trait. Indeed, even today, tomato breeders deliver this disease resistance gene to a hybrid variety from only one parent. This allows the remaining genetic drag to be masked.

Some phenotypes are determined by the genotype at one locus. These simple traits, like those studied by Gregor Mendel, fall in discontinuous categories such as green or yellow seeds. Most variation observed in nature, however, is continuous, like yield in field corn, or human blood pressure. Unlike simply inherited traits, continuous variation can be the result of polygenic inheritance. Loci that affect continuous variation are referred to as quantitative trait loci (QTLs). Variation in the phenotype of a quantitative trait is the result of the allelic composition at the QTLs and the environmental effect. The heritability of a trait is the proportion of the phenotypic variation attributed to the genetic variance. This ratio varies between 0 and 1.0. Thus, a trait with heritability near 1.0 is not greatly affected by the

environment. Those skilled in the art recognize the importance of creating commercial lines with high heritability agronomic traits because these cultivars will allow growers to produce a crop with uniform market specifications.

Consequently, mutations conferring improved agronomic traits are a powerful tool in the development of new and improved cultivars. Mutations are defined genetic alterations that do not require segregation from linked regions in order to avoid genetic drag. And again due to their genetic nature, contributions of mutations to a defined trait have high heritability. As explained in more detail below, however, the precise impact of a mutation or combination of mutations needs to be experimentally measured to understand the extent to which it depends on the relevant QTLs present in the germplasm in which they are tested. Understanding any QTL-dependent mutation impact is helpful, for instance, in cases where a mutation has a phenotypic effect in a heterotic genetic background that is significantly different in magnitude from the corresponding effect in plants with inbred depression.

Exploring the phenotypic effect of many stacked mutations, rather than single mutations, is more likely to result in finding mutations resulting in significant yield or quality

improvements. The genome of cultivated plants, for example, comprises a background system of complex molecular interactions. For a mutation to boost a trait its products need to fit into the complex, regulated downstream networks appropriately. If the genetic background changes then the effect of the mutation may change too. For this reason specific mutations sometimes fail to achieve the desired effect in all genetic backgrounds and environments. A change to a single component of a very complex system is unlikely to have a dramatic positive effect; several distinct alterations, on the other hand, as with a stack of mutations, are more likely to result in an enhanced or synergistic positive effect, and/or diminished negative features of a mutation-caused pleiotropic phenotype.

The methods described herein make it possible to produce and test in parallel a high number of mutations in different combinations. Any phenotypic feature could be affected by a large number of candidate mutations, and a much larger number of combinations or stacks mutations. But the phenotype of individual combinations, which may or may not turn out to be additive or even synergistic when compared to the phenotype of single mutations, is generally unpredictable, so testing a large number of combinations is necessary. Hence, the high-throughput methods described herein are useful for quickly sorting through large numbers and identifying the combinations of mutations that improve one or more phenotypic features. In other words, existing methods, such as random mutagenesis, would produce too rarely similar combinations as made possible by the methods presented here, and these combinations would be difficult to sort out as their effect would be entangled with the deleterious effects of other combinations. It is worth noting that while unpredictability of produced mutations with some CRISPR/Cas mutagenesis systems can actually be problematic for their application in other technological areas, it is actually advantageous for the methods presently described.

Generating new mutations, such as by random mutagenesis with mutagens or by wide crosses, while routinely accomplishable, is also a process that by its nature adds challenges to studying comparative performance of plants. This is because multiple unrelated mutations accumulate and generally create a range of mutant phenotypes in independently selected plants. Mutations causing genetic drag need to be segregated first to understand the potential impact of individual mutations or their combinations. Hence, compared to use of random mutagenesis, the procedures described here often require a limited number of backcrosses, if any, which cuts

down on the amount of labor necessary to make and characterize the materials, but more importantly provide results such that the relative performance of combined mutations can be reliably scored. Moreover, mutations or combinations are "recyclable", i.e. once made and characterized they are likely to find use in multiple seasons and experimental setups. For example, interesting mutations only need to be made once, and then they can be used repeatedly in combination with many other different mutations, and may need to be introgressed only once into any parental germplasm of interest. This feature is especially convenient for testing in elite germplasm because of the added effort sometimes required to introgress any mutation into a uniform and commercially relevant genetic background. In some aspects, the methods presented here maintain the benefits of random mutagenesis without the drawbacks. They enable testing in elite backgrounds of large number of combinations of only the most likely yield and quality impact candidate mutations. Since the impact of mutations can be dependent on the genetic background, testing in directly elite backgrounds is advantageous, by eliminating from consideration mutations of diminished impact in elite materials. Phenotyping of candidates is also convenient because of the uniform genetic background of the siblings that make up testing populations, which can be planted in proximity so that improved phenotypes can be easily scored.

In many cases, testing populations are made based on isogenic backgrounds so as to eliminate background genetic noise that would otherwise confound data interpretation. But in other cases, especially when the effect of a limited number of combined mutations is to be understood, the genetic background may be intentionally diverse. For example, a promising stack of mutations could be observed for performance in a segregating F2 population and

subsequent generations, thus allowing selection and production of parents capable to perform especially well in the presence of a specific combination of mutations.

Using the presented methods, a phenotypic measurement can be the yield of harvestable material under typical field cultivation conditions, i.e. without an intentionally applied selection pressure. This data is certainly relevant from a product performance perspective, for identifying undesirable interactions of stacked mutations, and when stacking mutations affecting different traits that cannot be revealed by a single assay. But in addition, while some mutations or stacks provide a survival or yield advantage under high selection pressure, they are known to otherwise have a negative impact when grown under typical cultivation conditions. Moreover, the populations of plants produced by the present methods are well-suited for comparative studies of related combinations. When testing side-by- side sibling plants that are otherwise genetically uniform but differ only with regard to having distinct combinations of a limited original pool of mutations, stacks of outstanding phenotypic impact can be readily identified. In some embodiments, populations produced according to the methods provided herein can be tested for field performance similarly to screening of segregating populations by plant breeders. This way, the effect of high numbers of combined mutations can be simultaneously observed, often in a commercially relevant, elite genetic background, which may be made up of defined heterotic groups and/or QTLs for specific traits, and so well-performing mutants that surpass

commercially relevant thresholds can be more easily identified. By using the methods described herein, useful combinations of mutations become self-revealing, circumventing the general unpredictability of the phenotype of stacked mutations.

Genomic targets

Genomic targets for mutation are contiguous chromosomal DNA regions generally encoding expressed sequences. Most often they are genes comprising a transcribed sequence which typically comprises a polypeptide-encoding sequence, and regulatory regions.

Regulatory region refers to a nucleic acid having nucleotide sequences that influence

transcription or translation initiation and rate, and stability and/or mobility of a transcription or translation product. Regulatory regions include, without limitation, promoter sequences, enhancer sequences, response elements, protein recognition sites, inducible elements, protein binding sequences, 5' and 3 ' untranslated regions (UTRs), transcriptional start sites, termination sequences, polyadenylation sequences, introns, and combinations thereof. A regulatory region typically comprises at least a core (basal) promoter. A regulatory region also may include at least one control element, such as an enhancer sequence, an upstream element or an upstream activation region (UAR).

Selection of genomic targets for mutation can be done rationally rather than randomly. For example, based on existing data, the targets can be known or inferred as likely to affect a trait, such as morphological development of a plant part that may increase yield or environmental stress resistance. The types of data useful in identifying genomic targets is dependent on the phenotype to be improved, but it may be from the location of quantitative trait loci (QTL), transgenic phenotypes caused by overexpression of sequences, participation in relevant signaling or metabolic pathways, involvement in relevant physiological process, or phenotypes of characterized mutations.

Genomic targets can be selected based on data from different species if needed, as functional homologs of a locus. Accordingly, one or more genomic targets in a species of interest can often be identified by sequence similarity to sequences in other species for which pertinent data exists. In addition to sequence similarity, conserved domains as defined by Pfam descriptions, synteny, and/or reciprocal BLAST results may be used in identifying suitable genomic targets. A functional homolog is a polypeptide that has sequence similarity to a reference polypeptide, and that carries out one or more of the biochemical or physiological function(s) of the reference polypeptide. A functional homolog and the reference polypeptide may be natural occurring polypeptides, and the sequence similarity may be due to convergent or divergent evolutionary events. As such, functional homologs are sometimes designated in the literature as homologs, or orthologs, or paralogs. The term "functional homolog" is sometimes applied to the nucleic acid or gene that contains a functionally homologous polypeptide.

Functional homologs can be identified by analysis of nucleotide and polypeptide sequence alignments. For example, performing a query on a database of nucleotide or polypeptide sequences can identify homologs of interest as a genomic target. Sequence analysis can involve BLAST, Reciprocal BLAST, or PSI-BLAST analysis of nonredundant databases using a reference sequence of defined interest. Amino acid sequence is, in some instances, deduced from the nucleotide sequence. Those polypeptides in the database that have greater than 40% sequence identity are candidates for further evaluation for suitability as genomic targets. Amino acid sequence similarity allows for conservative amino acid substitutions, such as substitution of one hydrophobic residue for another or substitution of one polar residue for another. If desired, manual inspection of such candidates can be carried out in order to narrow the number of candidates to be further evaluated. Manual inspection can be performed by selecting those candidates that appear to have domains present in genomic targets, e.g., conserved functional domains.

Conserved regions can be identified by locating a region within the primary amino acid sequence of a genomic target candidate polypeptide that is a repeated sequence, forms some secondary structure (e.g., helices and beta sheets), establishes positively or negatively charged domains, or represents a protein motif or domain. See, e.g., the Pfam web site describing consensus sequences for a variety of protein motifs and domains on the World Wide Web at and A description of the information included at the Pfam database is described in Sonnhammer et ah, Nucl. Acids Res., 26:320-322 (1998); Sonnhammer et ah, Proteins, 28:405-420 (1997); and Bateman et ah, Nucl. Acids Res., 27:260-262 (1999). Conserved regions also can be determined by aligning sequences of the same or related polypeptides from closely related species. Closely related species preferably are from the same family. In some embodiments, alignment of sequences from two different species is adequate.

Typically, polypeptides that exhibit at least about 40% amino acid sequence identity are useful to identify conserved regions. Conserved regions of related polypeptides exhibit at least 45% amino acid sequence identity (e.g., at least 50%, at least 60%, at least 70%, at least 80%, or at least 90% amino acid sequence identity). In some embodiments, a conserved region exhibits at least 92%, 94%, 96%, 98%, or 99% amino acid sequence identity.

Once a genomic target is selected, it typically contains many distinct residues that may be mutated by genome editing techniques. Creating mutations of a genomic target at many residues is desirable so as to generate essentially all the potential phenotypes from that target. The mutations are introduced generally 3-4 residues upstream of the Protospacer Adjacent Motif (PAM), which is for example 5'NGG3' for the Streptococcus pyogenes Cas9. Consequently, any PAM sequence on either strand of the genomic target can be used in guiding an adjacent

mutation. The residue to be mutated is often part of a polypeptide-coding region, although it could also be part of an intron or regulatory region. Mutating distinct residues of the same genomic target may be accomplished by introducing double or multiple mutations into the target with different gRNAs directed to mutate different residues, or by introducing single distinct individual mutations in different progenitor cells.

The phenotypes caused by individual mutations can range from indistinguishable from wild type to a most severe knockout effects. More rarely, gain of function mutations are observed. Phenotypes result from changes in function caused by the mutations. Depending on the nature of the target, many changes do not entirely obliterate (i.e. knockout) the function of the underlying wild type target, but may alter its expression level, its encoded polypeptide's affinity for substrate or for a protein complex to which it belongs, the cellular localization of an expression product, and/or its regulation in response to stimuli affecting its regulatory network. Mutations introduced close to the 5' terminus of a coding sequence are more likely to result in knockout phenotypes than mutations close to the 3' terminus. Consequently, when multiple distinct residues of a genomic target are selected, it is often desirable to select residues proximal to the 5' terminus and to the 3' terminus of the coding sequence.

In some cases, mutations can be directed to residues in conserved regions, so as to simultaneously mutate not only the main selected genomic target, but also paralogous genes sharing the conserved region. This approach can be fruitful, for example, if paralogous genes can substitute for each other to some extent. Alternatively, mutations can be directed to residues in unique regions to avoid mutating any paralogous genes.

Many genomic targets can be selected for mutation and testing. In general, it is useful to test all the different combinations of at least four targets simultaneously. But the number of targets in a development program can be much larger, i.e. 10, 15, 20, or 25 or more targets can be mutated at various residues and their combined phenotypic effect investigated. The choice of the multiple targets to be combined can be at least in part random, so as to enable observations of synergies between different phenotypes that might not be predictable. But, the combination of targets can also be at least in part non-random, so as to enhance the likelihood that the combined mutations would interact with each other and thus produce a new phenotype when combined. For example, many different targets can be selected based on their likelihood to produce a salt tolerant phenotype. Many mutations can be produced in each target, and different mutations of the targets can be recombined to identify those pairs or combinations that complement each other so as to perform exceptionally well on salt tolerance assays.

A gRNA can target one or more genes encoding a polypeptide necessary for elaboration of cell wall polysaccharides. Non-limiting examples of polypeptides necessary for elaboration of cell wall polysaccharides include polypeptides that function in the lignin pathway (e.g., phenylalanine ammonia-lyase (PAL), cinnamate 4-hydroxylase (C4H, 4-coumarate:coa ligase (4CL), p-coumarate 3-hydroxylase (C3H), p-hydroxycinnamoylcoa: quinate/shikimate p-hydroxycinnamoyltransferase (HCT), caffeoyl-coa omethyltransferase (CCOAOMT), cinnamoyl-coa reductase (CCR), ferulate 5-hydroxylase (F5H), caffeic acid o-methyltransferase (COMT), and cinnamyl alcohol dehydrogenase (CAD), and polypeptides that function in cellulose synthase. A gRNA can target one or more genes encoding polypeptides involved in hormone biosynthesis (e.g., repressors and/or critical enzymes). For example, reducing or eliminating the function of a repressor of a hormone biosynthesis pathway can be effective to increase hormone levels. In some cases, a repressor of a hormone biosynthesis pathway can be a corepressor. For example, reducing or eliminating the function of a polypeptide that brings about critical steps in hormone biosynthesis can be effective to decrease hormone levels. Non-limiting examples of polypeptides involved in hormone biosynthesis include polypeptides involved the gibberellin (GA) pathway, the brassinosteroids (BR) pathway, the indole-3-acetic acid (IAA) pathway, the jasmonic acid (J A) pathway, the abscisic acid (ABA) pathway, the salicylic acid (SA) pathway, the cytokinin pathway, and the ethylene pathway. Exemplary targets of the GA pathway include, for example, GA20-oxidase, GA3 -oxidase, GA2-oxidase, gibberellin insensitive dwarf (GID), and other polypeptides described in, for example, Park et al.

(WO2013/086499, published June 13, 2013). For example, reduction or elimination of a repressor of the GA pathway (e.g., GA2-oxidase) can be effective to activate the GA response. For example, reduction or elimination of an activator of the GA pathway (e.g., GA20-oxidase) can be effective to repress the GA response. In some embodiments, a gRNA can be designed to target a combination of one or more repressors and/or co-repressors of a hormone biosynthesis pathway and one or more polypeptides that bring about critical steps in hormone biosynthesis. A gRNA can target one or more genes encoding a polypeptide that represses cell division (e.g., cell cycle regulators). Non-limiting examples of polypeptides that repress cell division include cyclins (e.g., Arabidopsis CDC2aAt, CDC2bAt, CYCB 1 ; 1, and alfalfa CDC2fM and CYCB2;2, and their homologs in other species) and cyclin-dependent kinase (CDKs).

One or more gRNA-expressing transgenes can be used to reduce or eliminate function of a gene (e.g. an endogenous gene) in a manner that enhances biocontainment (e.g., prevent outflow of the transgene into nature). Non-limiting examples of genes that can be targeted with a gRNA include, genes encoding polypeptides causing sterility (e.g., polypeptide involved in seed development), genes encoding herbicide tolerance polypeptides, genes encoding pesticide tolerance (e.g., insect resistance) polypeptides, transgenes encoding polypeptides providing agronomic traits, and transgenes encoding polypeptides involved in cell wall conversion and digestion. A gRNA can target one or more genes encoding a polypeptide causing sterility. For example, a polypeptide causing sterility can be a polypeptide involved in seed development. Non-limiting examples of polypeptides involved in seed development include FIE, AP2, INO, ANT, the polypeptide encoded by the LEC2 gene, and HAP3-type CCAAT-box binding factor (CBF) subunit.

A gRNA can target one or more genes encoding an herbicide tolerance polypeptide. Herbicide tolerance is also sometimes referred to as herbicide resistance. Non-limiting examples of herbicide tolerance polypeptides include a polypeptide encoded by a polypeptide encoded by a phosphinothricin acetyl transferase (PAT) gene, a bialaphos resistance (BAR) gene, 5-enolpyruvyl-3-phosphoshikimate synthase (EPSPS), acetolactate synthase (ALS), acetyl coenzyme A carboxylase (ACCase), dicamba mono-oxygenase (DMO), aryloxyalkanoate dioxygenase- 12 (aad-12), and 4-hydroxyphenylpyruvate dioxygenase (HPPD).

A gRNA can target one or more genes encoding a pesticide tolerance polypeptide. For example, a pesticide tolerance polypeptide can be an insect resistance polypeptide. Nonlimiting examples of pesticide tolerance polypeptides include CrylAb, Cry 1 Ac, Cry 1 A.105, Cry IF, Cry2Ab, Cry3Bbl, Cry34Abl, Cry35Abl, mCry3A, and VIP3.

A gRNA can target one or more genes encoding a polypeptide conferring a desirable trait. For example, a desirable trait can be an agronomic trait. Non-limiting examples of agronomic traits include increased yield, drought tolerance, cold tolerance, tolerance to

environmental stresses, enhanced nitrogen use, and male sterility. Other desirable traits can include, for example, pathogen (e.g., virus, fungus, bacterium, and/or nematode) resistance, and product quality traits (e.g., delayed fruit ripening, altered amino acid profile, altered oil profile, modified seed storage proteins, enhanced floral characteristics for ornamentals, 5 increased solids in fruit).

Seq id nos 6-10 and 25-39 provide some examples of genomic targets that may be selected and either mutated in their respective species or first used to identify similar genomic targets in other species of interest.


Aspects of some embodiments relate to a transgenic plant (e.g., a parent plant or a progeny plant) that includes to at least one nucleic acid having a promoter operably linked to a gRNA sequence. "Operably linked" refers to the positioning of a regulatory region and a sequence to be transcribed in a nucleic acid so that the regulatory region is effective for regulating transcription of the sequence. Expressed gRNAs can target particular nucleic acid sequences (e.g., an endogenous gene) at which a Cas enzyme can induce a double stranded break. A gRNA can include a gRNA scaffold sequence and a gRNA targeting sequence, and can be designed to target a nucleic acid sequence within the genetic material of a plant (including the nuclear chromosomes, transgenic, choloroplastic, or mitochondrial sequences). A gRNA scaffold sequence can bind a Cas enzyme (e.g., Cas9) thus guiding the Cas enzyme to a target site at which a double stranded break is desired. See, e.g., Ran et al. (2013 Nat Protoc. 8(11):2281- 2308). A gRNA targeting sequence can be a nucleic acid sequence that can hybridize to a target sequence within the genetic material of a plant (e.g., a gene within a plant). In some cases, a

gRNA targeting sequence can hybridize to a coding or a noncoding strand of a target gene; thus, a gRNA targeting sequence can include a portion of a genomic target sequence or

complementary to a portion of a genomic target. Hybridization refers to a reaction in which two single stranded nucleic acid molecules or regions of molecules form a complex that is stabilized via hydrogen bonding between complementary bases of the nucleotide residues. A gRNA targeting sequence that hybridizes to a genomic target can be of any appropriate length that is sufficient to promote hybridization, a double stranded break, and double stranded break repair (e.g., nonhomologous end joining) at the desired site. In some cases, the gRNA targeting sequence can include a portion of a genomic target or the full length of a genomic target. A gRNA targeting sequence can be from about 5 to about 45 nucleotides in length (e.g., from about 5 to about 45, from about 8 to about 40, from about 10 to about 35, from about 13 to about 30, from about 15 to about 27, from about 17 to about 25, from about 18 to about 24, or from about 19 to about 23 nucleotides in length). For example, the gRNA targeting sequence can be at least 5, at least 8, at least 10, at least 13, at least 15, at least 17, at least 18, at least 19, or at least 20 nucleotides in length. For example, the gRNA targeting sequence can be no greater than 45, no greater than 40, no greater than 35, no greater than 30, no greater than 27, or no greater than 25 nucleotides in length. In some cases, the gRNA targeting sequence includes 20 nucleotides. The amount of sequence identity shared by a gRNA targeting sequence and a desired site in a genomic target can vary. For example, the amount of sequence identity can be at least 50%, 55%, 60%, 65%, 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 30 96%, 97%, 98%, 99% or 100% sequence identity. Methods for determining hybridization conditions (including complementarity and percent sequence identity) that can used as described herein include,

without limitation, those are described elsewhere (e.g., Sambrook et al. Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, 1989; and Ausubel et ah , Current Protocols In Molecular Biology, John Wiley & Sons, New York, 1987).

In some cases, the methods and materials provided herein (e.g., vectors) can include using multiple gRNAs directed to at least one target residue site within a gene (e.g., an endogenous gene) to reduce or eliminate function of the target gene upon mutagenesis. In some cases, a nucleic acid molecule can have at least one promoter operably linked to one gRNA-expressing sequence (e.g. , one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, or more gRNA-expressing sequences). In some cases, a nucleic acid molecule can have five gRNA-expressing transgenes. Multiple (e.g., two or more) gRNA-expressing transgenes provided herein can be directed to a single genomic target or can be designed to target multiple (e.g., two or more) genomic targets. In embodiments where multiple gRNA-expressing transgenes are directed to a single target gene, the gRNA-expressing transgenes can be directed to the same site within the genomic target, or the gRNA-expressing transgenes can be directed to different sites within the genomic target. In embodiments where multiple gRNA-expressing transgenes are directed to multiple genomic targets, the gRNA-expressing transgenes can each be directed to an independent genomic target.

Multiple (e.g., two or more) gRNAs directed to at least one target site within a gene(e.g. an endogenous gene) can be provided via a single nucleic acid molecule (e.g., in tandem expression cassettes) or can be provided via multiple nucleic acid molecules (e.g. on more than one expression cassette). In some cases, a nucleic acid molecule can have five gRNA-expressing transgenes provided via tandem expression cassettes. One or more gRNA-expressing transgenes provided herein can be used to reduce or eliminate function of a gene (e.g. an endogenous gene)

in a manner that improves plant health (e.g., to provide desirable agronomic traits). Non-limiting examples of genomic targets that can be altered with a gRNA include genes necessary for elaboration of cell wall polysaccharides, genes that are repressors or co-repressors of hormone biosynthesis pathways, genes that bring about critical steps in hormone biosynthesis, and genes that repress cell division.

Nucleic Acid Molecules

In some embodiments, a transgenic plant can be a parent plant including a nucleic acid molecule having a first promoter operably linked to at least one transgene. For example, a parent plant can include a nucleic acid molecule having a promoter (e.g., a ubiquitously expressing promoter, which may direct transcription by Pol III, such as the corn U6 (SEQ ID NO: 3) or Sorghum U3 (SEQ ID NO: 4) promoters) operably linked to at least one gRNA-sequence to be expressed, and a parent plant can include a second nucleic acid molecule having a second promoter (e.g. expressing at least in germline cells) operably linked to a Cas-encoding sequence. It is typically desired, but not always necessary, that the at least one transgene does not itself cause any phenotype in the parent plant, for example via insertional effects. Expression of both a gRNA and a Cas allows for the formation of a gRNA/Cas complex capable of introducing a double strand break in a target site within a genome (e.g., within a gene). The double stranded break can lead to introduction of at least one mutation in a genomic target such that the mutation confers a modified function of that genomic target. As mutations occur in germlines, progeny plants inherit the modified function of the target gene.

A promoter refers to a nucleic acid capable of driving expression of another nucleic acid (e.g., a coding nucleic acid). A promoter is operably linked to another nucleic acid when it is capable of driving expression of that nucleic acid fragment. The choice of promoter to be included in a nucleic acid molecule described herein depends upon several factors, including, but not limited to, efficiency, selectability, inducibility, desired expression level, and cell- or tissue-preferential expression. In different embodiments, the Cas-encoding sequence can be placed under the control of any of a number of promoters that are capable of directing expression in at least some progenitor cells of germline tissues, so that egg and pollen cells comprise the mutations.

Preparation of the nucleic acids disclosed herein can be accomplished using techniques in molecular biology, biochemistry, chromatin structure and analysis, computational chemistry, cell culture, recombinant DNA, and related fields. These techniques are described, for example, in Sambrook et al. Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor

Laboratory Press, 1989; and in Ausubel et al., Current Protocols In Molecular Biology, John Wiley & Sons, New York, 1987.

CRISPR-associated (Cas) genes

CRISPR/Cas systems are known in the art and can be engineered for directed genome editing. Cas genes encode RNA-guided DNA endonuclease enzymes capable of introducing a double strand break in a double helical nucleic acid sequence. Nucleases engineered to introduce single strand breaks can also be suitably adapted for use with the present invention. The Cas enzyme can be directed to make the double stranded break at a target site within a gene using a guide RNA. A Cas enzyme can be guided by a guide polynucleotide (e.g., a guide RNA) to recognize and introduce a sequence- specific double strand break at a site determined by the guide polynucleotide. A Cas enzyme can be from any appropriate species (e.g., an archaea or

bacterial species). For example, a Cas enzyme can be from Streptococcus pyogenes, Pseudomonas aeruginosa, or Escherichia coli. In some cases, a Cas enzyme can be a type I {e.g., type IA, IB, IC, ID, IE, or IF), type II (e.g., IIA, IIB, or IIC), or type III (e.g., IIIA or IIIB) Cas enzyme. The encoded Cas enzyme can be any appropriate homolog or Cas fragment in which the enzymatic function (i.e., the ability to introduce a sequence- specific strand breaks in a double helical nucleic acid sequence) is retained. In some cases, a Cas enzyme can be codon optimized for expression in particular cells, such as dicot or monocot plant cells. See, for example, the CRISPR/Cas profiles database available on the National Center for Biotechnology Information website (available at In some embodiments, a Cas gene is from Streptococcus pyogenes. Examples of Cas genes that can be used as described herein include, without limitation, Cas3, Cas4, Cas6, Cas8a, Cas8b, Cas8c, Cas9, CaslO, CaslOd, Cmr5, Cpfl (Zetsche et ah, 2015 "Cpfl Is a Single RNA Guided

Endonuclease of a Class 2 CRISPR-Cas System" Cell 163, 759-771), Csel, Csm2, Csn2, and Csyl genes. In some embodiments, a Cas gene is a Streptococcus pyogenes Cas9 gene (SEQ ID NO: 1).

Any appropriate CRISPR/Cas system can be used as described herein. Examples of CRISPR/Cas systems that can used as described herein include, without limitation, those are described elsewhere (e.g. U.S. Pat. Nos. 8,697,359; 8,771,945; 8,795,965; 8,865,406; 10

8,871,445; 8,889,356; 8,889,418; 8,895,308; 8,906,616; 8,932,814; 8,945,839; 8,993,233;

8,999,641; 9,115,348; U.S. Pat. App. Pub. Nos. 2011/0223638; 2014/0068797; 2014/0302563; 2014/0315985; 2015/0152398; 2015/0284697; and Schaeffer et al. 2015 Plant Sci. 240: 130-42).

Additional features that can be used to control and/or enhance the CRISPR/Cas system include, for example, protospacer adjacent motifs, spacers (e.g., target spacers), and termination signals (see, e.g., Mali et ah, 2013 Science 339:823-826). A gRNA-expressing transgene can include a protospacer adjacent motif (PAM) sequence. Without being bound by theory, it is believed that PAMs to be important for type I (e.g., type IA, IB, IC, ID, IE, or IF) and type II (e.g., IIA, IIB, or IIC) CRISPR-Cas systems, but are not necessary in type III (e.g., IIIA or IIIB) CRISPR-Cas systems. For example, it is believed that a type I or type II Cas enzyme will recognize and cleave a gene sequence having a PAM sequence at the 3 '-end. A PAM sequence can be on a coding strand or a noncoding strand of a target gene. A PAM sequence on a coding strand can be, for example, 5'-NGG-3' where N is any nucleotide followed by two guanine (G) nucleotides or 5'-NGA-3' where N is any nucleotide followed by a guanine (G) residue and an adenine (A) residue. A PAM sequence on a non-coding strand can be, for example, 5'-CCN-3' where N is any nucleotide following two cysteine (C) residues. A nucleic acid molecule having a gRNA expressing transgene as described herein can also include at least one target spacer. Thus, a target spacer corresponding to a sequence upstream of a PAM can be used to ensure binding of a gRNA to a target site within a gene and enable Cas enzyme activity at a nearby cleavage site within the gene.

Transgenic Plants and Methods of Making Transgenic Plants

In some embodiments this document relates to transgenic plants having at least one nucleic acid molecule described herein (e.g., having a promoter operably linked to at least one sequence to be transcribed). As used herein, a transgenic "plant" can constitute part or all of a whole plant. For example, a plant can include plant cells, explants, seed, plants grown from said seed, and grain having at least one nucleic acid molecule described herein. A transgenic plant also refers to progeny of an initial transgenic plant provided the progeny inherits a nucleic acid molecule described herein.

Transgenic Plants

A transgenic plant provided herein can be a parent plant including a nucleic acid molecule having a first promoter operably linked to at least one transgene. A parent plant can include any combination of promoters and transgenes described herein. For example, a first parent plant provided herein can include a first nucleic acid molecule having a first promoter operably linked to a first sequence, a second parent plant provided herein can include a second nucleic acid molecule having a second promoter operably linked to a second sequence, and so on. Preferably, expression of a first or a second transgene in a parent plant does not cause any phenotype in a parent plant. As such, a parent plant can be chosen based on the absence of any phenotype resulting from expression of the transgene. A first parent plant can include a first nucleic acid molecule as described herein. For example, a first parent plant can include a first nucleic acid molecule having a first promoter operably linked to at least one first transgene. The first promoter can be a ubiquitous promoter (e.g. a Pol III promoter). The first transgene can be a gRNA-expressing transgene.

A second parent plant can include a second nucleic acid molecule as described herein.

For example, a second parent plant can include a second nucleic acid molecule having a second promoter operably linked to a second transgene. The second promoter can be a regulated promoter (e.g., a tissue-specific promoter or a developmentally- specific promoter), or a ubiquitously expressing promoter. The second transgene can be a Cas-expressing transgene. In some embodiments, a second parent plant can include a second nucleic acid molecule having a developmentally- specific floral meristem Zm Zap l promoter (SEQ ID NO: 5) operably linked to at least one Cas9-encoding sequence (SEQ ID NO: 1). Expression of at least one Cas9-e encoding sequence in, for example, the floral meristems of the second parent plant may produce germline cells harboring different mutations, and thus a single plant can give rise to progeny with different mutations in the same genomic target. But in general, mutations can be kept in germ cells by Cas and gRNA co-expression in cellular progenitors, thus ensuring that the mutations become heritable. Accordingly, co-expression of the gRNA-expressing transgene and a Cas expressing transgene can be designed to occur so as to edit the cells of the gametophytes, the generative or sperms cells in the pollen, or the megaspore mother cell or the egg cell in the embryo sac, or the zygote.

A transgenic plant can also be a progeny resulting from a cross between a first parent plant and parent plant as described herein. Progeny include descendants of a particular plant or plant line. Progeny of an instant plant include seed formed on Fl, F2, F3, and subsequent generation plants, seeds formed on BCl, BC2, BC3, and subsequent generation plants, or seeds formed on F1BC1, F1BC2, F1BC3, and subsequent generation plants. Seed produced by a transgenic plant can be grown and then selfed (or outcrossed and selfed) to obtain seed homozygous for a mutation or transgene of interest. Progeny can include transgenic seed produced by crossing a first parent plant and second parent plant as described herein as well as transgenic plants grown from those transgenic seed.

Methods for Making Transgenic Plants

Nucleic acid molecules as described herein can be introduced into a plant or plant cell by any appropriate means in order to establish a transgenic plant. A plant or plant cell can be transformed by having a construct integrated into its genome, i.e., can be stably transformed. Stably transformed cells typically retain the introduced nucleic acid with each cell division. A plant or plant cell can also be transiently transformed such that the construct is not integrated into its genome. Transiently transformed cells typically lose all or some portion of the introduced nucleic acid construct with each cell division such that the introduced nucleic acid cannot be detected in daughter cells after a sufficient number of cell divisions. Both transiently transformed and stably transformed transgenic plants and plant cells can be used in the methods described herein.

When transiently transformed plant cells are used, a reporter sequence encoding a reporter polypeptide having a reporter activity can be included in the transformation procedure and an assay for reporter activity or expression can be performed at a suitable time after transformation. A suitable time for conducting the assay typically is about 1-21 days after transformation, e.g., about 1-14 days, about 1-7 days, or about 1-3 days. The use of transient assays is particularly convenient for rapid analysis in different species, or to confirm expression of a heterologous biomass composition-modulating polypeptide whose expression has not previously been confirmed in particular recipient cells.

Techniques for introducing nucleic acid molecules into monocotyledonous and dicotyledonous plants are known in the art, and include, without limitation, Agrobacterium mediated transformation, viral vector-mediated transformation, electroporation and particle gun transformation, e.g., U.S. Patents 5,538,880; 5,204,253; 5,591,616; 6,013,863; and 6,329,571. If a cell or tissue culture is used as the recipient tissue for transformation, plants can be regenerated from transformed cultures by techniques known to those skilled in the art.

Growing Transgenic Plants

Transgenic plants can be grown in a manner suitable for the species under consideration, either in a growth chamber, a greenhouse, or in a field. Transgenic plants can be bred as desired for a particular purpose, e.g., to introduce a recombinant nucleic acid into other lines, to transfer a recombinant nucleic acid to other species, or for further selection of other desirable traits. Alternatively, transgenic plants can be propagated vegetatively for those species amenable to such techniques.

Transgenic plants can be grown in suspension culture, or tissue or organ culture. For the purposes of this invention, solid and/or liquid tissue culture techniques can be used. When using solid medium, transgenic plant cells can be placed directly onto the medium or can be placed onto a filter that is then placed in contact with the medium. When using liquid medium, transgenic plant cells can be placed onto a flotation device, e.g., a porous membrane that contacts the liquid medium. A solid medium can be, for example, Murashige and Skoog (MS) medium containing agar and a suitable concentration of an auxin, e.g., 2,4-dichlorophenoxyacetic acid (2,4-D), and a suitable concentration of a cytokinin, e.g., kinetin.

It is often convenient to maintain Cas transgenes and gRNA transgenes in different parents, and then produce mutations by crossing the parents to obtain mutated seeds. gRNA-expressing transgenes can be stacked as needed. Then, crosses to the Cas transgene parents result in mutated progeny.

Mutations by transient transfection

In some embodiments the methods can be practiced at least in part without transgene expression. Accordingly, tissue culture materials such as protoplasts can be transfected with pre-assembled ribonucloeoproteins complexes of purified Cas and gRNA (see Woo et al., Nature Biotechnology 2015, 33: 1162- 1165). Plants regenerated from tissue culture often comprise the intended mutations. gRNAs can be mixed in different combinations before transfection to

produce candidates that combine the designed mutations as desirable. This technique can be especially useful in species where sexual crossing is difficult or impossible, as for Miscanthus x giganteus or Saccharum officianarum. pre-assembled gRNA-Cas9 ribonucloeoproteins.


The methods described herein can be applied to organisms capable of genetic

modification and sexual recombination. For example, the methods described herein can be applied to plants (e.g., plant species of importance to agriculture), fungi (e.g., yeast), protozoans, and animals (e.g., fish such as salmon or zebra fish, fruit flies, or earthworms). In some cases, the methods described herein can be applied to monocotyledonous and dicotyledonous plants and plant cell systems, including species from one of the following families: Acanthaceae, Alliaceae, Alstroemeriaceae, Amaryllidaceae, Apocynaceae, Arecaceae, Asteraceae, Berberidaceae, Bixaceae, Brassicaceae, Bromeliaceae, Cannabaceae, Caryophyllaceae, Cephalotaxaceae, Chenopodiaceae , Colchicaceae, Cucurbitaceae, Dioscoreaceae, Ephedraceae, Erythroxylaceae, Euphorbiaceae, Fabaceae, Lamiaceae, Linaceae, Lycopodiaceae, Malvaceae, Melanthiaceae , Musaceae, Myrtaceae, Nyssaceae, Papaveraceae, Pinaceae, Plantaginaceae, Poaceae,

Rosaceae, Rubiaceae, Salicaceae, Sapindaceae, Solanaceae, Taxaceae, Theaceae, or Vitaceae.

For example, suitable species may include members of the genus Abelmoschus, Abies, Acer, Agrostis, Allium, Alstroemeria, Ananas, Andrographis, Andropogon, Artemisia, Arundo, Atropa, Berberis, Beta, Bixa, Brassica, Calendula, Camellia, Camptotheca, Cannabis,

Capsicum, Carthamus, Catharanthus , Cephalotaxus , Chrysanthemum, Cinchona, Citrullus, Coffea, Colchicum, Coleus, Cucumis, Cucurbita, Cynodon, Datura, Dianthus, Digitalis, Dioscorea, Elaeis, Ephedra, Erianthus, Erythroxylum, Eucalyptus, Festuca, Fragaria,

Galanthus, Glycine, Gossypium, Helianthus, Hevea, Hordeum, Hyoscyamus, Jatropha, Lactuca, Linum, Lolium, Lupinus, Lycopersicon, Lycopodium, Manihot, Medicago, Mentha, Miscanthus, Musa, Nicotiana, Oryza, Panicum, Papaver, Parthenium, Pennisetum, Petunia, Phalaris, Phleum, Pinus, Poa, Poinsettia, Populus, Rauwolfia, Ricinus, Rosa, Saccharum, Salix,

Sanguinaria, Scopolia, Secale, Solanum, Sorghum, Spartina, Spinacea, Tanacetum, Taxus, Theobroma, Triticosecale, Triticum, Uniola, Veratrum, Vinca, Vitis, and Zea. In some embodiments, suitable species include Panicum spp., Sorghum spp., Miscanthus spp.,

Saccharum spp., Erianthus spp., Populus spp., Andropogon gerardii (big bluestem), Pennisetum purpureum (elephant grass), Phalaris arundinacea (reed canarygrass), Cynodon dactylon (^ermudagrass), Festuca arundinacea (Yall fescue), Spartina pectinata (prairie cord-grass), Medicago sativa (alfalfa), Arundo donax (giant reed), Secale cereale (rye), Salix spp. (willow), Eucalyptus spp. (eucalyptus), Triticosecale (triticum - wheat X rye) or bamboo.

Additional examples of suitable species include Helianthus annuus (sunflower), Carthamus tinctorius (safflower), Jatropha curcas (jatropha), Ricinus communis (castor), Elaeis guineensis (palm), Linum usitatissimum (flax), Brassica juncea, Beta vulgaris (sugarbeet), Manihot esculenta (cassava), Lycopersicon esculentum (tomato), Lactuca sativa (lettuce), Musa paradisiaca (banana), Solanum tuberosum (potato), Brassica oleracea (broccoli, cauliflower, Brussels sprouts), Camellia sinensis (tea), Fragaria ananassa (strawberry), Theobroma cacao (cocoa), Coffea arabica (coffee), Vitis vinifera (grape), Ananas comosus (pineapple), Capsicum annum (hot & sweet pepper), Allium cepa (onion), Cucumis melo (melon), Cucumis sativus (cucumber), Cucurbita maxima (squash), Cucurbita moschata (squash), Spinacea oleracea (spinach), Citrullus lanatus (watermelon), Abelmoschus esculentus (okra), Solanum melongena (eggplant), Papaver somniferum (opium poppy), Papaver orientale, Taxus baccata, Taxus

brevifolia, Artemisia annua, Cannabis sativa, Camptotheca acuminate, Catharanthus roseus, Vinca rosea, Cinchona officinalis, Colchicum autumnale, Veratrum californica, Digitalis lanata, Digitalis purpurea, Dioscorea spp., Andrographis paniculata, Atropa belladonna, Datura stomonium, Berberis spp., Cephalotaxus spp., Ephedra sinica, Ephedra spp., Erythroxylum coca, Galanthus wornorii, Scopolia spp., Lycopodium serratum (Huperzia serratd), Lycopodium spp., Rauwolfia serpentina, Rauwolfia spp., Sanguinaria canadensis, Hyoscyamus spp., Calendula officinalis, Chrysanthemum parthenium, Coleus forskohlii, Tanacetum parthenium, Parthenium argentatum (guayule), Hevea spp. (rubber), Mentha spicata (mint), Mentha piperita (mint), Bixa orellana, Alstroemeria spp., Rosa spp. (rose), Dianthus caryophyllus (carnation), Petunia spp. (petunia), Poinsettia pulcherrima (poinsettia), Nicotiana tabacum (tobacco), Lupinus albus

(lupin), Uniola paniculata (oats), bentgrass (Agrostis spp.), Populus tremuloides (aspen), Pinus spp. (pine), Abies spp. (fir), Acer spp. (maple), Hordeum vulgare (barley), Poa pratensis

(bluegrass), Lolium spp. (ryegrass) and Phleum pratense (timothy).

In some embodiments, a suitable species can be a wild, weedy, or cultivated Pennisetum species such as, but not limited to, Pennisetum alopecuroides, Pennisetum arnhemicum,

Pennisetum caffrum, Pennisetum clandestinum, Pennisetum divisum, Pennisetum glaucum, Pennisetum latifolium, Pennisetum macro stachyum, Pennisetum macrourum, Pennisetum orientale, Pennisetum pedicellatum, Pennisetum polystachion, Pennisetum polystachion ssp. Setosum, Pennisetum purpureum, Pennisetum setaceum, Pennisetum subangustum, Pennisetum typhoides, Pennisetum villosum, or hybrids thereof (e.g., Pennisetum purpureum x Pennisetum typhoidum).

In some embodiments, a suitable species can be a wild, weedy, or cultivated Miscanthus species and/or variety such as, but not limited to, Miscanthus x giganteus, Miscanthus sinensis, Miscanthus x ogiformis, Miscanthus floridulus, Miscanthus transmorrisonensis, Miscanthus oligostachyus, Miscanthus nepalensis, Miscanthus sacchariflorus , Miscanthus x giganteus 'Amuri' , Miscanthus x giganteus 'Nagara', Miscanthus x giganteus 'Illinois', Miscanthus sinensis var. 'Goliath' , Miscanthus sinensis var. 'Roland', Miscanthus sinensis var. 'Africa' , Miscanthus sinensis var. 'Fern Osten' , Miscanthus sinensis var. gracillimus, Miscanthus sinensis var. variegates, Miscanthus sinensis var. purpurascens, Miscanthus sinensis var. 'Malepartus' , Miscanthus sacchariflorus var. 'Robusta' , Miscanthus sinensis var. ' Silberfedher' (aka. Silver Feather), Miscanthus transmorrisonensis, Miscanthus condensatus, Miscanthus yakushimanum, Miscanthus var. 'Alexander' , Miscanthus var. 'Adagio' , Miscanthus var. 'Autumn Light', Miscanthus var. 'Cabaret', Miscanthus var. 'Condensatus', Miscanthus var. 'Cosmopolitan', Miscanthus var. 'Dixieland', Miscanthus var. 'Gilded Tower' (U.S. Patent No. PP14,743), Miscanthus var. 'Gold Bar' (U.S. Patent No. PP15,193), Miscanthus var. 'Gracillimus',

Miscanthus var. 'Graziella', Miscanthus var. 'Grosse Fontaine', Miscanthus var. 'Hinjo aka Little Nicky'™, Miscanthus var. 'Juli', Miscanthus var. 'Kaskade', Miscanthus var. 'Kirk Alexander', Miscanthus var. 'Kleine Fontaine', Miscanthus var. 'Kleine Silberspinne' (aka. 'Little Silver Spider'), Miscanthus var. 'Little Kitten', Miscanthus var. 'Little Zebra' (U.S. Patent No. PP13,008), Miscanthus var. 'Lottum', Miscanthus var. 'Malepartus', Miscanthus var. 'Morning Light', Miscanthus var. 'Mysterious Maiden' (U.S. Patent No. PP16,176), Miscanthus var. 'Nippon', Miscanthus var. 'November Sunset', Miscanthus var. 'Parachute', Miscanthus var. 'Positano', Miscanthus var. 'Puenktchen'(aka 'Little Dot'), Miscanthus var. 'Rigoletto', Miscanthus var. 'Sarabande', Miscanthus var. 'Silberpfeil' (aka.Silver Arrow), Miscanthus var. 'Silver stripe', Miscanthus var. 'Super Stripe' (U.S. Patent No. PP18,161), Miscanthus var.

'Strictus', ox Miscanthus var. 'Zebrinus' .

In some embodiments, a suitable species can be a wild, weedy, or cultivated sorghum species and/or variety such as, but not limited to, Sorghum almum, Sorghum amplum, Sorghum angustum, Sorghum arundinaceum, Sorghum bicolor (such as bicolor, guinea, caudatum, kafir, and durra), Sorghum brachypodum, Sorghum bulbosum, Sorghum burmahicum, Sorghum controversum, Sorghum drummondii, Sorghum ecarinatum, Sorghum exstans, Sorghum grande, Sorghum halepense, Sorghum interjectum, Sorghum intrans, Sorghum laxiflorum, Sorghum leiocladum, Sorghum macrospermum, Sorghum matarankense, Sorghum miliaceum, Sorghum nigrum, Sorghum nitidum, Sorghum plumosum, Sorghum propinquum, Sorghum

purpureosericeum, Sorghum stipoideum, Sorghum sudanensese, Sorghum timorense, Sorghum trichocladum, Sorghum versicolor, Sorghum virgatum, Sorghum vulgare, or hybrids such as Sorghum x almum, Sorghum x sudangrass or Sorghum x drummondii.

Thus, the methods described herein can be applied to a broad range of plant species, including species from the dicot genera Brassica, Carthamus, Glycine, Gossypium, Helianthus, Jatropha, Parthenium, Populus, and Ricinus; and the monocot genera Elaeis, Festuca, Hordeum, Lolium, Oryza, Panicum, Pennisetum, Phleum, Poa, Saccharum, Secale, Sorghum, Triticosecale, Triticum, and Zea. In some embodiments, a plant is a member of the species Panicum virgatum (switchgrass), Sorghum bicolor (sorghum, sudangrass), Miscanthus giganteus (miscanthus), Saccharum sp. (energycane), Populus balsamifera (poplar), Zea mays (corn), Glycine max (soybean), Brassica napus (canola), Triticum aestivum (wheat), Gossypium hirsutum (cotton), Oryza sativa (rice), Helianthus annuus (sunflower), Medicago sativa (alfalfa), Beta vulgaris (sugarbeet), or Pennisetum glaucum (pearl millet).

In certain embodiments, the methods described herein can be applied to hybrids of different species or varieties of a specific species {e.g., Saccharum sp. X Miscanthus sp., Sorghum sp. X Miscanthus sp., e.g., Panicum virgatum x Panicum amarum, Panicum virgatum x Panicum amarulum, and Pennisetum purpureum x Pennisetum typhoidum).

An elite plant line or elite plant variety can be an agronomically superior plant line that has resulted from many cycles of breeding and selection for superior agronomic performance. Generally, an elite variety is a collection of plants that has been selected for a particular characteristic or combination of characteristics or traits, uniform and stable in those

characteristics, and when propagated by appropriate means, retains those characteristics. An elite variety may have a high uniformity level at least with respect to specific genomic regions. For example, at least 90% of the individuals of an elite variety may exhibit a specific genotypic profile, as it may be detected and characterized with the respective molecular markers.

Numerous elite plant lines are available and known to those of skill in the art of breeding for any cultivated plants. Traits that may be considered to confer elitism include, without limitation, good lodging resistance, reduced bacterial infection susceptibility, good seed set, good pollen set, good roots, good cold germination, good combining ability, tolerance to pests, tolerance to disease, tolerance to drought, tolerance to salts or metals, uniform floral timing, good fertilizer use efficiency, high yield as an inbred, high yield as a hybrid, good plant height, and optionally herbicide resistance or tolerance. In some cases, an elite line or elite cultivar might not itself exhibit such traits, but rather it is considered elite because it exhibits the ability to serve as one parent of an elite hybrid.


In some aspects, the methods described herein are based in part on segregation of heterozygous mutations in sexual crossing. The recombination step often involves crossing of

two different plants, i.e., male and female, rather than self-fertilization of self-compatible plants. Typically, hybrids can be produced by preventing self-pollination of female parent plants (i.e., seed parents), and permitting pollen from male parent plants to fertilize female parent plant, and allowing Fl hybrid seeds to form on the female plants. Self-pollination of female plants can be prevented by physically emasculating the flowers at an early stage of flower development.

Alternatively, pollen formation can be prevented on the female parent plants using a form of male sterility. For example, male sterility can be cytoplasmic male sterility (CMS), nuclear male sterility, genetic male sterility such as temperature or photoperiod-sensitive genetic male sterility, molecular male sterility wherein a transgene or mutation inhibits microsporogenesis and/or pollen formation, or be produced by self-incompatibility. Female parent plants containing CMS are particularly useful. Some crop species such as corn, sorghum, canola, and rice have well known hybridization systems based on cytoplasmic male sterility (CMS). In embodiments in which the female parent plants are CMS, the male parent plants typically contain a fertility restorer gene to ensure that the Fl hybrids are fertile.

The parent plants can be grown as substantially homogeneous adjoining populations to facilitate natural cross-pollination from the male parent plants to the female parent plants. The Fl seed formed on the female parent plants can be selectively harvested by conventional means. One also can grow the two parent plants in bulk and harvest a blend of Fl hybrid seed formed on the female parent and seed formed upon the male parent as the result of self-pollination.

A hybridization system based on a two component design also can be adopted in species where CMS or physical emasculation options are not widely available. Accordingly, a line could be developed that is homozygous for a target transgene coding for the cytotoxic barnase sequence. A different line has an activator transgene with a DNA binding domain

complementary to upstream activating sequence of the barnase target and a transcription activating domain capable of driving tanscription, and driven by an anther specific promoter. The two lines are crossed to produce the female for the cross needed to produce the testing population. The male plant for the cross which produces the testing population, on the other hand, is homozygous for a barnase-inactivating barstar sequence. The barstar transgene could be either a target transgene for a two component system, possibly with the same UAS as the barnase transgene, or could be a direct fusion gene. Of course, other transgenes or mutations would also be present in the male and female progenitors of the testing population. Alternatively, similarly to the canola MS8/RF3 hybridization system, male sterility can be achieved with a barnase sequence driven by a tapetum- specific promoter (Mariani et al., Nature 357, 384-387, 1992), which can be used in conjunction with a linked herbicide tolerance gene for female propagation. Fertility can be restored when needed by crossing with a plant having a construct directing expression of barstar sequence in the same cells as the barnase. Suitable promoters may be found in the literature, including Kato et al., 2010 Plant Mol. Biol. Rep 28: 381-387, Luo et al., 2006 Plant Mol. Biol 62(3): 397-408, Gupta et al., 2007 Plant Cell Rep. 26(11): 1919-31, Liu et al., 2013 Planta 238(5): 845-57, and Goldberg et al., 1993 Plant Cell 5: 1217-1229.

Nevertheless, self-fertilization of a self-compatible species is also feasible to carry out the methods provided herein. Selfing of a plant with heterozygous mutations can provide by meiotic recombination needed for a testing population. Selfing of heterozygous materials may also be performed at an earlier or intermediate step of providing a testing population to produce plants with a genetic composition of heterozygous and/or homozygous mutations that may be desirable for phenotyping, producing, or propagating.

Testing populations

A population of individuals having different combinations of mutations needs to be made, and then phenotyped. There are many ways of making a suitable testing population.

In many cases, a Cas-expressing transgene is present in one parent of a cross, and gRNA-expressing transgenes are present in the other parent. It is often desirable for the Cas-expressing transgene to be homozygous in the parent, so the entire population formed by the cross accumulates mutations. The gRNA transgenes can also be homozygous in the parent, although all or some gRNA-expressing transgenes may also be heterozygous in certain experimental designs. The heterozygous transgenes will segregate in the progeny, and thus contribute to a population comprising individuals with either wild type or mutated genomic targets.

If the Cas transgene is expressed early in development, the progeny of the very first cross between a Cas-expressing transgene parent and a gRNAs-expressing parent could itself be a testing population. This is because relatively uniform genotypes are produced throughout the tissues of such individuals, so the phenotypes observed in this population are likely to be heritable. The individuals of the population will have different sets of mutations caused by the random NHEJ repair mechanism.

In most cases, it is desirable to produce a testing population that recombines mutations first generated in progenitor individuals. One reason for this is to "shuffle" the mutations and thus increase the diversity of the testing population, thus increasing the chances of seeing phenotypes caused by specific pairs or combinations of mutations. Also, especially if a high numbers of genomic targets are addressed, it is desirable to have a testing population with individuals comprising mutations only in different subsets of genomic targets. Another reason for producing testing populations by crossing individuals with the original germline mutations is to segregate transgenes away from the testing population. A Cas-expressing transgene may be problematic if present in certain testing populations as it may produce new unintended mutations.

In most embodiments, the testing population contains mutations recombined by meiotic segregation. Thus one or both parents have at least a subset of the mutations of interest in a heterozygous state, and they give rise to progeny, i.e. a testing population, with the expected recombination of mutations. In some embodiments, each mutation in a heterozygous state will segregate during meiosis, forming gametes either containing or free of the mutation. This meiotic segregation is used in the methods provided herein, so that parental plants carrying many mutations of interest generate progeny with many different combinations of the parental mutations.

The genetic background of a testing population is in many embodiments as homogenous as possible, so as to have as little individual to individual variation as possible, as this variation would interfere with the phenotype to be scored that is attributable to individual mutation combinations. Thus, transgenes may first be introgressed, if needed, in near isogenic lines, and then the various crosses could be planned to form the testing population. A testing population, however, is often the Fl seed of parents of complementary heterotic groups, as understanding the effect of mutations within the heterotic background of a commercially relevant hybrid is desirable. Either or both parent may comprise mutations, and the mutations may be made by gRNAs that are completely identical between the parents or progenitors of the parents, completely different, partly overlapping, or with at least a subset of the initial gRNAs directed to different residues of the same genomic targets.

There are many types of crosses that can produce testing populations. In many cases, forming heterozygous individuals by crossing individuals having germline mutations to wild type is appropriate. It is desirable, however, to have a testing population with individuals homozygous for at least some mutations so as to produce phenotypes of recessive alleles.

Consequently, selfing or sibling crossing is needed, or, if heterosis needs to be maintained, mutations can be generated independently or introgressed into complementary genetic backgrounds.

There are many ways in which parental plants having heterozygous mutations of interest can be obtained. For example, for self-compatible species, it is easy to make by selfing and selection a parent stock that is homozygous for the mutations of interest. Heterozygous mutations will then result by crossing homozygous plants with a plant null for the respective mutations. For self-incompatible species, fixing homozygous mutations in a propagating population is also feasible, and molecular characterization of individual progenitors would be especially helpful. Creating double haploids can also be useful, if feasible for a particular species, when needed to obtain plants homozygous for desired mutations.

In some embodiments testing populations are made by crossing parents with

heterogeneous mutant makeup. Crosses may be made randomly starting with parents of diverse but known mutation mixture composition. As long as pollination occurs randomly, the genetic structure of the progeny or testing population can be inferred from the distribution of mutations in the parents. This approach may be convenient in certain cases, such as when working with obligate outcrossing species or with populations of improvement rounds.

In some embodiments, the parent plants also can be homozygous for one or more mutations. In some embodiments, the parent plants are heterozygous at loci of interest, with both alleles mutated from the genomic target wild type.

In some embodiments, mutations may be present in both the male and female parents of the cross that makes the testing population. In other embodiments, all the heterozygous mutations may be present in a single parent. This approach is desirable when transformation of one parent is comparatively easy, so that introgression of mutations into a parent of a different genetic background is not necessary.


Populations of progeny plants can be screened and/or selected for those members of the population that have a trait or phenotype, or a combination of traits or phenotypes conferred by the particular combinations of mutations that is distinguishable from control plants. A control plant refers to a plant that does not contain one or more of the mutations in a plant of interest, but otherwise has the same or similar genetic background. A suitable control plant can be a non-mutant wild type plant, a non-mutant and optionally non-transgenic segregant from a

mutagenesis experiment, a plant that contains one or more mutations other than the one or more mutations of interest, or a plant that contains a subset of mutations. Phenotyping can be performed in a greenhouse and/or laboratory and/or in the field. In some embodiments, a population of plants can be selected that has improved heterosis, grain yield, tolerance to abiotic stress such as drought stress, osmotic stress, or nitrogen deficiency, soil aluminum, cold stress, frost stress, density stress, heat stress, oxidative stress, low light tolerance, herbicide stress, as well as improved water use efficiency, nitrogen use efficiency, phosphate use efficiency, seed oil or protein content, lignin content, biotic or pest resistance, biomass, chemical composition, plant architecture, flowering time, and/or biofuel conversion properties. In some cases, selection and/or screening can be carried out over multiple rounds of mutagenesis. Selection and/or screening can be carried out over one or more generations, and/or in more than one geographic location. In some cases, mutant plants can be grown and selected under conditions which induce a desired phenotype or are otherwise necessary to produce a desired phenotype in a mutant plant. In addition, selection and/or screening can be applied during a particular developmental stage in which the phenotype is expected to be exhibited by the plant. But, in many cases a phenotypic measure is yield of harvestable material under typical field cultivation conditions, i.e. without an intentionally applied selection pressure. Selection and/or screening can be carried out to choose those mutant plants having a statistically significant difference in yield (e.g., grain, vegetative biomass, or stem sucrose yield) relative to a control plant that lacks the combination of mutations. Selection and/or screening can be carried out to choose those mutant plants having a statistically significant difference in an abiotic stress tolerance level relative to a control plant that lacks the transgene. While the focus is most often on individuals with mutation

combinations exhibiting improved performance, it is sometimes useful to identify stacks of significantly impaired performance over a control. Identification of undesirable mutations can be useful in designing subsequent improvement rounds so as to eliminate or minimize their occurrence.

To test for density stress tolerance, the testing population can be planted at an excessive density for the respective genetic background controls, and yield of individual plants scored for identifying the best performing individuals (see, for example, Mansfield and Humm, 2014, Crop Science, 57: 157-173).

A heterotic group comprises a set of genotypes that perform well when crossed with genotypes from a different or complementary heterotic group. Inbred lines are classified into heterotic groups, and are further subdivided into families within a heterotic group, based on several criteria such as pedigree, molecular marker-based associations, and performance in hybrid combinations (see e.g. Smith at al. (1990) Theor. Appl. Gen. 80:833-840). For example for corn, the two most widely used heterotic groups in the United States are referred to as "Iowa Stiff Stalk Synthetic" (BSSS) and "Lancaster" or "Lancaster Sure Crop" (sometimes referred to as NSS, or iron-Stiff Stalk).

To test for nitrogen use efficiency, seeds of a testing population can be planted in a field using standard agronomic practices for the region, along with wild type controls of the same genetic background. Fertilizer is applied at about 50% of the optimal level for the respective location, so that yield of wild type plants is negatively impacted. See, Example 3.

Aside from pre-defined phenotypical observations to be made on testing populations such as those appropriate to screen for stress tolerances, the appearance of nearby planted negative controls can be useful in comparing to individuals in the testing population for observation of phenotypic differences that may be caused by some mutation combinations. Non-limiting examples of traits to be observed for example in corn include ear diameter, ear height, ear leaf length, ear leaf weight, ear length, ear position, ear number, grain color, kernel length, kernel number, kernel row arrangement, kernel row number, kernel type, kernel width, leaf length, leaf width, tassel size, tassel type, and uppermost ear shape, and others traits described, for example, in the Maize Traits for Fieldbooks. See the world wide web at

' 'cril . P/ Activity +2.1.2-1— hMaize+Traits+for+Fieldbooks . ' ' The methods provided can be used to generate a very large number of different combinations of mutations. But very large numbers can also have drawbacks, so in designing combinations it is often desirable to limit the number of combinations. A limit can sometimes be imposed by the need to replicate individual genotypes so to understand the statistical significance of the phenotypes observed, and as such this limit is correlated with the size of any designed study. But, a limited "unit" of related variability is also helpful in side-by-side comparisons. For example, a single parent having four gRNAs can generate two different mutations in each allele of the four genomic targets. If first crossed to the wild type and the progeny then selfed, the mutations can form 1296 different combinations in homozygous and heterozygous states.

Planting a population with having no more than this variability on a contiguous and identifiable plot helps by minimizing the environmental variability exposure and allowing for manageable comparative phenotyping. In other words, when related genotypes are replicated in a defined area, individuals can be readily examined for visually noticeable differences. As such, it is desirable to design variability units that occupy generally no more than about half a hectare or about one acre. For example, 1296 corn genotypes replicated 10-fold, i.e. about 13,000 plants, are typically planted on about one acre.

When seeking to first sort through the candidate mutations, it is preferable to make and phenotype a population of plants of a uniform genetic background if possible. However, a reduced number of candidates can be tested in variable genetic backgrounds. When the tested populations are sufficiently large, the interaction of different mutation combinations with known QTLs can thus be determined. Consequently, the methods provided herein can be used in conjunction with traditional breeding selections to produce cultivars with improved traits.


As described herein, plants that are identified as having an improved phenotypic feature can be genotyped using any methodology. Genotyping will often involve sequencing of the genomic target of the mutated materials of interest, i.e. at least around the residues targeted by gRNA used in mutagenesis, to determine the precise mutation introduced in specific individuals. Genotype refers to the combination of mutations present in an individual plant, which can be determined by a variety of methods known in the art, such as PCR with genomic target- specific primers or Southern blotting. Genotype can also refer to the combination of alleles that determines a characteristic or trait, and it can be indirectly characterized using markers or directly characterized by nucleic acid sequencing. Suitable markers include a genetic marker, or some other type of marker. The genotype can reflect the sequence of a portion of a chromosome, an entire chromosome, a portion of the genome, or of the entire genome. In some embodiments, leaf punches from individuals either to be selected for testing or identified as having an improved phenotypic feature can be genotyped. In some embodiments, seed chipping, in which the genetics of the seed can be assessed without destroying the seed, is used to select a subset of individuals from the progeny population for use as a testing. See, for example U.S. Patent No. 7,502,113. Accordingly, a population can be created that mixes a large number of mutations. Subsequently, as it may become desirable as informed by new performance data, a subpopulation comprising only a defined subset of mutations, and possibly lacking transgenes such as a Cas-expressing transgene, can be selected and studied. Or similarly, individuals from a large population can be eliminated from a study by genotyping plants before planting if they are deemed to contain mutation combinations that are undesirable.

Improvement rounds

Once a combination of mutations is identified by any means as having a desirable phenotypic performance, the improved materials can be subjected to additional rounds of improvement by adapting the methods used to identify the combination. In one type of improvement, the desired mutations are maintained in the background of all the plants of a testing population, and additional mutations combinations are also stacked. Mutation

combinations can be maintained in the genetic background by making a testing population using individuals related by lineage to individuals of selected phenotypes. Alternatively, the combined genomic targets identified can be maintained as subject of de novo mutation in new testing populations, which may generate additional pairs or combinations of mutations of interest. Some of the additional mutations can be second site mutations in the same genomic targets that are part of the originally identified combination of mutations.

In some embodiments, improvements may be made using the top performing materials from a phenotyped population. For example, the best individuals can be crossed to each other and their progeny phenotyped. When the diversity of original mutations is large, this approach may more quickly result in recognition of improved combinations. This approach works well when the testing population is made up of inbred lines or uniform true breeding populations. When the testing population is made up of hybrid plants, it is possible to make one or more corresponding populations by crosses to isogenic parents so as to cause similar mutation segregation as in the hybrid testing population.

The improvement rounds can be cycled as many times as needed to develop mutation combinations of incrementally enhanced performance in the respective assays or field conditions.

The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.


Example 1: Selecting salt tolerance targets

To rice make plants with enhanced salt tolerance, genomic targets comprising SEQ ID

NOs: 6-10 are selected. Mutations of these targets are expected to impact the soil salt sensitivity of mutant plants. Pairs or combinations of mutations in these targets are therefore likely to produce a higher resistance to salt phenotype.

Example 2: RNA targeting

SEQ ID NOs : 11-20 (which include the PAM 5'NGG3' for the Streptococcus pyogenes

Cas9) can serve as RNA targeting sequences for different residues of the genomic target represented by SEQ ID NO: 6.

Example 3: gRNA-expressing transgenes

A transformation vector comprising gRNAs expressing genes is made. The vector comprises five tandem expression cassettes, each made up of the promoter of SEQ ID NO: 3 operably linked to a gRNA sequence made up of a RNA targeting sequence of fused to a scaffolding sequence, and followed 3' by a Pol III terminator. The five expression cassettes of this vector have the target RNAs of SEQ ID Nos: 20-24 (which include the PAM 5'NGG3' for the Streptococcus pyogenes Cas9), designed to mutate genomic targets comprising SEQ ID NOs: 6-10. The vectors are used in Agrobacterium-mediated transformation of rice.

Example 4: Selecting corn ear morphology targets

To make corn plants with enhanced grain yield, genomic targets comprising SEQ ID NOs: 26-39 are selected. Mutations of these targets are expected to impact the ear morphology of mutant plants. Pairs or combinations of mutations in these targets are therefore likely to produce corn cobs with higher grain yield.


It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.