2-oxoglutarate-dependent dioxygenases and methods of use thereof

Abstract

Novel 2-oxoglutarate-dependent dioxygenases, particularly novel 14-beta- and 21 -hydroxylases and methods use thereof. Where the methods may result in the creation of steroidal compositions.

Description

[0001] 2-OXOGLUTARATE-DEPENDENT DIOXYGENASES AND METHODS OF USE THEREOF

[0002] By

[0003] Georg Jander

[0004] Gordon Younkin

[0005] This application claims priority under 35 U.S.C. §119(e) to U.S. Provisional Patent Application No. 63 / 712,838, filed October 28, 2024. The foregoing application is incorporated by reference herein.

[0006] FIELD OF THE INVENTION

[0007] This invention relates to the field of 2-oxoglutarate-dependent dioxygenases. More specifically, the invention provides novel 14P- and 21 -hydroxylases and methods of use thereof.

[0008] BACKGROUND OF THE INVENTION

[0009] Cardenolides are a group of steroidal plant specialized metabolites that have fascinated humans for centuries due to their inhibitory activity against animal Na+,K+-ATPases (Norn, et al. (2004) Dan. Medicinhist. Arbog., 119-132). They have evolved convergently in more than 12 plant lineages, including foxglove (Digitalis,- Plantaginaceae), dogbanes (Apocynaceae), and wallflower Erysimum,' Brassicaceae) (Agrawal, et al. (2012) New Phytol., 194:28-45). In these and other cardenolide-producing plants, cardenolides largely function in defense, serving as feeding and oviposition deterrents against insects and other herbivores (Renwick, et al. (1989) J. Chem. Ecol., 15:2161-2169; Zalucki, et al. (2001) Austral. Ecol., 26:547-555; Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733). While these compounds are toxic to humans in high doses, their biological activity has also made them useful in the treatment of heart conditions (Norn, et al. (2004) Dan. Medicinhist. Arbog., 119-132), and they have been studied more recently for their potential anticancer and antiviral properties (Newman, et al. (2008) Mol. Interv., 8:36-49; Wong, et al. (2018) Sci. Rep., 8:1-17). Despite their ecological and medical interest, the full biosynthetic pathway has not been described in any plant species.

[0010] Cardenolides are derived from steroid metabolism, having a steroid core with an unsaturated 5-member lactone ring on carbon 17, and are often modified via glycosylation, hydroxylation, and acylation (Agrawal, et al. (2012) New Phytol., 194:28-45). Importantly, all cardenolides carry a 14P-hydroxyl group, which is required for their activity against Na+,K+-ATPases (Schonfeld, et al. (1985) Naunyn. Schmiedebergs. Arch. Pharmacol., 329:414-426; Bose, et al. (1988) Br. J. Pharmacol., 93:453-461).

[0011] Cardenolide biosynthesis is thought to proceed through pregnane intermediates that are hydroxylated at carbons 14 and 21, with the 14-hydroxy group in the P configuration (pointing out of the page) (Kreis, W. (2017) Planta Med., 83:962-976). In contrast to earlier steps in the formation of these pregnane intermediates, which have been extensively studied in Digitalis and Erysimum (Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733; Carroll, et al. (2023) Nat. Commun., 14:4042; Klein, et al. (2021) Plant Cell Rep., 40:1631-1646; Leykauf, et al. (2023) Planta Med., 89:833-847; Younkin, G., bioRxiv (2024) doi.org / 10.1101 / 2024.04.10.588904; Klein, J. (2024) Plant& Cell Physiology 65:1500-1514), no candidate enzymes for the 14P- and 21 -hydroxylation have been proposed in any plant species. In animals, pregnane 21 -hydroxylation is an essential step in corticosteroid synthesis that is catalyzed by a cytochrome P450, CYP21B (White, et al. (1986) Proc. Natl. Acad. Sci., 83:5111-5115). Mutations to this gene are responsible for 21 -hydroxylase deficiency and associated adrenal hyperplasia (Amor, et al. (1988) Proc. Natl. Acad. Sci., 85: 1600-1604). However, no analog is known from plants. 14P-hydroxylation of steroids is not frequently reported across the tree of life, with most steroidal compounds being either A14-unsaturated or having a 14a substituent group. As such, no enzymes with steroid 14P-hydroxylase activity have been characterized. A cytochrome P450 capable of catalyzing the 14a-hydroxylation of steroids has been identified from the fungal pathogen Cochliobolus lunatus, although the regiospecificity of this enzyme is low (Chen, et al. (2019) Appl.

[0012] Microbiol. Biotechnol., 103:8363-8374; Suzuki, et al. (1993) Biochim. Biophys. Acta, 1203:215-223), and the stereochemistry is inverted relative to what is required for cardenolide biosynthesis. The predicted substrate for the 14P-hydroxylase, epipregnanolone 6, has a 14a-configured hydrogen atom, meaning that stereochemistry is inverted at that position. It has been suggested that this unusual inversion of stereochemistry at carbon 14 may involve a multi-step process proceeding through A8,14-unsaturated intermediates.

[0013] However, feeding of such intermediates in Digitalis purpurea supports a pathway where 14P-hydroxylation and stereochemical inversion occur in a single step (Deluca, et al. (1987) Zeitschrift fur Naturforsch., 42c:77-78).

[0014] Although there are no characterized plant enzymes with steroid 14P- and 21-hydroxylase activity, plant cytochromes P450 (P450s) and 2-oxoglutarate-dependent dioxygenases (2OGDs) are known to catalyze the oxidation of steroids and triterpenoids at other positions. P450s, which constitute the largest family of metabolic enzymes in plants and represent approximately 1% of plant protein-coding genes, catalyze diverse oxidation reactions in both primary and specialized metabolism (Nelson, et al. (2011) Plant J., 66: 194-211). They have long been known to act on varied triterpenoid substrates including a- and P- amyrin, cucurbitacins, and brassinosteroids (Yasumoto, et al. (2016) FEBS Lett., 590:533- 540; Zhou, et al. (2016) Nat. Plants 2:1-8; Ohnishi, et al. (2009) Phytochemistry 70:1918- 1929). Screens of P450s in the cardenolide-producing plants Digitalis lanala. Calotropis procera, and Erysimum cheiranthoides have identified several P450s that are involved in cardenolide biosynthesis (Carroll, et al. (2023) Nat. Commun., 14:4042; Kunert, et al. (2023) Nat. Plants 9:1607-1617; Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733). While these screens were not exhaustive, none of the P450s identified catalyze the 14P- or 21 -hydroxylation of cardenolide intermediates.

[0015] 2OGDs are the second largest enzyme family in plants and, like P450s, are involved in wide-ranging core and specialized metabolic pathways (Hagel, et al. (2018) Nat. Prod. Rep., 35:721-734). However, only more recently have 2OGDs been recognized for their importance in the oxidation of steroidal metabolites with the discoveries of a 16a-hydroxylase and a 23 -hydroxylase involved in steroidal glycoalkaloid metabolism in Solanum (Hagel, et al. (2018) Nat. Prod. Rep., 35:721-734; Nakayasu, et al. (2017) Plant Physiol., 175:120-133; Nakayasu, et al. (2020) Plant Cell Physiol., 61:21-28).

[0016] SUMMARY OF THE INVENTION

[0017] In accordance with the instant invention, novel 2-oxoglutarate-dependent dioxygenases (2OGDs) are provided. The instant invention provides cardenolide metabolism 5 (CARD5) and cardenolide metabolism 6 (CARD6). In certain embodiments, the CARD5 and CARD6 comprise a peptide or epitope tag. Compositions comprising CARD5 and / or CARD6 and a carrier are also provided. Nucleic acid molecules encoding CARD5 or CARD6 are also encompassed by the instant invention. In certain embodiments, the nucleic acid is contained within a vector, particularly an expression vector, such as a plasmid or viral vector.

[0018] In accordance with another aspect of the instant invention, methods of producing a compound, particularly a steroidal compound, with a 14P-hydroxyl are provided. In certain embodiments, the method comprises contacting a substrate, particularly a steroidal substrate, with CARD5. In certain embodiments, the method is performed in vitro (e.g., in solution with a carrier). In certain embodiments, the method comprises expressing a nucleic acid encoding CARD5 in a cell (e.g., mammalian (e.g., human), insect (e.g., Spodoptera frugiperda (e.g., Sf9, Sf21)), yeast (e.g., Saccharomyces cerevisiae, Pichia pastoris), or bacteria (e.g., E. coli)) and contacting the cell with the substrate. In certain embodiments, the method comprises expressing a nucleic acid encoding CARD5 in a plant or plant cell and contacting the plant or plant cell with the substrate. In certain embodiments, the plant or plant cell does not naturally produce cardenolides or is a species other than Erysimum cheiranthoides . In certain embodiments, the method further comprises expressing a nucleic acid encoding CARD1, CARD2, CARD3, and CARD4. In certain embodiments, the method further comprises isolating the compound with a 14P-hydroxyl. In certain embodiments, the synthesized compound is a cardenolide or cardiac glycoside.

[0019] In accordance with another aspect of the instant invention, methods of producing a compound, particularly a steroidal compound, with a 21 -hydroxyl or 21 -carbonyl are provided. In certain embodiments, the method comprises contacting a substrate, particularly a steroidal substrate (e.g., with a 14P-hydroxyl), with CARD6. In certain embodiments, the method is performed in vitro (e.g., in solution with a carrier). In certain embodiments, the method comprises expressing a nucleic acid encoding CARD6 in a cell (e.g., mammalian (e.g., human), insect (e.g., Spodoptera frugiperda (e.g., Sf9, Sf21 )), yeast (e.g., Saccharomyces cerevisiae, Pichia pastoris), or bacteria (e.g., E. coli)) and contacting the cell with the substrate. In certain embodiments, the method comprises expressing a nucleic acid encoding CARD6 in a plant or plant cell and contacting the plant or plant cell with the substrate. In certain embodiments, the plant or plant cell does not naturally produce cardenolides or is a species other than Erysimum cheiranthoides. In certain embodiments, the method further comprises expressing a nucleic acid encoding CARD1, CARD2, CARD3, and CARD4, and, optionally, CARD5. In certain embodiments, the method further comprises isolating the compound with a 21 -hydroxyl or 21 -carbonyl. In certain embodiments, the synthesized compound comprises a 14P-hydroxyl and a 21 -hydroxyl or 21 -carbonyl. In certain embodiments, the synthesized compound is a cardenolide or cardiac glycoside.

[0020] BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figures 1 A and IB show the identification of candidate 2-oxoglutarate dependent dioxygenases (2OGDs) for involvement in cardenolide biosynthesis. Figure 1 A provides the formation and transformation of pregnane intermediates in cardenolide biosynthesis in Erysimum cheiranthoides. Characterized enzymes are listed and the 14P- and 21-hydroxylation steps are indicated. Carbons 14 and 21 are labeled where relevant. Figure IB provides a screen of gene expression profiles for Erysimum 20GDs. Erche01g020322 and Erche03g034150 are indicated. FC: fold-change.

[0022] Figures 2A-2D provide a phylogenetic analysis of candidate genes. Nucleotide phylogenies inferred from coding sequences of 2-oxoglutarate-dependent dioxygenase (20GD) clades D0XC15 (Fig. 2A) and DOXC20 (Fig. 2C) from Arabidopsis thaliana Col-0 (At) and Erysimum cheiranthoides (Ec). Candidate genes for involvement in cardenolide biosynthesis are underlined and in bold. Select genes from other species: Brassica oleracea (Bo), Camelina sativa (Csa), Eutrema salsugineum (ES), Oryza sativa (Os), and Solarium lycopericum (SI). Scale bars indicate estimated number of substitutions per site, numbers at nodes are bootstrap support. Figures 2B and 2D: Microsynteny plots comparing loci containing candidate 20GDs in E. cheiranthoides (Ec) with syntenic regions in A. thaliana Col-0 (At). Pseudogenes are indicated with a dashed border. Figure 2E provides an aligned nucleotide coding sequences of Erche03g034150 (CARDENOLIDE METABOLISM 5,' CARDS,' SEQ ID NO: 1) from wildtype E. cheiranthoides and mutant lines. Card5-1 (SEQ ID NO: 2) and card5-2 (SEQ ID NO: 3) were generated with CRISPR / Cas9. Consensus sequence (SEQ ID NO: 4) also provided. Figure 2F provides an aligned nucleotide coding sequences of Erche01g020322 (CARDENOLIDE METABOLISM &, CARD6,' SEQ ID NO: 5)) from wildtype E. cheiranthoides and mutant lines. Card6 (SEQ ID NO: 6) was generated with CRISPR / Cas9. Consensus sequence (SEQ ID NO: 7) also provided. MultAlin (multalin.toulouse.inra.fr / multalin / ) was used to produce the alignments. Figure 2G provides nucleotide coding sequences of Erche03g034150 (CARDENOLIDE METABOLISM 5,' CARDS,' SEQ ID NO: 1), card5-l (SEQ ID NO: 2), card5-2 (SEQ ID NO: 3), and consensus sequence (SEQ ID NO: 4) along with the encoded amino acid sequences (ErcheO3gO3415O (CARDENOLIDE METABOLISM 5; CARD5; SEQ ID NO: 15), card5-l (SEQ ID NO: 16), card5-2 (SEQ ID NO: 17), and consensus sequence (SEQ ID NO: 18)). Figure 2H provides nucleotide coding sequences of Erche01g020322 (CARDENOLIDE METABOLISM 6,' CARD6,' SEQ ID NO: 5)), card6 (SEQ ID NO: 6), and consensus sequence (SEQ ID NO: 7) along with the encoded amino acid sequences (Erche01g020322 (CARDENOLIDE METABOLISM 6; CARD6; SEQ ID NO: 19)), card6 (SEQ ID NO: 20), and consensus sequence (SEQ ID NO: 21)).

[0023] Figures 3 A-3I provide the characterization of mutant lines. Normalized peak area for total cardenolides (Fig. 3A), pregnenolone (Fig. 3B), progesterone (Fig. 3C), compound 10 (Fig. 3D), and compound 11 (Fig. 3E) from wildtype Erysimum cheiranthoides (WT) and CRISPR / Cas9 mutants of Erche03g034150 (card5-l and card5-2, cyan) and Erche01g020322 (card6, magenta). N = 3 replicates per line. Letters indicate differences between groups, <0.05, one-way ANOVA with post-hoc Tukey’s HSD on log-transformed data. Error bars are ± s.d. Fig. 3F provides an extracted ion chromatogram (EIC) of

[0024] / ?? 7=299.23749, a fragment common to compounds 10 and 11. Full MS scans for compound 11 (Fig. 3G) and compound 10 (Fig. 3H), with hypothetical structure based on pattern of neutral losses. Hypothetical [M+H]+ and [M-sugar]+ peaks, which were not detected, are shown as grey dashed lines. Fig. 31 provides EIC of / ?? 7=375.2535, a fragment common to all digitoxigenin glycosides, following infiltration of leaves of mutant plants with cardenolide intermediates. Chromatograms are representative samples from N = 3 replicates. Figures 3 J and 3K provide tandem mass spectrometry (MSMS) spectrum for compound 11 from card6 mutant line. MSMS spectrum for compound 11, a glycosylated cardenolide intermediate that accumulates to high levels in the card6 Erysimum cheiranthoides line compared with epipregnanolone (Fig. 3 J) and digitoxigenin (Fig. 3K) standards. For compound 11, [M+H]+ is absent, and [M+Na]+ does not fragment well during MSMS. In order to facilitate comparison to digitoxigenin and epipregnanolone spectra, an MS fragment corresponding to the neutral loss of the glycoside was selected for MSMS analysis: [M-deoxyhexose-H2O+H]+(m / z=317.2481). Similarly, [M-H20+H]+ was selected for both digitoxigenin

[0025] ( / ?? 7=357.2430) and epipregnanolone (m / z=301.2531) standards. Peaks between m / z=50-250 are shown, with peaks shared between mirrored spectra within 0.005 Da.

[0026] Figures 4A-4H show the co-infiltration of CARD5 and CARD6 with cardenolide intermediates in Nicotiana benthamiana. Figures 4A-4H are extracted ion chromatograms (EIC) from representative samples and also provide barplots showing compounds detected following substrate co-infiltration or co-expression with upstream pathway enzymes CARD1, CARD2, CARD3, and CARD4. Figure 4G and 4H provide a diluted methanolic extract from card6 mutants was co-infiltrated with CARD6. Barplots show peak areas from N = 3 replicates per enzyme / substrate combination. Letters indicate differences between groups, <0.05, one-way ANOVA with post-hoc Tukey’s HSD on log-transformed data. Error bars are± s.d. Abbreviations: pregnenolone (preg), progesterone (prog), epipregnanolone (epipreg). Predicted molecular structures are displayed alongside chromatograms.

[0027] Compound 12 was validated by comparison with Cynanchum paniculatum extracts. Figures 4I-4L provide MSMS spectra for compounds 12-14 from substrate feeding upon coinfiltration with CARD5 in Nicotiana benthamiana. These three compounds are hypothesized to be 14P-hydroxy derivatives of pregnenolone (compound 12), progesterone (compound 13), and epipregnanolone (compound 14), differing only in the presence of a 3-oxo- or 3-hydroxy group, and saturation at carbon 5. 14P-hydroxypregnenolone (also called ketocalogenin) has been identified from Cynanchum paniculatum via purification and NMR. Consistent with the hypothesis that compound 12 is 14P-hydroxypregnenolone, compound 12 is also found in C. paniculatum extracts (Figs. 41 and 4J). Compounds 13 and 14 are predicted to differ from compound 12 by the presence or absence of a double bond. Therefore, some MSMS peaks are expected to be shifted by 2-4 Da between these spectra (Figs. 4K and 4L). For compound 12 from both Cynanchum paniculatum and TV. benthamiana (Fig. 4J) and compound 14 (Fig.

[0028] 4L), [M-H20+H]+ was selected for fragmentation because [M+H]+ was not detected, and [M+Na]+ did not fragment well. [M+H]+ was used as the precursor ion for compound 13 (Fig. 4J). Peaks shared between mirrored spectra within 0.005 Da. Figures 4M-4R provide MSMS spectra for compounds 15-21 from substrate feeding of CARD5 + CARD6 in Nicotiana benthamiana. These three compounds are hypothesized to be 14P-hydroxy, 21- oxo derivatives of pregnenolone (compounds 15 and 16), progesterone (compound 18 - no spectrum was collected for compound 17), epipregnanolone (compounds 19 and 20), differing only based on the presence of a 3 -oxo- or 3 -hydroxy group, and saturation at carbon 5. Compound 21 is the oxidation product of compound 11 by CARD6. Many peaks are expected to be shifted by 2-4 Da between these spectra, including the peaks 309 / 327 for compound 18, 311 / 329 for compounds 15 and 16, and 313 / 331 for compounds 19, 20, and 21. For compound 21 (Fig. 4R), the strong peak at 181 Da corresponds to a hexose. [M-NH4]+ was used as the precursor ion for fragmentation of all compounds because [M+H]+ was not detected, and [M+Na]+ did not fragment well.

[0029] Figures 5A-5F show substrate docking and analysis of residues involved in substrate binding. Fig. 5 A: The Arabidopsis AOP3 contains an extra 95 amino acids that are absent in CARD5. Fig. 5B: The AtAOP3 active site with Fe(II) and 2-oxoglutarate (2OG) bound. Glucoiberin was docked into the active site, and residues that coordinate the metal binding site and that comprise the active site but are variable between AtAOP3 and CARD5 are shown as sticks. Fig. 5C: Pregnenalone was docked into the CARD5 active site, and sticks are side chains are shown for the active site amino acids that are variable with AtAOP3. Figs. 5D and 5E: The Arabidopsis and Erysimum DA01 active sites with indole-3 -acetic acid (IAA), Mg, and 2OG have identical active site residues. Fig. 5F: The CARD6 active site with Fe(II), 2OG, and a diastereomer of 11 reveals the residues that are conserved and variable in comparison to EcDAOl.

[0030] Figure 6 provides the proposed catalytic mechanism for inversion of stereochemistry at carbon 14. The 2-oxoglutarate dependent dioxygenase active site, with iron bound to three amino acid residues and 2-oxoglutarate is displayed above epipregnanolone 6. After the first panel, only the portion of the substrate involved in the reaction is displayed. In this proposed reaction mechanism, Hisp is abstracted by the activated iron-oxo group, and Hua undergoes a 1,2-hydride shift to form a more stable tertiary radical intermediate.

[0031] DETAILED DESCRIPTION OF THE INVENTION

[0032] Cardenolide biosynthesis evolved convergently in many plant lineages, including wallflower (Erysimum^ Brassicaceae). The first steps of the biosynthetic pathway involving conversion of sterols to pregnane derivatives have been characterized in Erysimum and other plants, but several key enzymes remain elusive. Here, E. cheiranthoides 2OGDs were screened and the 14P- and 21 -hydroxylases that are required for cardenolide biosynthesis were identified. Erche03g034150 is designated as CARDENOLIDE METABOLISM 5 (CARD5), which is related to ALKENYL HYDROXYALKYL PRODUCING 1 (AOPI), and Erche01g023022 is designated as CARDENOLIDE METABOLISM 6 (CARD6), which is a duplicate of DIOXYGENASE FOR A UXIN OXIDA TION 1 (DAO I). Knockout mutants of both genes are deficient in cardenolide biosynthesis, instead accumulating pathway intermediates. These enzymes are active on cardenolide intermediates when expressed in Nicotiana benlhamiana. with CARD5 as the 14P-hydroxylase and CARD6 as the 21-oxygenase. Enzyme modeling and substrate docking identify key residues allowing shifts to substrate recognition during the neofunctionalization.

[0033] In accordance with the instant invention, 2-oxoglutarate-dependent dioxygenases (2OGDs) are provided. In certain embodiments, the 20GD is a hydroxylase. In certain embodiments, the 20DG is a 14P-hydroxylase (CARD5). In certain embodiments, the 20DG is a 21 -hydroxylase or 21 -oxygenase (CARD6). In certain embodiments, the 20GD is isolated. In certain embodiments, the 20GD comprises a polypeptide or epitope tag (e.g., His-tag (e.g., 6-8 histidines), FLAG tag (DYKDDDDK; SEQ ID NO: 8), glutathione-S-transferase (GST), green fluorescent protein (GFP), Myc tag (EQKLISEEDL; SEQ ID NO: 9), hemagglutinin tag (YPYDVPDYA; SEQ ID NO: 10), NE-tag (TKENPRSNQEESYDDNES; SEQ ID NO: 11), AviTag™ (GLNDIFEAQKIEWHE; SEQ ID NO: 12), Spot-Tag® (PDRVRAVSHWSS; SEQ ID NO: 13), orE-Tag (GAPVPYPDPLEPR; SEQ ID NO: 14)). The tag may be at the N-terminus and / or C-terminus of the protein.

[0034] The present invention provides a novel 14P-hydroxylase, Cardenolide Metabolism 5 (CARD5). In certain embodiments, CARD5 comprises an amino acid sequence having at least 75%, 80%, 85%, 90%, 95%, 97%, 99%, or 100% homology (identity) with the amino acid sequence encoded by SEQ ID NO: 1 or with amino acid sequence SEQ ID NO: 15. In certain embodiments, the CARD5 comprises SEQ ID NO: 15. In certain embodiments, the CARD5 is isolated. In certain embodiments, the CARD5 comprises a polypeptide or epitope tag (e.g., at the N-terminus and / or C-terminus).

[0035] The present invention provides a novel 21 -hydroxylase or 21 -oxygenase, Cardenolide Metabolism 6 (CARD6). In certain embodiments, CARD6 has 21 -hydroxylase activity. In certain embodiments, CARD6 has 21 -oxygenase activity. In certain embodiments, CARD6 comprises an amino acid sequence having at least 75%, 80%, 85%, 90%, 95%, 97%, 99%, or 100% homology (identity) with the amino acid sequence encoded by SEQ ID NO: 5 or with amino acid sequence SEQ ID NO: 19. In certain embodiments, the CARD6 comprises SEQ ID NO: 19. In certain embodiments, the CARD6 is isolated. In certain embodiments, the CARD6 comprises a polypeptide or epitope tag (e.g., at the N-terminus and / or C-terminus).

[0036] The 2OGDs of the instant invention may also be post-translationally modified. The 2OGDs may be post-translationally modified in a cell or in vitro.

[0037] Nucleic acid molecules encoding a 2OGD are also encompassed by the instant invention. Nucleic acid molecules encoding the 2OGD may be prepared by any method known in the art. In certain embodiments, the nucleic acid molecule is isolated. The nucleic acid molecules may be maintained in any convenient vector, particularly a plasmid or an expression vector. In certain embodiments, the 2OGD is CARD5. In certain embodiments, the 2OGD is CARD6. In certain embodiments, the nucleic acid molecule comprises a sequence having at least 75%, 80%, 85%, 90%, 95%, 97%, 99%, or 100% homology (identity) with SEQ ID NO: 1. In certain embodiments, the nucleic acid molecule comprises a sequence having at least 75%, 80%, 85%, 90%, 95%, 97%, 99%, or 100% homology (identity) with SEQ ID NO: 5.

[0038] Nucleic acid molecules encoding the 2OGD of the invention may be prepared, for example, by using recombinant DNA technology methods. The availability of nucleotide sequence information enables preparation of isolated nucleic acid molecules of the invention by a variety of means. For example, nucleic acid sequences encoding a 2OGD of the instant invention may be isolated from appropriate biological sources using standard protocols well known in the art.

[0039] Nucleic acid molecules of the present invention may be RNA or DNA (e.g., genomic DNA or cDNA) and may be single-stranded or double-stranded. In certain embodiments, the nucleic acid molecules are maintained in any convenient cloning vector. In certain embodiments, the nucleic acid molecules are maintained in a vector such as a plasmid or expression vector. The vectors may be suitable for propagation in E. coli host cell. In certain embodiments, the nucleic acids may be maintained in a vector suitable for expression in mammalian cells or plant cells. Vectors may comprise the regulatory elements necessary for expression of the DNA in the host cell positioned in such a manner as to permit expression of the DNA in the host cell. Such regulatory elements required for expression include, but are not limited to, promoter sequences, transcription initiation sequences, and enhancer sequences. In certain embodiments, the vector is a viral vector.

[0040] The 20GD of the present invention may be prepared in a variety of ways, according to known methods. The protein may be purified from appropriate sources (e.g., transformed plant cells or tissues which express the 20GD). The availability of nucleic acid molecules encoding the proteins enables production of the 20GD using in vitro expression methods known in the art. In certain embodiments, tagged fusion proteins comprising the 20GD can be generated (e.g., to facilitate purification and / or identification). Such tagged fusion proteins are encoded by part or all of a DNA molecule, ligated in the correct codon reading frame to a nucleotide sequence encoding a portion or all of a desired polypeptide or epitope tag (e g., His-tag (e.g., 6-8 histidines), FLAG tag (DYKDDDDK; SEQ ID NO: 8), glutathione-S-transferase (GST), green fluorescent protein (GFP), Myc tag (EQKLISEEDL; SEQ ID NO: 9), hemagglutinin tag (YPYDVPDYA; SEQ ID NO: 10), NE-tag (TKENPRSNQEESYDDNES; SEQ ID NO: 11), AviTag™ (GLNDIFEAQKIEWHE; SEQ ID NO: 12), Spot-Tag® (PDRVRAVSHWSS; SEQ ID NO: 13), orE-Tag (GAPVPYPDPLEPR; SEQ ID NO: 14)) which is inserted into a plasmid vector adapted for expression in a desired cell (e.g., plant cell). The tag may be at the N-terminus and / or C-terminus of the protein.

[0041] Compositions comprising at least one 2OGD and at least one carrier are also encompassed by the instant invention. In certain embodiments, the carrier is a pharmaceutically acceptable carrier. In certain embodiments, the 2OGD is isolated and / or substantially pure within the composition. In certain embodiments, the composition comprises CARD5 and CARD6. In certain embodiments, the composition further comprises CADR1, CARD2, CARD3, and / or CARD4.

[0042] Compositions comprising at least one 2OGD nucleic acid molecule and at least one carrier are also encompassed by the instant invention. In certain embodiments, the carrier is a pharmaceutically acceptable carrier. In certain embodiments, the 2OGD nucleic acid molecule is isolated and / or substantially pure within the composition. In certain embodiments, the composition comprises a CARD5 nucleic acid molecule and a CARD6 nucleic acid molecule. In certain embodiments, the composition further comprises nucleic acid molecules encoding CADR1, CARD2, CARD3, and / or CARD4 (e.g., encoding each of CADR1, CARD2, CARD3, and CARD4).

[0043] In accordance with the instant invention, methods of producing a compound with a 14P-hydroxyl are provided. In certain embodiments, the method comprises contacting a substrate with CARD5, thereby producing the compound with a 14P-hydroxyl. In certain embodiments, the CARD5 is isolated. In certain embodiments, the method is performed in vitro. In certain embodiments, the CARD5 is contained in a plant or plant cell extract (e.g., the extract of a plant or plant cell expressing CARD5 (e.g., transiently)). In certain embodiments, the method is performed in a cell, particularly a plant cell. In certain embodiments, the method comprises expressing (e.g., transiently) a nucleic acid encoding CARD5 in a cell such as a plant or plant cell and contacting the cell (e.g., plant or plant cell) with a substrate. In certain embodiments, the plant or plant cells are transformed using an Agrobacterium mediated transient plant transformation method. In certain embodiments, the cell (e.g., plant or plant cell) does not naturally produce cardenolides. In certain embodiments, the method further comprises isolating the synthesized / produced compound with a 14P-hydroxyl. In certain embodiments, the substrate is a steroid or steroidal compound. In certain embodiments, the substrate has a core structure comprising gonane (cyclopentanoperhydrophenanthrene). In certain embodiments, the substrate comprises the

[0044] structure:

[0045]

[0046] (optionally substituted and optionally saturated or unsaturated). In certain embodiments, the substrate comprises a hydrogen at carbon 14 of the steroid core structure. In certain embodiments, the substrate is selected from the group consisting of pregnenolone, progesterone, and epipregnanolone.

[0047] In accordance with the instant invention, methods of producing a compound with a 21 -hydroxyl or 21 -carbonyl are provided. In certain embodiments, the method comprises contacting a substrate with CARD6, thereby producing the compound with a 21 -hydroxyl or 21-carbonyl. In certain embodiments, the CARD6 is isolated. In certain embodiments, the method is performed in vitro. In certain embodiments, the CARD6 is contained in a plant or plant cell extract (e.g., the extract of a plant cell expressing CARD6 (e.g., transiently)). In certain embodiments, the method is performed in a cell, particularly a plant cell. In certain embodiments, the method comprises expressing (e.g., transiently) a nucleic acid encoding CARD6 in a cell (e.g., a plant or plant cell) and contacting the cell (e.g., plant or plant cell) with a substrate. In certain embodiments, the plant or plant cells are transformed using an Agrobacterium mediated transient plant transformation method. In certain embodiments, the cells (e.g., plant or plant cell) does not naturally produce cardenolides. In certain embodiments, the method further comprises isolating the synthesized / produced compound with a 21 -hydroxyl or 21 -carbonyl. In certain embodiments, the substrate is a steroid or steroidal compound. In certain embodiments, the substrate has a core structure comprising gonane (cyclopentanoperhydrophenanthrene). In certain embodiments, the substrate

[0048] comprises the structure:

[0049]

[0050] (optionally substituted and optionally saturated or unsaturated). In certain embodiments, the substrate comprises a hydroxyl at carbon 14 of the steroid core structure (e.g., 14P-hydroxyl).

[0051] In accordance with the instant invention, methods of producing a compound with a 14P-hydroxyl and a 21 -hydroxyl or 21 -carbonyl are provided. In certain embodiments, the method comprises contacting a substrate with CARD5 and CARD6, thereby producing the compound with a 14P-hydroxyl and a 21 -hydroxyl or 21 -carbonyl. In certain embodiments, the substrate is contacted with CARD5 and CARD6 at the same time (simultaneously). In certain embodiments, the substrate is first contacted with CARD5 and then contacted with CARD6. In certain embodiments, the CARD5 and / or CARD6 is isolated. In certain embodiments, the method is performed in vitro. In certain embodiments, the CARD5 and / or CARD6 is contained in a plant cell extract (e.g., the extract of a plant cell expressing CARD5 and / or CARD6 (e.g., transiently)). In certain embodiments, the method is performed in a cell, particularly a plant cell. In certain embodiments, the method comprises expressing (e.g., transiently) a nucleic acid encoding CARD5 and / or CARD6 in a plant or plant cell and contacting the plant or plant cell with a substrate. In certain embodiments, the plant or plant cells are transformed using an Agrobacterium mediated transient plant transformation method. In certain embodiments, the plant or plant cell does not naturally produce cardenolides. In certain embodiments, the method further comprises isolating the synthesized / produced compound with a 14P-hydroxyl and a 21 -hydroxyl or 21 -carbonyl. In certain embodiments, the substrate is a steroid or steroidal compound. In certain embodiments, the substrate has a core structure comprising gonane (cyclopentanoperhydrophenanthrene). In certain embodiments, the substrate comprises the structure:

[0052]

[0053] (optionally substituted and optionally saturated or unsaturated). In certain embodiments, the substrate comprises a hydrogen at carbon 14 of the steroid core structure. In certain embodiments, the substrate is selected from the group consisting of pregnenolone, progesterone, and epipregnanolone.

[0054] In accordance with the instant invention, methods of producing a compound with a 14P-hydroxyl and / or a 21-hydroxyl or 21-carbonyl are provided. In certain embodiments, the method comprises contacting a substrate with CARD1, CARD2, CARD3, CARD4, CARD5 and, optionally, CARD6, thereby producing the compound with a 14P-hydroxyl and / or a 21-hydroxyl or 21-carbonyl. In certain embodiments, the substrate is contacted with CARD1, CARD2, CARD3, CARD4, CARD5 and, optionally, CARD6 at the same time (simultaneously). In certain embodiments, the substrate is first contacted with CARD1, CARD2, CARD3, CARD4, and CARD5 and then contacted with CARD6. In certain embodiments, the CARD1, CARD2, CARD3, CARD4, CARD5 and / or CARD6 is isolated. In certain embodiments, the CARD1, CARD2, CARD3, CARD4, CARD5 and / or CARD6 is contained in a cell extract such as a plant cell extract (e.g., the extract of a plant cell expressing CARD1, CARD2, CARD3, CARD4, CARD5 and / or CARD6 (e g., transiently)). In certain embodiments, the method is performed in vitro. In certain embodiments, the method is performed in a cell, particularly a plant cell. In certain embodiments, the method comprises expressing (e.g., transiently) a nucleic acid encoding CARD1, CARD2, CARD3, CARD4, CARD5 and / or CARD6 in a plant or cell (e.g., plant cell) and contacting the plant or cell (e.g., plant cell) with a substrate. In certain embodiments, the plant or plant cells are transformed using an Agrobacterium mediated transient plant transformation method. In certain embodiments, the plant or cell (e.g., plant cell) does not naturally produce cardenolides. In certain embodiments, the method further comprises isolating the synthesized / produced compound. In certain embodiments, the substrate is a steroid or steroidal compound. In certain embodiments, the substrate has a core structure comprising gonane (cyclopentanoperhydrophenanthrene). In certain embodiments, the substrate

[0055] comprises the structure:

[0056]

[0057] (optionally substituted and optionally saturated or unsaturated). In certain embodiments, the substrate comprises a hydrogen at carbon 14 of the steroid core structure. In certain embodiments, the substrate is selected from the group consisting of pregnenolone, progesterone, and epipregnanolone.

[0058] The compounds synthesized or produced by the methods described herein are also encompassed by the instant invention. In certain embodiments, the compound is a steroid or steroidal compound. In certain embodiments, the compound is a 14P-hydroxyl steroid or steroidal compound. In certain embodiments, the compound is a 21-hydroxyl or 21-carbonyl steroid or steroidal compound. In certain embodiments, the compound is a steroid or steroidal comprising a 14P-hydroxyl and / or 21-hydroxyl or 21-carbonyl. In certain embodiments, the compound is a cardenolide. In certain embodiments, the compound is a

[0059]

[0060] Compositions comprising at least one compound synthesized or produced by the methods described herein and at least one carrier are also encompassed by the instant invention. In a particular embodiment, the compound is isolated and / or substantially pure within the composition. In certain embodiments, the carrier is a pharmaceutically acceptable carrier.

[0061] Definitions

[0062] The following definitions are provided to facilitate an understanding of the present invention.

[0063] The singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. With reference to nucleic acids of the invention, the term “isolated nucleic acid” is sometimes used. This term, when applied to DNA, refers to a DNA molecule that is separated from sequences with which it is immediately contiguous (in the 5' and 3' directions) in the naturally occurring genome of the organism from which it originates. For example, the “isolated nucleic acid” may comprise a DNA or cDNA molecule inserted into a vector, such as a plasmid or virus vector, or integrated into the DNA of a prokaryote or eukaryote. With respect to RNA molecules of the invention, the term “isolated nucleic acid” primarily refers to an RNA molecule encoded by an isolated DNA molecule as defined above. Alternatively, the term may refer to an RNA molecule that has been sufficiently separated from RNA molecules with which it would be associated in its natural state (i.e., in cells or tissues), such that it exists in a “substantially pure” form.

[0064] With respect to protein, the term “isolated protein” is sometimes used herein. This term may refer to a protein produced by expression of an isolated nucleic acid molecule of the invention. Alternatively, this term may refer to a protein which has been sufficiently separated from other proteins with which it would naturally be associated (e.g., so as to exist in “substantially pure” form). “Isolated” is not meant to exclude artificial or synthetic mixtures with other compounds or materials, or the presence of impurities that do not interfere with the fundamental activity, and that may be present, for example, due to incomplete purification, or the addition of stabilizers.

[0065] The term “substantially pure” refers to a preparation comprising at least 50-60% by weight of a given material (e.g., small molecule, nucleic acid, oligonucleotide, protein, etc.). More preferably, the preparation comprises at least 75% by weight, and most preferably 90-95% by weight of the given compound. Purity is measured by methods appropriate for the given compound (e.g. chromatographic methods, agarose or polyacrylamide gel electrophoresis, HPLC-MS analysis, and the like).

[0066] A “carrier” refers to, for example, a diluent, adjuvant, preservative (e.g., Thimersol, benzyl alcohol), anti-oxidant (e.g., ascorbic acid, sodium metabisulfite), solubilizer (e.g., polysorbate 80), emulsifier, buffer (e.g., Tris HC1, acetate, phosphate), antimicrobial, bulking substance (e.g., lactose, mannitol), excipient, auxiliary agent or vehicle with which an active agent of the present invention is administered. Carriers can be sterile liquids, such as water and oils, including those of petroleum, animal, vegetable or synthetic origin. Water or aqueous saline solutions and aqueous dextrose and glycerol solutions can be employed as carriers. “Pharmaceutically acceptable” indicates approval by a regulatory agency of the Federal or a state government or listed in the U.S. Pharmacopeia or other generally recognized pharmacopeia for use in animals, and more particularly in humans.

[0067] As used herein, the term “small molecule” refers to a substance or compound that has a relatively low molecular weight (e.g., less than 4,000, less than 2,000, particularly less than 1 kDa or 800 Da). Typically, small molecules are organic.

[0068] The term “vector” refers to a carrier nucleic acid molecule (e.g., RNA or DNA) into which a nucleic acid sequence can be inserted for introduction into a host cell where it will be replicated. An “expression vector” is a specialized vector that contains a gene or nucleic acid sequence with the necessary regulatory regions (e.g., promoter) needed for expression in a host cell.

[0069] The term “operably linked” means that the regulatory sequences necessary for expression of a coding sequence are placed in the DNA molecule in the appropriate positions relative to the coding sequence so as to effect expression of the coding sequence. This same definition is sometimes applied to the arrangement of coding sequences and transcription control elements (e.g. promoters, enhancers, and termination elements) in an expression vector. This definition is also sometimes applied to the arrangement of nucleic acid sequences of a first and a second nucleic acid molecule wherein a hybrid nucleic acid molecule is generated.

[0070] The following example is provided to illustrate certain embodiments of the invention. It is not intended to limit the invention in any way.

[0071] EXAMPLE MATERIALS AND METHODS

[0072] Plants and growth conditions

[0073] A genome-sequenced isolate of E. cheiranthoides (Ziist, et al. (2020) eLife 9:1-42), Arabidopsis Biological Resource Center (abrc.osu.edu) accession number CS29250, was used for all experiments. Plants were grown in Cornell Mix (by weight 56% peat moss, 35% vermiculite, 4% lime, 4% Osmocote slow release fertilizer [Scotts, Marysville, OH], and 1% Unimix [Scotts]) in Conviron (Winnipeg, CA) growth chambers with a constant temperature of 23°C, and 180 pM m'1photosynthetic photon flux density, and a 16:8 hour daymight cycle. RNA-sequencing, plant growth, cloning, and knockout of candidate genes

[0074] Raw RNA-sequencing reads from 48 Erysimum species (PRJNA563696; Ziist, et al. (2020) Elife 9:1-42) and A. cheiranthoides tissues (PRJNA1015726; Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733) were downloaded from the NCBI Short Read Archive. Read counts were quantified and analyzed as described (Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733), except that the E. cheiranthoides genome annotation v2.1 was used (Mirzaei, et al. (2024) Plants 13:466). Plant growth, cloning of candidate genes, and CRISPR / Cas9 knockout in Erysimum cheiranthoides were performed as described (Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733). Primers used for cloning of candidate genes and generation and screening of CRISPR / Cas9 mutants are provided in Table 1.

[0075]

[0076]

[0077] able 1: Primer and oligo sequences.

[0078] Substrate feeding and transient expression

[0079] Genes were transiently expressed in leaves of 4-week-old Nicotiana benthamiana plants. Agrobacterium tumefaciens growth and infiltration were as described (Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733). For pathway reconstruction, cardenolide biosynthetic enzymes CARDf CARD2, CARD3, and CARD4 were expressed together and with either CARD5 only or both CARD5 and CARD6. For substrate feeding experiments, CARD5 and CARD6 were infiltrated individually and as a pair, and separate plants expressing GFP were used as a negative control. A 200 pM solution of pregnenolone (Sigma- Aldrich, St. Louis, MO), progesterone (Sigma-Aldrich), or epipregnanolone (Cayman Chemical, Ann Arbor, MI) in 10 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer and 10 mM MgCh were infiltrated into leaves expressing candidate genes three days after A. tumefaciens infiltration, with three replicates for each gene / substrate combination and buffer-only controls. For CARD6, a 50% methanol extract of card6 knockout lines, diluted lOx into MES buffer, was infiltrated to test whether the intermediates accumulating in the mutant could serve as substrates. Tissue was collected four days after A. tumefaciens infiltration for LC-MS analysis.

[0080] E. cheiranthoides mutant lines were supplemented with intermediates in cardenolide biosynthesis to test for restoration of the pathway. A 200 pM solution of progesterone or 21-hydroxyprogesterone (Sigma-Aldrich) in MES buffer (see above), was infiltrated into the abaxial surface of young leaves of three-week old plants of card5-l and card6 mutant lines, with three replicates per substrate / mutant combination. Tissue was collected after 24 hours forLC-MS analysis.

[0081] Liquid chromatography-mass spectrometry (LC-MS) analysis

[0082] Methanolic extracts of E. cheiranthoides and A benthamiana leaves (Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733) were analyzed on an UltiMate™ 3000 UHPLC system coupled to a Q-Exactive™ hybrid quadrupole-orbitrap mass spectrometer (ThermoFisher Scientific). The instrument was fitted with a Supelco Titan™ C18 UHPLC Column (80A, 100 x 2.1 mm, particle size 1.9 pm; Sigma Aldrich). Injections of 2 pL were separated by a solvent gradient consisting of mobile phase A (water + 0.1% (v / v) formic acid) and mobile phase B (acetonitrile + 0.1% (v / v) formic acid): 0-0.5 minutes, hold at 2% B; 0.5- 10 minutes, linear gradient from 2%-97% B; 10-11.5 minutes, hold at 97% B, 11.5-13 minutes, hold at 2% B. All solvents were Optima LC / MS grade (ThermoFisher Scientific). The solvent flow rate was 0.5 mL / minute, the column oven was set to 40 °C, and the autosampler temperature was 15°C. The mass spectrometer was run in full scan positive ionization mode. Targeted MSMS spectra were collected with an isolation window of 2.0 m / z and normalized collision energy of 30%.

[0083] Untargeted analysis of LC-MS data was conducted using XCMS (Benton, et al.

[0084] (2010) Bioinformatics 26:2488-2489; Smith, et al. (2006) Anal. Chem., 78:779-787;

[0085] Tautenhahn, et al. (2008) BMC Bioinformatics 9:504) in R statistical software (R Core Team. R: A Language and Environment for Statistical Computing. (2020)) with the following functions and parameters. CentWave: snthresh 3, peakwidth c(2,7), ppm 2.5, noise 10000, prefilter c(3, 10000); MergeNeighboringPeaks: expandRt2; Obiwarp: binSize 0.01;

[0086] PeakDensity: minFraction 0.5, bw 2, binSize 0.01, and FillChromPeaks with default parameters. Once chromatographic peaks of interest were identified, peak areas were requantified in a targeted manner using a custom processing method in Xcalibur™ Software (ThermoFisher Scientific) using the following parameters: peak detection ICIS, smoothing points 1, baseline window 40, area noise factor 5, peak noise factor 15, tailing factor 2. Mass features for compounds 10-11 found in E. cheiranthoides mutants and compounds 12-21 from N. benthamiana assays are provided in Table 2. Authentic standards were not available for the structures of compounds 12-21, but a 70% methanolic extract of Cynanchum paniculatum (Paniculate Swallowwort Root; KHT Herbs & Goods, Sacramento, CA) was used to validate identification of compound 12 as 14P-hydroxypregn enol one.

[0087]

[0088] able 2: Masses and retention times for putative cardenolide intermediates.

[0089] Statistical and phylogenetic analysis

[0090] For statistical comparisons between compound abundances in mutant lines and transient expression assays, the aov and TukeyHSD functions in R statistical software were used for one-way ANOVA and post-hoc Tukey’s HSD tests. Log-transformed peak areas that had been normalized to the internal standard were used for analysis. Plots were made using MSnbase and multcomp View (Gatto, et al. (2012) Bioinformatics 28:288-289; Gatto, et al. (2021) J. Proteome Res., 20:1063-1069; Graves, et al. (2023) multcomp View: Visualizations of Paired Comparisons).

[0091] Sequences homologous to CARD5 and CARD6 were identified using BLAST against publicly available transcriptomes for Arabidopsis thaliana (Berardini, et al. (2015) Genesis 53:474-485) and Erysimum cheiranthoides (NCBI PRJNA563696; Ziist, et al. (2020) Elife 9: 1-42). For CARD5, a functionally characterized 2OGD from Solanum lycopersicum, S123DOX, was also included (Nakayasu, et al. (2020) Plant Cell Physiol., 61:21-28). In the case of CARD6. orthologs were retrieved from Oryza sativa (Kawahara, et al. (2013) Rice 6:4), Brassica oleracea. Camelina sativa, and Eutrema salsugineum (Tello-Ruiz, et al. (2022) Gramene: A Resource for Comparative Analysis of Plants Genomes and Pathways, in Plant Bioinformatics (ed. Edwards, D.) pages 101-131 (Springer, US)). Outgroups were selected from sister 2OGD clades based on a comprehensive phylogenetic analysis of plant 2OGDs (Kawai, et al. (2014) Plant J., 78:328-343). Nucleotide sequences were aligned using ClustalW (Sievers, et al. (2011) Mol. Syst. Biol., 7:539; Madeira, et al. (2022) Nucleic Acids Res., 50:W276-W279). Gene phylogenies were inferred using the IQ-TREE web server (Minh, et al. (2020) Mol. Biol. Evol., 37:1530-1534; Trifinopoulos, et al. (2016) Nucleic Acids Res., 44:W232-W235; Hoang, et al. (2018) Mol. Biol. Evol., 35:518-522) with default parameters, except bootstrap alignments were increased to 10,000.

[0092] Enzyme modeling and substrate docking

[0093] Three-dimensional structural models of each enzyme were obtained from AlphaFold (Jumper, et al. (2021) Nature 596:583-589) (AtDAOl : Q9XI75) or were generated using the amino acid sequence of the protein using ColabFold vl.5.5 (Mirdita, et al. (2022) Nat.

[0094] Methods 19:679-682). Metal ions and alpha-ketoglutarate were added to each protein using AlphaFill (Hekkelman, et al. (2023). Nat. Methods 20:205-213). Substrates (indole-3 -acetic acid, a diastereomer of 11, glucoiberin, and pregnenalone) were obtained from Zinc20 (zinc.docking.org), and were docked into enzyme active sites using AutoDock Vina (ver. 1.1.2) with grid box dimensions of 40 x 40 x 40 A and an exhaustiveness of 8 (Trott, et al. (2010) J. Comput. Chem., 31:455-461; Forli, et al. (2016) Nat. Protoc., 11:905-919).

[0095] Docking results were visualized in PyMOL (ver. 2.5.7) (pymol.org / 2 / ).

[0096] RESULTS

[0097] Identification of candidate 2OGDs

[0098] The expression patterns of Erysimum 2OGDs was examined to identify candidates for involvement in cardenolide biosynthesis. Because cardenolides are synthesized in E. cheiranthoides leaves and transported to the roots (Alani, et al. (2021) Phytochemistry 192: 112965), cardenolide biosynthetic genes were expected to be highly expressed in leaves compared to roots. Furthermore, cardenolide biosynthesis has been lost in Erysimum collinum (Ziist, et al. (2020) Elife 9:1-42). As such, low expression in A. collinum was expected compared to all other species of Erysimum. As for P450s (Younkin, et al. (2024) New Phytol. (2024) 242(6):2719-2733), 2OGDs were searched for following these expression patterns and two genes were identified, Erche01g020322 and Erche03g034150, as top candidates (Fig. IB). Notably, Erche03g034150 clustered with other cardenolide biosynthesis genes in a coexpression analysis across 48 Erysimum species (Younkin, G., bioRxiv (2024) doi.org / 10.1101 / 2024.04.10.588904). Phylogenetic analysis of candidate genes

[0099] To better understand the phylogenetic context of the candidate 20GDs and the activity of related enzymes, phylogenetic trees for each gene were inferred and microsynteny between the genomic regions of E. cheiranthoides and Arabidopsis thaliana containing these genes was examined. Erche01g020322 is a tandem duplicate of DIOXYGENASE FOR AUXIN OXIDATION 1 (EcDAOl) (Figs. 2A-2B), placing it within 20GD clade D0XC15 (Kawai, et al. (2014) Plant J., 78:328-343). AtDAOl and the closely related AtDAO2 have a critical role in auxin homeostasis through its oxidative deactivation (Porco, et al. (2016) Proc. Natl. Acad. Sci., 113:11016-11021; Zhang, et al. (2016) Proc. Natl. Acad. Sci., 113:11010- 11015; Zhao, et al. (2013) Dev. Cell 27: 113-122). In E. cheiranthoides , EcDAOl has been duplicated an additional time relative to A. thaliana (Figs. 2A-2B). This duplicate copy, Erche01g020322, sits on a long branch in the DAO phylogenetic tree, suggestive of rapid evolution of the gene sequence, with potential for neofunctionalization of the encoded protein.

[0100] Erche03g034150 belongs to the DOXC20 group of plant 2OGDs (Sievers, et al.

[0101] (2011) Mol. Syst. Biol. 7:539; Kawa, et al. (2014) Plant J., 78:328-343), and is nested within the AOP (akenyl hydroxyalkyl producing) clade. In A. thaliana, AtAOP2 and AtAOP3 have been implicated in the modification of aliphatic glucosinolates (Kliebenstein, et al. (2001) Plant Cell 13:68-693), with structural variation of the locus explaining variation in glucosinolate profiles among Arabidopsis accessions (Chan, et al. (2010) Genetics 185:991- 1007). AtAOPl, which is the most closely related A. thaliana gene to Erche03g034150 (Fig.

[0102] 2C), does not have a known function. Compared to the single copy of AOP1 in A. thaliana, which is contained within the GS-AOP locus (Kliebenstein, et al. (2001) Plant Cell 13:68-693), there are five copies in E. cheiranthoides that are scattered across three clusters on chromosomes 2 and 3 (Fig. 2D). It should be noted that chromosome 4 in 4. thaliana is syntenic with parts of chromosomes 2 and 3 in E. cheiranthoides (Ziist, et al. (2020) Elife 9:1-42), so this broad genomic distribution may be explained by tandem duplication followed by chromosomal fission.

[0103] Knockout of candidate genes

[0104] CRISPR / Cas9 knockouts of both candidates were generated to assess their potential roles in cardenolide biosynthesis. Two independent knockout mutants were produced for Erche03g034150 (card5-l and card5-2 , while only one was recovered for Erche01g020322 (card6). Sequences of the mutant lines are available in Figures 2E and 2F. Knockout lines for both genes were deficient in cardenolide biosynthesis, with a ~1, 000-fold reduction in total cardenolide-related peak area for both card5 lines, and a ~ 100-fold reduction for card6 (Fig. 3 A). In knockout mutants, the accumulation of pathway intermediates provides clues as to the role a gene plays in a pathway. In card5-l and card5-2, pregnenolone and progesterone accumulated to higher levels than in wildtype or card6 plants (Figs. 3B-3C), indicating that CARD5 may be the first hydroxylase to act on these pregnane intermediates. In card6. two apparently hydroxylated pregnane intermediates accumulate to very high levels (Figs. 3D-3F). The masses and fragmentation patterns of these compounds are consistent with either mono- (compound 11), or di- (compound 10) glycosylated hydroxyepipregnanolone (Figs. 3G, 3H, 3J, 3K). Notably, though compound 11 occurs at trace levels in wildtype plants, it is even less abundant in card5-l and card5-2 plants (Fig. 3E).

[0105] These results are consistent with a pathway where CARD5 acts first to hydroxylate pregnane intermediates, followed by a second hydroxylation by CARD6. However, the knockout lines alone do not confirm the regiospecificity of these enzymes. In order to address this question, mutant lines were fed with both progesterone and 21-hydroxyprogesterone. Progesterone did not restore cardenolide biosynthesis in either line, but 21 -hydroxyprogesterone did restore cardenolide biosynthesis in card6 (Fig. 31). This is indicative that CARD6 is required for steroid 21 -hydroxylation. Moreover, by process of elimination, the specific activity of CARD5 is the steroid 14P-hydroxylase.

[0106] Transient expression in Nicotiana benthamiana

[0107] In order to further investigate the activity of CARD5 and CARD6, the genes encoding these enzymes were transiently expressed in N. benthamiana leaves, potential substrates were co-infiltrated, and activity was monitored via LC-MS. CARD5 hydroxylated all substrates provided, including pregnenolone, progesterone, and epipregnanolone, to form compounds 12, 13, and 14 (Figs. 4A-4C, 4I-4L). Based on the results of substrate feeding to E. cheiranthoides mutant lines, it was predicted that these products are 14P-hydroxylated.

[0108] While authentic standards were not available for these compounds, 12 is found in Cynanchum paniculatum (Sugama, et al. (1986). Chem. Pharm. Bull. 34:4500-4507), which was used to confirm the identification of 12 as 14P-hydroxypregnenolone (Figs. 4I-4L). Consistent with evidence from the mutant lines that CARD5 catalyzes the first hydroxylation step, CARD6 activity was not directly detected on any of the substrates provided (Figs. 4A-4C). However, when CARD6 was expressed together with CARD5, consumption of compounds 12, 13, and 14 and formation of new compounds was observed, which were designated as compounds 15-20 (Figs. 4D-4F). Based on similar m / z, predicted molecular formula, and MSMS fragmentation, these compounds appear to be related, with compounds 15 and 16 likely retaining the A5-double bond from pregnenolone, compounds 17 and 18 retaining the a,P-unsaturated ketone from progesterone, and compounds 19 and 20 having a fully saturated steroid core (Figs. 4D-4F, 4M-4R). As expected, these compounds contain an extra oxygen atom relative to compounds 11-13, but their m / z is 2 Da less than would be expected for hydroxylated derivatives, indicating that the hydroxy group may have been further oxidized to a carbonyl. Furthermore, these compounds appear to be attached to a large substituent group with a predicted molecular formula C9H12O8, which might correspond to a hexose and a malonyl group. These substituents are presumably added by endogenous A benthamiana enzymes. Hypothetical structures for these compounds are provided in Figs. 4D-4F. For both CARD5 and CARD6, the compounds identified here were some of the most abundant products, but they were not the only products. The accumulation of other compounds with similar masses and retention times, perhaps representing additional hydroxylation or glycosylation of the intermediates described here, was observed in all cases.

[0109] Next, a methanolic extract from the card6 mutant line was infiltrated together with CARD6 into N. benthamiana to test whether the enzyme was active on the glycosylated intermediates found in card6 plants (Figure 2F). Interestingly, consumption of monoglycosylated compound 11 was observed, but not of the di -glycosylated compound 10 (Fig.

[0110] 4G), and saw the accumulation of compound 21, which was similar to compounds 15-20 (Figs. 4H, 4M-4R). Finally, CARD5 and CARD6 were expressed with the first four characterized genes from the E. cheiranthoides cardenolide biosynthesis pathway, which were designated CARDENOLIDE METABOLISM 1 - CARDENOLIDE METABOLISM 4 EcCYP87A126 (CARD I), EcS / iHSD (CARD2), Ec3KSI (CARD 3), and EcP5 R2 (CARD4) in an attempt to reassemble part of the pathway in a heterologous system. When only CARD1-CARD5 were present, accumulation of compounds 12 and 14 was observed and when CARD6 was included, compounds 15, 16, 19, and 20 were formed (Figs. 4A-4F).

[0111] Enzyme modeling and substrate docking

[0112] To understand the evolution of CARD5 and CARD6 in Erysimum, structural models of each protein was generated to visualize their active sites with their predicted substrates. When the global architecture of the monomers was compared, a large stretch of 95 amino acids between residues 176 and 271 was noted that are disordered in the AtA0P3 model and entirely absent in the CARD5 protein (Fig. 5A). In each of their active sites, the Fe(II) ion is coordinated with the canonical His / Asp / His triad and 2 -oxoglutarate (2OG) is bound (Figs.

[0113] 5B-5C). Glucoiberin (3-(methylsulfinyl)propylglucosinolate) was docked into the A0P3 active site, and pregnenalone was docked into the active site of CARD5, which allowed comparison of the putative active sites of these two enzymes. While a number of active site residues are conserved between the two enzymes (Pro 82, Phe 83, Leu 89, Ser 94, Arg 166, Met 168, Leu 183, His 186, Thr 187, Asp 188, Lys 189, His 243, Ala 257, Phe 259, Phe 288, His 296, and Arg 301 in CARD5), there are ten residues that vary. Two residues that have nonpolar side chains in AtAOP3 (Leu 85 and Phe 386) are also nonpolar in CARD5 (Phe 87 and Met 292, respectively). Notably, three polar residues in the AtAOP3 active site (Arg 162, Asn 355, and Tyr 389) are nonpolar residues in CARD5 (Met 164, Leu 261, and Phe 295, respectively). This presence of nonpolar active site residues in CARD5 active site may account for the activity with steroid substrates like pregnenalone.

[0114] The active site residues in the Arabidopsis DAO1 active site are entirely conserved in the A. cheiranthoides DAO1 (Figs. 5D-5E), which indicates that these residues are critical for DAOl’s role as the primary mechanism for IAA oxidative catabolism in plants (Zhang, et al. (2016) Proc. Natl. Acad. Sci., 113:11010-11015; Jin, et al. (2020) J. Struct. Biol.

[0115] 212: 107632). There are fewer notable differences in the active sites of EcDAOl and CARD6 with three active site variations that stand out as being important in the evolution of steroid recognition. The polar, charged Arg 283 in EcDAOl is instead a smaller polar Gin in the CARD6 active site, and the polar Gin 173 in EcDAOl is a nonpolar Pro 173 in CARD6 (Figs. 5E-5F). The nonpolar Met 287 in EcDAOl corresponds to Vai 287 in CARD6, which has a smaller nonpolar side chain. Taken together, these active site differences allow the binding of larger steroid substrates. The compound that is docked into the CARD6 active site, a diastereomer of 12, was oriented such that the hexose sugar was positioned away from 2OG and the Fe(II) ion, which would position the substrate for 21 -hydroxylation.

[0116] Identification of steroid 14[i- and 21-hydroxylases

[0117] Herein, two key enzymes in cardenolide biosynthesis were identified inE. cheiranthoides, CARD5 and CARD6, the steroid 14P- and 21-hydroxylases, respectively. Using gene knockouts, it was shown that both enzymes are required for cardenolide biosynthesis inE. cheiranthoides, and the order of hydroxylation, with 14P-hydroxylation occurring before 21 -hydroxylation, was clarified. Using these enzymes, it was possible to partially reconstitute cardenolide biosynthesis in TV. benthamiana. The discovery of these enzymes brings the engineering of the full cardenolide pathway in a heterologous system two steps closer and adds two new pregnane-modifying enzymes to the toolkit for production of other important steroidal compounds. CARD5 is of particular interest, as no other enzymes capable of steroid 14P-hydroxylation are known. The enzymes allow for chemo-enzymatic synthesis of diverse steroidal metabolites that can be useful in human health and treatment of disease (Zhao, et al. (2022) ACS Catal. 12:9839-9845).

[0118] Based on the cardenolide chemotype of card2, cards., and card4 mutants, which accumulate dehydrocardenolides (Younkin, G., bioRxiv (2024)

[0119] doi.org / 10.1101 / 2024.04.10.588904), it is unsurprising that CARD5 and CARD6 are active on multiple intermediates in cardenolide biosynthesis. The structure of the products of CARD6 in the transient expression experiments, compounds 15-20, is not definitively known. It is clear that the glycosylation of these compounds is catalyzed by endogenous N. benthamiana enzymes. It is possible that CARD6 hydroxylates compounds 12-14 at carbon 21, and an aldehyde or steroid dehydrogenase from N benthamiana further oxidizes these intermediates to compounds 15-20. Alternatively, CARD6 may catalyze the formation of these 21 -oxo-pregnanes, and cardenolide biosynthesis requires a specialized aldehyde dehydrogenase from E. cheiranthoides to form compound 8 (Fig. 1 A). In vitro assays with purified recombinant enzymes and structural elucidation of intermediates via NMR will clarify these possibilities.

[0120] On the mechanism for stereochemical inversion at C14

[0121] The stereochemical inversion at carbon 14 that apparently accompanies hydroxylation, which is critical for the biological activity of cardenolides, has been a matter of some interest (Kreis, et al. (1998) Planta Medica 64:491-499.; Zhao, et al. (2022) ACS Catalysis 12:9839-9845). The identification of the 14P-hydroxylase as a 2OGD facilitates speculation as to the catalytic mechanism that allows for deprotonation and subsequent hydroxylation to occur on opposite faces of the pregnane substrate. In the consensus reaction mechanism for 2OGDs, a Fe(IV)-oxo intermediate that abstracts a proton from the primary substrate, forming a substrate radical (Martinez, et al. (2015) J. Biol. Chem., 290:20702-20711; Kreis, et al. (1998) Planta Med., 64:491-499). For epipregnanolone, if this radical is formed via abstraction of Hisp, a 1,2-hydride shift of Hi 4a would form a more stable tertiary radical intermediate. Hydroxylation could then proceed from the same face of the molecule from which the proton was abstracted, while also resulting in the requisite inversion of stereochemistry at carbon 14 (Fig. 6). While this mechanism is speculative, such rearrangements during hydroxylation by 20GDs are not unprecedented. For example, in benzoxazinoid biosynthesis in maize, conversion of DIMBOA-Glc to TRIMBOA-Glc involves rearrangement of a methoxy group during hydroxylation by BX13, a 20GD (Handrick, et al. (2016) Plant Cell 28:1682-1700).

[0122] Gene duplication and neofunctionalization in the evolution of cardenolide biosynthesis Examination of the genomic context and sequence evolution of CARD5 and CARD6 provides concrete examples of the processes underlying the evolution of novel phytochemicals. In both cases, gene duplication was observed followed by shifts in substrate preference and activity, but these two new case studies provide insight into variation in this process. The case of CARD6 appears straightforward. In the ancestor of both Erysimum and Arabidopsis, DAO underwent a tandem duplication, resulting in DAO1 and DAO2. While there has likely been subfunctionalization between DAO1 and DAO2, both appear to primarily function in the oxidative inactivation of indole acetic acid (Zhang, et al. (2016) Proc. Natl. Acad. Sci., 113:11010-11015). In Erysimum, DAO1 was duplicated a second time (Figure 2A, B), with one copy undergoing a radical shift in substrate preference to accept hydroxypregnanes (likely 14P-hydroxypregnanes), eventually becoming CARD6. CARD6’s new role in cardenolide biosynthesis is likely the result of three variable active site residues (Pro 173, Gin 283, and Vai 287) that allow the enzyme to accommodate larger steroid substrates (Fig. 5E-5F).

[0123] The evolutionary history of CARD5 appears more complex. The GS-AOP locus in Arabidopsis has been the subject of substantial interest due to the relationship between structural variation at the locus and glucosinolate diversity (Kliebenstein, et al. (2001) Plant Cell 13:681-693; Chan, et al. (2010) Genetics 185:991-1007). The function of AtAOPl remains unknown, while AtAOP2 and AtAOP3 are involved in glucosinolate modification. The evolutionary history in Erysimum is further complicated by four additional duplications oiAOPl scattered across three clusters separated by several million base pairs (Figs. 2C-2D). Examination of synteny in this genomic region reveals evidence of substantial rearrangements and chromosomal fission (Ziist, et al. (2020) Elife 9: 1-42). At least one of these copies, CARD5, took on a new role essential to cardenolide metabolism. The functions of the other copies remain unknown in Erysimum, but it is possible that others may be involved in further steps in cardenolide biosynthesis or modification. In fact, two of the others, Erche03g038390 n<3Erche02g016180, are co-expressed with cardenolide biosynthesis genes (Younkin, G., bioRxiv (2024) doi.org / 10.1101 / 2024.04.10.588904). The close relationship of this clade to S123DOX (Fig. 2C) hints that this group of 2OGDs may readily accept steroidal substrates, but a more thorough investigation of activity across DOXC20 is needed before conclusions can be drawn. It is intriguing to find that closely related paralogous genes, AOP2IAOP3 and CAPD5, evolved key functions in the biosynthesis of structurally dissimilar defensive metabolites, glucosinolates and cardenolides respectively. Further investigation into the ancestral function of this clade and subsequent shifts in substrate preference will provide insights into how DOXC20 provided the raw genetic material for the evolution of diverse defensive metabolites.

[0124] Furthermore, the ability of CARD5 to act on pregnenolone indicates a minimal pathway for plant metabolites with cardiotonic activity. By expressing only CAPD1-CAPD5, mN. benlhamiana. the accumulation of compound 14, which is 14P-hydroxyepipregnanolone, is seen. CARD5 is also active on pregnenolone 2, so compound 12 could be produced by the expression of only CAPD1 and CAPD5. According to studies on the structure-activity relationship of steroidal compounds, these structures are sufficient for at least some inhibitory activity against Na+ / K+-ATPases (Schbnfeld, et al. (1985) Arch.

[0125] Pharmacol., 329:414-426). Production of low levels of pregnane metabolites such as pregnenolone and progesterone is widespread in plants (lino, et al. (2007) Phytochemistry 68:1664-1673; Lindemann, P. (2015) Steroids 103:145-152), so an increase in pregnane production coupled with the evolution of 14P-hydroxylase activity may be sufficient to provide an ecological defense against insects. After production of 14P-hydroxypregnanes is established, further modifications to this minimal structure, including lactone ring formation and glycosylation would strengthen and fine-tune biological activity. This model of a step-wise pathway evolution, with only small, biochemically accessible changes required for production of an active metabolite may help explain the recurrent evolution of cardenolides as defensive compounds across land plants and some animals (Agrawal, et al. (2012) New Phytol., 194:28-45).

[0126] Several publications and patent documents are cited throughout the specification in order to describe the state of the art to which this invention pertains. Each of these citations is incorporated herein by reference as though set forth in full.

[0127] While certain of the preferred embodiments of the present invention have been described and specifically exemplified above, it is not intended that the invention be limited to such embodiments. Various modifications may be made thereto without departing from the scope and spirit of the present invention, as set forth in the following claims.

Claims

WHAT IS CLAIMED IS:

1. An isolated cardenolide metabolism 5 (CARD5) protein, wherein said CARD5 comprises an amino acid sequence having at least 90% identity with the amino acid sequence encoded by SEQ ID NO: 1.

2. The CARD5 protein of claim 1 which comprises SEQ ID NO: 15.

3. The CARD5 protein of claim 1 or claim 2, further comprising a peptide or epitope tag.

4. An isolated cardenolide metabolism 6 (CARD6) protein, wherein said CARD6 comprises an amino acid sequence having at least 90% identity with the amino acid sequence encoded by SEQ ID NO: 5.

5. The CARD6 protein of claim 4 which comprises SEQ ID NO: 19.

6. The CARD6 protein of claim 4 or claim 5, further comprising a peptide or epitope tag.

7. A composition comprising a CARD5 protein and / or CARD6 protein and a carrier.

8. A nucleic acid molecule encoding the CARD5 protein of claim 1.

9. The nucleic acid molecule of claim 8, which comprises SEQ ID NO: 1.

10. A vector comprising the nucleic acid molecule of claim 8 or claim 9.

11. A nucleic acid molecule encoding the CARD6 protein of claim 4.

12. The nucleic acid molecule of claim 11, which comprises SEQ ID NO: 5.

13. A vector comprising the nucleic acid molecule of claim 11 or claim 12.

14. A method of producing a steroidal compound with a 14P-hydroxyl, said method comprising contacting a steroidal substrate with CARD5, thereby producing the steroidal compound with a 14P-hydroxyl.

15. The method of claim 14, comprising expressing a nucleic acid encoding CARD5 in a plant or plant cell and contacting said plant or plant cell with said steroidal substrate.

16. The method of claim 14, comprising expressing a nucleic acid encoding CARD5 in a cell and contacting said cell with said steroidal substrate.

17. The method of any one of claims 14-16, further comprising expressing a nucleic acid encoding CARD1, CARD2, CARD3, and CARD4.

18. The method of any one of claims 14-17, further comprising isolating the steroidal compound with a 14P-hydroxyl.

19. The method of any one of claims 14-18, wherein said steroidal compound with a 14P-hydroxyl is a cardenolide or cardiac glycoside.

20. A method of producing a steroidal compound with a 21 -hydroxyl or 21 -carbonyl, said method comprising contacting a steroidal substrate with CARD6, thereby producing the steroidal compound with a 21 -hydroxyl or 21 -carbonyl.

21. The method of claim 20, comprising expressing a nucleic acid encoding CARD6 in a plant or plant cell and contacting said plant or plant cell with said steroidal substrate.

22. The method of claim 20, comprising expressing a nucleic acid encoding CARD6 in a cell and contacting said cell with said steroidal substrate.

23. The method of any one of claims 20-22, further comprising expressing a nucleic acid encoding CARD1, CARD2, CARD3, and CARD4 and, optionally, CARD5.

24. The method of any one of claims 20-23, further comprising isolating the steroidal compound with a 21 -hydroxyl or 21 -carbonyl.

25. The method of any one of claims 20-24, wherein said steroidal compound with a 21-hydroxyl or 21 -carbonyl is a cardenolide or cardiac glycoside.

Images