PLANTS HAVING REDUCED LEVELS OF BITTER TASTE METABOLITES

Field of the invention

The present invention belongs to the field of agricultural biology. In particular the present invention relates to improved crops which have a reduced level of bitter taste secondary metabolites, more particularly a reduced level of guaianolide sequiterpene lactones. The invention provides means and methods for generating crops having a deficiency in kauniolide synthase.

Introduction to the invention

Industrial chicory (Cichorium intybus L. var. sativum) and witloof (Cichorium intybus L. var. foliosum) are traditional Belgian crops with an important economic value. Industrial chicory is mainly cultivated for the production of inulin, while witloof is a well-known leafy vegetable. The chicory root can contain up to 20% of storage carbohydrates, mainly inulin or, after hydrolysis, 18% of fructose and 2% of glucose. Inulin is not degraded in the stomach or in the small intestine, but is fermented in the colon so its caloric value is low. Furthermore, inulin has the properties of a dietary fiber and activates the growth beneficial bacteria in the colon. The inulin market size and forecast inulin market was valued at USD 1050 million in 2019 and is projected to reach USD 1.6 Million by 2027, growing at a CAGR of 6.2% from 2020 to 2027. Both industrial chicory and witloof are rich in nutritionally relevant compounds with beneficial effects on human health because of the biological and pharmacological activities of their specialized metabolites. However, their bitter taste limits the use of industrial chicory flour in food industry and negatively influences the consumer acceptability of witloof. Changing the bitterness would thus create new opportunities with a great economic impact. The specialized metabolites responsible for this bitter taste are sesquiterpene lactones (SLs). The SLs in the root are also co-isolated with inulin and then have to be subsequently removed with additional purification steps, increasing the cost of inulin isolation. Within the SLs, different classes are recognized, with the guaianolides being the most important with regard to bitterness. A total of 16 guaianolide SL metabolites could be detected in leaf extracts of C. intybus using UHPLC-HRMS (Kips, 2017). Known genes encoding enzymes involved in the guaianolide SL biosynthetic pathway are GERMACRENE A SYNTHASE (GAS), GERMACRENE A OXIDASE (GAO), COSTUNOLIDE SYNTHASE (COS) and KAUNIOLIDE SYNTHASE (KLS) (see Figure 1). WO2021/122982 discloses germacrene A synthase enzymes and chicory mutants thereof. A further understanding of the SL biosynthetic pathway is needed to unravel the formation of guaianolide SL metabolites. In the present invention genome-wide gene family annotation was combined with phylogenetic analysis and comprehensive transcriptome analysis of MeJA treated samples in three species (Chicorium intybus var. sativum, Chicorium intybus var. foliosum and Lactuca sativa). By using RNA-sequencing (RNA-seq) and differential gene expression (DGE) analysis new candidate KLS biosynthesis genes in the genome of C. intybus were identified. In a next step these KLS candidate genes were used in a heterologous gene expression assay in N. benthamiana to verify catalytic activity of the expressed enzymes. KLS genes showing an activity were knocked out in Chicorium intybus with the CRISPR/Cas9 genome editing technology. Surprisingly KLS mutant Chicorium plants did have 14 out of 16 SL metabolites completely eliminated as compared to the wild type plants. This is unexpected since the knock-out of germacrene synthase A (as shown in WO2021/122982) did not have such a drastic effect on the levels of SL metabolites. This means that the genetic ablation of biosynthetic enzymes in a complex metabolic pathway (here the biosynthesis pathway of SL secondary metabolites) leads to an unpredictable outcome and hence the strong reduction of guaianolide SL metabolites we have obtained in the instant invention could not be anticipated. Thus, the development of new crops having a loss in KLS genes leads to crops with a reduced level of SLs which may lead to cost savings in inulin extraction from inulin producing (root) crops and also for the production of less bitter (leaf) crops.

Figure legends

Figure 1: SL biosynthetic pathway with germacrene A synthase (GAS) catalyzing the cyclisation of farnesyl diphosphate (FPP) to germacrene A (GA); Oxidation of GA via germacra-l(10),4,ll(13)-trien-12-ol (GOH) and germacra-l(10),4,ll(13)-trien-12-al (GAL) into germacra-l(10),4,ll(13)-trien-12-oic acid (GAA) catalyzed by an NADPH-dependent single P450 enzyme, germacrene A oxidase (GAO); costunolide synthase (COS) enzyme catalyzing the hydroxylation at the C6 position of GAA to produce costunolide, a germacranolide SL, or hydroxylation at the C8 position of GAA by unknown enzyme; conversion of costunolide into ll(S),13-dihydrocostunolide by unknown enzyme; kauniolide synthase (KLS) enzyme catalyzing the conversion of costunolide to kauniolide, the first guaianolide SL; conversion of kauniolide and/or ll(S),13-dihydrocostunolide into guaianolide SL metabolites by unknown enzymes.

Figure 2: Phylogenetic analysis of the 27 selected C. intybus and corresponding L. sativa orthologous cytochrome P450s (CYPs) together with a heatmap of their response to MeJA (Iog2-fold change in gene expression after MeJA induction versus mock treatment) in industrial chicory (C), witloof (W) and lettuce (L). Gray boxes indicate non-expressed genes while MeJA-inducible genes with Iog2-fold change ≥ 1 are marked with a black outline. The phylogenetic tree from protein alignment was inferred with the UPGMA method. Bootstrap values (shown at the branching points) are based on 1000 replicates. KLS from T. parthenium (TpKLS) (AXG24152.1) was included for phylogenetic analysis. CYP71AV4 and CYP71BL3 were previously identified as functional CiGAO (ADF43080.1) and CiCOS (AEG79727.1), respectively. The candidate GAO, COS and KLS clade are indicated by the green, orange and red box, respectively.

Figure 3: Phylogenetic analysis of the six selected CiGAO candidates (green), four selected CiCOS candidates (orange) and ten selected CiKLS candidates (red) and the catalytically active GAO, COS and KLS proteins from different species (black): HaGAO from Helianthus annuus (ADF43082.1), TcGAO from Tanacetum cinerariifolium (AGO03789.1), TpGAO from Tanacetum parthenium (AHN62855.1), LsGAO from Lactuca sativa (ADF32078.1), CiGAO from Cichorium intybus (ADF43080.1), CcGAO from Cynara cardunculus (AIA09035.1), ScGAO from Saussurea costus (ADF43081.1), BsGAO from Barnadesia spinosa (ADF43083.1), TpKLS from Tanacetum parthenium (AXG24152.1), TcCOS from Tanacetum cinerariifolium (AGO03789.1), TpCOS from Tanacetum parthenium (AHN62856.1), CcCOS from Cynara cardunculus L. (AIA09038.1), LsCOS from Lactuca sativa (AEI59780.1) and CiCOS from Cichorium intybus (AEG79727.1). The phylogenetic tree from protein alignment was inferred with the UPGMA method. Bootstrap values (shown at the branching points) are based on 1000 replicates.

Figure 4: (A) Overlay of the El) (230.13 ± 0.05 Da) GC-MS chromatogram of N. benthamiana extracts co- agroinfiltrated with CiGASs, CiGAO and CiCOS (black chromatogram) or with CiGASs, CiGAO, CiCOS and TpKLS (red chromatogram). The red peak at ±30 min (indicated with a black arrow) is present in the N. benthamiana extracts agroinfiltrated with both CiCOS and TpKLS and absent in the N. benthamiana extracts agroinfiltrated with only CiCOS and EV. (B) Overlay of the EIC (230.13 ± 0.05 Da; 29.8 - 30.15 min) GC-MS chromatograms of all samples. The peak at 29.94 min corresponds with the red peak in panel A. The assays displayed in the black box indicate which agroinfiltrated N. benthamiana extracts produce this peak: CYP71BZ25, two alleles of CYP71BZ25 and three alleles of CYP71BZ25. The peak was not present in all other assays, indicated by all colored horizontal lines (C) Deconvoluted EI-MS spectrum of the peak eluting at 29.94 min. (D) Structure of kauniolide with the corresponding molecular weight (230.130131 Da)

Figure 5: Nucleotide alignments of paralogous CiKLS genes, visualizing CRISPR/Cas9 20 bp gene target sites and primers for HiPlex amplicon sequencing. Annotations: Gray = Exon; Colored = Forward 20 bp target sites and colored = Reverse 20 bp target sites; Dark green = Forward primer; Light green = Reverse primer (see Table 4).

Figure 6: alignment of functional KLS amino acid sequences, Tp: Tanacetum parthenium, conserved sequences present in TpKLS, CYP71BZ25, CYP71BZ26 and CYP71BZ27 are highlighted in bold and underlined. Detailed description of the invention

To facilitate the understanding of this invention a number of terms are defined below. Terms defined herein (unless otherwise specified) have meanings as commonly understood by a person of ordinary skill in the areas relevant to the present invention. As used in this specification and its appended claims, terms such as "a", "an" and "the" are not intended to refer to only a singular entity, but include the general class of which a specific example may be used for illustration, unless the context dictates otherwise. The terminology herein is used to describe specific embodiments of the invention, but their usage does not delimit the invention, except as outlined in the claims.

Sesquiterpene lactones are C15 terpenoids and constitute a major class of plant secondary metabolites with diverse chemical structures. They are present in plant species in the Acanthaceae, Apiaceae and Asteraceae with over 4000 different structures reported so far ² . Sesquiterpene lactones are classified in six bicyclic or tricyclic classes named guianolides, pseudoguaianolides, xanthanolides, eremophilanolides, eudesmanolides and germacranolides. Naturally occurring sesquiterpene lactones contain an a-methylene-y-butyrolactone or a, p-unsaturated cyclopentenone moiety. Guaianolides are defined by their special 5-7-5 tricyclic structure and these compounds are these compounds are most important with regard to bitterness. It would be desirable to have plants with a highly reduced or even absent production of guaianolides. Especially during the extraction of inulin from plants (such as chicory) guaianolides still have to be removed by additional purification steps, increasing the cost of insulin isolation. Inulin is a dietary fiber found in a variety of fruits, vegetables, and herbs, including wheat, onions, bananas, leeks, artichokes, and asparagus. Inulin is also a polymer of fructose molecule (fructan). Like other fructans, it is prebiotic, meaning that it nourishes the good bacteria in the gut. Moreover, the gut bacteria transform inulin and other prebiotics into short-chain fatty acids, which sustain colon cells and present several other health benefits. Inulin molecules link collectively in a way that the small intestine cannot break down, and instead travel to the lower gut and feed beneficial bacteria. Moreover, inulin is a starchy substance and is most commonly used for medicine obtained by soaking chicory roots in hot water.

Accordingly, in a first embodiment the present invention provides a plant which from the family Asteracea deficient in kauniolide synthase (KLS) activity, which plant has a gene disruption in one or more polynucleotides encoding for a kauniolide synthase polypeptide.

In another embodiment the invention provides a cultivated crop plant from the family Asteracea which crop plant is deficient in kauniolide synthase (KLS) activity, and which crop plant has a gene disruption in one or more polynucleotides encoding for a kauniolide synthase polypeptide. In a specific embodiment the plant KLS polypeptide sequence comprises SEQ ID NO: 25 and 26.

SEQ. ID NO: 25 and SEQ ID NO: 26 are amino acid boxes which are conserved between functional KLS synthase polypeptides. Figure 6 shows the alignment between functional KLS polypeptides.

In another embodiment a plant of the invention is selected from the genus Cichorium, Lactuca, Taraxacum, Scorzonera, Cynara, Helianthus, Parthenium or Artemisia.

In a specific embodiment the invention provides a plant which is Cichorium intybus or Cichorium endivia and wherein the KLS polypeptide sequences are depicted in SEQ ID NO: 14, 16, 18, 20, 22 and 24.

In another embodiment the invention provides a seed or plant cell derived from a plant as described herein before.

In another embodiment the invention provides a method for decreasing guaianolide sesquiterpene lactones in a plant from the family Asteraceae, the method comprising disrupting the expression of one or more polynucleotides in said plant encoding for a polypeptide with kauniolide synthase activity.

In another embodiment the invention provides a method for decreasing guaianolide sesquiterpene lactones in a plant from the family Asteraceae, the method comprising disrupting the expression of one or more polynucleotides in said plant encoding for a polypeptide with kauniolide synthase activity wherein the polypeptide comprises SEQ ID NO: 25 and 26.

In another embodiment the invention provides a method for decreasing guaianolide sesquiterpene lactones in a plant from the family Asteraceae, the method comprising disrupting the expression of one or more polynucleotides in said plant encoding for a polypeptide with kauniolide synthase activity wherein the polypeptide is depicted in SEQ ID NO: 14, 16, 18, 20, 22 or 24.

Thus, the activity of a kauniolide synthase protein (KLS protein) may be reduced or eliminated by inactivation (e.g. via gene disruption or gene mutagenesis) the gene encoding one or more KLS genes present in the plant genome. The disruption inhibits expression or activity of KLS compared to a corresponding control plant cell lacking the disruption. In one embodiment, the endogenous KLS gene comprises two, three or more endogenous KLS genes. Similarly, in another embodiment, in particular plants the endogenous KLS gene comprises three or more endogenous KLS genes. The wording "two or more endogenous KLS genes" or "three or more endogenous KLS genes" refers to two or more or three or more homologs of KLS but it is not excluded that two or more or three or more combinations of homologs of KLS are disrupted (or their activity reduced).

In another embodiment, the disruption step comprises insertion of one or more transposons, where the one or more transposons are inserted into the endogenous KLS gene(s). In yet another embodiment, the disruption comprises one or more point mutations in the endogenous KLS gene(s). The disruption can be a homozygous disruption in the KLS gene(s). Alternatively, the disruption is a heterozygous disruption in the KLS gene(s) In certain embodiments, when more than one KLS gene is involved, there is more than one disruption, which can include homozygous disruptions, heterozygous disruptions or a combination of homozygous disruptions and heterozygous disruptions.

Detection of expression products is performed either qualitatively (by detecting presence or absence of one or more product of interest) or quantitatively (by monitoring the level of expression of one or more product of interest) or alternatively measuring the enzymatic activity of the KLS polypeptide as is described in the appended examples section or measuring the absence of activity of the KLS polypeptide. In one embodiment, the expression product is an RNA expression product. Aspects of the invention optionally include monitoring an expression level of a nucleic acid, polypeptide as noted herein for detection of KLS or measuring the amount of kauniolide in a plant (presence of KLS activity) or in a population of plants or measuring the absence of guaianolide sesquiterpene lactones (absence of KLS activity).

Thus, many methods may be used to reduce or eliminate the activity of a KLS gene. More than one method may be used to reduce the activity of a single plant KLS gene. In addition, combinations of methods may be employed to reduce or eliminate the activity of two or more different KLS genes. Non- limiting examples of methods of reducing or eliminating the expression of a plant KLS are given below. In some embodiments of the present invention, a polynucleotide (such as an antisense polynucleotide) is introduced into a plant that upon introduction or expression, inhibits the expression of a KLS gene of the invention. The term "expression" as used herein refers to the biosynthesis of a gene product, including the transcription and/or translation of said gene product. For example, for the purposes of the present invention, an expression cassette capable of expressing a polynucleotide that inhibits the expression of a KLS polypeptide is an expression cassette capable of producing an RNA molecule that inhibits the transcription and/or translation of a KLS polypeptide of the invention. The "expression" or "production" of a protein or polypeptide from a DNA molecule refers to the transcription and translation of the coding sequence to produce the protein or polypeptide, while the "expression" or "production" of a protein or polypeptide from an RNA molecule refers to the translation of the RNA coding sequence to produce the protein or polypeptide.

As used herein, "polynucleotide" includes reference to a deoxyribopolynucleotide, ribopolynucleotide or analogs thereof that have the essential nature of a natural ribonucleotide in that they hybridize, under stringent hybridization conditions, to substantially the same nucleotide sequence as naturally occurring nucleotides and/or allow translation into the same amino acid(s) as the naturally occurring nucleotide(s). A polynucleotide can be full-length or a subsequence of a native or heterologous structural or regulatory gene. Unless otherwise indicated, the term includes reference to the specified sequence as well as the complementary sequence thereof. Thus, DNAs or RNAs with backbones modified for stability or for other reasons are "polynucleotides" as that term is intended herein. Moreover, DNAs or RNAs comprising unusual bases, such as inosine, or modified bases, such as tritylated bases, to name just two examples, are polynucleotides as the term is used herein. It will be appreciated that a great variety of modifications have been made to DNA and RNA that serve many useful purposes known to those of skill in the art. The term polynucleotide as it is employed herein embraces such chemically, enzymatically or metabolically modified forms of polynucleotides, as well as the chemical forms of DNA and RNA characteristic of viruses and cells, including inter alia, simple and complex cells.

As used herein, "nucleic acid" includes reference to a deoxyribonucleotide or ribonucleotide polymer in either single- or double-stranded form, and unless otherwise limited, encompasses known analogues having the essential nature of natural nucleotides in that they hybridize to single-stranded nucleic acids in a manner similar to naturally occurring nucleotides (e.g. peptide nucleic acids).

By "encoding" or "encoded," with respect to a specified nucleic acid, is meant comprising the information for transcription into an RNA and in some embodiments, translation into the specified protein. A nucleic acid encoding a protein may comprise non-translated sequences (e.g., introns) within translated regions of the nucleic acid, or may lack such intervening non-translated sequences (e.g., as in cDNA). The information by which a protein is encoded is specified by the use of codons. Typically, the amino acid sequence is encoded by the nucleic acid using the "universal" genetic code.

Significant advances have been made in the last few years towards development of methods and compositions to target and cleave genomic DNA by site specific nucleases (e.g., Zinc Finger Nucleases (ZFNs), Meganucleases, Transcription Activator-Like Effector Nucleases (TALENS) and Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated nuclease (CRISPR/Cas) with an engineered crRNA/tracr RNA), to induce targeted mutagenesis, induce targeted deletions of cellular DNA sequences, and facilitate targeted recombination of an exogenous donor DNA polynucleotide within a predetermined genomic locus. See, for example, U.S. Patent Publication No. 20030232410; 20050208489; 20050026157; 20050064474; and 20060188987, and International Patent Publication No. WO 2007/014275, the disclosures of which are incorporated by reference in their entireties for all purposes. U.S. Patent Publication No. 20080182332 describes use of non-canonical zinc finger nucleases (ZFNs) for targeted modification of plant genomes and U.S. Patent Publication No. 20090205083 describes ZFN-mediated targeted modification of a plant EPSPs genomic locus. Current methods for targeted insertion of exogenous DNA typically involve co-transformation of plant tissue with a donor DNA polynucleotide containing at least one transgene and a site-specific nuclease (e.g., ZFN) which is designed to bind and cleave a specific genomic locus of an actively transcribed coding sequence. This causes the donor DNA polynucleotide to stably insert within the cleaved genomic locus resulting in targeted gene addition at a specified genomic locus comprising an actively transcribed coding sequence. As used herein the term "zinc fingers," defines regions of amino acid sequence within a DNA binding protein binding domain whose structure is stabilized through coordination of a zinc ion.

A "zinc finger DNA binding protein" (or binding domain) is a protein, or a domain within a larger protein, that binds DNA in a sequence-specific manner through one or more zinc fingers, which are regions of amino acid sequence within the binding domain whose structure is stabilized through coordination of a zinc ion. The term zinc finger DNA binding protein is often abbreviated as zinc finger protein or ZFP. Zinc finger binding domains can be "engineered" to bind to a predetermined nucleotide sequence. Non- limiting examples of methods for engineering zinc finger proteins are design and selection. A designed zinc finger protein is a protein not occurring in nature whose design/composition results principally from rational criteria. Rational criteria for design include application of substitution rules and computerized algorithms for processing information in a database storing information of existing ZFP designs and binding data. See, for example, U.S. Pat. Nos. 6,140,081; 6,453,242; 6,534,261 and 6,794,136; see also WO 98/53058; WO 98/53059; WO 98/53060; WO 02/016536 and WO 03/016496.

A "TALE DNA binding domain" or "TALE" is a polypeptide comprising one or more TALE repeat domains/units. The repeat domains are involved in binding of the TALE to its cognate target DNA sequence. A single "repeat unit" (also referred to as a "repeat") is typically 33-35 amino acids in length and exhibits at least some sequence homology with other TALE repeat sequences within a naturally occurring TALE protein. See, e.g., U.S. Patent Publication No. 20110301073, incorporated by reference herein in its entirety.

The CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats)/Cas (CRISPR Associated) nuclease system. Briefly, a "CRISPR DNA binding domain" is a short-stranded RNA molecule that acting in concert with the CAS enzyme can selectively recognize, bind, and cleave genomic DNA. The CRISPR/Cas system can be engineered to create a double-stranded break (DSB) at a desired target in a genome, and repair of the DSB can be influenced by the use of repair inhibitors to cause an increase in error prone repair. See, e.g., Jinek et al (2012) Science 337, p. 816-821, Jinek et al, (2013), eLife 2:e00471, and David Segal, (2013) eLife 2:e00563).

Zinc finger, CRISPR and TALE binding domains can be "engineered" to bind to a predetermined nucleotide sequence, for example via engineering (altering one or more amino acids) of the recognition helix region of a naturally occurring zinc finger. Similarly, TALEs can be "engineered" to bind to a predetermined nucleotide sequence, for example by engineering of the amino acids involved in DNA binding (the repeat variable diresidue or RVD region). Therefore, engineered DNA binding proteins (zinc fingers or TALEs) are proteins that are non-naturally occurring. Non-limiting examples of methods for engineering DNA-binding proteins are design and selection. A designed DNA binding protein is a protein not occurring in nature whose design/composition results principally from rational criteria. Rational criteria for design include application of substitution rules and computerized algorithms for processing information in a database storing information of existing ZFP and/or TALE designs and binding data. See, for example, U.S. Pat. Nos. 6,140,081; 6,453,242; and 6,534,261; see also WO 98/53058; WO 98/53059; WO 98/53060; WO 02/016536 and WO 03/016496 and U.S. Publication Nos. 20110301073, 20110239315 and 20119145940.

A "selected" zinc finger protein, CRISPR or TALE is a protein not found in nature whose production results primarily from an empirical process such as phage display, interaction trap or hybrid selection. See e.g., U.S. Pat. No. 5,789,538; U.S. Pat. No. 5,925,523; U.S. Pat. No. 6,007,988; U.S. Pat. No. 6,013,453; U.S. Pat. No. 6,200,759; WO 95/19431; WO 96/06166; WO 98/53057; WO 98/54311; WO 00/27878; WO 01/60970 WO 01/88197 and WO 02/099084 and U.S. Publication Nos. 20110301073, 20110239315 and 20119145940.

In one embodiment, the polynucleotide encodes a zinc finger protein that binds to a gene encoding a KLS polypeptide, resulting in reduced expression of the gene. In particular embodiments, the zinc finger protein binds to a regulatory region of a KLS gene. In other embodiments, the zinc finger protein binds to a messenger RNA encoding a KLS polypeptide and prevents its translation. Methods of selecting sites for targeting by zinc finger proteins have been described, for example, in US6453242, and methods for using zinc finger proteins to inhibit the expression of genes in plants are described, for example, in US2003/0037355, each of which is herein incorporated by reference.

In another embodiment, the polynucleotide encoded a TALE protein that binds to a gene encoding a KLS polypeptide, resulting in reduced expression of the gene. In particular embodiments, the TALE protein binds to a regulatory region of a KLS gene. In other embodiments, the TALE protein binds to a messenger RNA encoding a KLS polypeptide and prevents its translation. Methods of selecting sites for targeting by zinc finger proteins have been described in e.g. Moscou MJ, Bogdanove AJ (2009) (A simple cipher governs DNA recognition by TAL effectors. Science 326:1501) and Morbitzer R, Romer P, Boch J, Lahaye T (2010) (Regulation of selected genome loci using de novo-engineered transcription activator-like effector (TALE)-type transcription factors. Proc Natl Acad Sci USA 107:21617-21622).

Additional methods for decreasing or eliminating the expression of endogenous genes in plants are also known in the art and can be similarly applied to the instant invention. These methods include other forms of mutagenesis, such as ethyl methanesulfonate-induced mutagenesis, deletion mutagenesis and fast neutron deletion mutagenesis used in a reverse genetics sense (with PCR) to identify plant lines in which the endogenous gene has been deleted. For examples of these methods see, Ohshima, et al, (1998) Virology 243:472-481; Okubara, et al, (1994) Genetics 137:867-874 and Quesada, et al, (2000) Genetics 154:421-436, each of which is herein incorporated by reference. In addition, a fast and automatable method for screening for chemically induced mutations, TILLING (Targeting Induced Local Lesions in Genomes), using denaturing HPLC or selective endonuclease digestion of selected PCR products is also applicable to the instant invention. See, McCallum, et al, (2000) Nat. Biotechnol 18:455- 457, herein incorporated by reference. Mutations that impact gene expression or that interfere with the function of the encoded protein are well known in the art. Insertional mutations in gene exons usually result in null-mutants. Mutations in conserved residues are particularly effective in inhibiting the activity of the encoded protein. Conserved residues of plant KLS polypeptides suitable for mutagenesis with the goal to eliminate KLS activity have been described.

Also single stranded DNA can be used to downregulate the expression of KLS genes. Methods for gene suppression using ssDNA are e.g. described in W02011/112570.

In yet another embodiment protein interference as described in the patent application W02007071789 (means and methods for mediating protein interference) can be used to downregulate a gene product. The latter technology is a knock-down technology which in contrast to RNAi acts at the post-translational level (i.e. it works directly on the protein level by inducing a specific protein aggregation of a chosen target). Protein aggregation is essentially a misfolding event which occurs through the formation of intermolecular beta-sheets resulting in a functional knockout of a selected target. Through the use of a dedicated algorithm it is possible to accurately predict which amino acidic stretches in a chosen target protein sequence have the highest self-associating tendency (Fernandez-Escamilla A. M. et al (2004) Nat Biotechnol 22(10): 1302-6. By expressing these specific peptides in the cells the protein of interest can be specifically targeted by inducing its irreversible aggregation and thus its functional knock-out.

In yet another embodiment the invention encompasses still additional methods for reducing or eliminating the activity of the KLS polypeptide. Examples of other methods for altering or mutating a genomic nucleotide sequence in a plant are known in the art and include, but are not limited to, the use of RNA:DNA vectors, RNA:DNA mutational vectors, RNA:DNA repair vectors, mixed-duplex oligonucleotides, self-complementary RNA:DNA oligonucleotides and recombinogenic oligonucleotide bases. Such vectors and methods of use are known in the art. See, for example, US5565350; US5731181; US5756325; US5760012; US5795972 and US5871984, each of which are herein incorporated by reference.

The term "expression" or "gene expression" means the transcription of a specific gene or specific genes or specific genetic construct. The term "expression" or "gene expression" in particular means the transcription of a gene or genes or genetic construct into structural RNA (rRNA, tRNA) or mRNA with or without subsequent translation of the latter into a protein. The process includes transcription of DNA and processing of the resulting mRNA product.

The term "introduction" or "transformation" as referred to herein encompass the transfer of an exogenous polynucleotide into a host cell, irrespective of the method used for transfer. Plant tissue capable of subsequent clonal propagation, whether by organogenesis or embryogenesis, may be transformed with a genetic construct of the present invention and a whole plant regenerated there from. The particular tissue chosen will vary depending on the clonal propagation systems available for, and best suited to, the particular species being transformed. Exemplary tissue targets include leaf disks, pollen, embryos, cotyledons, hypocotyls, mega-gametophytes, callus tissue, existing meristematic tissue (e.g., apical meristem, axillary buds, and root meristems), and induced meristem tissue (e.g., cotyledon meristem and hypocotyl meristem). The polynucleotide may be transiently or stably introduced into a host cell and may be maintained non-integrated, for example, as a plasmid. Alternatively, it may be integrated into the host genome. The resulting transformed plant cell may then be used to regenerate a transformed plant in a manner known to persons skilled in the art.

The transfer of foreign genes into the genome of a plant is called transformation. Transformation of plant species is now a fairly routine technique. Advantageously, any of several transformation methods may be used to introduce the gene of interest into a suitable ancestor cell. The methods described for the transformation and regeneration of plants from plant tissues or plant cells may be utilized for transient or for stable transformation. Transformation methods include the use of liposomes, electroporation, chemicals that increase free DNA uptake, injection of the DNA directly into the plant, particle gun bombardment, transformation using viruses or pollen and microprojection. Methods may be selected from the calcium/polyethylene glycol method for protoplasts (Krens, F.A. et al., (1982) Nature 296, 72-74; Negrutiu I et al. (1987) Plant Mol Biol 8: 363- 373); electroporation of protoplasts (Shillito R.D. et al. (1985) Bio/Technol 3, 1099-1 102); microinjection into plant material (Crossway A et al., (1986) Mol. Gen Genet 202: 179-185); DNA or RNA-coated particle bombardment (Klein TM et al., (1987) Nature 327: 70) infection with (non-integrative) viruses and the like. Transgenic plants, including transgenic crop plants, are preferably produced via Agrobacterium-mediated transformation. An advantageous transformation method is the transformation in planta. To this end, it is possible, for example, to allow the agrobacteria to act on plant seeds or to inoculate the plant meristem with agrobacteria. It has proved particularly expedient in accordance with the invention to allow a suspension of transformed agrobacteria to act on the intact plant or at least on the flower primordia. The plant is subsequently grown on until the seeds of the treated plant are obtained (Clough and Bent, Plant J. (1998) 16, 735-743). Methods for Agrobacterium-mediated transformation of rice include well known methods for rice transformation, such as those described in any of the following: European patent application EP1198985, Aldemita and Hodges (Planta 199: 612-617, 1996); Chan et al. (Plant Mol Biol 22 (3): 491 - 506, 1993), Hiei et al. (Plant J 6 (2): 271 -282, 1994), which disclosures are incorporated by reference herein as if fully set forth. In the case of corn transformation, the preferred method is as described in either Ishida et al. (Nat. Biotech. 14(6): 745-50, 1996) or Frame et al. (Plant Physiol 129(1): 13-22, 2002), which disclosures are incorporated by reference herein as if fully set forth. Said methods are further described by way of example in B. Jenes et al., Techniques for Gene Transfer, in: Transgenic Plants, Vol. 1 , Engineering and Utilization, eds. S.D. Kung and R. Wu, Academic Press (1993) 128-143 and in Potrykus Annu. Rev. Plant Physiol. Plant Mol. Biol. 42 (1991) 205-225). The nucleic acids or the construct to be expressed is preferably cloned into a vector, which is suitable for transforming Agrobacterium tumefaciens, for example pBinl9 (Bevan et al (1984) Nucl. Acids Res. 12-8711). Agrobacteria transformed by such a vector can then be used in known manner for the transformation of plants, such as plants used as a model, like Arabidopsis or crop plants such as, by way of example, tobacco plants, for example by immersing bruised leaves or chopped leaves in an agrobacterial solution and then culturing them in suitable media. The transformation of plants by means of Agrobacterium tumefaciens is described, for example, by Hofgen and Willmitzer in Nucl. Acid Res. (1988) 16, 9877 or is known inter alia from F.F. White, Vectors for Gene Transfer in Higher Plants; in Transgenic Plants, Vol. 1 , Engineering and Utilization, eds. S.D. Kung and R. Wu, Academic Press, 1993, pp. 15-38.

In addition to the transformation of somatic cells, which then have to be regenerated into intact plants, it is also possible to transform the cells of plant meristems and in particular those cells which develop into gametes. In this case, the transformed gametes follow the natural plant development, giving rise to transgenic plants. Thus, for example, seeds of Arabidopsis are treated with agrobacteria and seeds are obtained from the developing plants of which a certain proportion is transformed and thus transgenic [Feldman, KA and Marks MD (1987). Mol Gen Genet 208:1 -9; Feldmann K (1992). In: C Koncz, N-H Chua and J Shell, eds, Methods in Arabidopsis Research. Word Scientific, Singapore, pp. 274-289], Alternative methods are based on the repeated removal of the inflorescences and incubation of the excision site in the center of the rosette with transformed agrobacteria, whereby transformed seeds can likewise be obtained at a later point in time (Chang (1994). Plant J. 5: 551 -558; Katavic (1994). Mol Gen Genet, 245: 363-370). However, an especially effective method is the vacuum infiltration method with its modifications such as the "floral dip" method. In the case of vacuum infiltration of Arabidopsis, intact plants under reduced pressure are treated with an agrobacterial suspension [Bechthold, N (1993). CR Acad Sci Paris Life Sci, 316: 1 194-1 199], while in the case of the "floral dip" method the developing floral tissue is incubated briefly with a surfactant-treated agrobacterial suspension [Clough, SJ and Bent AF (1998) The Plant J. 16, 735-743], A certain proportion of transgenic seeds are harvested in both cases, and these seeds can be distinguished from non-transgenic seeds by growing under the above-described selective conditions. In addition, the stable transformation of plastids is of advantages because plastids are inherited maternally is most crops reducing or eliminating the risk of transgene flow through pollen. The transformation of the chloroplast genome is generally achieved by a process which has been schematically displayed in Klaus et al., 2004 [Nature Biotechnology 22 (2), 225-229], Briefly the sequences to be transformed are cloned together with a selectable marker gene between flanking sequences homologous to the chloroplast genome. These homologous flanking sequences direct site- specific integration into the plastome. Plastidal transformation has been described for many different plant species and an overview is given in Bock (2001) Transgenic plastids in basic research and plant biotechnology. J Mol Biol. 2001 Sep 21; 312 (3):425-38 or Maliga, P (2003) Progress towards commercialization of plastid transformation technology. Trends Biotechnol. 21, 20-28. Further biotechnological progress has recently been reported in form of marker free plastid transformants, which can be produced by a transient co-integrated maker gene (Klaus et al., 2004, Nature Biotechnology 22(2), 225-229).

The term "plant" as used herein encompasses whole plants, ancestors and progeny of the plants and plant parts, including seeds, shoots, stems, leaves, roots (including tubers), flowers, and tissues and organs, wherein each of the aforementioned comprise the gene/nucleic acid of interest. The term "plant" also encompasses plant cells, suspension cultures, callus tissue, embryos, meristematic regions, gametophytes, sporophytes, pollen and microspores, again wherein each of the aforementioned comprises the gene/nucleic acid of interest.

The plant of the invention may be a monocot or dicot. Preferably, the plant is of a species belonging to the Asteraceae family, such as of the subfamily Cichorioideae, optionally to the genus of Lactuca (e.g. Lactuca sativa), the genus of Taraxacum (e.g. Taraxacum officinale), the genus of Cichorium (e.g. Cichorium intybus, Cichorium endivia), the genus Tanacetum (e.g. Tanacetum parthenium) the genus Scorzonera (e.g. Scorzonera hispanica or Scorzonera humilis), the genus Cynara (e.g. Cynara scolymus), the genus Tragopogon (e.g. Tragopogon porrifolius) or the genus of Gazania. Optionally, the plant is of the subfamily Asteroideae, such as of the genus Heliantheae (e.g. Helianthus annuus or Helianthus tuberosus), the genus Parthenium (e.g. Parthenium argentatum), or the genus Artemisia (e.g. Artemisia annua). The plant may also be a plant of the Lamiaceae family, Vitaceae family and Cannabaceae family (e.g. see Nguyen et a!. Biochem Biophys Res Commun. 2016 Oct 28;479(4):622-627). The plant may be, or may be obtainable from, the Asteraceae family, preferably of the subfamily of Cichorioideae, preferably of the genus Cichorium, more preferably an Cichorium intybus or Cichorium endivia plant.

Optionally, the method of the invention further comprises a step for transferring the one or more modified KLS genes of the invention (the one or more nucleic acids of the invention) to offspring of the plant produced by the method of the invention, which may be performed by introgression. Breeding techniques for introgression are well known to one skilled in the art.

Preferably, the method of the invention results in a plant having reduced SL levels, compared to a control plant as defined herein.

In some embodiments, the plant cell according to the invention is non-propagating or cannot be regenerated into a plant.

All publications and patent applications in this specification are indicative of the level of ordinary skill in the art to which this invention pertains. All publications and patent applications are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated by reference.

Many modifications and other embodiments of the inventions set forth herein will come to mind to one skilled in the art to which these inventions pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the inventions are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Examples

1. Comprehensive screen for candidate genes of the SL biosynthetic pathway

1.1 Genome-wide annotation of the cytochrome P450 gene family in the C. intybus genome

The cytochrome P450 (CYP) family is known to be important in the conversion from FPP into kauniolide and thus was selected for genome-wide identification of all gene family members. The in house assembled and annotated reference genome sequence of C. intybus L. var. sativum was used to follow a two-step strategy. First, the known and putatively involved SL biosynthesis genes CiGAO, CiCOS and TpKLS were used to select CYP gene family members (HOMgroups) from a large number of species in the PLAZA 4.0 dicot database (Van Bel et al., 2018). Next, the previously described C. intybus SL biosynthesis genes and the PLAZA4.0 homologous CDS and protein sequences were used as queries for BLAST against the predicted C. intybus gene set and genome assembly (to uncover genes missed by gene prediction). Blast results were parsed to create a non-redundant list of C. intybus protein-coding gene models. During manual curation, all genome regions with a significant blast hit were searched for the corresponding open reading frame (ORF). Partial genes (smaller than 350 amino acids) or genes with obvious mutations in the coding sequence (premature stop codons, frameshifts and indels), or structural variants affecting the splicing donor or acceptor sites, leading to disruption of the protein coding capacity, were classified as pseudogenes. In total 396 CYP C. intybus gene models were manually annotated of which six putative paralogous CiGAO genes, four putative paralogous CiCOS genes and three putative paralogous CiKLS genes. Both C. intybus and L. sativa protein models were used for the creation of the CYP phylogenetic tree, visualized in Figure 2. Phylogenetic analysis combining catalytically active GAO, COS and KLS protein models from different species with the putative CiGAO, CiCOS and CiKLS genes is visualized in Figure 3.

1.2 Identification of MeJA-inducible genes using RNA-seq

The effect of MeJA on the expression of genes involved in early steps of the SL biosynthetic pathway (CiGASI, CiGASs, CiGAO and CiCOS) was evaluated with RT-qPCR analysis, showing the strongest upregulation after 6h MeJA treatment in most samples. Samples of industrial chicory 'OBOE', witloof 'Topmodel' and lettuce 'ZORBA', after 6h of mock or MeJA treatment were used for subsequent RNA- seq analysis in three biological replicates. The industrial chicory and witloof reads were mapped to the CDS sequences of C. intybus L. var. sativum (manually curated for CYP gene models), containing 53,507 annotated genes, while the lettuce reads were mapped to the CDS sequences of L. sativa, containing 37,829 annotated genes (Reyes-Chin-Wo et al., 2017) (Table 1). A total of 20,358 (38 % of 53,507) genes were expressed in industrial chicory, 20,024 genes (37 % of 53,507) were expressed in witloof, and 18,313 genes (48 % of 37,829) were expressed in lettuce (Table 1). Hereof, 3,384 (9 % of 39,495) industrial chicory genes, 2,696 (7 % of 37,509) witloof genes and 1,395 (8 % of 18,313) lettuce genes were statistically significant differentially expressed between MeJA and mock treatment (Table 1). Comparison between industrial chicory and witloof gene sets identified a total of 1,930 genes that were commonly differentially expressed upon MeJA treatment. In industrial chicory, witloof, and lettuce, the number of downregulated genes (MeJA vs. mock) was 1,489, 1,076 and 586, respectively and the number of upregulated genes (MeJA vs. mock) was 1895, 1620 and 809, respectively (Table 1). 1.3 Integration of DEG and phylogenetic analysis for the selection of SL biosynthetic candidate genes

A total of 320 and 354 genes were identified in the CYP superfamily in C. intybus and L. sativa, respectively. A total of 49 industrial chicory genes, 51 witloof genes and 33 lettuce genes were MeJA- inducible (Table 1). Phylogenetic analysis identified the clades with the previously described CiGAO, CiCOS and TpKLS genes (Liu et al., 2011, 2018). Taken together, these clades contain a total of 26 C. intybus and 23 L. sativa genes, of which 9 industrial chicory genes, 10 witloof genes and 9 lettuce genes were MeJA-inducible (Figure 2). Phylogenetic analysis further identified three subclades that each contained multiple additional paralogues of CiGAO, CiCOS and CiKLS genes.

A functional CiGAO (CYP71AV4) gene was previously cloned from C. intybus (Liu et al., 2011; Nguyen et al., 2010) and results revealed five close paralogues. CYP71AV27 was located on CiChr7, CYP71AV8 and CYP71AV26 were tandem duplicated genes located on CiChr7, and CYP71AV23 and CYP71AV25 were located on CiChr4 and were MeJA-inducible in both industrial chicory and witloof. In the same clade, six lettuce orthologues were identified of which four genes (Lsat_l_v5_gn_8_115261.1, Lsat_l_v5_gn_3_65300.1, Lsat_l_v5_gn_3_65241.1 and Lsat_l_v5_gn_5_46240.1) were MeJA- inducible. Based on the protein alignment and phylogenetic clustering of the CYP genes with the previously described CYP71AV4 (CiGAO) gene, a total of six genes in the genome of C. intybus were selected as candidate CiGAO paralogous genes (Figure 2; green).

Four putative paralogues were identified in the CiCOS clade: CYP71BL11, CYP71BL10, CYP71BL3 and CYP71BL12 of which all were located on CiChr5 in proximity of each other. CYP71BL3 revealed 99.8 % nucleotide identity (one SNP difference) to the previously described functional CiCOS gene (Liu et al., 2011) and CYP71BL10 and CYP71BL3 were MeJA-inducible in both industrial chicory and witloof (Figure 2). Within the same clade, three lettuce orthologues were identified, all of them were MeJA-inducible (Lsat_l_v5_gn_7_337801, Lsat_l_v5_gn_7_33721.1 and Lsat_l_v5_gn_4_113481.1). Based on the protein alignment and phylogenetic clustering of the CYP genes with the previously described CYP71BL3 (CiCOS) gene, a total of four genes in the genome of C. intybus were selected as candidate CiCOS paralogous genes (Figure 2; orange)

The clade containing TpKLS revealed a total of ten putative paralogues in C. intybus (Figure 2), divided in three subclades. The tandem duplicated gene pair CYP81BZ19 and CYP71BZ18 located on CiChr2 were both MeJA-inducible. The three tandem duplicated genes, CYP71BZ27, CYP71BZ25 and CYP71BZ26, were located on CiChr5 of which two genes were MeJA-inducible. In the subclade containing five genes, CYP71BZ22, CYP71BZ23 and CYP71BZ24 were tandem duplicated genes located on CiChr5 while CYP71BZ21 and CYP71BZ20 were tandem duplicated genes located on CiChr4, of which only CYP71BZ20 was MeJA-inducible in witloof. Within the clade, a total of eight lettuce orthologues were identified of which only two genes (Lsat_l_v5_gn_l_89541.1 and Lsat_l_v5_gn_7_33021.1) were MeJA-inducible. Based on the protein alignment and phylogenetic clustering of the CYP genes with the previously described TpKLS gene, a total of ten genes in the genome of C. intybus were selected as candidate CiKLS paralogous genes (Figure 2; red), of which none have previously been functionally characterized. Furthermore, other clades in proximity of the CiGAO, CiCOS and/or CiKLS clades were revealed, but none of these genes were MeJA-inducible.

Phylogenetic analysis of the six selected CiGAO candidates, four selected CiCOS candidates and ten selected CiKLS candidates was executed including the known catalytically active GAO, COS and KLS proteins from Helianthus annuus (sunflower), Tanacetum cinerariifolium (Dalmatian chrysanthemum), Tanacetum parthenium (feverfew), Lactuca sativa (lettuce), Cichorium intybus (chicory), Cynara cardunculus (artichoke), Saussurea costus (costus) and/or Bamadesia spinosa (clavelillo). This illustrates the distribution of the proteins over multiple tribes of the Asteraceae family, such as Cichorieae (chicory and lettuce), Cynareae (artichoke) and Barnadesieae (clavelillo).

Table 1. Overview of number of annotated genes, expressed genes, differentially expressed genes (DEG) between MeJA and mock treatment, MeJA downregulated and upregulated DEGs, and MeJA upregulated DEGs belonging to the CYP family (excluding pseudogenes), with total number of CYP genes in brackets, in industrial chicory, witloof and lettuce. RNA-seq reads from industrial chicory and witloof were mapped to the CDS sequences of C. intybus while lettuce RNA-seq reads were mapped to the CDS sequences of L. sativa. Upregulated genes have a Iog2-fold change > 1 while downregulated genes have a Iog2-fold change ≤ -1. The number of annotated genes of industrial chicory and witloof are equal because the same CDS gene set was used from the C. intybus var. sativum reference genome. Material & Methods to example 1

Gene identification and phylogenetic analysis

A novel, in house assembled and annotated reference genome sequence of C. intybus L. var. sativum (https://bioinformatics.psb.ugent.be/orcae/overview/Cicin2) (unpublished) was used for genome-wide identification of cytochrome P450 (CYP) gene family members, acting upstream in the SL biosynthetic pathway. First, the corresponding HOMgroups with all known orthologues of multiple species were extracted from the PLAZA dicots 4.0 database (https://bioinformatics.psb.ugent.be/plaza) (Van Bel et al., 2018). These coding DNA sequences (CDS) and protein sequences were used as queries to identify the C. intybus homologs using BLASTn and BLASTp searches (e-value <le-80), respectively, against the CDS and protein sequences of all 53,507 predicted C. intybus gene models. ORCAE (Sterck et al., 2012) was used for manual curation of the structural gene annotation. Each predicted gene model was manually curated, using available supporting data in ORCAE (such as RNA-seq read alignments, splicing sites, gapped alignments of de novo assembled transcripts, and blast hits of orthologues from closely related species). Predicted C. intybus proteins were further validated via multiple protein sequence alignment (MUSCLE) and CYP C. intybus genes leading to proteins smaller than 350 amino acids or without a C-terminal, were classified as pseudogenes. A tBLASTx was performed between the CDS of C. intybus and L. sativa (V8, 2016-01-20, id28333, https://lgr.genomecenter.ucdavis.edu/) in both directions (e-value <le-80) to identify the corresponding L. sativa orthologues. L. sativa genes were not manually curated but genes leading to proteins smaller than 350 amino acids were classified as pseudogenes. Proteins encoded by C. intybus and L. sativa genes (excluding the pseudogenes), were used for phylogenetic analysis (UPGMA tree) in Geneious 10.2.6 (http://www.geneious.com). The C. intybus CYP protein sequences were further classified by David R. Nelson (University of Tennessee, USA). The catalytically active GAO, COS and KLS proteins from different species were then extracted from NCBI (Eljounaidi et al., 2014; Frey et al., 2020; Liu et al., 2014, 2018; Ramirez et al., 2013) and further used for phylogenetic analysis together with the identified candidate CiGAO, CiCOS and CiKLS paralogous proteins.

MeJA treatment on seedlings

C. intybus var. foliosum 'Topmodel' and 'Van Tongelen', C. intybus var. sativum 'OBOE' and 'VL70' and L. sativa 'ZORBA' seeds were separately rinsed in demi water for 1 min, sterilized in 70 % EtOH for 1 min and sterilized in 1.5 % NaOCI + 0.02 % Tween20 for 15 min. Seeds were rinsed three times in sterile water for 1 min, air dried and placed in sterile plastic containers (145 mm x 100 mm x 60 mm) (Eco2nv, Zottegem, Belgium). Eight seeds from each species were placed on 145 mL solid growth medium (4.4 g.L" ¹ Murashige and Skoog (MS) + vitamins (Murashige & Skoog, 1962), 20 g.L ^-1 sucrose, 7 g.L ^-1 agar, pH = 5.8) topped with an autoclaved 14.5 x 10 cm 100 pM mesh (Prosep bvba, Zaventem, Belgium). The containers with seeds were kept in a growth chamber for 2 weeks at 23 ± 2°C under a 16h / 8h (light / dark) photoperiod at 40 pmol.m^.s ¹ photosynthetic active radiation (PAR). After two weeks of growth, the number of seedlings was reduced to four well grown seedlings per container and the mesh containing the seedlings was transferred to 145 mL solid mock growth medium (4.4 g.L ^-1 MS + vitamins, 20 g.L - s ¹ ucrose, 7 g.L - a ¹ gar, 0.0002 %Tween20, pH = 5.8) or 145 mL solid growth medium supplemented with 100 pM MeJA. After 2h, 6h or 24h of treatment, the four seedlings in each container were pooled and frozen in liquid nitrogen. A total of 120 samples (five plant varieties, two treatments, three time points, four biological replicates each) were harvested and grinded in liquid nitrogen.

RNA extraction and RT-qPCR analysis

RT-qPCR was performed to validate MeJA-induction of CiGASI (AF497999.1), CiGASs (AF498000.1), CiGAO (GU256644.1) and CiCOS (JF816041.1) candidate genes and to select the most suitable samples for RNA-seq. RNA was extracted from 150 to 200 mg grinded plant material, using the 3 % CTAB RNA extraction protocol (Luypaert et al., 2017). RNA extraction, quantification and RT-qPCR were executed as described in De Keyser et al. (2020). RNA was quantified using the Nanodrop ND1000 (Isogen Lifescience B.V.) and 200 ng μL ^-1 RNA was used for DNasel treatment with DNA-/ree™ (Ambion) at 37°C for 30 min. The RNA quality was verified on a subset of 16 samples using the Experion microfluidic capillary electrophoresis system (Bio-Rad) in combination with the RNA StdSens Chips (Bio-Rad). Next, cDNA synthesis was performed with the iScript™ cDNA synthesis kit (Bio-rad) starting from 820 ng of RNA. As a control for genomic DNA contamination, no reverse transcriptase (noRT) samples were included. Both cDNA and noRT samples were diluted 1/3 and stored at -20°C. Based on previous research, eleven reference genes were selected for industrial chicory and witloof (Delporte et al., 2015) and ten reference genes were selected for lettuce (Borowski et al., 2014; Sgamma et al., 2016), gene specific primers are listed in Table 8. All 24 cDNA samples of lettuce were used for reference gene validation in lettuce, while another similar cDNA sample set of industrial chicory and witloof was used for reference gene validation in industrial chicory and witloof. RT-qPCR with the reference genes was performed on these samples in a LightCycler480 (Roche). The geNorm (Vandesompele et al., 2002) module in qbase+ (Hellemans et al., 2007; Biogazelle) was used for reference gene selection. RT-qPCR primers targeting CiGASI, CiGASs, CiGAO and CiCOS were developed (Table 9) to amplify at least one paralogous gene copy. RT-qPCR analysis was done on the entire sample set (120 samples), all noRT samples and including no template controls (NTC) using both target and selected reference genes [TIP41 and PP2AA2 for industrial chicory and witloof (M-value = 0.124; CV-value = 0.043) and PP2AA3 and TIP41 for lettuce (M-value = 0.423; CV-value = 0.146)]. In case of amplification of noRTs, contamination with genomic DNA was considered to be negligible when the difference in quantification cycle (Cq) between noRT and the sample was above five cycles (Hellemans et al., 2007). Gene specific amplification efficiencies were determined using LinRegPCR (Ruijter et al., 2009; Table 9) and a normalization factor based on two validated reference genes was used for calculation of calibrated normalized relative quantities (CNRQ.) in qbase+ (Biogazelle). CNRQ values were analyzed in R v.3.4.3 (R Core Team, 2021) and were subjected to Kruskal-Wallis rank sum test at p<0.05, using Pairwise Wilcoxon rank sum test to determine differences between independent samples.

RNA-seq and DEG analysis

A total of 18 RNA samples were selected (three species, three biological replicates each per MeJA treatment versus mock) for RNA-seq analysis (Figure SI). Libraries were prepared and sequenced by the VIB Nucleomics Core facility (Leuven, Belgium) with SE-75 reads on an Illumina NextSeq500 instrument (SRA accession PRJNA738883). Adapters and low quality 3' ends were trimmed from the RNA-seq reads (with a quality threshold of 20) using Cutadapt (Martin, 2011). Next, reads were dereplicated using prinseq-lite (Schmieder & Edwards, 2011). Industrial chicory and witloof reads were mapped on the CDS sequences of the in house assembled and annotated reference genome of C. intybus L. var. sativum while lettuce reads were mapped on the CDS sequences of the annotated reference genome of L. sativa, using BWA-MEM (Li & Durbin, 2009) with default settings. Mapped reads were sorted using SAMtools (Li et al., 2009) and reads with mapping quality below 20 were discarded. Subsequent data analysis was done in R v.3.4.3 (R Core Team, 2021) using the package edgeR (Robinson et al., 2009). Genes which occurred in less than two samples with less than four Counts per Million reads mapped (CPM) were discarded. Normalization to scale the raw library sizes was done using the trimmed mean of M-values (to the reference) as proposed by Robinson and Oshiack (2010). To identify differentially expressed genes (DEGs), a negative binomial generalized log-linear model to the read counts for each gene was fitted and subsequently likelihood ratio tests were performed between the MeJA treatment and the mock per cultivar. Thresholds for further filtering of the DEGs were set at minimum 2-fold expression change (Iog2- fold change of 1) and an adjusted p-value (Benjamini-Hochberg multiple testing correction) cutoff of 0.05; and were further considered as statically significant DEGs between MeJA and mock treatment. A heatmap was constructed visualizing the gene expression ratio (Iog2-fold change) of MeJA over mock treated samples of industrial chicory, witloof and lettuce.

2.Transient expression assay in N. benthamiana for the reconstruction of the SL biosynthetic pathway

2.1 Isolation of the CDS of SL candidate genes

Multiple alleles of the CiKLS candidate genes were isolated using cDNA synthesized from RNA extracts of C. intybus plants of a range of varieties, tissues, growth conditions and treatments (Table 3). Table 3. Successfully isolated and sequence verified alleles of seven (out of ten) CiKLS candidate genes with corresponding putative gene clade and RNA isolation information (plant variety, tissue, growing condition and treatment). Underlined genes indicate 100 % protein identity with the protein sequence translated from the corresponding C. intybus reference gene coding sequence.

CYP71BZ25

SEQ ID NO: 13 depicts the nucleotide sequence of the gene CYP71BZ25 which encodes the KLS polypeptide depicted in SEQ. ID NO: 14. CYP71BZ26

SEQ ID NO: 15 and SEQ ID NO: 17 depict the nucleotide sequences of the gene CYP71BZ26 (two different alleles) which respectively encode the KLS polypeptides depicted in SEQ ID NO: 16 and SEQ ID NO: 18.

CYP71BZ27

SEQ ID NO: 19, 21 and 23 depict the nucleotide sequences of the gene CYP71BZ27 (three different alleles) which respectively encode the KLS polypeptides depicted in SEQ ID NO: 20, 22 and 24.

The nomenclature of the Cytochrome P450 sequences is described in Nelson DR (2009) Human Genomics, 'The Cytochrome P450 Homepage', Vol. 4, 59.

2.2 Production of SLs downstream of costunolide in N. benthamiana Full length coding sequences of twelve putative CiKLS CDSs (including multiple alleles) were expressed in N. benthamiana leaves by co-agroinfiltration with CiGASs, CiGAO (CYP71AV4) and CiCOS (CYP71BL3). Co-agroinfiltration of CiGASs, CiGAO and CiCOS was used as positive control for the production of costunolide, while co-agroinfiltration of CiGASs, CiGAO and CiCOS with TpKLS was used as positive control for the potential production of kauniolide, as described previously (Liu et al., 2018). The expression of the TpKLS coding sequence as well as the expression of CYP71BZ25 (SEQ ID NO: 13), the two alleles of CYP71BZ26 (SEQ. ID NO: 15 and 17) and the three alleles of CYP71BZ27 (SEQ ID NO: 19, 21 and 23) in N. benthamiana leaves, resulted in the observation of two peaks co-eluting at ± 30 min in their GC-MS chromatograms. After deconvolution of the peaks, one of them indeed showed an EI-MS and accurate mass corresponding to kauniolide. No such peak was observed in any of the other extracts (Figure 4). However, the need of a kauniolide standard arises to further verify the activity of TpKLS and the three paralogous CiKLS genes. Still, this indicates that the conversion of costunolide to kauniolide is most likely catalyzed by three paralogous CiKLS genes, some of which encoded by multiple functional alleles isolated from industrial chicory and/or witloof plants.

Material & Methods to example 2

Plant material

Plant materials of industrial chicory C. intybus var. sativum 'L9001' were provided by COSUCRA (Belgium), and witloof C. intybus var. foliosum 'Van Hamme' were provided by Nationale Proeftuin van witloof (Belgium). Roots of in vivo plants of the selected Cichorium varieties were rinsed with water, grated on the outside and cut into slices of 1 cm. The slices were rinsed for 1 min in 70 % ethanol, sterilized in 2.5 % NaOCI, and rinsed in sterile water. Next, the slices were cut into pieces of 1 - 2 cm ³ and transferred to solid plant medium (4.4 g.L ^-1 MS medium (Murashige & Skoog, 1962), 45 g.L ^-1 sucrose, 8 g.L ^-1 plant tissue culture agar No. 4 (Neogen, Michigan), pH 6) at 23 ± 2 °C under a 16 h / 8 h (light / dark) photoperiod at 40 μmol.m ^-2 .s ^{- 1} photosynthetic active radiation (PAR). After shoot induction, plants were transferred to new solid plant medium (4.4 g.L ^-1 MS medium + vitamins, 20 g.L ^-1 sucrose, 7 g.L ^-1 plant tissue culture agar No. 4, pH 6) and subcultured every six weeks. Two in vitro plants of each variety were transferred to pots (0; 9 cm) and grown in a peat based substrate (1.5 kg.m-3 fertilizer: 12N:14P:24K + trace elements, pH 5.0 - 6.5, EC 450 μS.cm-1, Van Israel, Geraardsbergen, Belgium) under greenhouse conditions (temperature ± 20 °C, 60 - 65 % relative humidity). Both in vitro plants and greenhouse plants (± 2 months after transfer) were treated with 100 μM MeJA solution by spraying the leaves till run-off and adding 20mL of MeJA solution to the agar or soil, respectively. Leaves and roots were harvested from all untreated industrial chicory and witloof plant varieties and from all six hours MeJA treated plant varieties and further grinded in liquid nitrogen.

For N. benthamiana infiltration assays, N. benthamiana was cultivated for 3 to 4 weeks in a plant growth chamber maintained at 25°C in a 14 h / 10 h (light / dark) photoperiod. Isolation and cloning of the CDS of CYP candidate genes from C. intybus

RNA was extracted from 150 to 200 mg grinded plant material, using the 3 % CTAB RNA extraction protocol (Luypaert et al., 2017). RNA was quantified using the Nanodrop ND1000 (Isogen Lifescience B.V.) and 200 ng μL ^-1 RNA was used for DNasel treatment with DNA-/ree™ (Ambion) at 37°C for 30 min. The cDNA synthesis was performed with the iScript™ cDNA synthesis kit (Bio-rad) starting from 820 ng of RNA and was used for gene isolation. Full length CDSs of the CiKLS candidate were amplified with Q5 High Fidelity Polymerase (New England Biolabs) using gene specific primers (Table 2). Successful amplified PCR amplification products were further used in an additional Q.5 High Fidelity Polymerase reaction using the gene specific primers flanked by AttB sites necessary for Gateway cloning. The PCR amplification products were run on a 1.5 % agarose gel and extracted and purified with GeneJET Gel Extraction Kit (Thermo Fisher Scientific). Previously identified CiGASs (AF498000.1) (Bouwmeester et al., 2002), CiGAO (GU256644.1) (Nguyen et al., 2010), CiCOS (JF816041.1) (Liu et al., 2011) and TpKLS (MF197559.1) (Liu et al., 2018) gene coding sequences were synthesized by Twist Bioscience (San Francisco, US). All amplified or synthesized isolates were subsequently cloned into the Gateway donor vector pDONR207 (Invitrogen) using Gateway™ BP clonase™ (Thermo Fisher Scientific) and sequence verified by Sanger sequencing (Eurofins) prior to insertion into the binary N. benthamiana expression vector pEAQ-HT-DESTl (Sainsbury et al., 2009) using Gateway™ LR clonase™ (Thermo Fisher Scientific).

Table 2 Gene specific primers for the amplification of full-length coding sequences in C. intybus using Q5

High Fidelity Polymerase.

CiKLS Gene Primer Forward Primer Reverse

CYP71BZ18 5'-ATGGACACACAAATAGCATT-3' 5'-TTAATTGTCGGGAGCATAAGTTG-3'

CYP71BZ19 5'-ATGAGCACCGAATTAGCATTCTC-3' 5'-TCAACTGTAAGAAGCATAAGAAG-3'

CYP71 BZ20 5' -ATGG AC ACTG ATATCACCTTCT-3' 5' -TC ATTGTTTTATTGTC A AA A-3'

CYP71BZ21 5'-ATGGACGCCGTCGCCGGTAT-3' 5'-TTAAAATGAAGAATATATAGTTG-3'

CYP71BZ22 5'-ATGGACGCCGATATCACCTTC-3' 5'-TCAAACTGAAAAATAGAGAGTTG-3'

CYP71BZ23 5'-ATGGACCTCGCTATCACCTAC-3' 5'-TCAAGACGAAGAATAGAGTGTCG-3'

CYP71BZ24 5'-ATGGACGCCGATATCACCTTC-3' 5'-TCAAAATGAAGAATATAGAGTTG-3'

CYP71BZ25 5'-ATGGCCATCGAAATGACCATC-3' 5'-TCAAAATGGAATAAAAACGGT-3'

CYP71BZ26 5'-ATGGTCATCGATATGACCATCGT-3' 5'-TCAAAACGGGATAAAAATTGTTG-3'

CYP71BZ27 5'-ATGGCCATCGATATGACCATCG-3' 5'-TCAAAACGGGATAAAAACTGTTG-3' N. benthamiana leaf infiltration

The recombinant N. benthamiana expression vectors were individually transformed into the Agrobacterium tumefaciens strain C58C1, carrying the pMP90 helper plasmid, by electroporation. Transformed A. tumefaciens were grown for 2 days in a shaking incubator (150 rpm) at 28°C in 5 mL yeast extract broth (YEB) medium, supplemented with 50 mg.L ^-1 rifampicin, 25 mg.L ^-1 kanamycin and 20 mg.L ^-1 gentamycin. After incubation, 500 mL of bacterial culture was used to inoculate 9.5mL of YEB medium supplemented with antibiotics and containing lOmM MES (pH 5.7) and 20 pM acetosyringone. After overnight incubation in a shaking incubator (150 rpm) at 28°C, cells were collected via centrifugation for 10 min at 4000 rpm and washed two times with 5 mL MQ-water. The cells were resuspended in 5 mL of infiltration buffer (100 pM acetosyringone, 10 mM MgCI2, and 10 mM MES, pH 5.7). The number of bacteria harvested for each construct was adjusted to a final OD600 of 1.0 after resuspension in the infiltration buffer. After 3 h incubation (150 rpm) at 28°C, the bacteria for transient co-expression were mixed. Each mixture was infiltrated to the abaxial side of three fully expanded leaves of 3- to 4-week-old N. benthamiana plants using 1 mL syringes. Infiltration with A. tumefaciens containing an empty pEAQ-HT-DESTl vector (EV) was executed as a negative control. The agroinfiltrated plants were incubated for 3 days in a growth chamber with conditions as previously described (see 3.2.1) until harvesting for metabolite analysis.

Metabolite extraction and GC-MS analysis

Agroinfiltrated N. benthamiana leaves were harvested and ground to a fine powder in liquid nitrogen. Around 100 mg of leaf material was extracted by vortexing with ImL of hexane for 10 min followed by centrifuging for 15 min at 3,000 rpm. The resulting organic extract was transferred to a new clean Eppendorf tube and evaporated to dryness under vacuum conditions. For GC-MS analysis, the residue obtained from metabolite extraction was trimethylsilylated using 10 μL of pyridine and 50 μL of N- methyl-N-(trimethylsilyl)trifluoroacetamine. GC-MS analysis was performed on a 7890B GC system and a 7250 GC/QTOF (Agilent). Analyses were performed on a VF-5ms capillary column (30m x 0.25 mm x

O.25 pm; Varian CP9013; Agilent) at a constant helium flow of 1.2 mL/min, using a 1 μL aliquot, which was injected in splitless mode. After injection, the oven was held at 70°C for 2 min, ramped to 210°C at a rate of 5°C/min, held at 210°C for 5 min, ramped to 320°C at a rate of 20°C/min, held at 320°C for 5 min, and finally cooled to 70°C at a rate of 50°C/min at the end of the run. The injector, the MS transfer line, the MS ion source, and the quadrupole were set to 280°C, 280°C, 230°C and 150°C, respectively. Full EI-MS spectra were generated for each sample by scanning the m/z range of 50 to 800 with a solvent delay of 15 min. Electron ionization energy was 70 eV. For relative quantification, the peak areas were calculated using the default settings of the Masshunter Quantitative Analysis software (version 2.6; NIST). Targeted analysis of costunolide in the agroinfiltrated N. benthamiana leaves was performed by comparing retention times and mass transitions with that of a costunolide standard (Sigma-Aldrich).

3. Development of a CRISPR/Cas9 methodology for Cichorium

3.1 CRISPR/Cas9 induced mutations in multiple putative paralogous KLS genes

The CRISPR/Cas9 genome editing workflow was used to target the multiple putative paralogous CiKLS genes in 'Van Hamme' protoplasts. A total of 25 regenerated plants, were acclimatized to the greenhouse and genetically characterized at all gene target loci, including 3 control plants (NCI) and 22 plants targeted in the CiKLS paralogous candidate genes. Out of the 22 CRISPR/Cas9 transfected and regenerated plants containing potential mutations, mutation analysis revealed 4 plants with at least one mutation in each CiKLS paralogous candidate gene (Table 5), making them the desired mutated plant genotypes for further SL metabolite profiling.

Table 5. Overview and number of CiKLS mutated plant genotypes.

WT / Indel = Heterozygous mutation; Indel = Single observed mutation; Indel / Indel = Compound heterozygous mutation; I = Insertion; D = Deletion, WT = Wild Type; '//' separates mutation data of two analyzed loci. Mutation types and plant genotypes with mutations leading to early stop codons, truncating the translated protein, are underlined. indicates no detected mutation.

Material and methods to example 3

Plant material

Plant materials of industrial chicory C. intybus var. sativum 'L9001' and 'V644' were provided by COSUCRA (Belgium), and witloof C. intybus var. foliosum 'Van Hamme' and 'Topmodel' were provided by Nationale Proeftuin van witloof (Belgium) and subcultured in vitro as previously described. In vitro plants were grown on solid plant medium (MS medium + vitamins, 20 g.L ^-1 sucrose, 7 g.L ^-1 plant tissue culture agar No. 4, pH 6) and subcultured every six weeks.

CRISPR/Cas9 vector construction

Two guide RNAs for the functionally characterized paralogous CiKLS genes were designed using Geneious 10.2.6 (http://www.geneious.com). The CRISPR/Cas9 gene target sequences and primers necessary for HiPlex amplicon sequencing are shown in Table 4 and visualized in Figure 5. Construction of the CRISPR/Cas9 expression vector is described in De Bruyn et al. (2020).

Table 4. Oligonucleotide sequences for the construction of 20 nucleotide gRNA sequences in CRISPR/Cas9 vectors and primer sequences for HiPlex amplicon sequencing of CiKLS target loci. Underlined = N20 guide RNA sequence, Italic = vector overhangs.

Protoplast isolation, transfection and regeneration

Cichorium protoplasts were isolated from young and healthy leaves from in vitro cultured plants as previously described (Deryckere et al. 2012). Protoplast suspensions were diluted to 500.000 protoplasts. mL ^-1 and 100 μL was added to a minimum of 10 pg and maximum a total of 20 pg vector material. Next, 120 μL PEG3350 solution (400 g.L - P ¹ EG3350, 72.8 g.L - m ¹ annitol, 23.6 g.L - C ¹ a(NO ₃ ) ₂ .4H ₂ O, pH 6) was added to the solution, gently mixed and samples were incubated in the dark for 10 min at room temperature. The transfection reaction was stopped by adding 600 μL of W5 medium (8.77 g.L ^-1 NaCI, 18.38 g.L - C ¹ aCI ₂ .2H ₂ O, 0.37 g.L ^-1 KCI and 0.9 g.L - g ¹ lucose, pH 5.8) and mixed by inverting the tubes five times. The samples were centrifuged for 5 min at 80 g in a swing out centrifuge (Eppendorf™ 5810R Centrifuge) and the supernatant was removed. Protoplasts transfected without vector (NCI) and protoplasts without the addition of both PEG and vector (NC2) were used as negative controls. After transfection, 600 μL of regeneration medium (% MS macro elements (without NH4NO3 and KNO3) (Murashige & Skoog, 1962) with Heller micro elements (Heller, 1953) and Morel & Wetmore vitamins (Morel & Wetmore, 1951), 18.3 mg.L ^-1 FeNA-EDTA, 750 mg.L ^-1 KCI, 100 mg.L ^-1 inositol, 750 mg.L ^-1 glutamine, 10 g.L ^-1 sucrose, 60 g.L ^-1 mannitol, 0.5 mg.L ^-1 NAA, 0.5 mg.L ^-1 BAP, pH 5.5) was added to the protoplast pellet and the protoplasts were regenerated into plants following the protocol described by Deryckere et al. (2012). After four to five months, the transfected shoots and respective control shoots were acclimatized for four weeks under a fog tunnel construction with plastic covering (temperature ± 25 °C, 70 - 80 % relative humidity). Afterwards, plantlets were transferred to pots (Ø; 9 cm) and grown in a peat based substrate (1.5 kg.m ^-3 fertilizer: 12N:14P:24K + trace elements, pH 5.0- 6.5, EC 450 pS.cm - ¹ , Van Israel, Geraardsbergen, Belgium) under greenhouse conditions (temperature ± 20 °C, 60 - 65 % relative humidity). Molecular analyses

Genomic DNA of all successfully regenerated plants (mutated and wild type) was extracted from ± 50 mg fresh leaf material using a CTAB method (Doyle & Doyle, 1990). Per sample, DNA concentration was measured using the Nanodrop ND1000 (Isogen Lifescience B.V.) and samples were diluted to obtain a final DNA concentration of maximum 40 ng.pl ¹ . Primers were designed for the identified (candidate) SL biosynthesis genes (Table 4) flanking the gRNA target site and the 100 - 150 bp amplicons were amplified using a highly multiplex (HiPlex) PCR reaction, while attaching sample-specific barcodes. Amplicons from all samples were pooled and ligated to Illumina TruSeq sequencing adapters using the KAPA Hyper prep PCR-free ligation kit according to manufacturer directions (Kapa Biosystems, USA). HiPlex amplification reactions and library preparations were performed by Floodlight Genomics LLC (Knoxville, TN, USA). The libraries were sequenced with 150 PE on a HiSeqBOOO instrument (Admera, USA). Forward and reverse reads were merged with PEAR (v0.9.8) (Zhang et al., 2014), sample-specific barcodes were used for sample demultiplexing with custom python scripts and sample-specific barcodes and linker sequences introduced during library preparation were removed. The following steps were performed per sample, and processed in parallel. Reads were sorted per gene by mapping (BWA-MEM with default parameters (Li & Durbin, 2009)) to the reference gene sequences, and the original fastq read files with all HiPlex reads per sample were split into subsets of reads per gene per sample using the readID. The gene-specific amplification primers were removed by trimming the reads with Cutadapt (Martin, 2011) and the remaining sequence window (defined as the entire sequence between the HiPlex primers per gene) was considered as an allele per gene. All unique read sequences, including any potential novel (non- reference) alleles originating from genome editing, were counted per gene per sample. After processing all samples, an integrated table was created listing all read counts per allele per gene per sample across the sample set. Next, the relative allele frequency was calculated as the number of reads per allele per gene per sample divided by the total number of reads per gene per sample. Analyzing plant material, low frequency alleles were removed using a minimal allele frequency threshold of 6 %. This frequency threshold was calibrated based on empirical observations of the distribution of allele frequencies of alternative (non-reference) sequences in wild type (non-mutated) loci, which were assumed to be derived from PCR artefacts, including sequence jumps; low-frequency sequencing errors, such as base calling errors inherent to Illumina short-read sequencing, read mapping errors, etc. 4.Guaianolide SL metabolite profiling of wild type and mutated Cichorium plants

4.1 SL metabolite profiling of mutated plant genotypes containing mutations in multiple paralogous SL candidate genes

Two of the four CiKLS mutated plant genotypes were sufficiently grown for successful SL metabolite profiling. Eight WT plants were analyzed as controls. The production of 14 out of 16 SL metabolites was successfully eliminated in mutated plant genotype M28 (Table 6 & Table 7) compared to the WT plants. This plant genotype contains a mutation in all three paralogous CiKLS gene which were functionally characterized, of which two paralogous genes contain a homozygous knockout mutation. These results indicate that the paralogous CiKLS genes are important SL biosynthesis genes for the production of the guaianolide SL metabolites in planta.

Table 6. SL metabolite composition (ng / mg dry weight (DW)) of the four standard SL metabolites in leaves of WT plants and plants containing a mutation in multiple putative paralogous CiKLS genes. Plant genotypes are described in Table 5. The mean +/- stdev is shown for duplicate measurements (biological replicates).

Table 7. Relative peak area of 12 SL metabolites in leaves of WT plants and plants containing a mutation in multiple putative paralogous CiKLS genes. Plant genotypes are described in Table 5. The mean ± stdev is shown for duplicate measurements. sl08a: deoxylactucin, sl08b: dihydrodeoxylactucin, sl09a: deoxylactucin glycoside, sl09b: dihydrodeoxylactucin glycoside, sllOa: deoxylactucin oxalate, sllOb: dihydrodeoxylactucin oxalate, sll2a: lactucin glycoside, sll2b: dihydrolactucin glycoside, sll3a: lactucin oxalate, si 13b: dihydrolactucin oxalate, sll5a: lactucopicrin oxalate and si 15b: dihydrolactucopicrin oxalate.

Material & Methods to example 4

Plant material

CRISPR/Cas9 genome edited genotypes were acclimatized to the greenhouse, as previously described.

Metabolite extraction and UHPLC-HRMS analysis

For metabolite profiling of the CRISPR/Cas9 genome edited plants and wild type control plants, the inner leaves were freshly harvested after 2 months of acclimatization. Each plant sample was immediately frozen in liquid nitrogen, then grinded in liquid nitrogen, freeze-dried (Gamma 1-20, Christ, ILVO P109, Belgium) and stored under vacuum conditions at -20°C till analysis. Both extraction and separation were based on the method developed by Kips (2017). To 50 mg of freeze-dried material, 1480 μL of extraction buffer (water + 0,1 % FA) and 74 μL of lOng/μL santonin internal standard was added. Samples were shaken for 15 min at 30 °C at a speed of 1400 rpm (Eppendorf thermomixer, Eppendorf AG, Hamburg, Germany) and subsequently centrifuged for 15 min at 20817 g. The supernatant was filtered over a 0.22 pM PVDF-filter (Millipore, Overijse, Belgium) and transferred to a vial for analysis. In order to separate the SL metabolites, ultra-high performance liquid chromatography (UHPLC) by means of an Acquity™ UPLC (Waters, Manchester, UK) was performed. A BEH C18 column (150 mm x 2.1 mm, 1.7 pm) was used for chromatographic separation (Waters). The mobile phase consisted of H ₂ O + 0.1 % formic acid (solvent A) and acetonitrile (ACN) + 0.1 % formic acid (solvent B). The gradient was initiated at 5 % B for 5 min, then linearly increased from 5 % to 53 % B in 20 min, held constant at 53 % for 1 min and finally set at 100 % B for 3 min. Afterwards, the initial conditions of 5 % B were re-equilibrated for 4 min prior to the next injection. The column temperature was set at 40 °C and the flow rate was 0.350 mL min ¹ . The injection volume was 5 μL. Detection of the SLs was performed by means of a Synapt G2-S (Waters) high resolution mass spectrometer (HRMS). The MS detector was operated in positive electrospray (ESI+) mode with a capillary potential of 1.5 kV. Source and desolvation temperatures were 120 °C and 500 °C, respectively. Gas flows were 800 L.h ¹ and 20 L.h ¹ for desolvation and cone gas, respectively. Data were acquired in MS _E mode with the collision energy at 4 eV in the low energy mode to determine the accurate mass, fragmentation spectra were obtained in the high energy mode using a collision energy ramp (8 - 40 eV). All compounds (sl08a: deoxylactucin, sl08b: dihydrodeoxylactucin, sl09a: deoxylactucin glycoside, sl09b: dihydrodeoxylactucin glycoside, sllOa: deoxylactucin oxalate, sllOb: dihydrodeoxylactucin oxalate, sllla: lactucin, slllb: dihydrolactucin, sll2a: lactucin glycoside, sll2b: dihydrolactucin glycoside, sll3a: lactucin oxalate, sll3b: dihydrolactucin oxalate, sll4a: lactucopicrin, sll4b: dihydrolactucopicrin, sll5a: lactucopicrin oxalate and sll5b: dihydrolactucopicrin oxalate) were identified based on the accurate mass and fragmentation pattern and reported as relative peak areas (area compound/area internal standard). Four compounds were quantified with reference standards: lactucin (sllla), dihydrolactucin (slllb), lactucopicrin (sll4a) and dihydrolactucopicrin (si 14b) while no standards were available for the other compounds. Data recording was achieved with MassLynx™ (v.4.1) while the integration was performed with TargetLynx™ (v. 4.1) (Waters). Mean and standard error values were calculated for each plant genotype of the relative peak areas of all 16 guaianolide SL metabolites extracted from the plants containing a mutation in the paralogous CiKLS genes. Significant SL metabolite changes between the wild type plants and the mutated plant genotypes were analyzed using Pairwise Wilcoxon Rank Sum Test (p < 0.05).

Sequence listing

SEQ ID NO: 1: nucleotide sequence of Cichorium intybus CYP71BZ18_1 gene SEQ. ID NO: 2 : amino acid sequence of Cichorium intybus CYP71BZ18_1 gene SEQ ID NO: 3: nucleotide sequence of Cichorium intybus CYP71BZ18_2 gene SEQ ID NO: 4: amino acid sequence of Cichorium intybus CYP71BZ18_2 gene SEQ ID NO: 5: nucleotide sequence of Cichorium intybus CYP71BZ19 gene SEQ ID NO: 6: amino acid sequence of Cichorium intybus CYP71BZ19 protein SEQ ID NO: 7: polynucleotide acid sequence of Cichorium intybus CYP71BZ20_l gene SEQ ID NO: 8: amino acid sequence of Cichorium intybus CYP71BZ20_l gene SEQ ID NO: 9: nucleotide sequence of Cichorium intybus CYP71BZ20_2 gene SEQ ID NO: 10: amino acid sequence of Cichorium intybus CYP71BZ20_2 gene SEQ ID NO: 11: nucleotide sequence of Cichorium intybus CYP71BZ23 gene SEQ ID NO: 12: amino acid sequence of Cichorium intybus CYP71BZ23 gene

SEQ. ID NO: 13: nucleotide sequence of Cichorium intybus CYP71BZ25 gene

SEQ ID NO: 14: amino acid sequence of Cichorium intybus CYP71BZ25 gene

SEQ ID NO: 15: nucleotide sequence of Cichorium intybus CYP71BZ26_1 gene SEQ ID NO: 16: amino acid sequence of Cichorium intybus CYP71BZ26_1 gene

SEQ ID NO: 17: nucleotide sequence amino acid sequence of Cichorium intybus CYP71BZ26_2 gene

SEQ ID NO: 18: amino acid sequence amino acid sequence of Cichorium intybus CYP71BZ26_2 gene

SEQ ID NO: 19: nucleotide sequence of Cichorium intybus CYP71BZ27_1 gene

SEQ ID NO: 20: amino acid sequence of Cichorium intybus CYP71BZ27_1 gene SEQ ID NO: 21: nucleotide sequence of Cichorium intybus CYP71BZ27_2 gene

SEQ ID NO: 22: amino acid sequence of Cichorium intybus CYP71BZ27_2 gene

SEQ ID NO: 23: nucleotide sequence of Cichorium intybus CYP71BZ27_3 gene

SEQ ID NO: 24: amino acid sequence of Cichorium intybus CYP71BZ27_3 gene

SEQ ID NO: 25: first conserved region in CY71BZ25, CYP71BZ26, CY71BZ27 and TpKLS amino acid sequences

SEQ ID NO: 26: second conserved region in CY71BZ25, CYP71BZ26, CY71BZ27 and TpKLS amino acid sequences

Table 9. Primers and PCR efficiency for the used candidate reference genes and CiGASs, CiGASI, CiGAO and CiCOS genes in RT-qPCR experiment.

References

Borowski, J. M., Galli, V., da Silva Messias, R., Perin, E. C., Buss, J. H., dos Anjos e Silva, S. D., & Rombaldi, C. V. (2014). Selection of candidate reference genes for real-time PCR studies in lettuce under abiotic stresses. Planta, 239, 1187-1200. https://doi.org/10.1007/s00425-014-2041-2

Bouwmeester, H. J., Kodde, J., Verstappen, F. W., Altug, I. G., de Kraker, J. W., & Wallaart, T. E. (2002). Isolation and characterization of two germacrene A synthase cDNA clones from chicory. Plant Physiol, 129(1), 134-144. https://doi.org/10.1104/pp.001024

De Bruyn, C., Ruttink, T., Eeckhaut, T., Jacobs, T., De Keyser, E., Goossens, A., & Van Laere, K. (2020). Establishment of CRISPR/Cas9 genome editing in witloof (Cichorium intybus var. foliosum). Frontiers, 2(December). https://doi.org/10.3389/fgeed.2020.604876

De Keyser, E., Desmet, L., Losschaert, M., & De Riek, J. (2020). A General Protocol for Accurate Gene Expression Analysis in Plants. Methods in Molecular Biology, 1015-1118. https://doi.org/10.1007/978-l-4939-9833-3_9

Delporte, M., Legrand, G., Hilbert, J.-L., & Gagneul, D. (2015). Selection and validation of reference genes for quantitative real-time PCR analysis of gene expression in Cichorium intybus. Frontiers in Plant Science, 6(651), 1-11. https://doi.org/10.3389/fpls.2015.00651

Deryckere, D., Eeckhaut, T., Van Huylenbroeck, J., & Van Bockstaele, E. (2012). Low melting point agarose beads as a standard method for plantlet regeneration from protoplasts within the Cichorium genus. Plant Cell Reports, 31(12), 2261-2269. https://doi.org/10.1007/s00299-012-1335-8

Doyle, J. ., & Doyle, J. . (1990). DNA isolation from small amount of plant tissue. Phytochemical Bulletin, 57, 13-15. https://doi.org/10.2307/4119796

Eljounaidi, K., Cankar, K., Comino, C., Moglia, A., Hehn, A., Bourgaud, F., Bouwmeester, H., Menin, B., Lanteri, S., & Beekwilder, J. (2014). Cytochrome P450s from Cynara cardunculus L. CYP71AV9 and CYP71BL5, catalyze distinct hydroxylations in the sesquiterpene lactone biosynthetic pathway. Plant Science, 223, 59-68. https://doi.Org/10.1016/j.plantsci.2014.03.007

Frey, M., Klaiber, L, Conrad, J., & Spring, O. (2020). CYP71BL9, the missing link in costunolide synthesis of sunflower. Phytochemistry, 177(January), 112430. https://doi.Org/10.1016/j.phytochem.2020.112430

Hellemans, J., Mortier, G., De Paepe, A., Speleman, F., & Vandesompele, J. (2007). qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biology, 8(2), 1-19. https://doi.org/10.1186/gb-2007-8-2-rl9 Heller, R. (1953). Recherche sur la nutrition minerale des tissus vegetaux cultives in vitro. Ann. Sci. Nat. Bot. Biol. Veg., 14, 1-123.

Kips, L. (2017). Characterization and processing of horticultural byproducts: a case study of tomato and Belgian endive roots.

Li, H., & Durbin, R. (2009). Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics, 25(14), 1754-1760. https://doi.org/10.1093/bioinformatics/btp324

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., & Durbin, R. (2009). The sequence alignment/map format and SAMtools. Bioinformatics, 25(16), 2078-2079. https://doi.org/10.1093/bioinformatics/btp352

Liu, Q.., Kashkooli, A. B., Manzano, D., Pateraki, I., Kolkman, P., Guallar, V., Vos, R. De, Maurice, C. R., Krol, S. Van Der, & Bouwmeester, H. (2018). Kauniolide synthase, a P450 with unusual hydroxylation and cyclization- elimination activity. Nature Communications, 9(4657), 1-13. https://doi.org/10.1038/s41467-018-06565-8

Liu, Q.., Majdi, M., Cankar, K., Goedbloed, M., Charnikhova, T., Verstappen, F. W. A., de Vos, R. C. H., Beekwilder, J., van der Krol, S., & Bouwmeester, H. J. (2011). Reconstitution of the costunolide biosynthetic pathway in yeast and Nicotiana benthamiana. PLoS ONE, 6(8), 1-12. https://doi.org/10.1371/journal.pone.0023255

Liu, Q.., Manzano, D., Tanic, N., Pesic, M., Bankovic, J., Pateraki, I., Ricard, L., Ferrer, A., de Vos, R., de Krol, S. van, & Bouwmeester, H. (2014). Elucidation and in planta reconstitution of the parthenolide biosynthetic pathway. Metabolic Engineering, 23, 145-153. https://doi.Org/10.1016/j.ymben.2014.03.005

Luypaert, G., Witters, J., Van Huylenbroeck, J., De Clercq, P., De Riek, J., & De Keyser, E. (2017). Induced expression of selected plant defence related genes in pot azalea, Rhododendron simsii hybrid. Euphytica, 213, 213-227. https://doi.org/10.1007/sl0681-017-2010-5

Martin, M. (2011). Cutadapt removes adapter sequences from high-throuhput sequencing reads. EMBnetJournal, 1-3. https://doi.Org/https://doi.org/10.14806/ej.17.l.200

Morel, G., & Wetmore, R. (1951). Fern callus tissue culture. American Journal of Botany, 38, 141-143.

Murashige, T., & Skoog, F. (1962). A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiologia Plantarum, 15, 473-497.

Nguyen, D. T., Go, J. C., Ikezawa, N., Macnevin, G., Kathiresan, M., Spring, O., & Ro, D. (2010). Biochemical conservation and evolution of germacrene A oxidase in Asteraceae. The Journal of Biological Chemistry, 285(22), 16588-16598. https://doi.org/10.1074/jbc.M110.111757

R Core Team. (2021). R: A language and environment for statistical computing. In R Foundation for Statistical Computing,, https://www.r-project.org/

Ramirez, A. M., Saillard, N., Yang, T., Franssen, M. C. R., Bouwmeester, H. J., & Jongsma, M. A. (2013). Biosynthesis of Sesquiterpene Lactones in Pyrethrum (Tanacetum cinerariifolium). PLoS ONE, 8(5), 1-13. https://doi.org/10.1371/journal.pone.0065030

Reyes-Chin-Wo, S., Wang, Z., Yang, X., Kozik, A., Arikit, S., Song, C., Xia, L., Froenicke, L., Lavelle, D. O., Truco, M.-J., Xia, R., Zhu, S., Xu, C., Xu, H., Xu, X., Cox, K., Korf, I., Meyers, B. C., & Michelmore, R. W. (2017). Genome assembly with in vitro proximity ligation data and whole-genome triplication in lettuce. Nature Communications, 8, 1-11. https://doi.org/10.1038/ncommsl4953

Robinson, M. D., McCarthy, D. J., & Smyth, G. K. (2009). edgeR: A Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics, 26(1), 139-140. https://doi.org/10.1093/bioinformatics/btp616

Robinson, M. D., & Oshiack, A. (2010). A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biology, 11, 1-9.

Ruijter, J. M., Ramakers, C., Hoogaars, W. M. H., Karlen, Y., Bakker, O., van den hoff, M. J. B., & Moorman, A. F. M. (2009). Amplification efficiency: Linking baseline and bias in the analysis of quantitative PCR data. Nucleic Acids Research, 37(6), 1-12. https://doi.org/10.1093/nar/gkp045

Sainsbury, F., Thuenemann, E. C., & Lomonossoff, G. P. (2009). pEAQ: versatile expression vectors for easy and quick transient expression of heterologous proteins in plants. Plant Biotechnology Journal, 7, 682-693. https://doi.Org/10.llll/j.1467-7652.2009.00434.x

Schmieder, R., & Edwards, R. (2011). Quality control and preprocessing of metagenomic datasets. Bioinformatics, 27(6), 863-864. https://doi.org/10.1093/bioinformatics/btr026

Sgamma, T., Pape, J., Massiah, A., & Jackson, S. (2016). Selection of reference genes for diurnal and developmental time-course real-time PCR expression analyses in lettuce. Plant Methods, 12(21), 1-9. https://doi.org/10.1186/S13007-016-0121-Y

Sterck, L., Bill iau, K., Abeel, T., Rouze, P., & Van de Peer, Y. (2012). ORCAE: online resource for community annotation of eukaryotes. Nat. Methods, 9(11), 1041.

Van Bel, M., Diels, T., Vancaester, E., Kreft, L., Botzki, A., Van de Peer, Yves Coppens, F., & Vandepoele, K. (2018). PLAZA 4.0: an integrative resource for functional, evolutionary and comparative plant genomics. Nucleic Acids Research, 46, 1190-1196.

Vandesompele, J., De Preter, K., Pattyn, I., Poppe, B., Van Roy, N., De Paepe, A., & Speleman, R. (2002).

Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology, 3(7), 1-12. https://doi.org/10.1186/gb-2002-3-7- research0034

Zhang, J., Kobert, K., Flouri, T., & Stamatakis, A. (2014). PEAR: A fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics, 30(5), 614-620. https://doi.org/10.1093/bioinformatics/btt593