Systems for inducing and analyzing genome-scale gene knockdown or knockout in plants

Abstract

The present invention relates to systems and methods for inducing and analyzing gene knockdown and / or knockout in plants at a large-scale. In particular, the present invention discloses Cell type-specific Multi-targeted AmiRNA or CRISPR (cMAC) function-scale toolbox, presenting a novel system for knockdown and / or knockout of gene families at the functional-scale, for example, at the transportome scale, and at the cell type level.

Description

SYSTEMS FOR INDUCING AND ANALYZING GENOME-SCALE GENE KNOCKDOWN OR KNOCKOUT IN PLANTSFIELD OF THE INVENTION

[0001] The present invention relates to systems and methods for inducing and analyzing gene knockdown and / or knockout in plants at a large-scale. In particular, the present invention discloses Cell type-specific Multi-targeted AmiRNA or CRISPR (cMAC) function-scale toolbox, presenting a novel system for knockdown and / or knockout of gene families at the functional-scale, for example, at the transportome scale, and at the cell type level.BACKGROUND OF THE INVENTION

[0002] For many decades, forward genetics has been a powerful approach for identifying genes and mutations that underlie phenotypes of interest in plants, including those that encode transporters. However, genetic analysis of transporter gene families has been precluded in many cases due to the lack of observable phenotypes in single-gene, loss-of- function mutations. Partial, complete, or conditional functional redundancy between different family members results in the absence of visible phenotypes in single -gene knockouts. Therefore, high-order mutants of redundant genes from the same gene family are needed to reveal or increase the visibility of phenotypes.

[0003] Ancient duplication events and a high retention rate of genes with similar sequences have contributed to the abundance of duplicated genes in plant genomes. For example, an average of 65% of plant genes are paralogous, ranging from 46% in the moss Physcomitrella patens to 84% in the apple Malus domestica. Of the approximately 25,000 genes in Arabidopsis, 75% belong to gene families with at least two members. This robust genetic redundancy allows plant adaptation to the ever-changing environment. Genetic redundancy is tightly connected to mechanistic cellular networks. For example, the NPF transporter family contains 53 genes and the ABC transporter family contains 130 genes in Arabidopsi . These genes exhibit redundant phenotypes despite the somewhat diverse substrate specificities of the encoded proteins.

[0004] In order to identify novel redundant genes, overexpression strategies have often been used. However, such approaches can lead to pleiotropic phenotypes that may not be relevant to the biological activity of the studied protein. Therefore, it is of high priority to generate visible loss-of-function mutations in a forward-genetics manner. Significant progress has been made in recent years using genome-scale RNA interference methods and artificial microRNAs (amiRNAs) or gene editing using clustered, regularly interspaced short palindromic repeats (CRISPR). The amiRNAs are generated using an endogenous miRNA backbone as a template, replacing the target recognition sequence with a custom sequence.

[0005] The CRISPR system comprises an RNA-protein complex consisting of a singleguided RNA (sgRNA) and a CRISPR-associated protein 9 (Cas9) endonuclease. This complex is guided by the sgRNA through Watson-Crick base pairing to a target DNA sequence. Upon cleavage by Cas9, the target locus undergoes insertion / deletion (indel) of few base pairs (bp) at the repaired locus, leading to frameshift mutations and the knock out of the targeted locus. Both amiRNA precursors and sgRNAs can be computationally designed to target a specific group of potentially redundant genes.

[0006] As a tool to overcome the obstacle of functional redundancy, a computationally designed library of 22,000 amiRNAs, termed Phantom, was generated; each amiRNA in the library targets two to five potentially redundant genes identified by shared sequence homology. The Phantom library was divided into 10 sub-libraries; each containing 1,505- 4,082 amiRNAs targeting a functionally defined protein class. By overcoming functional redundancy and bypassing the genetic linkage of multiple knockdowns, the Phantom library, driven by a constitutive 35S promoter, allows rapid mapping of hundreds of candidate lines without the need for laborious map-based cloning. Use of this library resulted in the identification of dozens of novel redundant genes, including several previously unknown hormone transporters (Hauser F et al., Plant Cell. 2013 Aug;25(8):2848-63).

[0007] Notably, the Phantom amiRNAs are driven by the strong constitutive 35S promoter. This has several drawbacks: First, expression from the 35S promoter often results in a decay of amiRNA-mediated silencing over generations, lethal lines, and pleiotropic effects. Second, only 1,777 of amiRNAs in the Phantom library target transporter genes. Third, its algorithm design allowed amiRNA to target genes from the same family but did not account for thephylogenetic relationships among family members. Consequently, in many cases, the designed amiRNAs targeted only distantly related genes, reducing the probability of targeting genes with overlapping functions.

[0008] WO 2023 / 199308 to inventors of the present invention discloses the employment of CRISPR genomic libraries to construct large knockout mutant libraries for forward-genetic screens and gene function analysis, enabling the exploration of gene function. A genetic toolbox called Multi-knock integrates forward genetics with dynamically targeted genomescale CRISPR tools to address the issue of masked phenotypic variation resulting from functional redundancy. The pipeline of this approach operates as follows: firstly, a computational pipeline based on the Multi-Knock design algorithms is employed to design sgRNAs capable of simultaneous knockout of multiple genes per family. The selected sgRNA candidates are then synthesized and cloned into plasmids. Following transformation into plants, a screen is conducted to identify CRISPR lines containing a desired phenotype. Positive lines are subsequently subjected to further characterization to identify the inserted sgRNA, the cleaved genes, and any phenotypes that may rise from further screens.

[0009] Following this approach, a total of 59,129 multi-targeting sgRNAs were designed, synthesized, and cloned into a genome-editing intronized Cas9 vector, creating ten functional sub-libraries targeting 16,152 genes in Arabidopsis. Transformation of Arabidopsis plants with 5,635 sgRNAs specifically targeting the Arabidopsis transportome resulted in over 3,500 CRISPR lines. The screen successfully recovered multiple known phenotypes, validating the efficacy of the approach. Moreover, this study unveiled novel transporters whose functions had remained concealed due to genetic redundancy. The Multi-knock approach, which involves screening the entire genome, is suitable for general screens. However, when the research question revolves around specific biological processes, such as understanding the mechanism of nutrient absorption, a more targeted approach is necessary.

[0010] In order to address these shortcomings and to allow the identification of redundancy in more specific biological processes, such as within the transportome components, there is a need for more advance, precise systems for targeting closely homologous genes within gene families and in the context of specific tissue and cell type.SUMMARY OF THE INVENTION

[0011] The present invention provides systems and methods that allow the identification of redundant plant genes, for example, transportome components. According to certain aspects, the present invention provides a Cell type-specific Multi-targeted AmiRNA or CRISPR library (cMAC library) enabling a large-scale screening of redundant genes. According to certain aspects, the present invention provides a Cell type-specific Multi-targeted AmiRNA or CRISPR (cMAC) transportome-scale toolbox, presenting a novel system for knockdown and / or knockout of gene families at the transportome scale and at the cell type level. In some embodiments, the cMAC library contains 5,565 amiRNAs or 2,859 sgRNAs, each designed to target two to eight closely homologous genes within transporter gene families in a variety of combinations.

[0012] It is now disclosed that the genetic toolbox presented herein allows scientists to overcome functional redundancy in a forward-genetic manner and, for the first time, allows targeting specific genes, e.g. transporter genes, from defined families in a tissue-specific manner. The AmiRNA or CRISPR libraries described herein are used for knockdown or knockout multiple closely related genes in specific cell types.

[0013] Unexpectedly, the methods described herein enable implementation of high throughput screening and analysis of functional redundancy, while utilizing specific promoters, enhancing and accelerating the discovery of novel gene functions.

[0014] Advantageously, the methods described herein enable targeted and controlled knockout or knockdown of gene activity at the tissue or cell specific level, allowing the discovery of gene-phenotype associations that may otherwise be masked by unrelated or nonspecific effects. For instance, certain gene knockouts can impact the growth and development of plants in a manner that hinders the identification of the gene’s specific function at later stages or within particular tissues.

[0015] According to certain aspects the present invention provides a system for inducing knockdown or knockout of multiple genes in a plant, comprising: at least one Cell typespecific Multi-targeted AmiRNA or CRISPR library (cMAC library) comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more amiRNA sequencesor one or more sgRNA sequences, wherein each of the amiRNA or sgRNA sequences targets multiple genes.

[0016] According to some embodiments, the AmiRNA or CRISPR library comprises at least 50, 100, 200, 300, 400, 500 or more polynucleotides, each encoding one or more unique sgRNA sequences or unique amiRNA sequences.

[0017] According to certain embodiments, each of the polynucleotide encodes a single amiRNA sequence or a single sgRNA sequence, wherein the single amiRNA or sgRNA targets multiple genes.

[0018] According to certain embodiments, each of the polynucleotide encodes at least two amiRNA sequences or sgRNA sequences, wherein each of the amiRNA or sgRNA targets multiple genes.

[0019] According to some embodiments, each of the sgRNA or amiRNA sequences targets between 2-15 genes, 2-10 genes or 2-8 genes.

[0020] According to some embodiments, the library is an amiRNA library.

[0021] According to some embodiments, the library is a CRISPR library.

[0022] According to some embodiments, the expression of the polynucleotide encoding the sgRNA sequence or amiRNA sequence is regulated by a tissue-specific promoter or a developmental-stage specific promoter.

[0023] According to some embodiments, the promoter is a tissue-specific promoter selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root epidermis promoter, cortex promoter, root cortex promoter, shoot promoter, and root promoter. Each possibility represents a separate embodiment of the invention.

[0024] According to some embodiments, the library comprises a plurality of polynucleotides each encoding an sgRNA sequence or amiRNA sequence regulated by the same tissuespecific promoter. According to other embodiments, the library comprises at least two polynucleotides each encoding an sgRNA sequence or amiRNA sequence regulated by a different tissue-specific promoter. According to additional embodiments, the librarycomprises a plurality of polynucleotides encoding an sgRNA sequences or amiRNA sequences regulated by at least 2, 3, 4, 5, 6, 7, 8, 9, 10, or more different tissue-specific promoters.

[0025] According to some embodiments, the amiRNA or CRISPR library is configured to ensure that at least two different sgRNAs or amiRNAs are inserted into each plant. According to some embodiments, the library comprises a plurality of polynucleotides, each encoding a plurality of different sgRNA or amiRNA.

[0026] According to some embodiments, each of the encoded plurality of different amiRNA or sgRNA targets a different set of genes. According to some embodiments, each of the encoded plurality of different amiRNA or sgRNA targets a different set of genes within a single gene family. According to other embodiments, each of the encoded plurality of different amiRNA or sgRNA targets the same set of multiple genes. According to some embodiments, each of the encoded plurality of different amiRNA or sgRNA targets at least one same gene within a gene set.

[0027] According to other embodiments, the library comprises a plurality of polynucleotides, each encoding a plurality of sgRNAs or amiRNAs having the same sequence.

[0028] According to some embodiments, the plurality of sgRNA or amiRNA are encoded by a polynucleotide present on a single DNA vector. According to other embodiments, the plurality of sgRNA or amiRNA are encoded by polynucleotides present on different DNA vectors.

[0029] According to some embodiments, the library comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10 or more polynucleotides encoding sgRNA or amiRNA sequences, wherein the expression of each is regulated by a different tissue-specific promoter selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root cortex promoter, shoot promoter, and root promoter. Each possibility represents a separate embodiment of the invention.

[0030] According to some embodiments, the promoter is selected from the group consisting of SIG6, ARSK1, PGP4, CO2, KST1, SCR, S18, CORI3, IQD22, SUC2, ML1, and UBQ10.

[0031] According to some embodiments, the system comprises a plurality of libraries, each library comprising a plurality of polynucleotides each encoding one or more sgRNA or amiRNA sequences, wherein the expression of the plurality of polynucleotides within each library is regulated by a different tissue-specific promoter.

[0032] According to some embodiments, the library further comprises one or more amiRNA sequences or one or more sgRNA sequences, wherein each of the amiRNA or sgRNA sequences targets a single gene.

[0033] According to some embodiments, the amiRNA sequences or sgRNA sequences target transportome genes.

[0034] According to some embodiments, the amiRNA sequences or sgRNA sequences target at least one gene member comprised within a gene set. According to certain embodiments, the gene set is selected from the group consisting of mTACT-CP, mTACT-APC, mTACT- MFS, mTACT-DMT, mTACT-MATE, mTACT-ABC, mTACT-PA, and mTACT-UF. Each possibility represents a separate embodiment of the invention.

[0035] According to some embodiments, the plant is selected from the group consisting of a wild plant, a crop plant, and a model plant.

[0036] According to certain embodiments, the crop plant is selected from the group consisting of rice, maize, wheat, barley, tomato, potato, eggplant, sesame, soybean, sunflower, canola, sugarcane, alfalfa, millet, Leguminosae (bean, pea), flax, lupinus, rapeseed, tobacco, and cotton. Each possibility represents a separate embodiment of the present invention.

[0037] According to some embodiments, the model plant is Arabidopsis.

[0038] According to certain embodiments, the plant is a genetically modified plant. According to certain alternative embodiments, the plant is a non-genetically modified plant.

[0039] According to some embodiments, the library comprises 5,565 amiRNA sequences targeting a plurality of genes within Arabidopsis gene set, wherein the gene set comprises

[0040] According to some embodiments, the library comprises 2,859 sgRNA sequences targeting a plurality of genes within Arabidopsis gene set, wherein the gene set comprises 707 Arabidopsis transporter genes.

[0041] According to some embodiments, the library is a combined AmiRNA and CRISPR library.

[0042] The present invention further comprises, according to certain aspects, a method of identifying at least one phenotypic effect of gene knockdown or knockout in a plant, the method comprising: transforming the system described herein into a plurality of plants, thereby producing a plant population, wherein each plant of the population comprises at least two knocked-down or knocked-out genes; examining the phenotype of each plant in the plant population to identify at least one modified phenotype; and identifying the at least two knocked-down or knocked-out genes associated with the at least one modified phenotype.

[0043] According to certain embodiments, identifying the at least one modified phenotype is performed by comparing the phenotype of the plant comprising the system of the invention to the phenotype of a corresponding plant not comprising said system and grown under the same growth conditions. According to some embodiments, identifying the at least one modified phenotype is performed by comparing the phenotype of the plant comprising the system of the invention to the phenotype of a corresponding plant as is known in the art.

[0044] The present invention further comprises, according to certain aspects, a method for identifying multiple members within at least one gene set underlying a phenotype in a plant, the method comprising:(i) transforming the system described herein into a population of plants, thereby producing a plant population wherein each plant of the population comprises at least one sgRNA or amiRNA targeting multiple gene members;(ii) screening the plant population for at least one selected phenotype;(iii) selecting at least one plant showing the at least one selected phenotype; and(iv) identifying in the selected plant at least one sgRNA or amiRNA targeting the multiple-gene members, thereby identifying said multiple gene members underlying said selected phenotype.

[0045] According to certain aspects, the present invention provides a method for identifying multiple members within at least one gene set underlying a phenotype, the method comprising:(i) clustering coding sequences within genetic data of a plant species to sequence clusters, each cluster representing a gene set;(ii) producing a CRISPR library comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more unique sgRNAs, wherein each of the sgRNAs targets a plurality of gene members comprised within the gene set, wherein each of said sgRNA expression is controlled by at least one tissue and / or developmental stage-specific promoter;(iii) transforming the library into a plurality of plants, thereby producing a plant population wherein each plant of the population comprises at least one sgRNA targeting multiple gene members;(iv) screening the plant population for at least one selected phenotype;(v) selecting at least one plant showing the at least one selected phenotype; and(vi) identifying in the selected plant at least one sgRNA targeting the multiple-gene members; thereby identifying said multiple gene members underlying said selected phenotype.

[0046] According to some embodiments, each plant of the population comprises a plurality of sgRNAs targeting multiple gene members

[0047] According to some embodiments, each of the sgRNAs targets between 2-15, 2-10 or 2-8 gene members comprised within the gene set.

[0048] According to some embodiments, the promoter is a tissue specific promoter as described hereinabove. In some embodiments, the tissue specific promoter is selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root cortex promoter, shoot promoter, and root promoter.

[0049] According to certain aspects, the present invention provides a method for identifying multiple members within at least one gene set underlying a phenotype, the method comprising:(i) clustering coding sequences within genetic data of a plant species to sequence clusters, each cluster representing a gene set;(ii) producing an amiRNA library comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more unique amiRNAs, wherein each of the amiRNAs targets a plurality of gene members comprised within the gene set, wherein each of said amiRNA expression is controlled by at least one tissue and / or developmental stage-specific promoter;(iii) transforming the library into a plurality of plants, thereby producing a plant population wherein each plant of the population comprises at least one amiRNA targeting multiple gene members;(iv) screening the plant population for at least one selected phenotype;(v) selecting at least one plant showing the at least one selected phenotype; and(vi) identifying in the selected plant at least one amiRNA targeting the multiple-gene members; thereby identifying said multiple gene members underlying said selected phenotype.

[0050] According to some embodiments, each of the amiRNA targets between 2-15, 2-10 or 2-8 gene members comprised within the gene set.

[0051] According to some embodiments, each plant of the population comprises a plurality of amiRNAs targeting multiple gene members.

[0052] According to some embodiments, the promoter is a tissue specific promoter as described herein above.

[0053] According to certain further aspects, the present invention provides an isolated polynucleotide comprising a tissue specific promoter operably linked to one or more amiRNA encoding sequences or one or more sgRNA encoding sequences, wherein each of the encoded amiRNAs or sgRNAs targets multiple plant genes.

[0054] According to certain embodiments, the plant genes are members of a gene set. According to certain embodiments, the members of the gene set share a certain structural homology. According to certain additional or alternative embodiments, the members of the gene set participate in a plant functional pathway.

[0055] The tissue specific promoters are as described hereinabove.

[0056] A plant or a plurality of plants each comprising at least one polynucleotide comprising a tissue specific promoter operably linked to one or more amiRNA encoding sequences or one or more sgRNA encoding sequences, wherein each of the encoded amiRNA or sgRNA targets multiple plant genes are also encompassed within the teachings of the present invention.

[0057] According to certain embodiments, each plant of the plurality of plants comprises a same polynucleotide having a same configuration of the tissue specific promoter and one or more amiRNA encoding sequence or sgRNA encoding sequence.

[0058] According to certain embodiments, each plant of the plurality of plants comprises a different polynucleotide with the same tissue specific promoter and different configuration of the one or more amiRNA encoding sequence or sgRNA encoding sequence.

[0059] According to certain embodiments, the plurality of plants comprises plants with different polynucleotides, wherein the different polynucleotides comprise a variety of tissue specific promoters and a same or different configuration of the one or more amiRNA encoding sequences or sgRNA encoding sequences.

[0060] It is to be understood that any combination of each of the aspects and the embodiments disclosed herein is explicitly encompassed within the disclosure of the present invention.

[0061] These and further aspects and features of the present invention will become apparent from the detailed description, examples and claims which follow.DESCRIPTION OF THE DRAWINGS

[0062] Figures 1A-1B. Expression pattern of cell type-specific promoters used to drive multi-targeted amiRNA libraries. Fig. 1A - Shown are confocal or stereoscope images forpMLl:H2B-GFP (shoot epidermis), pIQD22:GUS-mCi trine (palisade mesophyll), pSCR:YFP (endodermis and bundle sheath), pSUC2:YFP (phloem companion cells), pCORI3:GUS-mCitrine (spongy mesophyll), pKSTl:GFP (guard cells), pS18:YFP (xylem), pPGP4:NLS-YFP (root epidermis and cortex), pCO2:YFP (cortex), pUBQ10:YFP (constitutive), pSIG6:GUS (shoot), and pARSKI :GUS (root). Shown are cell wall dye propidium iodide and chlorophyll auto-fluorescence (in pIQD22 and pC0RI3). pMLl, pSCR, pSUC2, pKSTl, pS18, pPGP4, pCO2, and pUBQlO seedlings are 5-days-old, pSIG6 and pARSKI seedlings are 12-days-old, and pIQD22 and pCORI3 are 17-days-old. Scale bars = 50 pm for pMLl, pIQD22, pSCR, pSUC2, pCORI3, pKSTl, pS18, pPGP4, pCO2, and pUBQlO; scale bar = 1 mm for pARSKI; and scale bar = 2 mm for pSIG6. Fig. IB - Illustrations of the cell type-specific expression patterns driven by promoters used in this study to drive multi-targeted amiRNA expression. Illustrations are of leaf tissue (upper left), root cross-section (lower left), root tip (center) and whole plant (right).

[0063] Figure 2A-2C. Cloning of cell type-specific multi-targeted amiRNA library. Fig. 2A - Illustration of the cell type-specific amiRNA library cloning. Shown are the construction steps involving Golden Gate sites and cell type-specific promoter insertion into destination vectors, followed by amiRNA library amplification and ligation into the destination vector using the Golden Gate reaction. Fig. 2B - Deep sequencing results of amiRNA distribution for the indicated libraries. Coverage indicates the detected amiRNAs from the 1,777 amiRNAs synthesized in the transporter-specific (LIO) Phantom library. Skew indicates equal distribution demonstration. Fig. 2C - Illustration of amiRNA library transformation into Arabidopsis for genetic screens. Each plant incorporates a single amiRNA that targets multiple genes from the same family.

[0064] Figures 3A-3B. The differentially-expressed nutrient-deficiency transporters (DENTs) families. Fig. 3A - Upset plots illustrating the distribution of DENTs across various subsets of nutrient deficiencies. The seven rows at the bottom represent the seven different nutrients. For example, the right-most column represents a subset of nutrients containing N, P and S. The vertical bars at the top display the number of differentially expressed genes specifically associated with each subset of nutrient deficiencies, focusing on the top 14 subsets of deficiencies for all DENTs. The horizontal bars at the bottom-left indicate the total number of genes that were differentially expressed for each nutrient. Fig. 3B - Sizedistribution of gene families and subfamilies across all genes that have at least one DENT. The x-axis represents the size of each gene family (number of genes), while the y-axis displays the total number of families of that size.

[0065] Figures 4A-4C. Overview of the cMAC sgRNA library. Fig. 4A - Overview of the sgRNA post -processing. The number of sgRNAs before and after each stage are shown in the gray rectangles. The dark gray rectangles show the type of sgRNAs removed in each stage. Fig. 4B - Number of targeted genes per sgRNA. Each bar represents the number of genes targeted by a single sgRNA, and the y-axis represents the number of sgRNAs with that number of targeted genes. Fig 4C - The targeting coverage of each gene.

[0066] Figures 5A-5O. Construction of multi-targeted nutrient-specific multiplexed CRISPR libraries. Fig. 5A - Overview of the number of targeted genes and sgRNAs in the single sgRNA library. Fig. 5B - Number of genes targeted by sgRNAs in the single sgRNA library. Fig. 5C - Number of mismatches per sgRNA in the single sgRNA library. Fig. 5D - Total sgRNAs in each nutrient sub-library. Fig. 5E - Overview of the number of targeted genes and multiplexes in the multiplexed library. Fig. 5F - Number of genes targeted by multiplexes in the multiplexed library. Fig. 5G - Number of mismatches per multiplexed in the multiplexed library. Fig. 5H - Number of multiplexes associated with specific nutrients in the multiplexed library. Fig. 51 - Number of multiplexes associated with a specific or with a number of nutrients in the multiplexed library. FIGs. 5J-5K - Representative family network (Fig. 5J) and overview of all sgRNAs gene target networks (Fig. 5K) in the multiplexed CRISPR library. Genes can be targeted by multiple multiplexes, and each multiplex can target multiple genes at once. Shown are single multiplexes as well as single genes as indicated. FIGs. 5E-5N - sgRNAXmultiplex distributions of the libraries based on deep sequencing results; pRPS5:N used for ubiquitous expression of the single sgRNA nitrate library (Fig. 5E); pRPS5:K used for ubiquitous expression (inherited mutations) of the single sgRNA potassium library (Fig. 5M); pRPS5:Multiplex used for ubiquitous expression (inherited mutations) of the multiplex library (Fig. 5N). Coverage indicates the percentage of sgRNAs detected relative to the total number of synthesized sgRNAs. Skew indicates equal distribution demonstration. Fig. 50 - Comparison between NGS results of pRPS 5: Multiplex E.coli and pRPS 5: Multiplex Agro.

[0067] Figures 6A-6E. Cell type-specific multi-targeted CRISPR library designed to target putative nutrient transporters. Fig. 6A - Illustration of the cell type-specific CRISPR vector that was generated in this study. The vector includes tissue-specific promoter cloning sites, an intronized Cas9 and library cloning sites. Fig. 6B - The tissue-specific promoters used in this study: pSCR; endodermis and bundle sheath, pCORI3; spongy mesophyll, pSIG6; shootspecific promoter, pSUC2; phloem companion cells, pKSTl; guard cells, pPGP4; root epidermis and cortex, pMLl; shoot epidermis and pARSKl; root-specific promoter. For pSIG6: Scale bar = 200 pM; for pSUC2, and pARSKl: Scale bar = 100 pM; pSCR, pCORI3, and pMLl: Scale bar = 50 pM; and for pPGP4: scale bar = 20 pM. FIGs. 6C-6D - sgRNA\multiplex distributions of the libraries based on deep sequencing results; pPGP4:Fe used for root epidermis and cortex-specific expression of the single sgRNA iron library (Fig. 6C); pARSKl :Multiplex used for root-specific expression of the multiplexed library (Fig. 6D). Coverage indicates the percentage of amiRNAs detected relative to the total number of amiRNAs theoretically synthesized. Skew indicates equal distribution demonstration. Fig. 6E - Shown are representative images of 2-month-old plant transformed with pPGP4:sg- IRT1,2 and pPGP4:sg-SULTRl.1,1.2 with respective control plants. Chromatograms of the knockout genes show the types of mutations. Dashed line rectangular represents the cell typespecific knockout. Dashed boxes marks the mixed peaks in the chromatogram sequence, sgRNA positions are indicated.

[0068] Figures 7A-7G. cMAC multi-targeted amiRNA library design. Fig. 7A - The amiRNA library was filtered based on free energy, number of amiRNAs per internal node, sequence similarity, and Golden Gate sites. Fig. 7B - Number of genes targeted per amiRNA for the entire cMAC library. Fig. 7C - Number of amiRNAs per internal node in the phylogenetic tree for the entire cMAC library. Fig. 7D - Number of genes targeted per total genes for each sub-library. CP, channels and porins; APC, amino acid / polyamine / organo- cation, cation carriers, and mitochondrial carriers; MFS, major facilitator superfamily; DMT, drug / metabolite transporter group; MATE, multi-drug and toxic compound extrusion and other carrier transporters; ABC, ATP binding cassette family; PA, primary active transporters; UF, unknown function. Fig. 7E - Number of amiRNAs in each sub-library. The total number of amiRNA is 5,565. Fig. 7F - Network relationship between amiRNAs (triangles) and target genes (ellipses) for the cMAC-ABC amiRNA library. Fig. 7G -Example of a typical cMAC network of ABCB family members (ellipse) targeted by amiRNAs (triangles).

[0069] Figures 8A-8B. Construction and validation of cMAC sub-libraries. Fig. 8A - Illustration of syntheses of 5,565 unique amiRNAs, each targeting two to eight genes from the same transporter gene family. amiRNA library and sub-library amplification are performed using unique adaptors designed to match amiRNAs from a particular transporter family. Numbers of amiRNAs in each sub-library are indicated. Fig. 8B(l-9) - amiRNA distributions for the amiRNA library and sub-libraries based on deep sequencing results. pS18:cMAC and pSUC2:cMAC were used for the expression of the entire library (5,565 amiRNAs) driven by S18 and SUC2 promoters, respectively. pUBQ10:cMAC-CP, pUBQ10:cMAC-MFS, pUBQ10:cMAC-MATE, and pUBQ10:cMAC-ABC were used for ubiquitous expression of the sub-libraries CP, MFS, MATE, and ABC, respectively; pARSKI :cMAC-MFS was used for root expression of the MFS sub-library and pSUC2:cMAC-MFS and pSUC2:cMAC-DMT were used for phloem companion cells- specific expression of the MFS and DMT sub-libraries, respectively. Coverage indicates the percentage of amiRNAs detected relative to the total number of amiRNAs theoretically synthesized. Skew indicates equal distribution demonstration.

[0070] Figures 9A-9E. pSUC2:cMAC-DMT reveals a putative activity of 3 DMTs in regulating leaf senescence and physiology. Fig. 9A - Shoot and leaf phenotypes of 40-day- old Col-0 and pSUC2:miR-3UAfT plants. Scale bars = 1 cm. Fig. 9B - Phylogenetic tree based on amino-acid sequences of the Arabidopsis DMT family coding sequence. Dark gray dots indicate family members putatively targeted by pSUC2:miR-3UAfT. FIGs. 9C-9E - Eeaf area (Fig. 9C), Fv / Fm, measurement of the maximum quantum yield of PSII (Fig. 9D), and water content measurements (Fig. 9E) of 49-day-old Col-0 and pSUC2:miR-3UAfT plants. Significance was evaluated by Students t-test, ** p value < 0.005 **** p value < 0.0001, n > 12.

[0071] Figures 10A-10B. Design of the mTACT amiRNA library. Fig 10A - Number of genes targeted by each amiRNA in each mTACT sub-library. Fig. 10B - Number of amiRNAs per internal node in the phylogenetic clade of the gene family.

[0072] Figure 11. Number of genes targeted per amiRNA in each mTACT sub-library. Number of genes targeted by individual amiRNAs in each amiRNA sub-library. The number of genes in an individual internal node was set between two and eight.

[0073] Figure 12. Number of amiRNAs per internal node of a gene family. The number of amiRNAs targeting individual internal nodes in each mTACT sub-library. The maximal number was eight, which was chosen in the filtering process.

[0074] Figure 13. Number of amiRNAs targeting each individual gene. Data are shown for each mTACT amiRNA sub-library. The value zero represents genes that are not targeted.

[0075] Figure 14. Schematic illustration of network relationships between amiRNAs (triangles) and target genes (ellipses) in mTACT sub-libraries.

[0076] Figures 15A-15B. Characterization of pSUC2:mTACT-DMT line. Fig 15A - Shoot and leaf phenotypes of 26-day-old Col-0 and pSUC2:miR-3UAfT plants. Scale bars = 1 cm. Senescence is not yet detected in 26-day-old plants. Fig. 15B - Relative color shades of 49- day-old Col-0 and pSUC2:miR-3UAfT ordered by RGB score. Statistical significance was determined by two-way ANOVA (p < 0.05). Images of pSUC2:miR-3UAfT leaf show visible difference compared with the wild type plants.DETAILED DESCRIPTION OF THE INVENTION

[0077] The present invention provides systems and methods that allow the identification of redundant plant genes, for example, transportome components. According to certain aspects, the present invention provides a Cell type-specific Multi-targeted AmiRNA or CRISPR library (cMAC library) enabling a large-scale screening of redundant genes.

[0078] The need to accelerate crop breeding programs has never been greater, as the world population is exponentially increasing, the climate is changing, and resources are limited. Breeding relies on genetic variation. However, altering many phenotypes by introducing genetic variation in a single gene is challenging due to large gene families with high functional redundancy.

[0079] The term “plant” is used herein in its broadest sense. It includes, but is not limited to, any species of woody, herbaceous, perennial, or annual plant. It also refers to a plurality ofplant cells that are largely differentiated into a structure that is present at a stage of the plant development capable of producing a whole plant and optionally a crop.

[0080] In plants, both developmental processes and response to the environment conditions are spatiotemporally regulated by an assembly of signaling molecules such as hormones and secondary metabolites. However, the characterization of transported signaling molecules and their translocation mechanisms is difficult due to the functional redundancy of plant genomes and shortcomings in methodologies.

[0081] The present invention discloses the development of cMAC, a multi-targeted CRISPR or amiRNA, cell type-specific genome-scale toolboxes, which can be used to reveal phenotypic plasticity in plants. These tools are based on a large set of artificial sgRNA and microRNAs (amiRNAs), each designed to optimally target multiple members of a particular gene family encoding family proteins. According to certain exemplary embodiments, the toolbox includes 5,565 amiRNAs or 2800 thousands of sgRNAs, targeting -80% of the Arabidopsis transportome.

[0082] According to some embodiments of the invention, the amiRNA or sgRNA library is driven under 12 cell type-specific promoters, allowing the design of spatial-specific genetic screens. The cMAC is further divided into eight sub-libraries of amiRNAs targeting a functionally defined protein class. A proof-of-concept screen validated the cMAC approach of the invention by revealing novel phenotypes and associating the identified phenotypes with poorly studied genes. According to some embodiments, the cMAC also includes 2,859 single-guide RNAs (sgRNAs) targeting 707 transporters assumed to be involved in the uptake of macro- and micronutrients, facilitating the design of functionality-specific screens.

[0083] With the ability to overcome functional redundancy in a functionally-scale, for example in a transportome-scale, and further in cell type-specific manner, the cMAC toolbox allows the plant research and breeding communities to study previously hidden genetic factors and agricultural traits.

[0084] According to certain aspects, the present invention provides a system for inducing knockdown or knockout of multiple genes in a plant, comprising: at least one Cell typespecific Multi-targeted AmiRNA and / or CRISPR library (cMAC library) comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more amiRNAsequences and / or one or more sgRNA sequences, wherein each of the amiRNA and / or sgRNA sequences targets multiple genes.

[0085] According to some embodiments, the AmiRNA and / or CRISPR library comprises at least 20, 50, 100, 200, 300, 400, 500, 1000, 1500, 2000 or more polynucleotides, each encoding one or more unique sgRNA sequences and / or one or more unique amiRNA sequences.

[0086] According to some embodiments, each of the sgRNA and / or amiRNA sequences target between 2-15 genes. According to some embodiments, each of the sgRNA and / or amiRNA sequences targets between 2-6, 2-7, 2-8, 2-9, 2-10, 2-11, 2-13, 2-14 or 2-15 genes.

[0087] According to some embodiments, the library is an amiRNA library. According to some embodiments, the library is a CRISPR library. According to other embodiments, the library is a combined amiRNA and CRISPR library.

[0088] According to some embodiments, the expression of the polynucleotide encoding the sgRNA sequence and / or amiRNA sequence is regulated by a tissue-specific promoter.

[0089] According to these embodiments, the polynucleotide comprises a tissue specific promoter operably linked to the one or more sgRNA encoding sequences and / or amiRNA encoding sequences.

[0090] According to some embodiments, the expression of the polynucleotide encoding the sgRNA sequence and / or amiRNA sequence is regulated by a developmental-stage specific promoter. According to these embodiments, the polynucleotide comprises a developmental - stage specific promoter operably linked to the one or more sgRNA encoding sequences and / or amiRNA encoding sequences.

[0091] As used herein, a coding nucleic acid sequence is “operably linked” to a regulatory sequence (e.g., promoter) if the regulatory sequence is capable of exerting a regulatory effect on the coding sequence linked thereto. As used herein, the term “promoter” refers to a region of DNA which lies upstream of the transcriptional initiation site of a gene to which RNA polymerase binds to initiate transcription of RNA. The promoter controls where (e.g., which portion of a plant) and / or when (e.g., at which stage or condition in the lifetime of anorganism) the gene is expressed. A promoter can be an endogenous or a heterologous promoter with respect to the gene (polynucleotide) controlled thereby.

[0092] According to some embodiments, the promoter is a tissue-specific promoter selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root-specific cortex promoter, shoot promoter, and root promoter. Each possibility represents a separate embodiment of the invention.

[0093] According to some embodiments, the library comprises a plurality of polynucleotides each encoding sgRNA sequence or amiRNA sequence, wherein the expression of the plurality of polynucleotides is regulated by a same tissue-specific promoter. According to other embodiments, the library comprises at least two polynucleotides each encoding sgRNA sequence or amiRNA sequence the expression of which is regulated by a different tissuespecific promoter. According to additional embodiments, the library comprises polynucleotides encoding sgRNA sequence or amiRNA sequence regulated by at least 2, 3, 4, 5, 6, 7, 8, 9, 10, or more different tissue-specific promoters.

[0094] According to some embodiments, the AmiRNA or CRISPR library is configured to ensure that at least two different sgRNAs or amiRNAs are inserted to each plant. In some embodiments, the library comprises DNA vectors comprising two polynucleotides, each encoding a different sgRNA or amiRNA. In other embodiments, ensuring that at least two different sgRNAs or amiRNAs are inserted to each plant is done by inserting two different DNA vectors, each comprising a different selection marker and sgRNA or amiRNA.

[0095] According to some embodiments, the library comprises a plurality of polynucleotides, each encoding a plurality of different sgRNA or amiRNA.

[0096] According to some embodiments, each of the encoded plurality of different amiRNA or sgRNA target a different set of genes. According to some embodiments, each of the encoded plurality of different amiRNA or sgRNA target different set of genes within a single gene family. According to other embodiments, each of the encoded plurality of different amiRNA or sgRNA target the same set of multiple genes. According to some embodiments,each of the encoded plurality of different amiRNA or sgRNA target at least one same gene within a gene set.

[0097] According to other embodiments, the library comprises a plurality of polynucleotides, each encoding a same plurality of sgRNAs or amiRNAs.

[0098] It is to be understood that “(is) regulated by”, where referring to sgRNA or amiRNA, means regulation of the expression of said sgRNA or amiRNA sequences by a certain promoter.

[0099] According to some embodiments, the system comprises a plurality of libraries, each library comprising a plurality of polynucleotides encoding sgRNA or amiRNA sequences, wherein the expression of the plurality of polynucleotides within each library is regulated by a different tissue-specific promoter.

[0100] According to certain embodiments, the library comprises a plurality of sub-libraries, each sub-library comprising polynucleotides encoding sgRNA or amiRNA sequences regulated by a different tissue-specific promoter.

[0101] According to some embodiments, the library comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10 or more polynucleotides encoding sgRNA or amiRNA sequences, wherein the expression of each is regulated by a different tissue-specific promoter selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root cortex promoter, shoot promoter, and root promoter. Each possibility represents a separate embodiment of the invention.

[0102] According to some embodiments, the promoter is selected from the group consisting of SIG6, ARSK1, PGP4, CO2, KST1, SCR, S18, CORI3, IQD22, SUC2, ML1, and UBQ10. Each possibility represents a separate embodiment of the invention.

[0103] According to some embodiments, the library comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10 or more polynucleotides encoding sgRNA or amiRNA sequences, each is regulated by a different tissue-specific promoter selected from the group consisting of SIG6, ARSK1,PGP4, C02, KST1, SCR, S18, C0RI3, IQD22, SUC2, ML1, and UBQ10. Each possibility represents a separate embodiment of the invention.

[0104] According to some embodiments, the amiRNA sequences or sgRNA sequences target transportome genes.

[0105] According to some embodiments, the amiRNA sequences or sgRNA sequences target at least one gene member comprised within a gene set. According to certain embodiments, the gene set is selected from the group consisting of mTACT-CP, mTACT-APC, mTACT- MFS, mT ACT -DMT, mT ACT-MATE, mTACT -ABC, mTACT -PA, and mTACT -UF. Each possibility represents a separate embodiment of the invention.

[0106] According to some embodiments, the plant is selected from the group consisting of a wild plant, an agricultural crop plant or a cultivar, a genetically modified plant and a non- genetically modified plant.

[0107] As used herein, the terms “cultivated plant”, “crop plant”, “agricultural plant", or "plant of agronomic importance", are used interchangeably, and include plants that are cultivated by humans for food, feed, fiber, construction, fuel purposes and more. The term encompasses a whole plant, a grafted plant, ancestor(s) and progeny of the plants and plant parts (also referred to herein as “a portion”), including seeds, shoots, stems, roots (including tubers), rootstock, scion, and plant cells, tissues, and organs. The plant may be in any form, including suspension cultures, embryos, meristematic regions, callus tissue, leaves, gametophytes, sporophytes, pollen, and microspores. Plants that are particularly useful in the methods of the invention include all plants which belong to the kingdom Plantae, in particular monocotyledonous and dicotyledonous plants, including a fodder or forage legume, ornamental plants, and food crops.

[0108] According to certain embodiments, the crop plant is selected from the group consisting of rice, maize, wheat, barley, tomato, potato, eggplant, sesame, soybean, sunflower, canola, sugarcane, alfalfa, millet, Leguminosae (bean, pea), flax, lupinus, rapeseed, tobacco, and cotton. Each possibility represents a separate embodiment of the present invention.

[0109] According to some embodiments, the model plant is Arabidopsis.

[0110] According to some embodiments, the amiRNA sequences target at least one gene member comprised within a gene set, the gene set comprising 5,565 amiRNA sequences targeting 81.7% of Arabidopsis transportome.

[0111] According to some embodiments, the sgRNA sequences target at least one gene member comprised within a gene set, the gene set comprising 2,859 sgRNA sequences targeting 707 Arabidopsis transporter genes.

[0112] According to some embodiments, the library comprises a plurality of polynucleotides encoding different amiRNAs or sgRNAs, each targeting a single gene.

[0113] According to some embodiments, at least two of the amiRNAs or sgRNAs are different amiRNAs or sgRNAs targeting same plurality of genes.

[0114] According to some embodiments, the library is a combined AmiRNA and CRISPR library.

[0115] The present invention further comprises, according to certain aspects, a method of studying phenotypic effects of gene knockdown or knockout in a plant, comprising: transforming the system described herein into the plant and measuring resulting phenotypic effects.

[0116] The present invention further provides, according to certain aspects, a method for identifying multiple members within at least one gene set underlying a phenotype in a plant, comprising transforming the system described herein into the plant and measuring resulting phenotypic effects.

[0117] According to certain aspects, the present invention provides a method of identifying at least one phenotypic effect of gene knockdown or knockout in a plant, the method comprising:(i) transforming the system described herein into a plurality of plants, thereby producing a plant population, wherein each plant of the population comprises at least two knocked-down or knocked-out genes;(ii) examining the phenotype of each plant in the plant population to identify at least one modified phenotype; and(iii) identifying the at least two knocked-down or knocked-out genes associated with the at least one modified phenotype.

[0118] According to certain embodiments, identifying the at least one modified phenotype is performed by comparing the phenotype of the plant comprising the system of the invention to the phenotype of a corresponding plant not comprising said system and grown under the same growth conditions. According to some embodiments, identifying the at least one modified phenotype is performed by comparing the phenotype of the plant comprising the system of the invention to the phenotype of a corresponding plant as is known in the art.

[0119] The present invention further comprises, according to certain aspects, a method for identifying multiple members within at least one gene set underlying a phenotype in a plant, the method comprising:(i) transforming the system described herein into a population of plants, thereby producing a plant population wherein each plant of the population comprises at least one sgRNA or amiRNA targeting multiple gene members;(ii) screening the plant population for at least one selected phenotype; selecting at least one plant showing the at least one selected phenotype; and(iii) identifying in the selected plant at least one sgRNA or amiRNA targeting the multiple-gene members, thereby identifying said multiple gene members underlying said selected phenotype.

[0120] According to certain aspects, the present invention provides a method for identifying multiple members within at least one gene set underlying a phenotype, the method comprising:(i) clustering coding sequences within genetic data of a plant species to sequence clusters, each cluster representing a gene set;(ii) producing a CRISPR library comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more unique sgRNAs, wherein each of the sgRNAs targets a plurality of gene members comprised within the gene set, wherein each of said sgRNA expression is controlled by at least one tissue and / or developmental stage-specific promoter;(iii) transforming the library into a plurality of plants, thereby producing a plant population wherein each plant of the population comprises at least one sgRNA targeting multiple gene members;(iv) screening the plant population for at least one selected phenotype;(v) selecting at least one plant showing at least one selected phenotype; and(vi) identifying in the selected plant at least one sgRNA targeting the multiple-gene members; thereby identifying said multiple gene members underlying said selected phenotype.

[0121] According to some embodiments, each of the sgRNAs targets between 2-15, 2-10 or 2-8 gene members comprised within the gene set.

[0122] According to some embodiments, the promoter is a tissue specific promoter as described herein above. In some embodiments, the tissue specific promoter is selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root cortex promoter, shoot promoter, and root promoter.

[0123] According to certain aspects, the present invention provides a method for identifying multiple members within at least one gene set underlying a phenotype, the method comprising:(i) clustering coding sequences within genetic data of a plant species to sequence clusters, each cluster representing a gene set;(ii) producing an amiRNA library comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more unique amiRNAs, wherein each of the amiRNAs targets a plurality of gene members comprised within the gene set, wherein each of said amiRNA expression is controlled by at least one tissue and / or developmental stage-specific promoter;(iii) transforming the library into a plurality of plants, thereby producing a plant population wherein each plant of the population comprises at least one amiRNA targeting multiple gene members;(iv) screening the plant population for at least one selected phenotype;(v) selecting at least one plant showing at least one selected phenotype; and(vi) identifying in the selected plant at least one amiRNA targeting the multiple-gene members; thereby identifying said multiple gene members underlying said selected phenotype.

[0124] According to some embodiments, each of the amiRNA targets between 2-15, 2-10 or 2-8 gene members comprised within the gene set.

[0125] According to some embodiments, the promoter is a tissue specific promoter as described herein above.

[0126] The polynucleotides described herein encode an sgRNA sequence or amiRNA sequence that specifically targeting two or more plant genes, in particular transportome- related genes.

[0127] According to certain further aspects, the present invention provides an isolated polynucleotide comprising a tissue specific promoter operably linked to one or more amiRNA encoding sequences or one or more sgRNA encoding sequences, wherein each of the encoded amiRNAs or sgRNAs targets multiple plant genes.

[0128] According to certain embodiments, the plant genes are members of a gene set. According to certain embodiments, the members of the gene set share a certain structural homology. According to certain additional or alternative embodiments, the members of the gene set participate in a plant functional pathway.

[0129] The tissue specific promoters are as described hereinabove.

[0130] According to certain further aspects, the present invention provides a plant or a plurality of plants, each comprising at least one polynucleotide comprising a tissue specific promoter operably linked to one or more amiRNA encoding sequences or one or moresgRNA encoding sequences, wherein each of the encoded amiRNAs or sgRNAs targets multiple plant genes.

[0131] According to certain embodiments, each plant of the plurality of plants comprises a polynucleotide having the same configuration of the tissue specific promoter and one or more amiRNA encoding sequences or sgRNA encoding sequences.

[0132] According to certain embodiments, each plant of the plurality of plants comprises a polynucleotide having the same tissue specific promoter and a different configuration of the one or more amiRNA encoding sequences or sgRNA encoding sequences.

[0133] According to certain embodiments, the plurality of plants comprises plants comprising different polynucleotides, the different polynucleotides comprise different tissue specific promoters and the same or different configuration of the one or more amiRNA encoding sequences or sgRNA encoding sequences.

[0134] The term “transportome” is used herein as known in the art and refers to the entire set of transport proteins or transport-related proteins within a plant. These proteins are responsible for transferring molecules, ions, and other substances across cellular membranes, both within cells and across different cellular compartments.

[0135] The term “library” as used herein refers to a collection of similar sized DNA fragments, a collection that includes several different items. “Library” or “sub-library” are interchangeable and depend on the context.

[0136] As used herein, the term "genetic redundancy" refers to the existence of multiple different genes performing the same or similar biological function, and that inactivation of only one, or even several of these genes but not all, has little to no effect on the phenotype.

[0137] As used herein, the term "a plurality" refers "at least two", typically more than two.

[0138] The term “CRISPR library” is used herein as known in the art and refers to a collection of single guide RNAs (sgRNAs) designed to target various genes. The sgRNAs guide the Cas9 or another CRISPR-associated protein to specific locations in the genome, enabling the CRISPR system to modify or edit the genomic sequence. The library may optionally include a Cas9 or another CRISPR-associated protein.

[0139] According to certain embodiments, the library is a CRISPR library further comprising a nucleic acid sequence encoding an RNA-guided DNA endonuclease enzyme. According to certain exemplary embodiments, the endonuclease is Cas9.

[0140] As used herein, the terms "single guide RNA", "sgRNA" and "gRNA" are used herein interchangeably and refer to a piece of RNA that function as guides for RNA- or DNA- targeting enzymes, which they form complexes with. The targeting specificity of the CRISPR / Cas system is determined by a short sequence (e.g., 20-nt) at the 5' end of the gRNA. The desired target sequence must precede the protospacer adjacent motif (PAM). After base pairing of the gRNA to the target, Cas mediates a double strand break about 3-nucleotides (nt) upstream of PAM. The term sgRNA may also refer to a DNA encoding the sgRNA, depending on the context and as is known in the art.

[0141] The term “amiRNA library" are used herein to describe a collection of constructs comprising polynucleotides encoding amiRNAs.

[0142] The term “amiRNAs” or “artificial microRNAs” are used herein as known in the art and refers to engineered molecules that mimic the structure and function of naturally occurring microRNAs (miRNAs) but are designed to target specific genes of interest. miRNAs are small, non-coding RNAs that regulate gene expression by binding to complementary sequences in messenger RNAs (mRNAs), leading to the degradation of the mRNA or inhibition of its translation into protein.

[0143] The term "gene family" refers to a group of related genes that share a common ancestor. The present invention further encompasses a population of plants, plant cells, or host cells that comprises the library or libraries described herein. The libraries described herein are transformed into a cell culture or tissue culture, from which a population of plants, each containing at least one polynucleotide from the library, is grown. The transformation is as known in the art.

[0144] The polynucleotides of the library are contained within DNA constructs or plasmids suitable for introduction into plants. Numerous plant transformation vectors and methods for transforming plants are available. Plant transformation vectors that may be used in the present invention include, for example, T-DNA vectors or plasmids, which are suitable for use in Agrobacterium-mediated transformation methods.

[0145] The methods of the invention may include a step of introducing the library’s polynucleotides into a population of plants. By “introducing” is intended presenting to the plant the polynucleotide construct in such a manner that the construct gains access to the interior of a cell of the plant. The methods of the invention do not depend on a particular method for introducing a library’s polynucleotides to a plant, only that the polynucleotide construct gains access to the interior of at least one cell of the plant. Methods for introducing polynucleotide constructs into plants are known in the art including, but not limited to, stable transformation methods, transient transformation methods, and virus-mediated methods. By “stable transformation” is intended that the polynucleotide construct introduced into a plant integrates into the genome of the plant and is capable of being inherited by progeny thereof. By “transient transformation” is intended that a polynucleotide construct introduced into a plant does not integrate into the genome of the plant.

[0146] For the transformation of plants and plant cells, the library’s polynucleotides of the invention are inserted using standard techniques into any vector known in the art that is suitable for expression of the nucleotide sequences in a plant or plant cell. The selection of the vector depends on the preferred transformation technique and the target plant species to be transformed. Methodologies for constructing plant expression cassettes and introducing foreign nucleic acids into plants are generally known in the art and have been previously described. For example, foreign DNA can be introduced into plants, using tumor-inducing (Ti) plasmid vectors. Other methods utilized for foreign DNA delivery involve the use of PEG mediated protoplast transformation, electroporation, microinjection, and biolistics or microprojectile bombardment for direct DNA uptake.

[0147] The method of transformation depends upon the plant cell to be transformed, stability of vectors used, expression level of gene products and other parameters. Other non-limiting examples for suitable methods of introducing nucleotide sequences into plant cells and subsequent insertion into the plant genome include microinjection, electroporation, Agrobacterium-mediated transformation, direct gene transfer, and ballistic particle acceleration.EXAMPLESReference is now made to the following examples, which together with the above descriptions, illustrate the invention in a non-limiting fashion.Example 1: Constructing cell type and tissue-specific vectors to drive multi -targeted amiRNAs

[0148] Traditional forward genetic screens typically fail to identify homologous genes with overlapping functions. In these cases, silencing multiple gene family members is required to uncover loss-of-function phenotypes. The Phantom transportome library, which was designed to target homologous transporter genes in Arabidopsis, has several drawbacks including the decay of amiRNA-mediated silencing over generations, lethality, and pleiotropic effects that interfere with phenotype interpretation. To address these challenges, the present disclosure relates to the construction of a new library of amiRNAs to target multiple members of each transporter gene family across the Arabidopsis genome in a cell type-specific manner. The presently disclosed novel tool allows the design of spatial forwardgenetics screens, expanding our ability to identify signaling molecules produced in one cell type that act in a distal tissue, an approach not available to the plant genetic community thus far.

[0149] To allow library-level, cell type-specific amiRNA expression, we constructed a constitutive promoter vector (pUBQ) and 11 distinct cell type-specific promoter vectors: shoot epidermis (pMLl), palisade mesophyll (pIQD22), endodermis and bundle sheath (pSCR), phloem companion cells (pSUC2), spongy mesophyll (pCORI3), guard cell (pKSTl), xylem (pS18), root-specific epidermis and cortex (pPGP4), root cortex (pCO2), shoot (pSIG6), and root (pARSKI )(Table 1). The expression patterns of these promoters were confirmed using NLS-YFP, YFP, H2B-GFP, GFP, or GUS reporters (Fig. 1). Outside of the leaf, pCORI3 and pIQD22 drove additional expression in root and stem cell types. All promoters were cloned into compatible vectors that allows single-step highly efficient amiRNA transformation. It was expected that expression of the transportome-specific amiRNA libraries under different cell type-specific promoters would result in less silencing over generations and decreased probability of lethal and pleiotropic effects compared to the 35S lines.Example 2: Generating cell type-specific, multi-targeted amiRNA libraries

[0150] To construct the cell type-specific promoter vectors, first, destination vectors were generated with open Golden Gate restriction sites for the amiRNA library cloning (Fig. 2A). Next, the Phantom transporter-specific amiRNA sub-library, containing 1,777 amiRNAs targeting transporter families in bulk, were cloned into destination vectors with tissuespecific promoters using primers containing Golden Gate restriction sites at the termini (Fig. 2A). To verify the representation and equal distribution of the amiRNAs in the cloned libraries, we deep-sequenced the pUBQ, pSIG6, pKSTl, pC02, pSUC2 and pPGP4 libraries. The data showed coverage that ranged from 86% to 96.6% of the 1,777 amiRNAs in the original Phantom transportome library and skewness between the values 0.322 to 1.139 (Fig. 2B). In addition, there was high variation between the amiRNAs distribution within the libraries (Fig. 2B). Non-equal cloning is not optimal as, ideally, each amiRNA should be screened at a similar frequency.Example 3: CRISPR based system - The design of the multi-targeted DENTs, an sgRNA library targeting differentially-expressed nutrient transporters

[0151] To identify potential genes involved in nutrient uptake, we incorporated data from an RNA-Sequencing (RNA-Seq) study that examined the expression levels of different genes in Arabidopsis roots upon nutrient deficiency. The study utilized hydroponically grown Arabidopsis plants of the Columbia (Col-0) ecotype at 4 weeks of age. These plants were exposed to either low nutrient levels or normal nutrient levels for a duration of 4 days. RNA was extracted from the roots and subjected to paired-end RNA-Seq analysis to determine changes in gene expression levels. In total, 33,558 annotated genes were analyzed, and 5,609 genes showed significant differential expression under nutrient deficiencies, as indicated by adjusted P values (q values) < 0.05. Considering that transporters are well-known to play a vital role in nutrient uptake processes, we intersected the differentially expressed genes with the list of transporters obtained from the ARAMEMNON database. This intersection revealed 452 distinct differentially-expressed nutrient-deficiency transporters (DENTs). Among these genes, 72.8% exhibited differential expression specifically under N deficiency, with 40.9% exclusively displaying differential expression under N deficiency (Fig. 3).

[0152] The gene annotations, genome assembly, and gene families for Arabidopsis thaliana were obtained from PLAZA 5.0., resulting in a total of 114 gene families, harboring one or more DENTs. For very large families, finding efficient sgRNAs that target all family members may be difficult. In addition, distant homologs are less likely to be functionally- redundant. Hence, large gene families were divided into homologous subfamily clusters of eight genes or less. Out of the 114 gene families, 47 harbored more than eight genes and were partitioned into subfamilies. Subsequently, subfamily clusters that did not contain any DENTs were discarded. This pre-processing procedure resulted in 201 homologous families and subfamilies containing one or more DENTs.

[0153] Following the family extraction, sgRNAs were designed using a multi -targeting sgRNA design tool CRISPys (Hyams et al. J Mol Biol (2018) 430, 2184-2195), and they were summarized into a list of 2859 sgRNAs (Fig. 4A), targeting 707 genes in total, including all 452 DENTs and their closely-related homologs. The majority of sgRNAs (59.3%) were designed to target two homologs, while 6.9% had the capacity to target four or more genes (Fig. 4B). Singleton sgRNAs represented 20.2% of the total sgRNA pool. Most genes (37.4% of the DENTs, 30.7% of all genes) were covered by eight sgRNAs, as expected from setting eight as the number of top sgRNAs per internal node (Fig. 4C). 3.5% of all genes were covered by twenty or more sgRNAs, with up to 27 sgRNAs targeting a single gene.

[0154] Each sgRNA targets 2 to 8 genes, with an average number of 2.04 genes (Fig. 5A- 5B). In cases where a gene target did not have a direct family member, we generated a sgRNA that is designed to target only 1 gene, holding to 20.7% of the cases. The number of mismatches of the sgRNAs in the library is between 0 to 2.5, with an average number of 0.706 mismatches (Fig. 5C). In our efforts to increase the flexibility of the trait-specific forward genetics tool, the sgRNAs were classified into 7 subgroups according to different nutrient transporter genes. Each sgRNA was associated with one or more nutrient groups, thus creating the following seven sub-libraries: Nitrate (N): 2,296 sgRNAs targeting 329 genes; Potassium (K): 1,591 sgRNAs targeting 215 genes; Phosphate (P): 548 sgRNAs targeting 68 genes; Calcium (Ca): 544 sgRNAs targeting 62 genes; Iron (Fe): 251 sgRNAs targeting 30 genes; Magnesium (Mg): 197 sgRNAs targeting 25 genes; and Sulfur (S): 177 sgRNAs targeting 30 genes (Fig. 5D).Example 4: Designing multiplexed and multi-targeted CRISPR library

[0155] Thus far, our previous Mulit- Knock library, as well as libraries developed by others, possessed a single sgRNA. We envisioned that having a multiplexed library of 2 sgRNA on a single vector should have a significant impact on the overall tool efficiency in terms of gene coverage, number of mismatches, and, thus, cleavage rate. To generate the library, we developed a new algorithmic code that computed combinatorial factors that would allow targeting different genetic combinations from the same monophyletic cluster.

[0156] The multiplexed library design was then applied to the putative nutrient transporter gene list we used for the single sgRNA library. The new multiplexed library contained 1,599 multiplexes, comprised of 1,826 unique sgRNAs, that were designed to target a total of 694 genes. In the multiplexed library, each multiplex targets between 2-13 genes, with an average of 4.36 target genes (Fig. 5E-5F). Furthermore, the multiplexed library has reduced the average number of mismatches compared to the single sgRNA library, with an average of 0.347 number of mismatches (Fig. 5G). Each multiplex in the library was associated with one or more nutrient groups: Nitrate (N): 1,370 sgRNAs; Potassium (K): 948 sgRNAs; Phosphate (P): 399 sgRNAs; Calcium (Ca): 395 sgRNAs; Iron (Fe): 158 sgRNAs; Magnesium (Mg): 151 sgRNAs; and Sulfur (S): 149 sgRNAs (FIGs. 5H-5I).

[0157] The network relationship of the multiplexes targeting genes in the library highlights the robustness and coverage of the library design (FIGs. 5J-5K). Thus, our multiplexed approach allows us to incorporate two unique sgRNAs into every plant, addressing functional redundancy more comprehensively. This approach not only magnifies the efficiency of our gene editing efforts but also allows us to target more genes with fewer vectors, therefore plants.Subsequently, we synthesized and cloned the single sgRNA sub-libraries and the multiplexed library under the RPS5 constitutive promoter: pRPS5:K, pRPS5:N, pRPS5:P, pRPS5:S, pRPS5:Mg, pRPS5:Ca, pRPS5:Fe and pRPS5.-Multiplex. To verify the representation and equal distribution of the sgRNAs in the cloned libraries, we deep-sequenced them. The deep sequencing results show an equal representation of the sgRNAs and multiplexes in the library (Fig. 5N) and further that the sub-libraries contain only the sgRNAs that belong to the defined families as expected (FIGs. 5L-M). The data showed coverage that ranged from 98.69% to100% and skewness between the values of -1.458 to 0.607 (FIGs. 5L-5M). To verify that the transition to Agrobacterium does not affect the library coverage and distribution, we sequenced pRPS5:Midtiplex library and compared it to E. coli. We obtained similar results between libraries, with an R2= 0.9 (Fig. 50), suggesting a robust cloning system and verifying that the tool is ready for plant transformation.Example 5: Construction of cell type-specific CRISPR libraries

[0158] Stable and inherited mutations, driven by a constitutive or egg -cell specific promoter, are the common and straightforward way to create mutations and analyze a given phenotype. However, in several cases, it can cause drawbacks. Lethal and sterile plants, or general pleiotropic effects, are examples of such pitfalls. Therefore, to address these shortcomings, we generated a cell-type-specific CRISPR system that would allow us to design spatial genetics screens. For example, expanding our ability to identify signaling molecules produced at one cell type and act in a distal tissue or creating a system that addresses a specific function such as nutrient uptake from the root epidermis. To generate the tissuespecific cloning vectors for the CRISPR Cas9 libraries, we constructed new cloning vectors containing the following properties (Fig. 6A). First, we cloned golden gate cloning sites that will allow easy cloning of promoters of interest. Downstream to those sites, there is intronized Cas9, which has 12 introns integrated into the maize codon-optimized Cas9. These introns were shown to have a dramatic positive effect on the Cas9 genome editing efficiency in Arabidopsis. Downstream to the intronized Cas9, the vector contains Bsal sites that are used for cloning the CRISPR library (or for any sgRNA of interest). For this study, we have chosen 8 tissue-specific promoters: pARSKl ; expressed in the root, pKSTl ; expressed in the guard cells, pSCR expressed in the endodermis and bundle sheath, pSIG6 expressed in the shoot, pSUC2'. expressed in the phloem companion cells, pCOREA. expressed in the spongy mesophyll, pPGP4'. expressed in the root cortex and epidermis, and pMLE, expressed in the shoot epidermis (Fig. 6B). These promoters allow tissue-specific Cas9 expression, leading to cell type-specific DNA mutations in plants using CRISPR. As a proof of concept, we cloned two libraries under tissue-specific promoters: pPGP4:Fe and pARSKl :Midtiplex. The deep sequencing results show an equal representation of the sgRNAs in the libraries, the data showed coverage that ranged from 98.12% to 100% and skewness between the values of - 1.458 to 0.954 (Fig. 6C-6D).

[0159] The two cloned tissue-specific libraries were transformed into the background of Col- 0 plants as one bulk transformation. 500 T1 resistant plants from pPGP4:Fe were selected using Basta. Several T1 plants from the pPGP4:Fe library (root- specific epidermis and cortex promoter) have shown a striking physiological phenotype displaying defective shoot growth. For example, we isolated an independent line with pale, bleached and small-size shoots. Extracting DNA, amplifying the sgRNA cassette and sequencing revealed that this sgRNA putatively targets IRT1 and IRT2 (Fig. 6E). This phenotype resembles the irtl mutant, while irt2 single mutant does not have a visible loss of function phenotype. The double mutant irtl / irt2 has not been reported so far. Sequencing the root DNA indicated for cell typespecific mutations (Fig. 6E). This DNA cleavage did not take place in the shoot. The results demonstrate our ability to knockout multiple genes at a library level, tissue-specific manner. The screen has also recovered the known nutrient transporters SULTR genes, as the pPGP4:sg-SULTRl;l,l;2 exhibited a shoot growth inhibition phenotype (Fig. 6E). SULTR1.1 and SULTR1.2 are known sulfate transporters responsible for sulfate uptake from the soil. pPGP4:sg-SULTRl;l,l;2 showed tissue -specific knockout using sanger sequencing results (Fig. 6E).Example 6: cMAC: next-level multi-targeted amiRNA and CRISPR transportome library in Arabidopsis

[0160] To further improve the ability to knock down multiple homologous genes of the transportome simultaneously and to address that some amiRNAs were over-represented by several folds compared to others in the Phantom-derived libraries, we constructed the cMAC- library of amiRNAs designed to target multiple members of each transporter gene family across the Arabidopsis transportome. The cMAC -library has the following improvements relative to the Phantom transportome library: 1) an expanded amiRNA arsenal of thousands of new multi-targeted amiRNAs, 2) an improved algorithm that accounts for homologous relationships within gene families, thus increasing the probability that the designed amiRNAs will silence functionally redundant genes, and 3) targeted sub-libraries that allow exclusive amplification of amiRNAs targeting defined transporter gene families. We envisioned that cMAC would allow, for the first time, robust, high-quality, tissue-specific transportome analysis.

[0161] To construct a library of amiRNAs that target multiple members of each gene family in the Arabidopsis transportome, we obtained a list of genes with putative transporter function, as annotated in the ARAMEMNON and PLAZA databases. After excluding singleton genes (i.e., genes with no family members) and genes without coding sequence data, the ARAMEMNON databased includes 1,158 genes from 83 families, and the PLAZA database includes 1,325 genes from 189 families. Together the two classifications include 1,471 genes that encode putative transporters. We then designed a set of amiRNAs that optimally target multiple members of each gene family while accounting for the similarity among family members (see Methods). Briefly, we first reconstructed the gene tree of each gene family. In these trees, homologous subgroups of genes that are closely related to each other are placed close together. We then traversed over the internal nodes of the tree and designed the optimal amiRNAs for each subgroup (i.e., the list of genes that are the descendants of the specified internal node) using the WMD3 program and according to the hybridization energy of the amiRNAs, while excluding amiRNAs with likely off-target activity (Fig. 8A). The maximal number of amiRNAs per internal node was set to 8.

[0162] The cMAC library antisense and sense sequences were then introduced in silico into the miRNA159a backbone sequence. amiRNA159a has previously been shown to be non- mobile, and efficiently silence gene expression in several studies in Arabidopsis. The entire amiRNA library was divided into eight different sub-groups using unique adaptors. These adaptors were added to allow amplification of a particular group of amiRNAs targeting a specific family or functional class (Tables 4 and 5). The groups include: channels and porins (CP); amino acid / polyamine / organo-cation, cation carriers, and mitochondrial carriers (APC); major facilitator superfamily (MFS); drug / metabolite transporter group (DMT); multi -drug and toxic compound extrusion and other carrier transporters (MATE); ATP binding cassette family (ABC); primary active transporters (PA); unknown function (UF). The resulting concatenated sequences (including backbone, antisense, sense, and adaptors) were removed if the sequence contained a Bsal restriction enzyme recognition site since Bsal was used for Golden Gate amiRNA library cloning.

[0163] Following the filtering procedure, the in-silico library contained 5,565 amiRNA sequences, targeting 1,202 genes, 81.7% of the putative transporter-encoding genes in Arabidopsis (Table 4). Notably, 269 genes predicted to encode transporters (18.3 %) are nottargeted by any amiRNA (Table 4). Each amiRNA targets two to eight genes from the same family (Fig. 7B, Fig. 10, Fig. 11). Moreover, each internal node in the phylogenetic tree was targeted up to eight times (Fig. 7C, Fig. 10B, Fig. 12). Most of the amiRNAs target two genes with a gradual decrease in the number of amiRNAs targeting three or more genes up to eight, which was set as a cut-off. Each gene is also targeted by more than one amiRNA, thus enabling various genetic combinations (FIGs. 10-13). Furthermore, in each sub-group, the number of genes targeted is between 77 and 205 genes (Fig. 7D, Table 4), and each subgroup contains between 402 and 858 amiRNAs (Fig. 7E, Table 4). The network relationship of the amiRNAs targeting genes in the library highlights the robustness and the coverage of the library design (Fig. 7F-7G, Fig. 14).Example 7: Constructing the cell type-specific cMAC toolbox

[0164] To construct the cMAC libraries driven by cell type-specific promoters, we cloned the entire cMAC library (5,565 amiRNAs) under control of two different cell type-specific promoters S18 and SUC2 (xylem and phloem companion cells respectively). In addition, we cloned four representative sub-libraries under the constitutive UBQ10 promoter, two representative sub-libraries under control of the promoter SUC2, and one representative sublibrary under the root-specific promoter ARSK1 (Fig. 8A). To confirm that there is a complete and equal representation of amiRNA distributions and that each cMAC sub-library contains expected amiRNAs, we deep-sequenced the cloned libraries. The deep sequencing results showed 100% coverage for all libraries, and a bell-shaped representation of the amiRNAs in the library (Skew = 0.113 - 0.713) (Fig. 8B).

[0165] Next, we transformed Arabidopsis of the Col-0 background with DMT cMAC sublibrary driven by the phloem companion cell-specific promoter SUC2 (pSUC2:cMAC- DMT). T1 seeds were selected for BASTA resistance, and seeds from 180 T2 plants were collected individually. 25 T2 seeds were sown on plates, shifted to soil after one week and grown under long-day conditions. In a phenotypic screen of T2 plants from the pSUC2:cMAC-DMT library focused on shoot growth, leaf color, and morphology, we identified a line with a striking physiological phenotype displaying senesced leaves starting from day 30 compared to Col-0 (Fig. 9A). The phenotype showed dominant segregation in T2 plants, suggesting that it is driven by the amiRNA and not by an off-target loss-of-functionmutation. This amiRNA line putatively targets three closely related genes from the Nucleotide Sugar Transporter (NST) family (part of the Drug / Metabolite Transporter (DMT) super-family): AT2G25520 (UAfT4), AT4G32390 (UAfT3), AT5G25400 (UAfTl) (Fig. 9B, Table 3). Hereafter, we refer to this line as pSUC2:miR-3UAfT.

[0166] Using a phenomics system, we evaluated shoot morphology, photosynthesis, and plant color shade parameters under normal growth conditions for the pSUC2:miR-3UAfT line. Compared to wild-type, the shoot area of pSUC2:miR-3UAfT was smaller (Fig. 9C), and the leaf shape was sharper and thinner (Fig. 15A). Maximum PSII quantum yield (Fv / Fm) measurements, indicating of plant photosynthesis performance, implied that the photosynthesis rate of pSUC2:miR-3UAfT was lower compared to the control (Fig. 9D), and water content measurements indicated reduced water content in pSUC2:miR-3UAfT plants (Fig. 9E). Relative color shade analysis showed that the color of pSUC2:miR-3UAfT leaves is visibly different compared with wild type plants (Fig. 15B). Interestingly, the three UAfT transporters, together with UAfT2 (AT5G11230), were characterized as Golgi-localized UDP-arabinofuranose (UDP-Araf) transporters in Arabidopsis. Proteoliposome-based transport assays showed that the UAfT transporters could efficiently transport UDP-Araf in vitro. Notably, we did not verify the on-target activity of pSUC2:miR-3UAfT by repeating the phenotype using independent lines. Generating a triple mutant using combined T-DNA lines, CRISPR, or an intendant amiRNA line would allow validation that the phenotype is not driven by an off-target genetic factor.

[0167] L- Arabinose (Ara) is a plant-specific monosaccharide found in different cell wall- associated molecules. It exists in two different ring forms, the furanose (Araf) and the pyranose (Arap). Ara is initially synthesized as a form of UDP-Arap (Kotake et al., 2016. Journal of Plant Research 129: 781-792.). The synthesized UDP-Arap is transported from the Golgi lumen to the cytosol, where a UDP-Ara mutase (UAM) interconverts UDP-Arap to UDP-Araf. UDP-Araf is then transported back into the Golgi lumen, via specific UAfT transporters to be incorporated into Ara-containing cell wall-associated. While UAfT overexpression lines showed a strong phenotype, the loss-of-function of the single mutants did not present a major change in the abundance of monosaccharides in dry seeds or growth and physiology-related phenotypes. As such, it has been hypothesized that functional redundancy likely plays a role in the UAfT family. Here, our cMAC system successfullyovercame these limitations and identified a phenotypic defect when homologous UAfT family members were collectively targeted in a specific cell-type.Discussion

[0168] cMAC is a genetic toolbox of amiRNAs and sgRNAs, each designed to optimally target multiple members of a particular gene family encoding transporter proteins. In total, the cMAC toolbox includes 5,565 amiRNAs targeting 81.7% of the transportome, and 2,859 sgRNAs targeting 707 genes with possible involvement in nutrient uptake. Expression of the amiRNAs or sgRNAs is driven ubiquitously or using cell type-specific promoters, allowing the design of spatial-specific genetic screens. cMAC can be implemented into any forward- genetic screen. For example, one may transform cMAC into the wild-type background and carry out "classic" phenotypic screens. These screens are likely to reveal previously unreported redundant phenotypes, as we demonstrated here. In addition, we speculate that cMAC could be used in combination with more complex forward genetics screens by introducing the amiRNA library into a fluorescent reporter background or into a mutant or over-expression phenotype for a suppressor screen.

[0169] The cMAC toolbox provides several important improvements over classic genetics and the previously described Phantom amiRNA library. First, the new amiRNA target or sgRNA target combinations that are part of cMAC increase the probability of family knockdown and knockout phenotypes.

[0170] The cMAC CRISPR libraries include single and multiplex sgRNA libraries that allow optimal knockout of multiple closely related genes in specific cell types. In our quest to enhance the efficiency and precision of gene editing in plants and to further overcome the challenge of functional redundancy, we engineered a cutting-edge multiplex library. This approach involved incorporating two unique sgRNAs into every plant, allowing it to address functional redundancy more comprehensively. This approach not only magnifies the efficiency of our gene editing efforts but also allows us to target more genes with fewer vectors / plants and significantly expand the gene coverage and cleavage efficiency at a library scale. Using the improved algorithm allowed us to design the first trait-specific libraries. As a proof of concept, we generated nutrient transporters-specific libraries. Understanding the dynamics of nutrient uptake, transport, assimilation, and their biological interactions iscrucial for developing crops with greater nutrient efficiency and improved productivity. In addition to the broad multiplexed library that targets all 694 DENTs, we designed 7 sublibraries, one for each of the N, K, P, Ca, Fe, Mg, and S nutrient transporters. These sublibraries allow scientists to flexibly design targeted genetic but in large-scale. The ability to drive the CRISPR libraries under a cell type-specific promoter is an essential addition to the toolbox, as it allows studies specific processes in the plant. For example, despite this molecular-level understanding of Fe transport, there are significant gaps in the knowledge of how Fe travels from the soil to leaf chloroplasts and seeds. It is currently unclear how Fe is exported from the epidermis cells, how Fe bypasses the endodermis cell file with the Casparian strip and suberin barriers, how Fe exits the pericycle cells to translocate into the xylem, and how Fe enters the chloroplast outer envelope. The cell type-specific libraries we generated in this study can allow studying those specific questions. Furthermore, the tissuespecific system bypasses issues of lethal phenotypes caused by sgRNAs expressed under the control of a constitutive promoter.

[0171] The cMAC amiRNA library includes 5,565 amiRNAs not present in the Phantom library. Each amiRNA or sgRNA was designed to target several genes from the same family, and numerous amiRNAs or sgRNAs target each gene. Such combinations are ideal for overcoming genetic linkage. Second, the bottom-up amiRNA and sgRNA plotting for each node used in library design maximized the likelihood that redundant functions would be revealed. This design was based on the premise that closely related genes with similar protein sequences will share an overlapping function. Therefore, genes from the same family that do not share a close node on the phylogenetic tree are not targeted by the same amiRNA. Third, the entire amiRNA library was divided into eight sub-libraries, targeting specific functional protein classes. These sub-libraries will allow scientists to flexibly design genetic screens and thus enhance the success rates. Fourth, the ability to drive the amiRNA library under a cell type-specific promoter is an essential addition of cMAC. Cell-file-specific expression of the amiRNA library allows particular processes in the plant to be tackled. The tissue-specific system bypasses issues of lethal phenotypes caused by amiRNAs expressed under the control of the 35S promoter. Finally, the cMAC amiRNA and CRISPR screening strategies feature several advantages over the classical forward-genetic approach. The amiRNAs are dominant in their activity, and therefore the phenotypes are visible at the T1 generation. In addition,the amiRNA or sgRNA can be sequenced to identify putative targets without need for laborious map-based cloning.

[0172] There are limitations to the cMAC approach. Using our strategy, thousands of amiRNAs can be expressed under the control of a tissue-specific promoter. Although miRNAs primarily function in a cell-autonomous manner, meaning they do not freely move in the plant over long- or short-distances, several reports have shown that specific miRNAs can move and that this movement is required for plant growth and survival. It has been suggested that macromolecules have the ability to move from their site of synthesis along the plant, which might allow them to function at a distinct target site. Such movement might trigger a systemic response throughout the plant via a non-cell-autonomous activity. A set of experiments addressing whether miRNA mobility is developmentally regulated have been conducted recently. We specifically selected miRNA 159a as a backbone for cMAC since it was not found to be mobile. To rule out the possibility of amiRNA movement in a specific context, an amiRNA targeting GFP or GUS expressed under the respective cell type-specific promoter in the genetic background of ubiquitously expressed GUS and GFP reporters could be performed. Just as with any other forward and reverse genetic approach, it is essential to validate the on -target activity by analyses of independent lines. Validation experiments may involve different amiRNAs targeting the same set of genes, CRISPR, or a combination of T- DNA lines. However, the latter might be a problem if the target genes are genetically linked.

[0173] In summary, the cMAC toolbox developed here allows scientists to overcome the functional redundancy of plants in a forward-genetic manner and will allow, for the first time, simultaneous and specific targeting of transporter genes from specified families. cMAC provides a unique toolbox for the scientific community that can be used to study the transport mechanisms of various signaling molecules in plants. The cMAC design and pipeline can be implemented at the genome-scale to study redundant families beyond the plant transportome.Materials and methods

[0174] Plant material and growth conditionsAll Arabidopsis thaliana lines used in this work are in Colombia background (Col-0 ecotype, Salk Institute La Jolla, CA, USA). Sterilized seeds were plated on Murashige & Skoog (MS) x 0.5 (Duchefa Biochemic) medium containing 1% sucrose and 0.8% plant agar(Duchefa Biochemic), pH was adjusted to 5.6-5.8 with 1 M KOH. For transgenic plant selection, antibiotics were added to a final concentration of 50 pg / ml kanamycin (Duchefa Biochemic), 100 pg / ml spectinomycin, 150 pg / ml gentamycin (Duchefa Biochemic), and 30 pg / ml hygromycin (Bio-Gold). Plates with seeds were stratified for 48 hours at 4°C then transferred to growth chambers (Percival CU41L5) at 21 °C, 100-120 pEm-2S-l light intensity under long-day conditions (16 h light / 8 h dark). For seed production, plant transformation, crossing, and soil pot assays, seeds were sown on wet soil. Plants were grown in growth rooms under long-day conditions at 21 °C. The following constructs were previously described: pMLl:H2B-GFP, pKSTl:GFP, pSUC2:YFP, pCO2:YFP, pSCR:YFP, pS18:YFP, pUBQ10:YFP.

[0175] Seed sterilizationSeeds were sterilized by vapor-phase sterilization (chlorine fumes) for 2.5 hours in Eppendorf tubes in the presence of 100 ml of 11% sodium hypochlorite and 5 ml of 32% hydrochloric acid in a sealed desiccator.

[0176] Bacterial material and growth conditionAll bacteria were grown on LB agar media: 20 g of LB and 15 g bacteriological agar (DIFCO) was added to 1 L doubly distilled water and autoclaved for 20 minutes at 120 °C. Antibiotics were added according to the specific resistances of bacteria at final concentrations of 50 pg / ml kanamycin, 100 pg / ml ampicillin, 30 pg / ml hygromycin, 100 pg / ml spectinomycin, 25 pg / ml gentamycin, 10 pg / ml tetracycline, and 25 pg / ml rifampicin. Plasmids were multiplied in chemically competent E. coli strain DH5a and extracted with a GenElute plasmid mini extraction kit (Sigma-Aldrich) following the manufacturer’s protocol.

[0177] Agrobacterium transformationElectro-competent Agrobacterium tumefaciens strain GV3101 was incubated on ice with 100 ng plasmids for 2 minutes, then electroporated in a MicroPulser (BIO-RAD) (2.2 Kv, 5.8 ms). Bacteria were transferred immediately to 1 ml liquid LB and shaken for 2 hours at 28 °C. Subsequently, bacteria were plated on LB agar plates containing the relevant antibiotics for 2 days at 28 °C.

[0178] Plant DNA extraction and PCRA “crude” DNA extraction method was used to extract plant genomic DNA to be used as a template for the PCR for sequencing and genotyping purposes. A few young leaves from Arabidopsis thaliana (about 100 mg) were placed in a 2 ml round-tip Eppendorf tube and frozen in liquid nitrogen. The leaves were crushed using a tissue-lyser to a thin powder and were homogenized with 400 pl DNA extraction buffer (200 mM Tris-HCL, pH 7.5-8.0, 25 mM EDTA, 250 mM NaCl, 0.5% SDS). The tubes were vortexed for 5 seconds and centrifuged for 1 minute at 13,000 rpm in an Eppendorf mini centrifuge. The supernatant was transferred to a new tube and DNA was precipitated with 300 pl isopropanol and incubated for 5 minutes at room temperature, followed by centrifugation at 13,000 rpm for 5 minutes at room temperature. The pellet was washed with 400 pl 70% EtOH and centrifuged for 1 minute at 13,000 rpm at room temperature, and the DNA pellet was dried and resuspended in 100 pl doubly distilled water. DNA amplification for sequencing and cloning was done by PCR in a Sensoquest labcycler using the Taq Ready Mix (HyLabs) following the manufacturer’ s protocol.

[0179] Histochemical GUS stainingFor histochemical detection of GUS activity, plant tissues were incubated for approximately 16 hours at 37 °C in 100 mM sodium phosphate buffer (pH 7.0) containing 0.1% Triton X-100, 1 mM 5-bromo-4-chloro-3-indolyl-P-D-glucuronic acid cyclohexylammonium salt (Sigma- Aldrich), 2 mM potassium ferricyanide, and 2 mM potassium ferrocyanide. Tissues were immersed in 70% ethanol until transparent. GUS- stained tissues were imaged using a Zeiss Stemi 2000-C stereomicroscope (Jefferson et al., 1987). Images were captured using ZEN software (Zeiss).

[0180] Stomatai conductance measurementsStomatai conductance to water was measured using a LI-COR LI-600 system on 30- day-old plants. Plants were grown in growth chambers under long-day conditions at 21 °C. Measurements were performed between 10:00 a.m. and 1:00 p.m.

[0181] Thermal imagingThermal images were obtained using an FLIR T6xx series instrument. The camera was placed vertically ~50 cm above the plants. Leaf temperature was quantified using a customized region-of-interest tool implemented according to the manufacturer’s instructions. Measurements were performed on 30-day-old plants.

[0182] Plant phenotypingFor leaf area measurements, plants were grown on soil, one plant per pot, in a growth chamber. A Nikon D5300 digital camera with a Nikon 60 mm Macro Lens was used to photograph 28-day-old plants. The surface area was measured using ImageJ software (http: / / rsbweb.nih.gov / ij / index.html). Petiole length measurements were performed on 28- day-old plants using a ruler. Petiole angles were quantified using ImageJ software (http: / / rsbweb.nih.gov / ij / index.html). Measurements were performed on 30-day-old plants.

[0183] Phylogenetic tree of the Arabidopsis GLR and DMT familiesProtein sequences for Arabidopsis thaliana GLR and DMT family members were retrieved from TAIR (https: / / www.arabidopsis.org). Phylogenetic relationships were defined using (http: / / www.phylogeny.fr / ) and visualized with FigTree software (http: / / tree.bio.ed.ac.uk / software / figtree / ).

[0184] Confocal imagingSeedlings were stained in 10 mg / L propidium iodide for 1 minute, rinsed, and mounted in water. Seedlings were imaged on a Zeiss LSM 780 laser scanning confocal microscope with the laser set at 514 nm for YFP and propidium iodide excitation. Emission filters used were 517-570 nm. Image analysis and signal quantification were done using ZEN lite 2012 software.

[0185] Tissue-specific promoter cloningTissue-specific promoters were amplified from genomic DNA or from plasmids containing the specific promoters using Phusion high fidelity Taq polymerase (New England Biolabs), following the manufacturer’s protocol. The amplification was carried out using primers containing restriction sites (Table 2) necessary for cloning into the destination plasmids pGREEN (pG0229-T) and pBIN PLUS.

[0186] Library cloningThe promoters used in this study for library construction were previously described and characterized: pPGP4, pARSKl, pSIG6, pKSTl were amplified from genomic DNA; pSUC2, pCO2, pS18, pCORI3, pIQD22 , pMLl, pSCR and pUBQlO were amplified from plasmids.

[0187] A plasmid library expressing a pool of amiRNAs was designed and synthesized. The amiRNA library was cloned into pGREEN or pBIN PLUS containing one of the cell typespecific promoters (SIG6, ARSK1, PGP4, CO2, KST1, SCR, SI 8, CORI3, IQD22, SUC2, ML1, or UBQ10) using the Golden Gate method. An aliquot of 2 pl of Golden Gate products was transformed into 50 pl E. coli DH5a competent cells by heat shock reaction at 42 °C for 30-60 seconds. For each construct, we did 18 transformations. After 1 hour of shaking in 37 °C, samples from three tubes were pooled, and the bacteria were plated on 145 / 20 mm LB plates containing 50 mg / ml kanamycin resulting in six plates for each construct. Cells were grown overnight in a 37 °C incubator. Bacteria from each plate were scraped into 1 L of sterile LB media with kanamycin, and cultures shaken at 37 °C overnight. For plasmid extraction, we used The QIAprep Spin Maxiprep (Qiagen 20-12162). Next, aliquots of 1 pl of DNA were transformed into the Agrobacterium pSoup bacteria system (Hellens et al., 2000, Plant Molecular Biology 42: 819-832) by electroporation. For each promoter, 18 transformations were performed. After 2 hours of shaking at 30 °C, the Agrobacterium pSoup were plated on 145 / 20 mm LB plates containing 50 pg / ml kanamycin, 10 pg / ml tetracycline, 25 pg / ml gentamycin, and 25 pg / ml rifampicin. Three tubes were pooled, resulting in six plates for each construct. Bacteria were grown overnight at 30 °C. Next, each construct was transformed into 6 trays of Col-0 as described above. T1 seeds were collected and sown on soil. Beginning when plants were two weeks old, they were sprayed with 0.1% BASTA every 3 days, 3 times in total times.

[0188] Deep sequencingFor deep sequencing analysis of the L10 transporter library, we started from the pUBQlO / pSIG6 / pKSTl / pCO2 / pSUC2 / pPGP4:L10 plasmids DNA. By PCR, we created an amplicon for the sense strand of miRNA319a using the following primers: forward, gcggtagacaaattggatc (SEQ ID NO: 44); reverse, caagaaaattggaatacaaaagagag (SEQ ID NO:45). For deep sequencing analysis of the cMAC transporter library, we started from the pG0229-T:cMAC plasmids DNA. By PCR, we created an amplicon for the sense strand of miRNA159a using the following primers: forward, gagctttaacttgcccttta (SEQ ID NO: 46); reverse, aagaaaaataaaaaatagagaaggtg (SEQ ID NO: 47). Following amplification, the PCR product was purified using the NucleoSpin Gel and PCR Clean-up system (MACHEREY- NAGEL), and samples were sequenced by Novogene. Deep sequencing data was analyzed using Python. Numbers of reads per amiRNA sequence were determined using the Biopython package. The results were analyzed using Excel.

[0189] amiRNA library constructionCoding sequences for each gene family were obtained from the TAIR 11 database. The nucleotide sequences were then translated to amino acids, and these were used as input to MAFFT to construct a multiple sequence alignment. MAFFT was executed with the algorithm parameter set to “auto” and the maxiterate parameter set to 1000. The resulting alignments were used for maximum likelihood reconstruction of the corresponding gene trees with PhyML, using the LG+I+G model and four categories of the gamma distribution. Each internal node in the reconstructed gene trees induces a subfamily of homologous genes, such that subfamilies induced by internal nodes that are further away from the root represent more closely related genes. We then traversed over the internal nodes of the tree and applied the WMD3 program to obtain a list of potential amiRNA sequences for each subfamily of genes of size n<l 1. WMD3 potentially outputs a large number of potential amiRNA. To keep only the most efficient amiRNAs, the list of amiRNA was filtered according to the following criteria, such that at most x = 8 amiRNAs are selected per internal node:

[0190] (1) amiRNAs that target less than two genes with free energy fraction above the threshold value Q were filtered (Q is computed relative to the predicted efficiency of a perfect match). Based on previous studies (Li et al., 2013), Q was set to 0.75.

[0191] (2) For each internal node, the amiRNAs were first sorted according to the number of targets with a free energy fraction of at least Q, and then according to the score obtained from the WMD3 output results. Then, in each subtree the amiRNAs were added iteratively from the root to the leaves (in a top-down fashion) until reaching a predefined number of amiRNAs per internal node (here, x=8), or until none were available. An amiRNA was addedif either of the two conditions was fulfilled: (a) The sequence of the candidate amiRNA differed by at least m=2 bases from the sequences of all amiRNAs that had been chosen thus far (either in the examined internal node or its ancestral nodes), (b) The sequence of the candidate amiRNA did not fulfill the first condition, but its WMD3 score was better than all other similar amiRNAs.

[0192] (3) For internal nodes with less than x = 8 selected amiRNAs, an additional procedure was applied to add amiRNAs that target a subgroup of genes in the subfamily (but are not induced by other internal nodes in the tree). To this end, for each gene family, WMD3 was executed on all the target genes with the "musthave" parameter set to 2, allowing to design amiRNAs for all possible sets of at least two genes. The candidate amiRNAs were ascribed to the internal node representing the most recent common ancestor (MRCA) of the targeted genes, ranked as in step (2) and added until the total number of amiRNAs per internal node is x = 8 . However, this option could provide candidate amiRNAs that target genes that are too distant from each other. To this end, we first clustered the targeted genes to monophyletic groups. For each pair of clusters, A and B, the distance between them was defined aswhere d(i,j) is the number of internal nodes between genes i and j. An amiRNA was added if the distance between all clusters was at most 3.

[0193] All amiRNAs included in the PHANTOM, LIO library were filtered.All amiRNAs (sense sequences) and antisense sequences that contained the Bsal restriction site sequence or its reverse complement were removed.

[0194] Next, sense and antisense strands were incorporated into the backbone sequence of amiRNA 159a, and adaptors were attached according to functional class. For each class, unique forward and reverse primers were designed. The oligonucleotides were synthesized by TWIST Bioscience.

[0195] Network constructionThe network graphs were created using Cytoscape

[0196] BioinformaticsDNA sequences alignment was performed using the SnapGene alignment software(http: / / www.snapgene.com / ) and the NCBI BLAST tool(https: / / blast.ncbi.nlm.nih.gov / Blast.cgi / ). Gene sequences and relevant information were obtained from the Arabidopsis Information Resource (https: / / www.arabidopsis.org / ).Tables

[0197] Table 1. Promoters used in this study to drive cell type-specific mTACT. Shown are promoter names, locus IDs, and tissues of expression.

[0198] Table 2. List of primers and restriction enzymes used in this study for library construction. Listed are primers and restriction enzymes used in this research for amplifying and cloning the indicated promoters. Lower-case bold letters are nucleotides at restriction sites, and upper-case letters are promoter sequence.

[0199] Table 3. List of targeted genes in pSUC2:miR-3UAJT lines. Shown are mTACT lines found in the screen, amiRNA sequences and putative targeted genes.

[0200] Table 4. Number of targeted genes and number of amiRNAs in each mTACT amiRNA sub-library. Shown are the number of targeted genes, the number of genes in the sub-family, and the number of amiRNAs in each sub-library. CP, channels and porins; APC, amino acid / polyamine / organo-cation, cation carriers, and mitochondrial carriers; MFS, major facilitator superfamily; DMT, drug / metabolite transporter group; MATE, multi-drug and toxic compound extrusion and other carrier transporters; ABC, ATP binding cassette family; PA, primary active transporters; and UF, unknown function.

[0201] Table 5. List of adaptors and primers of the mTACT library. Adaptors and primers used for amplification of the mTACT amiRNA library and sub-libraries are listed. Lowercase bold letters are Golden Gate (Bsal) sites, and upper-case bold letters are adaptor sequence.

[0202] Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims.

Claims

CLAIMS1. A system for inducing knockdown or knockout of multiple genes in a plant, comprising: at least one Cell type-specific Multi-targeted AmiRNA or CRISPR library (cMAC library) comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more amiRNA sequences or one or more sgRNA sequences, wherein each of the amiRNA or sgRNA sequences targets multiple genes.

2. The system of claim 1, wherein the AmiRNA or CRISPR library comprises at least 50, 100, 200, 300, 400, 500 or more polynucleotides, each encoding one or more unique sgRNA sequences or unique amiRNA sequences.

3. The system of any one of claims 1-2, wherein each of the sgRNA or amiRNA sequences targets between 2-15 genes.

4. The system of any one of the preceding claims, wherein expression of the polynucleotide encoding the sgRNA sequence or amiRNA sequence is regulated by a tissue-specific promoter or a developmental-stage specific promoter.

5. The system of claim 4, wherein the promoter is a tissue-specific promoter selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root-specific cortex promoter, shoot promoter, and root promoter.

6. The system of any one of the preceding claims, wherein the library comprises a plurality of polynucleotides encoding one or more sgRNA sequences or amiRNA sequences regulated by at least 2, 3, 4, 5, 6, 7, 8, 9, 10, or more different tissue-specific promoters.

7. The system of any one of the preceding claims, wherein the amiRNA or CRISPR library is configured to ensure that at least two different sgRNA sequences or amiRNA sequences are inserted to each plant.

8. The system of any one of the preceding claims, wherein the library comprises a plurality of polynucleotides, each encoding a plurality of different sgRNA sequences or amiRNA sequences.

9. The system of claim 8, wherein each of the encoded plurality of different amiRNA sequences or sgRNA sequences targets a different set of genes.

10. The system of claim 8, wherein each of the encoded plurality of different amiRNA sequences or sgRNA sequences targets the same set of multiple genes.

11. The system of any one of the preceding claims, wherein said system comprises a plurality of libraries, each library comprising a plurality of polynucleotides encoding one or more sgRNA or amiRNA sequences, wherein the expression of the plurality of polynucleotides within each library is regulated by a different tissue-specific promoter.

12. The system of any one of the preceding claims, wherein expression of the sgRNA sequences or amiRNA sequences is regulated by a promoter selected from the group consisting of SIG6, ARSK1, PGP4, CO2, KST1, SCR, S18, CORI3, IQD22, SUC2, ML1, and UBQ10.

13. The system of any one of the preceding claims, wherein the sgRNA sequences or amiRNA sequences target transportome genes.

14. The system of any one of the preceding claims, wherein the amiRNA sequences or the sgRNA sequences target at least one gene member comprised within a gene set.

15. The system of any one of the preceding claims, wherein the plant is selected from the group consisting of a crop plant, a model plant, and a wild plant.

16. The system of claim 15, wherein the plant is a crop plant selected from the group consisting of rice, maize, wheat, barley, tomato, potato, eggplant, sesame, soybean, sunflower, canola, sugarcane, alfalfa, millet, Leguminosae, flax, lupinus, rapeseed, tobacco, and cotton.

17. The system of claim 15, wherein the model plant is Arabidopsis.

18. The system of claim 17, wherein the library comprises 5,565 amiRNA sequences targeting a plurality of genes within Arabidopsis gene set, wherein the gene set comprises19. The system of claim 17, wherein the library comprises 2,859 sgRNA sequences targeting a plurality of genes within Arabidopsis gene set, wherein the gene set comprises 707 Arabidopsis transporter genes.

20. A method for identifying at least one phenotypic effect of gene knockdown or knockout in a plant, the method comprising:(i) transforming the system according to any one of claims 1 to 19 into a plurality of plants, thereby producing a plant population, wherein each plant of the population comprises at least two knocked-down or knocked-out genes;(ii) examining the phenotype of each plant in the plant population to identify at least one modified phenotype; and(iii) identifying the at least two knocked-down or knocked-out genes associated with the at least one modified phenotype.

21. The method of claim 20, wherein identifying the at least one modified phenotype is performed by comparing the phenotype of the plant comprising the system according to any one of claims 1 to 19 to the phenotype of a corresponding plant not comprising said system and grown under the same growth conditions or by comparing the phenotype of the plant comprising said system to the phenotype of a corresponding plant as is known in the art.

22. A method for identifying multiple members within at least one gene set underlying a phenotype, the method comprising:(i) clustering coding sequences within genetic data of a plant species to sequence clusters, each cluster representing a gene set;(ii) producing a CRISPR library comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more unique sgRNA sequences, wherein each of the sgRNA sequences targets a plurality of gene members comprised within the gene set, wherein the expression of each of said sgRNA sequences is controlled by at least one tissue-specific and / or developmental stage-specific promoter;(iii) transforming the library into a plurality of plants, thereby producing a plant population wherein each plant of the population comprises at least one sgRNA sequence targeting multiple gene members;(iv) screening the plant population for at least one selected phenotype;(v) selecting at least one plant showing the at least one selected phenotype; and(vi) identifying in the selected plant at least one sgRNA sequence targeting the multiple -gene members; thereby identifying said multiple gene members underlying said selected phenotype.

23. A method for identifying multiple members within at least one gene set underlying a phenotype, the method comprising:(i) clustering coding sequences within genetic data of a plant species to sequence clusters, each cluster representing a gene set;(ii) producing an amiRNA library comprising a plurality of polynucleotides, wherein each polynucleotide encodes one or more unique amiRNA sequences, wherein each of the amiRNA sequences targets a plurality of gene members comprised within the gene set, wherein the expression of each of said amiRNA sequences is controlled by at least one tissue-specific and / or developmental stage-specific promoter;(iii) transforming the library into a plurality of plants, thereby producing a plant population wherein each plant of the population comprises at least one amiRNA sequence targeting multiple gene members;(iv) screening the plant population for at least one selected phenotype;(v) selecting at least one plant showing at least one selected phenotype; and(vi) identifying in the selected plant at least one amiRNA sequence targeting the multiple -gene members; thereby identifying said multiple gene members underlying said selected phenotype.

24. An isolated polynucleotide comprising a tissue specific promoter operably linked to one or more amiRNA encoding sequences or one or more sgRNA encodingsequences, wherein each of the encoded amiRNAs or sgRNAs targets multiple plant genes.

25. The isolated polynucleotide of claim 24, wherein the plant genes are members of a gene set.

26. The isolated polynucleotide of claim 25, wherein the members of the gene set share a certain structural homology.

27. The isolated polynucleotide of claim 25, wherein the members of the gene set participate in a plant functional pathway.

28. The isolated polynucleotide of any one of claims 24-27, wherein the promoter is selected from the group consisting of shoot epidermis promoter, palisade mesophyll promoter, endodermis promoter, bundle sheath promoter, phloem companion cells promoter, spongy mesophyll promoter, guard cell promoter, root-specific epidermis promoter, cortex promoter, root cortex promoter, shoot promoter, and root promoter.

29. A plant comprising the isolated polynucleotide of any one of claims 24-28.

30. A plurality of plants, wherein each plant of the plurality of plants comprises the isolated polynucleotide of any one of claims 24-28.

31. The plurality of plants of claim 30, wherein each plant of said plurality of plants comprises a same polynucleotide having a same configuration of the tissue specific promoter and one or more amiRNA encoding sequence or sgRNA encoding sequence.

32. The plurality of plants of claim 30, wherein each plant of said plurality of plants comprises a different polynucleotide with the same tissue specific promoter and different configuration of the one or more amiRNA encoding sequence or sgRNA encoding sequence.

33. The plurality of plants of claim 30, wherein said plurality of plants comprises plants with different polynucleotides, wherein the different polynucleotides comprise a variety of tissue specific promoters and a same or different configuration of the one or more amiRNA encoding sequences or sgRNA encoding sequences.

Images