Four enzymes control natural variation in the steroid core ofErysimumcardenolides

12 Plants commonly produce families of structurally related metabolites with similar defensive 13 functions. This apparent redundancy raises the question of underlying molecular mechanisms 14 and adaptive benefits of such chemical variation. Cardenolides, a class defensive compounds 15 found in the wallflower genus Erysimum (L., Brassicaceae) and scattered across other plant 16 families, show substantial structural variation, with glycosylation and hydroxylation being 17 common modifications of a steroid core, which itself may vary in terms of stereochemistry and 18 saturation. Through a combination of chemical mutagenesis and analysis of gene coexpression 19 networks, we identified four enzymes involved in cardenolide biosynthesis in Erysimum that 20 work together to determine stereochemistry at carbon 5 of the steroid core: Ec3βHSD, a 3β-21 hydroxysteroid dehydrogenase, Ec3KSI, a ketosteroid isomerase, EcP5βR2, a progesterone 5β-22 reductase, and EcDET2, a steroid 5α-reductase. We biochemically characterized the activity of 23 these enzymes in vitro and generated CRISPR/Cas9 knockout lines to confirm activity in vivo. 24 Cardenolide biosynthesis was not eliminated in any of the knockouts. Instead, mutant plants 25 accumulated cardenolides with altered saturation and stereochemistry of the steroid core. 26 Furthermore, we found variation in carbon 5 configuration among the cardenolides of 44 species 27 of Erysimum, where the occurrence of some 5β-cardenolides is associated with the expression 28 and sequence of P5βR2. This may have allowed Erysimum species to fine-tune their defensive 29 profiles to target specific herbivore populations over the course of evolution. 30 31 SIGNIFICANCE STATEMENT 32 Plants use an array of toxic compounds to defend themselves from attack against insects and 33 other herbivores. One mechanism through which plants may evolve more toxic compounds is 34 through modifications to the structure of compounds they already produce. In this study, we 35 show how plants in the wallflower genus Erysimum use four enzymes to fine-tune the structure 36 of toxic metabolites called cardenolides. Natural variation in the sequence and expression of a 37 single enzyme called progesterone 5β-reductase 2 partly explains the variation in cardenolides 38 observed across the Erysimum genus. These alterations to cardenolide structure over the course 39 of evolution suggests that there may be context-dependent benefits to Erysimum to invest in one 40 cardenolide variant over another. 41 42 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 3

43 It is well established that plants synthetize specialized compounds with toxic properties to 44 defend themselves from herbivore attack (1–3), with the presence or absence of specific 45 compounds shaping herbivore community structure (4). However, due to physiological and 46 phylogenetic constraints, plants usually only produce a few, structurally related toxic 47 compounds. Because of this limitation, structural diversity within metabolite classes and 48 modifications to existing metabolites play an essential role in regulating herbivore resistance. In 49 fact, there is substantial evidence that small, biochemically accessible modifications to the 50 structure of existing metabolites are a critical mechanism through which plants evolve new, more 51 potent defenses in response to herbivore pressure. For example, in the furanocoumarin-producing 52 lineages of the Apiaceae, a single enzyme transforms umbelliferone to xanthotoxin, which is far 53 more toxic and restricted in its occurrence (5). An analysis of Inga foliar metabolomes proposed 54 a model where differences in chemical diversity between species were best explained by 55 regulatory changes to enzymes in existing biosynthetic pathways, allowing plants to rapidly alter 56 their chemical defenses during evolution, facilitating adaptation and divergence between closely 57 related plant lineages (6). 58 Cardenolides, which inhibit Na+/K+-ATPases in animals and evolved repeatedly in 59 diverse plant lineages (7), exhibit variation in the stereochemical configuration of carbon 5 of 60 their steroid core (Figure 1). Fixation of carbon 5 configuration likely occurs during a series of 61 steps involving repeated oxidation and reduction of pregnane intermediates. In plants, as in 62 animals, such reactions are mostly catalyzed by hydroxysteroid dehydrogenases, either from the 63 short chain dehydrogenase/reductase (SDR) or the aldo-keto reductase (AKR) families (8–10), 64 though progesterone-5β-reductases (P5βRs) also play a role in plants (11–13). 65 The relative activity of these enzymes on cardenolide intermediates may help explain the 66 natural variation seen in cardenolide configuration at carbon 5 across plant lineages. For 67 example, 5β-cardenolides are commonly found in Digitalis (Plantaginaceae) (14), whereas many 68 members of the Apocynaceae family accumulate 5α-cardenolides (15). Less commonly reported 69 are unsaturated cardenolides such as Δ5,6-unsaturated xysmalogenin and Δ4,5-unsaturated 70 canarigenin (Figure 1B). Xysmalogenin has been identified in some members of the 71 Apocynaceae, including Gomphocarpus sinaicus Boiss. (16), Asclepias curassavica L. (17), 72 Periploca sepium Bunge (18), as well as in Isoplexis spp. (Plantaginaceae, sometimes classified 73 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 4 Digitalis). (19). Canarigenin has also been reported from Isoplexis spp. (19, 20) and Convallaria 74 majalis L. (Asparagaceae) (21). In Erysimum, both 5α- and 5β-cardenolides occur across the 75 genus, including co-occurrence of both in many species (22, 23). At least one species, Erysimum 76 x allionii (formerly Cheiranthus allionii), also accumulates Δ4-cardenolides (24). The functional 77 implications of this structural variation at carbon 5 has been the subject of some research in both 78 medical and ecological contexts, with substantial impact on target-site binding, inhibitory 79 activity, and toxicity. However, much remains unclear in this regard (25–27). 80 Figure 1. Variation in the cardenolide steroid core. 81 (A) Carbon numbering scheme for the steroid core of 82 pregnanes and cardenolides. (B) Four structural classes 83 of cardenolides. Variable position of double bond or 84 stereochemical configuration of carbon 5 is highlighted 85 in blue. One example of each of the four structural 86 classes of cardenolides is shown. They may be modified 87 via hydroxylation, glycosylation, etc. 88 89 90 91 92 93 94 95 96 Enzymes such as hydroxylases and glycosyltransferases that modify pathway end 97 products are often cited as a driver of structural diversity within classes of defensive metabolites, 98 e.g., in the well-studied glucosinolate and benzoxazinoid pathways (28, 29). By contrast, we 99 provide an example of core pathway enzymes that control structural diversity of pathway end 100 products. Specifically, we show that the structure of cardenolides produced by the genus 101 Erysimum depends on the sequence and expression of four key enzymes influencing the 102 stereochemistry and saturation at carbon 5 of the steroid core. 103

104 Mutant screens 105 Two independent ethyl methanesulfonate (EMS)-mutagenized lines with altered cardenolide 106 content were characterized, and the causal mutations identified via bulked segregant analysis 107 A 2 3 4 5 10 1 6 7 8 9 14 13 12 11 15 16 17 18 19 20 21 O 5β-cardenolides digitoxigenin 5α-cardenolides uzarigenin Δ4-cardenolides canarigenin Δ5-cardenolides xysmalogenin B HO OH O O H H HO OH O O H H HO OH O O H H H HO OH O O H H H .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 5 (BSA). EMS mutant line #635 accumulates low levels of the cardenolides found in wildtype E. 108 cheiranthoides (Figure 2A; Table S1) and instead accumulates peaks with cardenolide-like 109 fragmentation and a m/z two Daltons less than digitoxigenin glycosides, suggesting that they may 110 be Δ4- or Δ5-unsaturated cardenolides (Figures 1B). This phenotype mapped to a region on 111 chromosome 7 (Figure 2B) containing a G148E missense mutation in Erche07g001535 112 (Ec3βHSD), a gene encoding a 3β-hydroxy-Δ5-steroid dehydrogenase (3βHSD) (Figure 2C). 113 Enzymes of this type are thought to be required for cardenolide biosynthesis (30–32). 114 115 Figure 2. Identification of candidate genes for cardenolide biosynthesis in Erysimum 116 cheiranthoides. (A) Cardenolide abundances in wildtype (WT) E. cheiranthoides and ethyl 117 methanesulfonate (EMS) mutant lines. Cardenolides in mutant #454 have the same mass as 118 cardenolides found in WT plants but elute at different retention times. Presumed 119 dehydrocardenolides are identified by a characteristic genin at m/z=373.2379. N = 4, error bars 120 indicate ± s.d., letters are p<0.001, ANOV A with post-hoc Tukey’s HSD on log-transformed peak 121 areas. (B) Mapping results of bulked segregant analysis for mutant #635. Alternate (mutant) 122 allele frequency, smoothed over 1 Mbp segments, is plotted across the eight E. cheiranthoides 123 chromosomes. In plants with a mutant chemotype (red), mutant alleles dominate in the latter half 124 of chromosome seven. (C) Within this region is a 3β-hydroxysteroid dehydrogenase (Ec3βHSD) 125 with a G148E missense mutation in mutant #635 plants. (D) In mutant #454 plants, there are two 126 EMS-induced mutations in the coding region of EcP5βR2. For mapping results for mutant #454, 127 see Figure S1. Exons are shown as blue rectangles, introns as blue lines, and untranslated regions 128 Total cardenolide−related peak area 0 1 2 3 4 Normalized peak area a c b b a b c b a 5β−cardenolides5α−cardenolides dehydrocardenolides WT mutant #454 mutant #635 A 1 2 3 4 5 6 7 8 wildtype 635 mutant Alternate allele frequency in wildtype and 635 mutant, smoothed Alternate allele frequency (%) 0.0 0.2 0.4 0.6 0.8 1.0 B DC Chromosome Steroid biosynthesis Steroid related, uncharacterized Confirmed role in cardenolide biosynthesis Other metabolic enzyme glycosyltransferase EcCYP716A418 EcCYP87A126 Erche07g001535 (Ec3βHSD) Erche07g017160 (Ec3KSI) Erche07g010950 (EcDET2) E ATG TAG Ec3βHSD CACCGGACCAC mutant #635 CACCGAACCAC Chr07: 19,548,921 bp 19,547,574 bp TAAATG EcP5βR2 TCAAAGACCTGG...TCATGGGAATTC mutant #454 TCAAAAACCTGG...TCATGAGAATTC Chr06: 16,478,737 bp 16,477472 bp .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 6 as yellow rectangles. (E) Cardenolide-related gene coexpression cluster from an analysis of 129 transcript abundances in 48 Erysimum species. EcCYP716A418 and EcCYP87A126 are 130 cytochrome P450s involved in cardenolide biosynthesis in E. cheiranthoides. Other candidates 131 for involvement in cardenolide biosynthesis include a 5α-reductase (EcDET2), Ec3βHSD, and a 132 3-ketosteroid isomerase (Ec3KSI). 133 134 EMS mutant #454 was described by Mirzaei et al. (33), with the cardenolide phenotype 135 being linked to a locus on chromosome 6 (Figure S1). However, no causal mutation was 136 identified in that study. We re-evaluated the cardenolide phenotype and found that mutant #454 137 plants accumulate compounds with the same mass as digitoxigenin glycosides, but with shifted 138 retention times (Figure S2). We hypothesized that these peaks represent uzarigenin glycosides 139 with 5α stereochemistry (Figure 1B, 2A; Table S1). This led us to the identification of 140 Erche06g007150 (EcP5βR2), which encodes a progesterone-5β-reductase (P5βR). This group of 141 enzymes catalyzes the stereospecific reduction of α,β-unsaturated ketones including 142 progesterone, methylvinylketone, and 2-cyclohexene-1-one (34). EcP5βR2 contained two amino 143 acid mutations (R184K and G201R) at the genetically linked locus on chromosome 6 (Figure 144 2D). We therefore hypothesized that in the absence of a functional P5βR, a steroid 5α-reductase 145 (5αR) acts on progesterone, resulting in the accumulation of 5α-cardenolides. 146 Gene coexpression analysis 147 Gene coexpression analysis across 48 Erysimum species revealed a cluster of 28 148 coexpressed genes related to steroid metabolism and cardenolide biosynthesis (Figure 2E). Two 149 genes, EcCYP87A126 and EcCYP716A418, are involved in cardenolide biosynthesis (35), and 150 one, Ec3βHSD, was identified in EMS mutant #635. Eight genes in the cluster encode enzymes 151 that are directly involved in core sterol or isoprenoid metabolism (Table S2). The remaining 152 genes are considered candidates for involvement in cardenolide biosynthesis. Of note is a gene 153 encoding a short-chain dehydrogenase/reductase (SDR), Erche07g017160 (Ec3KSI). The closest 154 Arabidopsis ortholog, AT2G33630, is annotated as having a 3β-hydroxysteroid-155 dehydrogenase/isomerase domain (IPR002225)(36). An additional candidate from the 156 coexpression cluster is a steroid 5αR, Erche07g010950 (EcDET2), which is involved in 157 brassinosteroid biosynthesis (37) and may also be involved in 5α-cardenolide biosynthesis. 158 Functional characterization of candidate enzymes 159 We examined the role of candidate enzymes for involvement in cardenolide biosynthesis by 160 functionally characterizing the recombinant purified proteins in vitro. Recombinant Ec3βHSD 161 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 7 had steroid-3-dehydrogenase activity on pregnenolone 2 to form isoprogesterone 3 in the 162 presence of NAD+. We also saw some production of progesterone 4, implying that the enzyme 163 may additionally have Δ5,4 isomerase activity (Figure 3A, S3; Table S3). As plant 3βHSD 164 enzymes typically do not possess Δ5,4 isomerase activity (31, 38, 39), we independently checked 165 for isomerase activity by supplying the enzyme directly with isoprogesterone 3, but we did not 166 see an increase in isomerization to progesterone 4 relative to a negative control (Figure 3B, S3; 167 Table S3). 168 169 Figure 3. Characterization of candidate cardenolide biosynthetic enzymes from Erysimum 170 cheiranthoides. Assays are in vitro with purified recombinant enzymes except where noted. (A) 171 Conversion of pregnenolone to isoprogesterone and progesterone by 3β-hydroxysteroid 172 dehydrogenase (Ec3βHSD) and 3-ketosteroid isomerase (Ec3KSI). (B) Conversion of 173 isoprogesterone to progesterone by Ec3βHSD and Ec3KSI. Under assay conditions, ~50% of the 174 peak area is attributable to progesterone even in the negative control. (C) Conversion of 175 progesterone to 5β-pregnane-3,20-dione by progesterone 5β-reductase 2 (EcP5βR2). (D) In N. 176 benthamiana, significantly more 5α-pregnane-3,20-dione is produced when progesterone is 177 pregnanolone 0 20 40 60 80 100Normalized peak area Nb_4g2680Nb_677Nb_467g17Nb_467g15Nb_477Nb_4677 b c b b b a progesterone normalized peak area HO H O H H H O H O H H H Ec3βHSD EcCYP87A126 Ec3KSI HO O H H H O O H H H O O H H H O O H H H H EcDET2 pregnenolone 2 isoprogesterone 3 progesterone 4 5β-pregnane-3,20-dione 55α-pregnane-3,20-dione 7 Ec3βHSD epipregnanolone 6 EcCYP87A126 Ec3βHSD Ec3KSI EcP5βR2 + + + + + - - + + - + + + - + + - - + + + + - + Ec3βHSD Ec3KSI - - + - - + + + - - + - - + + + Ec3βHSD Ec3KSI - + - - - + C F HO H O H H H isopregnanolone 8 Ec3βHSD NAD+ NADH NADH NAD+ NADH NAD+ NADPH NADP+ pregnenolone + +++ + +++ isoprogesterone +++ A D B E Ec3βHSD Ec3KSI - + - - - + 5β-pregnane-3,20-dione +++ EcP5βR2 - + progesterone ++ 5α-pregnane-3,20-dione - - - - + - - ++ - - GFP EcDET2 progesterone log10 normalized isoprogesterone peak area Normalized peak area 0 10 100 1000 negative controlEc3βHSD P<0.001 log10 normalized isoprogesterone peak area Normalized peak area 0 10 100 1000 negative controlEc3βHSD Ec3KSI b a b log10 normalized isoprogesterone peak area Normalized peak area 0 10 100 1000 negative controlEcP5βR P=0.013 Nicotiana benthamiana isoprogesterone normalized peak area5β-pregnane-3,20-dione normalized peak area progesterone:isoprogesterone peak area ratio Percent conversion to progesterone 0 5 10 15 Proportion converted to progesterone negative controlEc3βHSD Ec3KSI b b a -+ - - - + ++ + epi- or isopregnanolone normalized peak area epi- or isopregnanolone normalized peak area EcP5βR NAPD+ NADPH 5α-pregnane-3,20-dione normalized peak area log10 normalized isoprogesterone peak area Normalized peak area 0 50 100 negative controlEc3βHSD Ec3KSIEc3βHSD + Ec3KSI b a b a log10 normalized progesterone peak area normalized peak area 0 5000 10000 15000 negative controlEc3βHSD Ec3KSIEc3βHSD + Ec3KSI c b c a log10 normalized 5apregnane320dione peak area Normalized peak area 0 1000 2000 GFP + progEcDET2 + bufferEcDET2 + prog b c a .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 8 coinfiltrated with 5α-reductase (EcDET2) compared to a negative control. (E) Reduction of 5β-178 pregnane-3,20-dione and 5ɑ-pregnane-3,20-dione to epipregnanolone or a stereoisomer by 179 Ec3βHSD but not Ec3KSI. (F) Production of epipregnanolone or stereoisomer when 180 EcCYP87A126, Ec3βHSD, Ec3KSI, and EcP5BR2 are coexpressed in N. benthamiana. For all 181 assays: N = 3 replicates per enzyme, except in (D) where N=5. Error bars indicate ± s.d., letters 182 are P<0.001, ANOV A with post-hoc Tukey’s HSD. P-values above bars are from Student’s t-test. 183 Statistics were performed on log-transformed peak areas. Negative controls used a purified 184 recombinant 2-oxoglutarate dioxygenase not otherwise discussed in this study. LCMS 185 chromatograms and MSMS spectra associated with all assays are provided in Figure S3. 186 187 We also tested Ec3βHSD for the ability to catalyze the reverse reaction, the reduction of 188 the 3-keto group of 5β-pregnane-3,20-dione 5 or 5α-pregnane-3,20-dione 7 to the 3β-hydroxyl in 189 epipregnanolone 6 or isopregnanolone 8. When Ec3βHSD was supplied with 5β-pregnane-3,20-190 dione 5 or 5α-pregnane-3,20-dione 7 as a substrate and NADH as a cofactor, we saw formation 191 of a product with m/z=319.2637 in both cases, which is consistent with epipregnanolone 6 and 192 isopregnanolone 8, but we were unable to separate these two products and the epipregnanolone 193 standard chromatographically (Figure 3E, S3; Table S3). Therefore, the exact stereochemical 194 configuration of these predicted products is not confirmed. 195 Recombinant Ec3KSI converts isoprogesterone 3 to progesterone 4 in vitro (Figure 3B, 196 S3). We also tested Ec3KSI for 3βHSD activity, but no activity was observed for either the 197 oxidation or reduction reaction (Figure 3A,E, S3; Table S3). Furthermore, when we combined 198 Ec3βHSD and Ec3KSI in a single reaction and supplied pregnenolone and NAD+, we observed 199 consumption of isoprogesterone and increased formation of progesterone relative to the same 200 reaction containing only Ec3βHSD (Figure 3A; Table S3). We therefore identified Ec3KSI as a 201 3-ketosteroid isomerase and showed that it works in concert with Ec3βHSD to convert 202 pregnenolone 2 to progesterone 4. When supplied with progesterone 4 and NADPH, recombinant 203 EcP5βR2 catalyzes the formation of 5β-pregnane-3,20-dione 5 (Figure 3C, S3; Table S3). 204 EcDET2 is membrane-bound, complicating purification of the recombinant protein. We 205 therefore coinfiltrated EcDET2 with progesterone in Nicotiana benthamiana leaves. Although 206 the 5α-reduction of progesterone 4 is catalyzed by endogenous enzymes in N. benthamiana 207 leaves even in the GFP control, more 5α-pregnane-3,20-dione 7 is produced when EcDET2 is 208 present, consistent with previous studies showing that DET2 orthologs can use progesterone 4 as 209 a substrate (40) (Figure 3D, S3; Table S3). 210 Production of epipregnanolone in N. benthamiana 211 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 9 To test whether the identified enzymes can work together in planta to produce intermediates in 212 cardenolide biosynthesis, we coexpressed the sterol side chain cleaving enzyme that initiates 213 cardenolide biosynthesis, EcCYP87A126 (35, 41) together with Ec3βHSD, Ec3KSI, and 214 EcP5βR2 in leaves of N. benthamiana and observed the production of a compound with the same 215 m/z and retention time as epipregnanolone 6, although we cannot rule out that it may be a 216 stereoisomer (Figure S3). A small amount of epipregnanolone 6 was detected as long as 217 EcCYP87A126 was present, but significantly more epipregnanolone 6 was produced if 218 Ec3βHSD, Ec3KSI, and EcP5βR2 were all included (Figure 3F; Table S3). 219 Knockouts of candidate genes have altered cardenolide profiles 220 To assess the role of candidate enzymes in cardenolide biosynthesis in vivo, we generated two 221 independent CRISPR/Cas9 knockout lines for all candidate genes (Figures S4-S8 for sequences 222 of mutants). With the exception of det2 knockout lines, which had the characteristic dwarf 223 phenotype of brassinosteroid biosynthetic mutants (37) (Figure 4A), none of the mutant lines 224 displayed obvious growth phenotypes. We analyzed cardenolide profiles via UPLC-MS, using 225 both methanolic extracts of intact cardenolides and cardenolide extracts subjected to mild acidic 226 conditions, which resulted in hydrolysis of the sugar moieties. This allowed us to directly 227 compare the cardenolide genins produced by each mutant line. For Δ4-cardenolides (canarigenin 228 glycosides), a water molecule is eliminated under acidic conditions to produce 3,5-229 anhydroperiplogenin (Figure 4C)(19, 20), which we used as a proxy for Δ4-cardenolide 230 abundance. While we did not have an authentic standard for uzarigenin, xysmalogenin, 231 canarigenin, 3,5-anhydroperiplogenin, cannogenol, or cannogenin, MSMS spectra together with 232 pathway logic allowed us to identify these compounds with reasonable confidence. To further 233 confirm our identification of uzarigenin, we performed acid hydrolysis on cardenolide extracts 234 from uzarigenin-containing Calotropis procera leaves (Figure S9)(42). 235 Based on a principal component analysis (PCA) of intact cardenolides in the mutant lines, 236 we confirmed a causal relationship between the genetically linked mutations in EMS mutants 237 #454 and #635 and their cardenolide phenotypes (Figure 4B; Table S1). Hydrolyzed extracts of 238 wildtype E. cheiranthoides and det2 single mutants were dominated by the 5β-cardenolide series: 239 digitoxigenin, cannogenol, cannogenin, and strophanthidin. All other mutant lines had altered 240 cardenolide profiles relative to wildtype, but cardenolide production was not eliminated in any of 241 the mutants. 3bhsd and 3ksi lines had similar cardenolide profiles, with both lines accumulating 242 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 10 lower levels of 5β-cardenolides compared to wildtype and containing Δ5-cardenolides with 243 m/z=373.2379, which are not found in wildtype E. cheiranthoides (Figure 4C, S10; Table S4). 244 Although the chemotype of 3bhsd and ksi lines was qualitatively very similar, the reduction in 245 5β-cardenolide abundance was more severe in 3bhsd plants (Figure S11; Table S5). One possible 246 explanation for this is a degree of functional overlap between these enzymes. To test whether 247 Ec3βHSD and Ec3KSI have redundant roles in cardenolide biosynthesis, we crossed 3bhsd-1 and 248 3ksi-1 lines to generate 3bhsd/3ksi double mutants. The double mutants had a cardenolide profile 249 very similar to 3bhsd plants (Figure S11; Table S5), suggesting that Ec3βHSD and Ec3KSI have 250 distinct roles. If the two enzymes were redundant, we would expect an additive effect on the 251 cardenolide phenotype in the double mutant. 252 253 Figure 4. Characterization of Erysimum cheiranthoides cardenolide biosynthesis mutants. 254 (A) Photo of E. cheiranthoides wildtype (WT), a steroid 5α-reductase (det2) knockout, and a 255 DET2/progesterone 5β-reductase (det2/p5br) double mutant, which display a dwarf phenotype 256 characteristic of brassinosteroid biosynthesis mutants. (B) Principal component analysis of intact 257 cardenolides detected by UPLC-MS in E. cheiranthoides WT and mutant lines: 3β-258 hydroxysteroid dehydrogenase (3bhsd), 3-ketosteroid isomerase (3ksi). Ethyl methane-sulfonate 259 (EMS) mutants #454 and #635 cluster with corresponding CRISPR/Cas9 mutant lines. Arrows 260 A B C O H O H H H EcCYP87A126 HO O H H H O O H H H O O H H H HO OH O O H H H HO OH O O H H HO OH O O H H HO OH O O H H H O O H H H H 5β-cardenolides5α-cardenolidesΔ4-cardenolidesΔ5-cardenolides Ec3βHSD Ec3KSI EcP5βR2 EcDET2 WT det2 det2/p5br2 −2 0 2 4 6 −4 −2 0 2 PCA of all peaks (normalized by internal standard) PC1 (49.1%) PC2 (34%) wildtype 3bhsd−1 3bhsd−2 mutant #635 3ksi−1 3ksi−2 p5br2−1 p5br2−2 mutant #454 det2−1 det2−2 det2/p5br2−1 det2/p5br2−2 → → → 5β−cardenolides 5α−cardenolides dehydrocardenolides OH O O hydrolytic conditions 3,5-anhydroperiplogenin Peak area attributable to each genin Proportion of total peak area WT 3bhsd 3ksi p5br2 det2−1 det2/p5br2−2 1 2 1 2 1 2 0.00 0.25 0.50 0.75 1.00 HO H OH O O HO H H HO H OH O O O H H HO OH OH O O O H H xysmalogenin canarigenin uzarigenin digitoxigenin cannogenol cannogenin strophanthidin .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 11 indicate loadings for individual cardenolide mass features and are grouped into structural classes 261 by color. (C) Relative peak area of cardenolide genins following acid hydrolysis of E. 262 cheiranthoides mutant lines (N = 3 plants per line). For Δ5-cardenolides (grey; xysmalogenin), 263 5α-cardenolides (red; uzarigenin), and 5β-cardenolides (blue; digitoxigenin, cannogenin, 264 cannogenol, and strophanthidin), the cardenolide genins were detected directly via UPLC-MS. 265 Δ4-cardenolides undergo dehydration under acid conditions, so 3,5-anhydroperiplogenin was 266 detected as a proxy for canarigenin (yellow). 267 268 Acid hydrolysis confirmed the digitoxigenin-glycoside isomers in p5br2 plants to be 269 uzarigenin glycosides, with 5α configuration (Figure 4C). Additionally, p5br2 lines accumulated 270 the same m/z=373.2379 peaks observed in 3bhsd and 3ksi lines, which we observed to be a mix 271 of Δ4- and Δ5-cardenolides following hydrolysis (Figure 4C; Table S4). Notably, cardenolide 272 hydroxylation (resulting in the derived cardenolide genins cannogenol, cannogenin, and 273 strophanthidin) is eliminated in p5br2 plants, suggesting that the cardenolide hydroxylases 274 expressed in E. cheiranthoides act only on 5β-cardenolides (Figure 4C, S10). Although these 275

clearly demonstrate the involvement of EcP5βR2 in cardenolide biosynthesis, we also 276 investigated a paralogous gene, EcP5βR1 (Erche02g027660), for potential involvement in the 277 pathway. EcP5βR1 acted on progesterone 4 to produce 5β-pregnane-3,20-dione 5 when co-278 infiltrated in N. benthamiana (Figure S12). However, in a EcP5βR1 knockout line, the 279 cardenolide profile was unchanged (Figure S13; Table S6). 280 In order to test whether 5α-cardenolide production in p5br2 mutants was mediated by 281 EcDET2, we generated CRISPR/Cas9 knockouts of EcDET2 in the p5br2-1 background. In 282 p5br2/det2 double mutants, production of cardenolides with a fully reduced steroid core was 283 nearly eliminated, and was replaced by accumulation of a mix of Δ4- and Δ5-cardenolides 284 (Figure 4C, Table S4), confirming that EcDET2 acts as a 5αR in cardenolide biosynthesis in the 285 absence of a functional EcP5βR2. 286 Natural variation in progesterone 5β-reductase activity across the genus Erysimum 287 We conducted a survey of the cardenolide genins across the genus by subjecting methanolic leaf 288 extracts from 44 species of Erysimum to acid hydrolysis. The following cardenolide genins were 289 detected: the 5β-cardenolides digitoxigenin, cannogenol, cannogenin, and strophanthidin; the 5α-290 cardenolide uzarigenin; the Δ4-cardenolide canarigenin; the Δ5-cardenolide xysmalogenin; and 291 an isomer of cannogenol that we speculate may be its 5α-isomer. With the exception of E. 292 collinum, which does not produce detectable levels of cardenolides, both uzarigenin and 293 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 12 strophanthidin were detected in hydrolyzed extracts of all species. Digitoxigenin, cannogenol, 294 and cannogenin were relatively rare, being mostly restricted to the monophyletic clade 295 containing E. cheiranthoides, E. sylvestre, and two closely related species of uncertain 296 taxonomic identity (43)(Figure 5A). 3,5-anhydroperiplogenin (canarigenin) and xysmalogenin 297 were also detected at low levels in some species (Figure 5A; Table S7). 298 299 Figure 5. Gene expression and cardenolide configuration across 44 species of Erysimum. (A) 300 Expression of cardenolide biosynthesis genes and detection of eight cardenolide genins 301 following acid hydrolysis mapped against an Erysimum species phylogeny (Züst et al 2020). 302 Grey indicates very low expression (<50 counts per million reads). Expression for each gene is 303 normalized separately. * indicates a frameshift mutation in P5BR2-1. Cardenolide genin 304 abundances are displayed as percent of total LC-MS peak area of all genins, mean of n=1-3 305 replicates per species. No data are displayed for E. collinum due to very low levels of 306 cardenolides. (B) Location and sequence of frame-shift mutations in the coding region of 307 EcP5bR2 orthologs from Erysimum sp. 1 (ER1), E. pieninicum (PIE), E. hungaricum (HUN), E. 308 hieraciifolium (HIE), and E. virgatum (VIR). Regions translated out of frame are indicated in 309 red, with in-frame stop codons bold and underlined. (C) Amino acid phylogeny of Erysimum 310 P5bR2 proteins for which a full sequence could be recovered from the transcriptomes. Activity 311 of selected enzymes were tested via transient expression and co-infiltration of progesterone in 312 Nicotiana benthamiana leaves. Numbers at nodes indicate bootstrap support from 10,000 313 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 <50 Erche04g002680 Erche07g001535 Erche07g017160 Erche02g027660 Erche06g007150 Erche07g010950 MED NEV BAS FIZ NER RUS MEX SCO BIC LAG AMO MEZ FRA CAP ALI CSS CRA COL DIF AND RHA WIT ODO PSE VIR HIE ER1 HUN PIE ER2 PUL CUS MIC CRE KOT ECE SYL ER4 ER3 NAX CHR WIC INC REP 0 200 400 600 800 1000 0.2 ASP_NO_STOP VIR2 SYL ECE_2g27660 PIE2_R CRP_R_DN12743_i3 PSE_R RHA_R HUN_R RHA3 PUL2_R HIE1 CRP_R ECE_6g7150 SUF KOT_R CSS 97 42 86 90 76 71 84 36 82 99 64 53 91 0.2 ASP_NO_STOP VIR2 SYL ECE_2g27660 PIE2_R CRP_R_DN12743_i3 PSE_R RHA_R HUN_R RHA3 PUL2_R HIE1 CRP_R ECE_6g7150 SUF KOT_R CSS 97 42 86 90 76 71 84 36 82 99 64 53 91 0.2 ASP_NO_STOP VIR2 SYL ECE_2g27660 PIE2_R CRP_R_DN12743_i3 PSE_R RHA_R HUN_R RHA3 PUL2_R HIE1 CRP_R ECE_6g7150 SUF KOT_R CSS 97 42 86 90 76 71 84 36 82 99 64 53 91 Erysimum sp. 1 E. kotschyanum E. crassipes E. rhaeticum E. pseudorhaeticum Erysimum sp. 2 E. pulchellum E. hieraciifolium E. virgatum Erysimum sp. 1 E. pieninicum E. hungaricum E. cheiranthoides (EcP5βR2) Erysimum sp. 4 E. sylvestre Erysimum sp. 3 E. cheiranthoides (EcP5βR1) cannogenol(5β) cannogenolisomer (presumed 5ɑ)uzarigenin (5ɑ) canarigenin (Δ4) E. mediohispanicum E. nevadense E. bastetanum E. fitzii E. nervosum E. ruscinonense E. merxmuelleri E. scoparium E. bicolor E. lagascae E. amoenum E. menziesii E. franciscanum E. capitatum E. allionii E. crassipes E. crassicaule E. collinum E. diffusum E. andrzejowskianum E. rhaeticum E. wittmannii E. odoratum E. pseudorhaeticum E. virgatum E. hieraciifolium Erysimum sp. 1 E. hungaricum E. pieninicum Erysimum sp. 2 E. pulchellum E. cuspidatum E. microstylum E. crepidifolium E. kotschyanum E. cheiranthoides E. sylvestre Erysimum sp. 4 Erysimum sp. 3 E. naxense E. cheiri E. wilczekianum E. incanum E. repandum C A Active on progesterone in Nicotiana benthamiana Inactive (truncated protein) Not tested (truncated protein) Not tested (full length protein) ATG TAA EcP5βR2 TGTGTTT...TAA ER1-1 TGT--TT...TAA EcP5βR2 GGATATCGGGA...GGGAGAGAGA-----------AA...TGA PIE/HUN GG-------GA...GGGAGAGAGA-----------AA...TGA HIE GGATATCGGGA...GG-AGAGAGA-----------AA...TGA VIR GGATATCGGGA...GGGAGAGAGATGGGAGAGAGAAA...TGA B CYP87A1263βHSD3KSIP5βR1P5βR2DET2cardenolide genins ** ** Erche04g002680 Erche07g001535 Erche07g017160 Erche02g027660 Erche06g007150 Erche07g010950 MED NEV BAS BAE FIZ NER RUS MEX SCO BIC LAG AMO MEZ FRA CAP ALI CSS CRA COL DIF AND RHA WIT ODO MAJ PSE VIR HIE ER1 HUN PIE ER2 PUL CUS MIC CRE KOT ECE SYL ER4 ER3 NAX CHR WIC INC REP 0 5e−01 1 1.5 2 2.5 30 1000normalized expression * P5βR2-1 frameshift mutation xysmalogenin (Δ5) * digitoxigenin (5β) cannogenin (5β) strophanthidin (5β) P5βR2-2P5βR2-1 Peak area attributable to each genin Proportion of total peak area MED NEV BAS FIZ NER RUS MEX SCO BIC LAG AMO MEZ FRA CAP ALI CSS CRA COL DIF AND RHA WIT ODO PSE VIR HIE ER1 HUN PIE ER2 PUL CUS MIC CRE KOT ECE SYL ER4 ER3 NAX CHR WIC INC REP 0.00 0.25 0.50 0.75 1.00 REP INC WIC CHR NAX ER3 ER4 SYL ECE KOT CRE MIC CUS PUL ER2 PIE HUN ER1 HIE VIR PSE ODO WIT RHA AND DIF COL CRA CSS ALI CAP FRA MEZ AMO LAG BIC SCO MEX RUS NER FIZ BAS NEV MED Inf −Infvalue Erche04g002680Erche07g001535Erche07g017170Erche06g007150Erche07g010950digitoxigeninuzarigeninxysmalogenindianhydroperiplogenin .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 13 replicates and scale bar indicates estimated substitutions per site. Multiple sequence alignment 314 underlying P5bR2 protein phylogeny is available in Figure S14. 315 316 We next examined P5βR2 and DET2 sequences and expression levels across the 317 Erysimum genus to better understand how they interact to determine relative levels of 5α- and 318 5β-cardenolides. Of the three P5βR sequences found in the E. cheiranthoides genome, only two 319 have orthologs that are expressed in the species included in this study. EcP5βR1 orthologs are 320 uniformly expressed across the genus, but based on the E. cheiranthoides p5br1 knockouts, they 321 are unlikely to be involved in 5β-cardenolide biosynthesis. By contrast, EcP5βR2 orthologs are 322 only expressed at greater than 50 counts per million reads (CPM) in 15 of the 44 species 323 examined (Figure 5A; Table S8), and in five of these species, P5βR2 contains a frameshift 324 mutation (Figure 5B, S14). Based on protein phylogeny, Erysimum P5βR2 proteins can be 325 further classified into two clades, P5βR2-1 and P5βR2-2, with EcP5βR2 belonging to the 326 P5βR2-1 clade (Figure 5C). We cloned P5βR2 orthologs in both clades from six Erysimum 327 species, as well as EcP5βR1, and assessed activity via co-infiltration with progesterone 4 in N. 328 benthamiana. All full-length P5βR2 proteins were capable of converting progesterone 4 to 5β-329 pregnane-3,20-dione 5 in planta, while truncations resulted in a loss of activity (Figure 5C, S12). 330 Erysimum sp. 1 was the only species to express both P5βR2-1 and P5βR2-2, but the expressed 331 P5βR2-1 encodes a non-functional protein. Among species examined, the expression of a 332 functional P5βR2-1 was required for production of the 5β-cardenolides digitoxigenin and 333 cannogenin. 334 Orthologs of Ec3βHSD, EcKSI, and EcDET2 are expressed across the genus, including in 335 Erysimum collinum, where CYP87A126, the first gene in the pathway, is not expressed and very 336 low levels of cardenolides are produced (43) (Figure 5A), suggesting that they may have roles in 337 steroid metabolism outside of cardenolide biosynthesis. Close examination of the transcriptome 338 data revealed that some species express more than one distinct transcript of the genes examined 339 here, despite the E. cheiranthoides genome containing only one copy in the case of Ec3βHSD 340 and EcDET2. However, the nature of transcriptomic data makes it difficult to assess genomic 341 copy number, and it is unclear whether transcriptomic sequence variation is due to polypoidy in 342 some species, gene duplication and sequence divergence, or allelic variation of a single locus. 343 Phylogenetic analysis of E. cheiranthoides cardenolide biosynthesis genes 344 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 14 We inferred phylogenetic trees for Ec3βHSD, EcKSI, EcP5βR, and EcDET2 to better understand 345 their relationship to characterized genes from other species. All species examined had more than 346 one Ec3βHSD-like gene, with the exception of E. cheiranthoides. The three orthologs from E. 347 crepidifolium (44), two from D. lanata (30–32), and two from A. thaliana, AtSDR5 348 (AT2G47140) and AtSDR3 (AT2G47130) (10), have been shown to accept cardenolide 349 intermediates in vitro, but in planta evidence for involvement in cardenolide biosynthesis only 350 exists for Dl3βHSD1 (30) (Figures 6A, S15). Of the species included in the KSI gene tree, only 351 E. cheiranthoides contained more than one copy (Figures 6B, S16). To our knowledge, no 352 enzymes from this group have been biochemically characterized prior to this study. 353 354 Figure 6. Phylogenetic analysis of Erysimum cheiranthoides cardenolide biosynthetic genes. 355 Nucleotide phylogenies for cardenolide biosynthetic genes from E. cheiranthoides and selected 356 Active on cardenolide intermediates, or in vivo evidence for involvement in vivo evidence suggests not involved Not tested Non-cardenolide producing species Cg_12956 Cg_5349 EcDET2 (Erche07g010950) AtDET2 (AT2G38050) Dl2934 Mp0170s0018 0.2 Cg_12956 Dl2934 Cg_5349 Erche07g010950 AtDET2_AT2G38050 M_polymorpha_0170s0018 57 46 96 0.09 Ecre_5540 Ecre_9065 AtSDR5_AT2G47140 AT2G47150 Ecre_7213 Erche07g001535 AtSR3_AT2G471430 AtSDR4_AT3G29250 AT3G29260 Dl3BHSD2 Dl3BHSD1 AT3G51680 AT2G47120 98 79 97 97 100 49 62 90 94 98 Dl3βHSD1 Dl3βHSD2 AtSDR5 (AT2G47140) Ec3βHSD (Erche07g001535) EcreHSD1 AT2G47150 AT2G47120 EcreHSD2 AtSDR3 (AT2G47130) AtSDR4 (AT3G29250) AT3G29260 EcreHSD3 AtSDR2 (AT3G51680) 0.2 Erche06g011020 Dl_DN4564 Dl_DN477 Erche06g007150 Ecre_P5BR2 DlP5BR2 AT5G58750 AT4G24220_VEP1 Ecre_P5BR1 Erche02g027660 Mp0191s0002 Dl_DN67489 DlP5BR1 57 99 100 100 96 49 100 99 99 100 81 Dl_DN67489 AtP5βR2 (AT5G58750) EcP5βR2 (Erche06g007150) EcP5βR3 (Erche06g011020) EcreP5βR2 Mp0191s0002 Dl_4564 DlP5βR2 AtP5βR1 (AT4G24220) EcP5βR1 (Erche02g027660) EcreP5βR1 Dl_DN477 DlP5βR1 Erche04g028310 Ec3KSI (Erche07g017160) AT2G33630 Dl_DN10245 Cg_010433 AT1G47290 0.2 AT2G33630 AT1G47290 Erche04g028310 Erche07g017170 Dl_DN10245 Cg010433 98 53 88 A B C D active on pregnenolone in vivo evidence active on isoprogesterone in vivo evidence active on progesterone in vivo evidence active on progesterone in vivo evidence .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 15 other species for which functional work has been done. (A) 3β-hydroxysteroid dehydrogenase 357 (3βHSD) and related short-chain dehydrogenases (SDR), (B) 3-ketosteroid isomerase (3KSI), 358 (C) progesterone 5β-reductase (P5βR), and (D) steroid 5α-reductase (DET2). Experimental 359 evidence for activity on cardenolide intermediates or in vivo evidence for involvement in 360 cardenolide biosynthesis, either in this or previous studies, are marked by black squares. 361 Enzymes for which in vivo evidence suggests the enzyme is not involved in cardenolide 362 biosynthesis are marked by red squares. Species included: Arabidopsis thaliana (At/AT), 363 Calotropis gigantea (Cg), Erysimum cheiranthoides (Ec/Erche), Erysimum crepidifolium (Ecre), 364 Digitalis lanata (Dl), and Marchantia polymorpha (Mp). Numbers at nodes indicate bootstrap 365 support from 10,000 replicates and scale bar indicates estimated substitutions per site. Multiple 366 sequence alignments underlying these phylogenies are provided in Figures S15-S18. 367 368 P5βR genes fall into two clades arising from an ancient duplication event (33.8% amino 369 acid identity between EcP5βR1 and EcP5βR2). Members of both clades have been shown to act 370 on progesterone in vitro (13, 34, 45, 46), but only DlP5βR1, which is a more closely related to 371 EcP5βR1, has a confirmed role in 5β-cardenolide biosynthesis (47). EcP5βR2, which we show to 372 be required for 5β-cardenolide biosynthesis in E. cheiranthoides, belongs to the other clade 373 (Figures 6C, S17). EcDET2, which is required for 5α-cardenolide biosynthesis in E. 374 cheiranthoides, is a single copy gene for most species examined, and it accepts progesterone as a 375 substrate even in species that do not make cardenolides, such as A. thaliana and S. lycopersicum 376 (40). Intriguingly, C. gigantea, which accumulates 5α-cardenolides, contains two copies of this 377 gene (Figures 6D, S18). 378

379 Identification E. cheiranthoides cardenolide biosynthesis genes 380 In this study, we identified and biochemically characterized four cardenolide biosynthetic 381 enzymes from E. cheiranthoides. Of these, Ec3βHSD and EcP5βR2 belong to enzyme families 382 that have been speculated, or shown, to be involved in cardenolide biosynthesis in other species 383 (13, 30, 31, 45, 47). However, the mutant lines generated in this study provide critical in vivo 384 evidence for their role in cardenolide synthesis, and the lack of a cardenolide phenotype for 385 EcP5βR1 knockouts highlights the potential disparity between in vitro activity and in vivo 386 function. 387 We additionally demonstrate the capacity of some enzymes to assume multiple roles in 388 related metabolic pathways. The dwarf phenotype of the det2 mutant lines in this study confirms 389 that DET2 is required for brassinosteroid production in E. cheiranthoides. However, the lack of 390 cardenolides with a fully saturated ring system in the p5br2/det2 double mutants show that 391 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 16 EcDET2 can also contribute to 5α-cardenolide synthesis. Because wildtype E. cheiranthoides 392 primarily produces 5β-cardenolides, it is perhaps unsurprising that it does not have a dedicated 393 copy of DET2 for cardenolide biosynthesis that could be regulated independently or have altered 394 kinetics. In other Erysimum species that naturally accumulate higher levels of 5α-cardenolides, 395 DET2 may have been duplicated, although we were unable to confirm this with the available 396 transcriptomes. 397 The involvement of a dedicated ketosteroid isomerase in cardenolide biosynthesis has 398 been the subject of substantial research (38, 39), with others speculating that isomerization is 399 catalyzed by 3βHSD, as is the case in animals (8, 48). Activity-guided fractionation of Digitalis 400 protein extracts revealed that isomerization was catalyzed by an enzyme distinct from 3βHSD, 401 but the protein sequence of the putative 3-ketosteroid isomerase was not identified. We identified 402 Ec3KSI, a member of short chain dehydrogenase/reductase family 42E (SDR42E), as 403 responsible for catalyzing the isomerization of isoprogesterone 3 to progesterone 4 (Figure 3) 404 while lacking 3βHSD activity. Bacterial ketosteroid isomerases have been extensively 405 characterized (49), but less is known about eukaryotic enzymes with KSI activity that lack 406 3βHSD activity. We demonstrated with CRISPR/Cas9-generated knockout lines that Ec3KSI is 407 involved in cardenolide biosynthesis. However, even in the absence of Ec3KSI, 5β-cardenolides 408 are produced, albeit in lower quantities than occur in wildtype plants. It is possible that some 409 isomerization occurs non-enzymatically, or another enzyme, possibly Ec3βHSD, can partially 410 compensate for the loss of KSI activity in the mutant lines. 411 Still in question is the extent to which plant 3βHSDs also possess KSI activity. While 412 there has been at least one report of a tomato 3βHSD with KSI activity, our assays were 413 inconclusive, as we saw isomerization when Ec3βHSD was supplied with pregnenolone 2 and 414 NAD+, but not when Ec3βHSD is supplied with isoprogesterone 3. This implies either that the 415 isomerization by Ec3βHSD must accompany oxidase activity, the isomerization observed during 416 the 3βHSD assay occurred non-enzymatically, or isomerization catalyzed by Ec3βHSD is slow. A 417 comprehensive analysis of 3βHSD activity across plants, with careful attention to the 418 spontaneous isomerization of isoprogesterone 3, is warranted to differentiate these possibilities. 419 Natural variation in P5βR expression and sequence influences cardenolide stereochemistry in 420 Erysimum 421 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 17 We observed substantial interspecific variation in the accumulation of 5α-, 5β-, and dehydro-422 cardenolides across the Erysimum genus. Although cardenolides of each type had been reported 423 previously in at least one Erysimum species (22, 24, 50), other studies focused primarily on 5β-424 cardenolides (45, 51). We found that the 5β-cardenolides digitoxigenin and cannogenin were 425 restricted to a monophyletic clade containing E. cheiranthoides, E. sylvestre, and two closely 426 related species of uncertain taxonomic identity (43), and that their occurrence coincided with 427 expression of a functional P5βR2-1. In one monophyletic group of five Erysimum species, 428 P5βR2-1, while expressed, contains a frameshift mutation, corresponding with the apparent loss 429 of digitoxigenin and cannogenin synthesis. Several other species, including E. crassipes, E. 430 rhaeticum, and E. kotschyanum, express P5βR2-2 at high levels, but we did not detect 431 digitoxigenin or cannogenin glycosides in these species. We hypothesize that the P5βR2-1 clade 432 is specialized for involvement in 5β-cardenolide biosynthesis, whereas P5βR1 and P5βR2-2 433 assume other roles in plant metabolism. The fact that these enzymes are expressed and active on 434 cardenolide intermediates, but seem to be uninvolved in the pathway, points to the possibility of 435 substrate channeling or compartmentalization of cardenolide biosynthesis. 436 By contrast, strophanthidin, a more hydroxylated 5β-cardenolide, was ubiquitous across 437 the genus, and its occurrence did not depend on expression of P5βR2-1 in most species. The 438 occurrence of C5-hydroxylated 5β-cardenolides in the absence of P5βR2-1 implies the existence 439 of a P5βR-independent pathway for strophanthidin biosynthesis in Erysimum. For example, it is 440 possible that inversion of carbon 5 stereochemistry occurs during C5-hydroxylation, and in fact a 441 similar stereochemical inversion is thought to occur during C14-hydroxylation in cardenolide 442 biosynthesis (52). Alternatively, we cannot exclude the possibility that one of the other expressed 443 P5βR proteins forms digitoxigenin as an intermediate which is entirely converted by cardenolide 444 hydroxylases to strophanthidin. Interestingly, the E. cheiranthoides p5br2 mutant lacks 445 strophanthidin, suggesting that such a P5βR-independent strophanthidin pathway, if it exists, 446 may have been lost in this species. 447 Previous studies have reported digitoxigenin in several of the species for which we failed 448 to detect it in our assays (22, 24). This discrepancy may be explained in a number of ways. First, 449 we only sampled single accessions or seed batches per species, which may underestimate 450 potentially substantial intraspecific variation in plant chemistry. Second, many early studies used 451 paper chromatography or other low-resolution chromatographic techniques that would render 452 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 18 correct stereochemical assignment difficult or impossible, and even high-resolution methods may 453 miss differences in stereochemistry without appropriate reference material. For example, in a 454 previous study profiling intact cardenolides in the same species studied here, the E. 455 cheiranthoides clade exhibited a distinct chemotype with limited overlap in non-hydroxylated 456 cardenolides compared to the other species examined, which may have been a reflection of a 457 difference in carbon 5 stereochemistry (43). In particular, the E. cheiranthoides clade 458 accumulated a set of unique mono- and diglycosides of digitoxigenin, cannogenol, and 459 cannogenin, whereas most other Erysimum species accumulated an isomeric set of putative 5α-460 cardenolides with shifted HPLC retention times. Consistent with our results, no equivalent 461 pattern was apparent for strophanthidin glycosides, and no 5α-isomers of strophanthidin have 462 been described for any Erysimum species (22, 43). 463 Because core pathway enzymes are apparently active regardless of carbon 5 464 stereochemistry, switching between production of digitoxigenin and uzarigenin-glycosides only 465 requires alteration to the expression or sequence of a single enzyme, P5βR2-1. Control of the 466 production of unsaturated cardenolides, which do not occur at high levels in any of the species in 467 this study, appears to be somewhat more cryptic, with potential for expression level, gene 468 duplication, or protein-protein interactions to play a role. An analogous process controls steroidal 469 glycoalkaloid diversity in Solanum, where expression of GAME25, a SDR related to Ec3βHSD, 470 controls saturation of steroidal glycoalkaloids across the genus (53). Along with overall polarity, 471 stereochemical configuration is a major contributor to variation in toxicity and deterrent activity 472 of cardenolides (25–27, 54). As such, altering cardenolide stereochemical configuration may be a 473 relatively simple evolutionary mechanism through which Erysimum fine-tunes its defensive 474 profile to the most pervasive herbivores in a given ecological context. This theory is somewhat 475 borne out by the findings of Mirzaei et al. (33), who showed that an E. cheiranthoides P5βR2 476 mutant line, which produces primarily 5α-cardenolides, was more resistant to Trichoplusia ni 477 (cabbage loopers) and Myzus persicae (green peach aphids), but extracts from 5β-cardenolide-478 producing wildtype plants had a greater inhibitory effect on porcine Na+,K+-ATPase in vitro. 479 This work represents a step forward in our understanding of the biosynthesis of medically 480 important cardiac glycosides and provides insight into molecular mechanisms through which 481 cardenolide structural variation may be controlled. The apparently modular nature of the 482 cardenolide pathway, where the presence or activity of individual enzymes alters the pathway 483 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 19 end products, has broad implications for future engineering of the pathway in heterologous 484 systems for research or medical purposes. Such flexibility in pathway assembly may allow for 485 rapid production and testing of varied cardenolide structures for biomedical applications. 486 Furthermore, the mutant E. cheiranthoides lines presented here will facilitate investigation of the 487 functional and ecological implications of carbon 5 configuration in cardenolides. Insect feeding 488 assays and field experiments with these mutant plants would illuminate herbivore preferences 489 and provide further insight into the selective pressures that may have shaped the cardenolide 490 profiles observed in nature today. 491

492 Plant growth, cloning, expression, and knockout of candidate genes 493 Plant growth, cloning of candidate genes, transient expression in Nicotiana benthamiana, and 494 CRISPR/Cas9 knockout in Erysimum cheiranthoides were performed as described previously 495 (35). For the genus-wide experiment, lyophilized tissue collected during a previous study and 496 stored at -20 °C (43) was used. Primers used for cloning of candidate genes and for generation 497 and screening of CRISPR/Cas9 mutants are provided in Table S9. 498 Mutagenic screens 499 Ethyl methanesulfonate (EMS) mutagenesis was modified from Mirzaei et al., 2020 (33). Ten 500 grams of E. cheiranthoides seeds were soaked at 4 oC overnight in 100 mL of 100 mM phosphate 501 buffer, pH 7.5. The buffer was decanted, and the seeds were resuspended in 100 mL of fresh 502 phosphate buffer with 0.6% EMS (Sigma-Aldrich, St. Louis, MO). The seeds were shaken at 23 503 oC for six hours, washed twenty times with deionized water, and grown in twenty 25x25 cm flats 504 to maturity. M2 seeds were pool-harvested from each flat at two timepoints. To screen for 505 mutants in cardenolide biosynthesis, 32 plants from each pool (in total, 1120 plants) were grown 506 for four weeks. Approximately 30 mg of leaf tissue was harvested from each plant for UHPLC-507 MS analysis. Plants that showed divergent cardenolide phenotypes were backcrossed to 508 wildtype, and F2 progeny were used for bulked segregant analysis (BSA) as described previously 509 (33). 510 Coexpression networking analysis 511 Raw RNA-sequencing reads from 48 Erysimum species (43) were downloaded from the NCBI 512 Short Read Archive (SRP225657) and were pseudoaligned to the transcriptome associated with 513 E. cheiranthoides genome v2.1 (NCBI: PRJNA563696)(55, 56) using kallisto (57) with default 514 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 20 parameters, yielding transcript counts, which were filtered to retain transcripts with more than 10 515 counts in at least 10 samples. Filtered counts were used for the mr2mods gene coexpression 516 analysis pipeline using default parameters (58). 517 Protein expression and purification 518 For protein purification, genes were inserted into the ChampionTM pET300-NT-DEST plasmid 519 (ThermoFisher Scientific, Waltham, MA), before being transformed into Rosetta(DE3) E. coli 520 (MilliporeSigma, St. Louis, MO). Single colonies were picked from LB agar plates (100 μg/mL 521 carbenicillin; 20 μg/mL chloramphenicol) to inoculate 10 mL liquid LB cultures with the same 522 antibiotics. Cultures were grown overnight at 37 °C and 225 rpm in a I2500 Incubator Shaker 523 incubator (Eppendorf, Hamburg Germany). After 18 hours, 6 mL of the culture was transferred 524 to 250 mL TB medium (59) containing the same antibiotics, and cells were grown under the 525 same conditions until OD600=0.6. Isopropyl β-d-1-thiogalactopyranoside (IPTG, ThermoFisher 526 Scientific) was added to a final concentration of 1 mM to induce protein expression, and cultures 527 were incubated at 25 °C and 225 RPM for 6 hours before being placed on ice and centrifuged at 528 5,000 rcf and 4 °C for 15 minutes in a Sorvall RC5C Plus centrifuge (ThermoFisher Scientific). 529 Cells were resuspended in 40 mL of lysis buffer, consisting of 40 mM Tris pH8, 20 mM 530 imidazole (Sigma-Aldrich, St. Louis, MO), 500 mM NaCl (ThermoFisher Scientific), 10% (v/v) 531 glycerol, and 1% (v/v) Tween 20 (Sigma-Aldrich). The cell suspension was lysed by freezing in 532 liquid N2 and thawing on ice twice, followed by sonication using a Branson Sonifier 250 while 533 still on ice, four times in 10-second intervals, with a 30-second rest between each interval. The 534 cell lysate was centrifuged for 45 minutes at 4 °C at 13,000 rcf. Following centrifugation, the 535 supernatant was loaded onto a column containing 1 mL of Ni-NTA resin (Invitrogen, Waltham, 536 MA) that had been previously equilibrated with 1 column volume of lysis buffer. After all 537 supernatant had passed through the column, 1 column volume of wash buffer (50 mM Tris pH 538 8.0, 20 mM imidazole, 500 mM NaCl, 10% (v/v) glycerol) was passed through the column. 539 Finally, 0.5 mL elution buffer (50 mM Tris pH 8.0, 250 mM imidazole, 500 mM NaCl, 10% v/v 540 glycerol) was loaded onto the column, and the flow-through was collected in a 2 mL 541 microcentrifuge tube (Laboratory Products Sales, Rochester, NY). Protein concentration was 542 measured using a NanoDrop One (ThermoFisher Scientific). 543 In vitro enzyme assays 544 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 21 Steroid-3β-hydroxysteroid dehydrogenase (3βHSD) and 3-ketosteroid isomerase (3KSI) assay 545 conditions were adapted from previous studies (39, 53). One hundred μl reactions contained 4 546 mM KPO4 buffered at pH 6.5 and 1 μg purified enzyme. For the 3βHSD oxidation, final 547 concentrations of 150 μM NAD+ (Sigma-Aldrich) and 10 μM pregnenolone (Sigma-Aldrich) 548 were used. To test for 3βHSD reductase activity, 150 μM NADH (Sigma-Aldrich) and 10 μM 5β-549 pregnane-3,20-dione (aablocks, San Diego, CA) or 5α-pregnane-3,20-dione (Sigma-Aldrich) 550 were used. For the KSI assay, the same conditions were used, but 10 μM isoprogesterone (TLC 551 Pharmaceutical Standards, Newmarket, ON) was used and NAD+/NADH were omitted. 552 Progesterone 5β-reductase (P5βR) assays were adapted from Herl et al. (13) and Sonawane et al. 553 (53). 100 μl reactions contained final concentrations of 4 mM KPO4 pH 7.2, 150 μM NADPH 554 (Cayman Chemical, Ann Arbor, MI, USA), 10 μM progesterone (Sigma-Aldrich), and 1 μg 555 purified enzyme. 556 Reactions were incubated for 1 hour at 28 °C for the 3βHSD and 3KSI assays, and at 37 557 °C for the P5βR assay. All assays were terminated by addition of 100 μl 100% methanol 558 (ThermoFisher Scientific) containing 15 μg/mL ouabain (Sigma-Aldrich) as an internal standard, 559 centrifuged at 17,000 rcf in an Eppendorf 5417R Centrifuge at 4 °C, and transferred to vials 560 (ThermoFisher Scientific) for UHPLC-MS analysis. 561 Metabolite extraction 562 Metabolites were extracted from fresh tissue of E. cheiranthoides and N. benthamiana as 563 described previously (35). Where indicated, metabolite extracts were subjected to acid hydrolysis 564 to isolate cardenolide genins using a protocol adapted from Schaller & Kries (19). In brief, 700 565 μL 100% methanol for two leaf 14 mm leaf disks of fresh E. cheiranthoides tissue or 750 μL 566 95% (v/v) methanol per 20 mg of lyophilized tissue in the genus-wide experiment, was used to 567 extract metabolites for 30 minutes at 25 °C. After centrifugation for three minutes at 17,000 rcf, 568 700 μL supernatant was transferred to a fresh microcentrifuge tube. Twenty μL of 6 M 569 hydrochloric acid was added to each tube, and samples were incubated for 18 hours at 28 °C. 570 Hydrolysis was terminated with 200 μL saturated sodium phosphate solution, and samples were 571 extracted twice with 200 μL chloroform. The organic (lower) phase was evaporated to dryness in 572 a Savant SpeedVacTM SC110 (Thermo Fisher Scientific). Samples were resuspended in 50 μL 573 methanol and centrifuged for 10 minutes at 17,000 rcf before being transferred to glass mass 574 spectrometry vials for UHPLC-MS analysis. 575 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 22 Liquid chromatography-mass spectrometry (LC-MS) analysis 576 All samples were analyzed on an UltiMate 3000 UHPLC system coupled to a Q-Exactive hybrid 577 quadrupole-orbitrap mass spectrometer (Thermo Fisher Scientific, Waltham, MA). The 578 instrument was fitted with a Supelco TitanTM C18 UHPLC Column (80Å, 100 x 2.1 mm, particle 579 size 1.9 μm; Sigma Aldrich). Injections of 2 μL were separated by a solvent gradient consisting 580 of mobile phase A (water + 0.1% (v/v) formic acid) and mobile phase B (acetonitrile + 0.1% 581 (v/v) formic acid). A 13-minute method was used for analysis of non-hydrolyzed samples: 0-0.55 582 minutes, hold at 2% B; 0.5-10 minutes, linear gradient from 2%-97% B; 10-11.5 minutes, hold 583 at 97% B, 11.5-13 minutes, hold at 2% B. A longer solvent gradient was used in hydrolysis 584 experiments: 0-5 minutes, hold at 2% B; 5-22 minutes, linear gradient from 2%-97% B; 22-23.5 585 minutes, hold at 97% B, 23.5-25 minutes, hold at 2% B. All solvents were Optima LC/MS grade 586 (Thermo Fisher Scientific). The solvent flow rate was 0.5 mL/minute, the column oven was set 587 to 40 °C, and the autosampler temperature was 15 °C for all methods. The mass spectrometer 588 was run in full scan positive ionization mode. Targeted MSMS spectra were collected with an 589 isolation window of 2.0 m/z and normalized collision energy of 30%. 590 LC-MS peak areas were quantified using a custom processing method in XcaliburTM 591 Software (ThermoFisher Scientific) using the following parameters: peak detection ICIS, 592 smoothing points 1, baseline window 40, area noise factor 5, peak noise factor 15, tailing factor 593 2. Mass features used for quantification are provided in Table S10 for in-tact cardenolides, Table 594 S11 for hydrolyzed cardenolides, and Table S12 for cardenolide intermediates from in vitro and 595 N. benthamiana assays. 596 Statistical and phylogenetic analysis 597 The following functions in R statistical software (60) were used for statistical tests, which were 598 performed on log-transformed LC-MS peak areas, normalized to an internal standard: aov, 599 TukeyHSD, and t.test. Plots were made using MSnbase (61, 62), multcompView (63), and 600 pheatmap (64). 601 Sequences homologous to Ec3βHSD, EcKSI, EcP5βR2, and EcDET2 were identified 602 using BLAST against publicly available transcriptomes for Arabidopsis thaliana (65), Calotropis 603 gigantea (66), Digitalis lanata (48) (NCBI PRJNA923725), Marchantia polymorpha (67), and 604 other Erysimum species (43) (NCBI PRJNA563696), and were aligned using ClustalW (68, 69). 605 Gene phylogenies were inferred using IQ-TREE web server (70–72) with default parameters, 606 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 23 except bootstrap alignments were increased to 10,000. Raw data underlying all figures are 607 available in the Supporting Information. 608 609 ACKNOWLEDGMENTS 610 We thank Tobias Krug for assistance with laboratory assays. This research was funded by United 611 States Department of Agriculture award 2020-67013-30896, US National Science Foundation 612 award 1645256, and an award from the Triad Foundation to GJ; a Chemistry Biology Interface 613 Training Program fellowship under National Institutes of Health/National Institute of General 614 Medical Sciences (T32GM138826) and a US National Science Foundation Graduate Research 615 Fellowship (DGE–2139899) to GCY; a Swiss National Science Foundation grant (PCEFP3-616 194590) to TZ; and a Summer Undergraduate Research Fellowship from the American Society 617 of Plant Biologists and a Rawlings Cornell Presidential Research Scholar award to MLA. 618 619 COMPETING INTERESTS 620 None declared. 621 622 AUTHOR CONTRIBUTIONS 623 GCY , MLA, and GJ designed the research; GCY, MLA, and TZ performed the research; TZ 624 contributed critical plant material; GCY and MLA analyzed data; GCY , TZ, and GJ wrote and 625 edited the manuscript. 626 627 DATA A V AILABILITY 628 The raw data that support the findings of this study are available in the Supporting Information. 629 Seeds from mutant lines will be made available from the Arabidopsis Biological Resource 630 Center. Due to high mortality and poor seed set, DET2 knockout lines are not available from the 631 ABRC. 632 633 CITATIONS 634 1. G. Fraenkel, The Raison d’Etre of Secondary Plant Substances. Science (80-. ). 129, 635 1466–1470 (1959). 636 2. P. R. Ehrlich, P. H. Raven, Butterflies and Plants : A Study in Coevolution. Evolution (N. 637 Y). 18, 586–608 (1964). 638 3. P. Feeny, Defensive Ecology of the Cruciferae. Ann. Missouri Bot. Gard. 64, 221–234 639 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 24 (1977). 640 4. H. V. Cornell, B. A. Hawkins, Herbivore responses to plant secondary compounds: A test 641 of phytochemical coevolution theory. Am. Nat. 161, 507–522 (2003). 642 5. M. Berenbaum, Toxicity of a furanocoumarin to armyworms: A case of biosynthetic 643 escape from insect herbivores. Science (80-. ). 201, 532–534 (1978). 644 6. D. L. Forrister, et al., Diversity and divergence: evolution of secondary metabolism in the 645 tropical tree genus Inga. New Phytol. 237, 631–642 (2023). 646 7. A. A. Agrawal, G. Petschenka, R. A. Bingham, M. G. Weber, S. Rasmann, Toxic 647 cardenolides: Chemical ecology and coevolution of specialized plant-herbivore 648 interactions. New Phytol. 194, 28–45 (2012). 649 8. W. L. Miller, R. J. Auchus, The molecular biology, biochemistry, and physiology of 650 human steroidogenesis and its disorders. Endocr. Rev. 32, 81–151 (2011). 651 9. T. J. Bach, M. Rohmer, Isoprenoid Synthesis in Plants and Microorganisms (2013). 652 10. P. Lindemann, Steroidogenesis in plants - Biosynthesis and conversions of progesterone 653 and other pregnane derivatives. Steroids 103, 145–152 (2015). 654 11. J. Munkert, et al., Progesterone 5β-reductase genes of the Brassicaceae family as function-655 associated molecular markers. Plant Biol. 17, 1113–1122 (2015). 656 12. K. Schmidt, et al., PRISEs (progesteron 5b-reductase and/or iridoid synthase-like 1,4-657 enone reductases): Catalytic and substrate promiscuity allows for realization of multiple 658 pathways in plant metabolism. Phytochemistry 156, 9–19 (2018). 659 13. V. Herl, G. Fischer, F. Mu, W. Kreis, Molecular cloning and heterologous expression of 660 progesterone 5b-reductase from Digitalis lanata Ehrh. Phytochemistry 67, 225–231 661 (2006). 662 14. W. Kreis, The Foxgloves (Digitalis) Revisited. Planta Med. 83, 962–976 (2017). 663 15. S. Wen, et al., Cardenolides from the Apocynaceae family and their anticancer activity. 664 Fitoterapia 112, 74–84 (2016). 665 16. H. El-Askary, J. Hölzl, S. Hilal, E. S. El-Kashoury, A comparative study of the 666 cardenolide content of different organs of Gomphocarpus sinaicus. Phytochemistry 38, 667 1181–1184 (1995). 668 17. T. Warashina, K. Shikata, T. Miyase, S. Fujii, T. Noro, New cardenolide and acylated 669 lignan glycosides from the aerial parts of Asclepias curassavica. Chem. Pharm. Bull. 56, 670 1159–1163 (2008). 671 18. J. Xu, K. Takeya, H. Itokawa, Pregnanes and cardenolides from Periploca sepium. 672 Phytochemistry 29, 344–346 (1990). 673 19. F. Schaller, W. Kreis, Cardenolide genin pattern in Isoplexis plants and shoot cultures. 674 Planta Med. 72, 1149–1156 (2006). 675 20. S. Spengel, E. Hauser, H. H. Linde, A. X. Vaz, K. Meyer, Die Glykoside der Blstter von 676 Isoplexis canariensis (L.) G. Don. Helv. Chim. Acta 50, 1893–1911 (1967). 677 21. V. K. Saxena, P. K. Chaturvedi, A novel cardenolide, canarigenin-3-O-α-L-678 rhamnopyranosyl-(1->5)-O-β-D-xylofuranoside, from the rhizomes of Convallaria 679 majalis. J. Nat. Prod. 55, 39–42 (1992). 680 22. I. F. Makarevich, K. V. Zhernoklev, T. V. Slyusarskaya, G. N. Yarmolenko, Cardenolide-681 containing plants of the family Cruciferae. Chem. Nat. Compd. 30, 275–289 (1994). 682 23. J. A. Moore, C. Tamm, T. Reichstein, Die Glykoside der Goldlacksamen, Cheiranthus 683 Cheiri L. Helv. Chim. Acta 37, 755–770 (1954). 684 24. I. F. Makarevich, Cheiranthus allioni--A unique cardenolide-bearing plant. Chem. Nat. 685 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 25 Compd. 28, 265–271 (1992). 686 25. J. Nielsen, Host plant selection of monophagous and oligophagous flea beetles feeding on 687 crucifers. Entomol. Exp. Appl. 24, 362–369 (1978). 688 26. G. Petschenka, et al., Relative selectivity of plant cardenolides for Na+/K+-ATPases from 689 the monarch butterfly and non-resistant insects. Front. Plant Sci. 9, 1–13 (2018). 690 27. C. Tamm, “New aspects of cardiac glycosides: The stereochemistry of the glycosides in 691 relation to biological activity” in Proceedings of the First International Pharmacological 692 Meeting, B. Uvnās, W. Wilbrandt, P. Lindgren, Eds. (The Macmillan Company, 1963), 693 pp. 11–26. 694 28. S. Chhajed, et al., Glucosinolate Biosynthesis and the Glucosinolate–Myrosinase System 695 in Plant Defense Shweta. Agronomy 10 (2020). 696 29. S. Zhou, A. Richter, G. Jander, B. Thompson, T. Road, Beyond Defense : Multiple 697 Functions of Benzoxazinoids in Maize Metabolism Special Focus Issue – Mini Review. 698 Plant Cell Physiol. 59, 1528–1537 (2018). 699 30. T. Leykauf, et al., Overexpression and RNAi-mediated Knockdown of Two 3β-hydroxy-700 Δ5-steroid dehydrogenase Genes in Digitalis lanata Shoot Cultures Reveal Their Role in 701 Cardenolide Biosynthesis. Planta Med. 89, 833–847 (2023). 702 31. V. Herl, J. Frankenstein, N. Meitinger, F. Müller-Uri, W. Kreis, Δ5-3β-hydroxysteroid 703 dehydrogenase (3βHSD) from Digitalis lanata. Heterologous expression and 704 characterisation of the recombinant enzyme. Planta Med. 73, 704–710 (2007). 705 32. S. Seidel, W. Kreis, E. Reinhard, Δ5-3β-hydroysteroid dehydrogenase/Δ5-Δ4-ketosteroid 706 isomerase (3β-HSD), a possible enzyme of cardiac glycoside biosynthesis, in cell cultures 707 and plants of Digitalis lanata EHRH. Plant Cell Rep. 8, 621–624 (1990). 708 33. M. Mirzaei, et al., Less Is More: a Mutation in the Chemical Defense Pathway of 709 Erysimum cheiranthoides (Brassicaceae) Reduces Total Cardenolide Abundance but 710 Increases Resistance to Insect Herbivores. J. Chem. Ecol. 46, 1131–1143 (2020). 711 34. J. Munkert, et al., Iridoid synthase activity is common among the plant progesterone 5β-712 reductase family. Mol. Plant 8, 136–152 (2015). 713 35. G. C. Younkin, et al., Cardiac glycosides protect wormseed wallflower (Erysimum 714 cheiranthoides) against some, but not all, glucosinolate- adapted herbivores. New Phytol. 715 (2024) https:/doi.org/10.1111/nph.19534. 716 36. T. Paysan-Lafosse, et al., InterPro in 2022. Nucleic Acids Res. 51, D418–D427 (2023). 717 37. S. Fujioka, et al., The Arabidopsis deetiolated2 mutant is blocked early in brassinosteroid 718 biosynthesis. Plant Cell 9, 1951–1962 (1997). 719 38. N. Meitinger, et al., The catalytic mechanism of the 3-ketosteroid isomerase of Digitalis 720 lanata involves an intramolecular proton transfer and the activity is not associated with 721 the 3β-hydroxysteroid dehydrogenase activity. Tetrahedron Lett. 57, 1567–1571 (2016). 722 39. N. Meitinger, D. Geiger, T. W. Augusto, R. Maia De Pádua, W. Kreis, Purification of Δ5-723 3-ketosteroid isomerase from Digitalis lanata. Phytochemistry 109, 6–13 (2015). 724 40. F. Rosati, et al., 5α-Reductase activity in Lycopersicon esculentum: Cloning and 725 functional characterization of LeDET2 and evidence of the presence of two isoenzymes. J. 726 Steroid Biochem. Mol. Biol. 96, 287–299 (2005). 727 41. M. Kunert, et al., Promiscuous CYP87A enzyme activity initiates cardenolide 728 biosynthesis in plants. Nat. Plants 9, 1607–1617 (2023). 729 42. A. Pandey, V. Swarnkar, T. Pandey, P. Srivastava, Transcriptome and Metabolite analysis 730 reveal candidate genes of the cardiac glycoside biosynthetic pathway from Calotropis 731 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 26 procera. Sci. Rep., 1–14 (2016). 732 43. T. Züst, et al., Independent evolution of ancestral and novel defenses in a genus of toxic 733 plants (Erysimum, Brassicaceae). Elife 9, 1–42 (2020). 734 44. J. Munkert, M. Ernst, F. Müller-Uri, W. Kreis, Identification and stress-induced 735 expression of three 3β- hydroxysteroid dehydrogenases from Erysimum crepidifolium 736 Rchb. and their putative role in cardenolide biosynthesis. Phytochemistry 100, 26–33 737 (2014). 738 45. J. Munkert, P. Bauer, E. Burda, F. Müller-Uri, W. Kreis, Progesterone 5β-reductase of 739 Erysimum crepidifolium: CDNA cloning, expression in Escherichia coli, and reduction of 740 enones with the recombinant protein. Phytochemistry 72, 1710–1717 (2011). 741 46. P. Bauer, et al., Highly conserved progesterone 5 b-reductase genes (P5bR) from 5b-742 cardenolide-free and 5b-cardenolide-producing angiosperms. Phytochemistry 71, 1495–743 1505 (2010). 744 47. J. Klein, et al., RNAi-mediated gene knockdown of progesterone 5β-reductases in 745 Digitalis lanata reduces 5β-cardenolide content. Plant Cell Rep. 40, 1631–1646 (2021). 746 48. E. Carroll, B. R. Gopal, I. Raghavan, M. Mukherjee, Z. Q. Wang, A cytochrome P450 747 CYP87A4 imparts sterol side-chain cleavage in digoxin biosynthesis. Nat. Commun. 14, 748 4042 (2023). 749 49. N.-C. Ha, G. Choi, K. Y. Choi, B.-H. Oh, Structure and enzymology of Δ5-3-ketosteroid 750 isomerase. Curr. Opin. Struct. Biol. 11, 674–678 (2001). 751 50. D. M. Piatak, P.-F. L. Tang, P. D. Sørensen, Constituents of Erysimum inconspicuum. 752 Two sulfur-containing lactone compounds. J. Nat. Prod. 48, 424–428 (1985). 753 51. T. Züst, M. Mirzaei, G. Jander, Erysimum cheiranthoides, an ecological research system 754 with potential as a genetic and genomic model for studying cardiac glycoside 755 biosynthesis. Phytochem. Rev. 17, 1239–1251 (2018). 756 52. M. E. Deluca, A. M. Seldes, E. G. Gros, The 14ß-HydroxyIation in the Biosynthesis of 757 Cardenolides in Digitalis purpurea. The Role of 3ß-Hydroxy-5ß-pregn-8(14)-en-20-one. 758 Zeitschrift für Naturforsch. 42c, 77–78 (1987). 759 53. P. D. Sonawane, et al., Short-chain dehydrogenase/reductase governs steroidal specialized 760 metabolites structural diversity and toxicity in the genus Solanum. Proc. Natl. Acad. Sci. 761 115, E5419–E5428 (2018). 762 54. K. Repke, New developments in cardiac glycoside structure-activity relationships. Trends 763 Pharmacol. Sci. 6, 275–278 (1982). 764 55. S. R. Strickler, A. F. Powell, L. A. Mueller, T. Zust, G. Jander, NCBI BioProject ID 765 PRJNA563696. Rapid and independent evolution of ancestral and novel chemical 766 defenses in a genus of toxic plants (Erysimum, Brassicaceae) (2019). 767 56. M. Mirzaei, et al., Aphid resistance segregates independently of cardiac glycoside and 768 glucosinolate content in an Erysimum cheiranthoides ( wormseed wallflower ) F2 769 population. bioRxiv (2024) https:/doi.org/https://doi.org/10.1101/2024.01.11.575310. 770 57. N. L. Bray, H. Pimentel, P. Melsted, L. Pachter, Near-optimal probabilistic RNA-seq 771 quantification. Nat. Biotechnol. 34, 525–527 (2016). 772 58. J. H. Wisecaver, et al., A global coexpression network approach for connecting genes to 773 specialized metabolic pathways in plants. Plant Cell 29, 944–959 (2017). 774 59. , Terrific Broth (TB) Medium (2015) https:/doi.org/10.1101/pdb.rec085894. 775 60. R Core Team, R: A Language and Environment for Statistical Computing (2020). 776 61. L. Gatto, K. Lilley, MSnbase - an R/Bioconductor package for isobaric tagged mass 777 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 27 spectrometry data visualization, processing and quantitation. Bioinformatics 28, 288–289 778 (2012). 779 62. L. Gatto, S. Gibb, J. Rainer, MSnbase, efficient and elegant R-based processing and 780 visualisation of raw mass spectrometry data. bioRxiv (2020). 781 63. S. Graves, H.-P. Piepho, L. S. with help from Sundar Dorai-Raj, multcompView: 782 Visualizations of Paired Comparisons (2023). 783 64. R. Kolde, pheatmap: Pretty Heatmaps (2019). 784 65. T. Z. Berardini, et al., The Arabidopsis information resource: Making and mining the 785 “gold standard” annotated reference plant genome. Genesis 53, 474–485 (2015). 786 66. G. M. Hoopes, et al., Genome assembly and annotation of the medicinal plant Calotropis 787 gigantea a producer of anticancer and antimalarial cardenolides. G3 Genes, Genomes, 788 Genet. 8, 385–391 (2018). 789 67. M. K. Tello-Ruiz, P. Jaiswal, D. Ware, “Gramene: A Resource for Comparative Analysis 790 of Plants Genomes and Pathways” in Plant Bioinformatics, D. Edwards, Ed. (Springer US, 791 2022), pp. 101–131. 792 68. F. Sievers, et al., Fast, scalable generation of high-quality protein multiple sequence 793 alignments using Clustal Omega. Mol. Syst. Biol. 7 (2011). 794 69. F. Madeira, et al., Search and sequence analysis tools services from EMBL-EBI in 2022. 795 Nucleic Acids Res. 50, W276–W279 (2022). 796 70. B. Q. Minh, et al., IQ-TREE 2: New Models and Efficient Methods for Phylogenetic 797 Inference in the Genomic Era. Mol. Biol. Evol. 37, 1530–1534 (2020). 798 71. J. Trifinopoulos, L. T. Nguyen, A. von Haeseler, B. Q. Minh, W-IQ-TREE: a fast online 799 phylogenetic tool for maximum likelihood analysis. Nucleic Acids Res. 44, W232–W235 800 (2016). 801 72. D. T. Hoang, O. Chernomor, A. Von Haeseler, B. Q. Minh, L. S. Vinh, UFBoot2: 802 Improving the ultrafast bootstrap approximation. Mol. Biol. Evol. 35, 518–522 (2018). 803 804 805 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 28 Supporting Information 806 807 Article title: Four enzymes control natural variation in the steroid core of Erysimum cardenolides 808 809 Authors: Gordon C. Younkin1,2, Martin L. Alani1, Tobias Züst3, Georg Jander1§ 810 811 The following Supporting Information is available for this article: 812 Figure S1 BSA results for mutant #454 813 Figure S2 EICs of mutant #454 and p5br2-1 cardenolides 814 Figure S3 MSMS spectra of cardenolide intermediates 815 Figure S4 Sequences of 3bhsd mutants 816 Figure S5 Sequences of 3ksi mutants 817 Figure S6 Sequences of p5br2 mutants 818 Figure S7 Sequences of det2 mutants 819 Figure S8 Sequences of p5br1 mutants 820 Figure S9 EICs and MSMS spectra of cardenolide genins 821 Figure S10 Genin abundances from hydrolysis Erysimum cheiranthoides mutant cardenolides 822 Figure S11 Cardenolide abundances in 3bhsd/3ksi double mutants 823 Figure S12 EICs from P5βR enzyme activity assays in Nicotiana benthamiana 824 Figure S13 Cardenolide abundances in p5br1 mutants 825 Figure S14 Multiple sequence alignment of Erysimum P5βR proteins 826 Figure S15 Multiple sequence alignment of 3βHSD coding sequences from selected species 827 Figure S16 Multiple sequence alignment of 3KSI coding sequences from selected species 828 Figure S17 Multiple sequence alignment of P5βR2 coding sequences from selected species 829 Figure S18 Multiple sequence alignment of DET2 coding sequences from selected species 830 Table S1 Cardenolide abundances in Erysimum cheiranthoides mutant lines 831 Table S2 Coexpression cluster of cardenolide biosynthesis genes 832 Table S3 Cardenolide intermediate abundances from enzyme assays 833 Table S4 Cardenolide genin abundances in Erysimum cheiranthoides mutant lines 834 Table S5 Cardenolide abundances in 3bhsd/3ksi double mutants 835 Table S6 Cardenolide abundances in p5br1 mutant 836 Table S7 Cardenolide genin abundances in Erysimum species 837 Table S8 Expression of cardenolide biosynthetic genes in Erysimum species 838 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 29 Table S9 Sequences of primers 839 Table S10 m/z and retention times used for quantifying cardenolides in LCMS data 840 Table S11 m/z and retention times used for quantifying cardenolide genins in hydrolysis 841 experiments 842 Table S12 m/z and retention times used for quantifying cardenolide intermediates in in vitro 843 assays844 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 30 845 Figure S1. Bulked segregant analysis (BSA) from Erysimum cheiranthoides mutant #454. 846 Mutant #454 was generated via ethyl methanesulfonate (EMS) mutagenesis. Alternate (mutant) 847 allele frequency, smoothed over 1 Mbp segments, is plotted across the eight E. cheiranthoides 848 chromosomes. In plants with a mutant chemotype (red), mutant alleles dominate in the latter half 849 of chromosome seven. Within this region is a progesterone reductase (EcP5βR2) with two 850 missense mutations (R184K and G201R) in mutant #454 plants. The cardenolide phenotype 851 associated with mutant #454 and the BSA results displayed here were first described in Mirzaei 852 et al. 2020. The causal mutation at the linked locus was first described in this study. 853 854 855 856 Figure S2. Extracted ion chromatograms (EIC) for cardenolides in wildtype Erysimum 857 cheiranthoides, mutant #454, and p5br2-1. EIC for m/z 375.2535, a fragment common to 858 digitoxigenin and uzarigenin glycosides, resulting from the neutral loss of all sugar moieties, 859 leaving only the genin intact. In mutant #454 and p5br2 mutant lines, we observe cardenolides 860 with the same mass as those found in wildtype plants, but they elute at different retention times, 861 suggesting that they may be structural isomers. 862 ECHEv2.0ch01 ECHEv2.0ch02 ECHEv2.0ch03 ECHEv2.0ch04 ECHEv2.0ch05 ECHEv2.0ch06 ECHEv2.0ch07 ECHEv2.0ch08 wildtye 454 mutant Alternate allele frequency in wildtype and 454 mutant, smoothed Alternate allele frequency (%) 0.0 0.2 0.4 0.6 0.8 1.0 4.9e+07 p5br2−1 3.8e+07 mutant #454 4.2 4.4 4.6 4.8 5.0 5.2 6.7e+06 wildtype EIC (m/z = 375.2535) Retention time (minutes) .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 31 863 Figure S3. Extracted ion chromatograms (EIC) and MSMS spectra from products of in 864 vitro or Nicotiana benthamiana transient expression assays compared with authentic 865 standards. EIC (A) of isoprogesterone and progesterone formed by supplying Ec3βHSD or/and 866 Ec3KSI with pregnenolone and NAD+ in vitro. (B) MSMS spectra of progesterone. High-quality 867 isoprogesterone MSMS spectrum could not be collected from in vitro assays due to low signal. 868 (C) EIC of conversion of isoprogesterone to progesterone in vitro. (D) EIC of 5β-pregnane-3,20-869 dione formed by supplying EcP5βR2 with progesterone and NADPH in vitro, and corresponding 870 MSMS spectrum (E). (F) MSMS spectrum of 5ɑ-pregnane-3,20-dione formed by EcDET2 871 following coinfiltration with progesterone in leaves of Nicotiana benthamiana. EIC (G) and 872 MSMS spectra of epipregnanolone or diastereomer formed by supplying Ec3βHSD with 5ɑ-873 pregnane-3,20-dione (H) or 5β-pregnane-3,20-dione (I) and NADH in vitro. 874 100 150 200 250 300 Mirrored MSMS spectra m/z Relative intensity 81.0701 85.0649 95.0855 107.0854 135.1165147.1164 159.1164 161.1322 171.1166 173.1323175.1478185.1318199.1474 281.2253 282.2293 299.2362 85.065 135.1167 147.1167159.1166161.1322 241.1945 281.226 282.2292 299.2365 317.2468 −100 −50 0 50 100 EcP5βR2 5β−pregnane−3,20−dione standard 100 150 200 250 300 Mirrored MSMS spectra m/z Relative intensity 135.1165 189.1634 283.2415 301.252 135.1166 283.2417 301.2525 −100 −50 0 50 100 Ec3βHSD + 5α−pregnane−3,20−dione epipregnanolone standard 100 150 200 250 300 Mirrored MSMS spectra m/z Relative intensity 135.1165 283.2416 284.2448 301.2518 135.1166 147.1167161.1323 187.148 283.2417 284.2452 301.2525 −100 −50 0 50 100 Ec3βHSD + 5β−pregnane−3,20−dione epipregnanolone standard B F H 100 150 200 250 300 Mirrored MSMS spectra m/z Relative intensity 97.0648 109.0647 315.2318 97.0648 109.0647 315.2314 −100 −50 0 50 100 Ec3βHSD + Ec3KSI + pregnenolone progesterone standard [M+H]+ (m/z 315.2324) 100 150 200 250 300 Mirrored MSMS spectra m/z Relative intensity 85.065 281.2264 299.2362 317.2089 318.2409 81.0701 85.065 95.0856 159.1167 161.1323 175.1479 241.1948 281.226 299.2366 −100 −50 0 50 100 EcDET2 + progesterone 5α−preganane−3,20−dione standard [M+H]+ (m/z 317.2481) [M-H2O+H]+ (m/z 301.2531)[M-H2O+H]+ (m/z 301.2531) 1.2e+08 Ec3βHSD 1.2e+08 Ec3KSI 6.8 7.0 7.2 7.4 7.6 7.8 1.2e+08 negative control EIC (m/z = 315.2324) Retention time (minutes) A 8.4e+05 EcP5βR2 8.4e+05 negative control 6e+06 5α−pregenane−3,20−dione 7.4 7.6 7.8 8.0 8.2 1.4e+07 5β−pregenane−3,20−dione EIC (m/z = 317.2481) Retention time (minutes) EcP5βR2 + progesterone negative control + progesterone C D G Ec3βHSD + 5β-pregnane-3,20-dione Ec3βHSD + 5ɑ-pregnane-3,20-dione 1.8e+06 1.8e+06 1.8e+06 7.0 7.2 7.4 7.6 7.8 8.0 6.2e+06 epipregnanolone EIC (m/z = 301.2531) Retention time (minutes) negative control + 5β-pregnane-3,20-dione epipregnanolone negative control + isoprogesterone 2.8e+06 Ec3βHSD 2.8e+06 Ec3KSI 2.8e+06 Ec3βHSD + Ec3KSI 2.8e+06 negative control 6.8 7.0 7.2 7.4 7.6 7.8 8.6e+06 standard EIC (m/z = 315.2324) Retention time (minutes) isoprogesterone progesterone negative control + pregnenolone E I Ec3βHSD + isoprogesterone Ec3KSI + isoprogesterone progesterone isoprogesterone Ec3βHSD + pregenenolone Ec3KSI + pregenenolone Ec3βHSD + Ec3KSI pregenenolone [M+H]+ (m/z 299.2362) + progesterone .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 32 875 Figure S4. Aligned nucleotide coding sequences of Erysimum cheiranthoides 3βHSD from 876 WT and mutant lines. Mutant #30 was generated via chemical mutagenesis with ethyl 877 methanesulfonate (EMS), and 3bhsd-1 and 3bhsd-2 lines were generated with CRISPR/Cas9. 878 Abbreviations: Erysimum cheiranthoides (Ec), wildtype (WT), 3β-hydroxysteroid 879 dehydrogenase (3βHSD). Sequences of gRNAs used for generation of these lines are available in 880 Table S1. MultAlin (http://multalin.toulouse.inra.fr/multalin/) was used to produce the alignment. 881 882 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 33 883 Figure S5. Aligned nucleotide coding sequences of Erysimum cheiranthoides 3KSI from WT 884 and mutant lines. 3ksi-1 and 3ksi-2 lines were generated with CRISPR/Cas9. Abbreviations: 885 Erysimum cheiranthoides (Ec), wildtype (WT), 3ketosteroid isomerase (3KSI). Sequences of 886 gRNAs used for generation of these lines are available in Table S1. MultAlin 887 (http://multalin.toulouse.inra.fr/multalin/) was used to produce the alignment. 888 889 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 34 890 Figure S7. Aligned nucleotide coding sequences of Erysimum cheiranthoides DET2 from 891 WT and mutant lines. det2-1 and det2-2 lines were generated with CRISPR/Cas9 in the WT 892 background. det2/p5br2-1 and det2/p5br2-2 lines were generated with CRISPR/Cas9 in the 893 p5br2-1 background. Abbreviations: Erysimum cheiranthoides (Ec), wildtype (WT), 894 progesterone 5β-reductase (P5βR), steroid 5α-reductase (DET2). Sequences of gRNAs used for 895 generation of these lines are available in Table S1. MultAlin 896 (http://multalin.toulouse.inra.fr/multalin/) was used to produce the alignment. 897 898 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 35 899 Figure S8. Aligned nucleotide coding sequences of Erysimum cheiranthoides P5βR1 from 900 WT and mutant lines. The p5br1-1 line was generated with CRISPR/Cas9. Abbreviations: 901 Erysimum cheiranthoides (Ec), wildtype (WT), progesterone 5β-reductase (P5βR). Sequences of 902 gRNAs used for generation of these lines are available in Table S1. MultAlin 903 (http://multalin.toulouse.inra.fr/multalin/) was used to produce the alignment. 904 905 906 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 36 907 Figure S9. Extracted ion chromatograms and MSMS spectra for cardenolide genins. 908 Selected extraction ion chromatograms (EIC) and MSMS spectra for cardenolide genins from 909 hydrolyzed extracts of Erysimum cheiranthoides wildtype (WT) and mutant leaves, compared to 910 authentic standards where available. (A) EIC of digitoxigenin and uzarigenin [M+H]+, which are 911 stereoisomers and are separated by retention time. An uzarigenin standard was not available; 912 hydrolyzed Calotropis procera leaf extract, which are known to contain uzarigenin, was used 913 instead. (B) Presumed xysmalogenin [M+H]+ is abundant in p5br2/det2 double mutants 914 compared to WT leaves. 3,5-anhydroperiplogenin [M+H]+ (C) is formed when canarigenin 915 glycosides are subjected to acidic conditions in p5br2/det2 double mutants. (D-G) Corresponding 916 MSMS spectra. Authentic standards were not available for xysmalogenin (F) or 3,5-917 anhydroperiplogenin (G). Instead, they are compared to a digitoxigenin MSMS spectrum. The 918 spectra are similar, but some peaks are shifted by 2 Daltons, providing further evidence that these 919 peaks represent dehydrocardenolides. (H) Cannogenol [M+Na]+ in E. cheiranthoides, and an 920 isomer in E. pieninicum that may be the 5α conformation. (I) Strophanthidin [M+Na]+ in E. 921 cheiranthoides compared to an authentic standard. 922 923 100 150 200 250 300 350 Mirrored MSMS spectra m/z Relative intensity 81.07 95.0855105.0697 107.0854 109.101 121.1009 133.1008 135.1165 145.101147.1166 159.1164 161.1322 185.1323187.1479 231.1381 257.1527 293.2251 321.2203 339.231 340.2343 357.2415 81.07 95.0854 107.0854121.1009 135.1165 147.1165161.1321 187.1478 231.1374 257.1529 293.2256 321.2207 339.2311 340.2347 357.2416 −100 −50 0 50 100 WT digitoxigenin standard 100 150 200 250 300 350 Mirrored MSMS spectra m/z Relative intensity 79.0543 81.07 93.069895.0855 105.0696 107.0853 121.1008 133.1008 135.1165 145.1009147.1164159.1165 161.132 185.132 187.1477 199.1477201.1631 205.1219 231.1374 279.2098 293.2256 321.2207 339.2312 340.2342 357.2413 358.2456 81.07 93.07 95.0855 105.0699 107.0854 119.0851 121.101 131.0854 133.1009 135.1165 145.1009147.1166159.1163 161.1321 185.1324 187.1479 199.1473 201.1632 205.1219 231.1376 245.153257.1532 279.2097 293.2259 321.2205 322.2237 339.2314 340.2347 357.2421 358.245 373.2375 374.2408 −100 −50 0 50 100 p5br2−1 Calotropis 50 100 150 200 250 300 350 Mirrored MSMS spectra m/z Relative intensity 67.0546 71.0494 81.07 93.0698 95.0854 103.0752 105.0697 107.0852 109.1011121.1009131.0853 133.101 135.1165145.1008 159.1167 261.2206 319.2055 337.2159 357.26 81.07 95.0854 107.0854121.1009 135.1165 147.1165161.1321 187.1478 231.1374 257.1529 293.2256 321.2207 339.2311 340.2347357.2416 −100 −50 0 50 100 p5br2/det2−2 digitoxigenin standard 100 150 200 250 300 350 Mirrored MSMS spectra m/z Relative intensity 81.07 95.0854105.0696107.0853 109.1009119.0852131.0853 133.1009 137.0957 145.1009 147.1166 157.101 159.1165 161.1322 185.1319 291.2094 319.2051 337.2155 339.2313 355.2261 81.07 95.0854 107.0854121.1009 135.1165 147.1165161.1321 187.1478 231.1374 257.1529 293.2256 321.2207 339.2311 340.2347 357.2416 −100 −50 0 50 100 p5br2/det2−2 digitoxigenin standard digitoxigenin [M+H]+ m/z 375.2529, RT 12.81 m/z m/z m/z m/z uzarigenin [M+H]+ m/z 375.2529, RT 12.93 presumed xysmalogenin [M+H]+ m/z 373.2379, RT 12.30 presumed 3,5-anhydroperiplogenin [M+H]+ m/z 355.2273, RT 15.99 A B C D E F G 5.3e+06 WT 6.8e+07 p5br2−1 2e+08 digitoxigenin 12.6 12.8 13.0 13.2 13.4 2.3e+06 Calotropis EIC (m/z = 375.2529) Retention time (minutes) digitoxigenin uzarigenin 3.4e+07 p5br2/det2−2 12.0 12.4 12.8 3.4e+07 WT EIC (m/z = 373.2379) Retention time (minutes) presumed xysmalogenin 4.1e+07 p5br2/det2−2 15.8 16.2 16.6 4.1e+07 WT EIC (m/z = 355.2273) Retention time (minutes) presumed 3,5-anhydroperiplogenin 1.6e+06 E. cheiranthoides 9.0 10.5 12.0 5.3e+05 E. pieninicum EIC (m/z = 413.2304) Retention time (minutes) 2.8e+07 E. cheiranthoides 9.0 10.0 9.2e+06 strophanthidin EIC (m/z = 427.2097) Retention time (minutes) H I presumed cannogenol cannogenol isomer .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 37 924 Figure S10. Normalized peak areas for cardenolide genins from hydrolyzed leaf extracts of 925 Erysimum cheiranthoides mutant lines. (A) Digitoxigenin, (B) cannogenol, (C) cannogenin, 926 (D) strophanthidin, (E) total 5β-cardenolides (sum of digitoxigenin, cannogenol, cannogenin, and 927 strophanthidin), (F) 5α-cardenolides (uzarigenin), (G) Δ4-cardenolides (dianhydroperiplogenin 928 as a proxy for canirigenin), (H) Δ5-cardenolides (xysmalogenin). Abbreviations: wildtype (WT), 929 3β-hydroxysteroid dehydrogenase (3bhsd), 3-ketosteroid isomerase (3ksi), progesterone 5β-930 reductase (p5br), and steroid 5α-reductase (det2). Error bars are ± s.d. Letters indicate P<0.05, 931 one-way ANOV A with post-hoc Tukey’s HSD test. 932 933 Δ5−cardenolides 0 20 40 60 80 100 120Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 e a ab bc c de de e d Δ4−cardenolides 0 20 40 60 80 100 120Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 b b b b b b b b a 5α−cardenolides 0 20 40 60 80 100Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 b b b b b a a b b 5β−cardenolides 0 20 40 60 80 100Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 a d d c c e e b e A B C D E F G H Digitoxigenin 0 20 40 60 80 100Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 b c c c c c c a c cannogenol 0 20 40 60 80 100Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 a cd cd c c d d b d cannogenin 0 20 40 60 80 100Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 a c c b b d d b d strophanthidin 0 20 40 60 80 100Normalized peak area WT 3bhsd 3ksi p5br det2 det2/p5br 1 2 1 2 1 2 a d d c c e e b e .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 38 934 Figure S11. Cardenolide analysis from Ec3βHSD/Ec3KSI double mutants. (A) PCA of 935 cardenolides detected in Erysimum cheiranthoides wildtype, 3bhsd, 3ksi, and 3bhsd/3ksi mutant 936 lines. Normalized peak area of (B) total 5β-cardenolides and (C) total dehydrocardenolides. (D) 937 Ratio of dehydrocardenolide peak area to 5β-cardenolide peak area. Abbreviations: wildtype 938 (WT), 3β-hydroxysteroid dehydrogenase (3bhsd), 3-ketosteroid isomerase (3ksi). N=3 plants per 939 line. Error bars are ± s.d. Letters indicate P<0.05, one-way ANOV A with post-hoc Tukey’s HSD 940 test. 941 942 −6 −4 −2 0 2 −1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 PCA of all peaks (normalized by internal standard) PC1 (81.9%) PC2 (7.7%) → → wildtype 3bhsd−1 3ksi−2 3bhsd/3ksi 5β−cardenolides dehydrocardenolides A dehydrocardenolides 0 2 4 6 8 10 12 Normalized peak area WT 3bhsd −1 3ksi −2 3bhsd/3ksi b a a a 5β−cardenolides 0 1 2 3 4 5 6 Normalized peak area WT 3bhsd −1 3ksi −2 3bhsd/3ksi a c b c Ratio of dehydrocardenolides to 5 βcardenolides 0 2 4 6 8 10 12 14 Normalized peak area WT 3bhsd −1 3ksi −2 3bhsd/3ksi c a b a Ratio of dehydrocardenolide to 5β-cardenolide peak area B C DPeak area ratio .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 39 943 Figure S12. Coinfiltration of Erysimum progesterone 5β-reductases and progesterone in 944 Nicotiana benthamiana leaves. Erysimum progesterone 5β-reductases (P5βR) cloned into 945 pEAQ-HT-DEST1 were infiltrated into leaves of N. benthamiana (Nb), followed by infiltration 946 of progesterone after three days. Extracted ion chromatograms (EIC) at m/z=317.2481 show 947 production of 5β-pregnane-3,20-dione. When progesterone is infiltrated into N. benthamiana 948 leaves with no co-infiltrated enzyme, 5α-pregnane-3,20-dione is produced by endogenous N. 949 benthamiana enzymes. (A) P5βR2 orthologs from selected species of Erysimum. E. 950 hieraciifolium and Erysimum sp. 1 P5βR2 proteins are truncated by a premature stop codon and 951 are non-functional in this assay. (B) P5βR1 and P5βR2 from E. cheiranthoides (Ec). The 952 retention time disparity between the two experiments is due to a shorter LCMS method used in 953 panel B. 954 955 5.2e+06 Nb+ progesterone 5.2e+06 E. kotschyanum 5.2e+06 E. crassipes 5.2e+06 E. rhaeticum 5.2e+06 E. pulchellum 5.2e+06 E. hieraciifolium 5.2e+06 Erysimum sp. 1 5.2e+06 E. cheiranthoides 3.1e+07 5β−pregnane−3,20−dione 7.4 7.5 7.6 7.7 7.8 7.9 8.0 4.7e+06 5α−pregnane−3,20−dione EIC (m/z = 317.2481) Retention time (minutes) 2.7e+06 EcP5βR1 2.7e+06 EcP5βR2 2.6e+06 5β−pregnane−3,20−dione 5.5 5.6 5.7 5.8 5.9 6.0 6.1 6.2 5.3e+06 5α−pregnane−3,20−dione EIC (m/z = 317.2481) Retention time (minutes) A B .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 40 956 Figure S13. Cardenolide analysis from EcP5βR1 mutants. (A) PCA of cardenolides detected 957 in Erysimum cheiranthoides wildtype (WT) and p5br1 (progesterone 5β-reductase 1) mutant 958 line. (B) Normalized peak area of total cardenolides. All cardenolides in this experiment are also 959 found in WT E. cheiranthoides and are presumed to be 5β-cardenolides. No dehydrocardenolides 960 were detected. No differences were detected between groups (one way ANOV A: P = 0.957). 961 N=3. Error bars are ± s.d. 962 963 A B −4 −2 0 2 4 −1.5 −1.0 −0.5 0.0 0.5 1.0 PCA of all peaks (normalized by internal standard) PC1 (85.7%) PC2 (8.4%) → wildtype p5br1−1 5β−cardenolides 5β−cardenolides 0 20 40 60 80 100 120Normalized peak area WT p5br1 −1 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 41 Figure S14. Multiple sequence alignment of progesterone 5β-reductases (P5βR) from 964 Erysimum species. All orthologs of EcP5βR2 that could be recovered from transcriptome data 965 are included, and EcP5βR1 (ECE_2g27660) is included as an outgroup. Translated proteins were 966 aligned using Clustal Omega. Species included: E. crassipes (CSS), E. cheiranthoides (ECE), 967 Erysimum sp. 1 (ER1), Erysimum sp. 2 (ER2), Erysimum sp. 3 (ER3), Erysimum sp. 4 (ER4), E. 968 hieraciifolium (HIE), E. hungaricum (HUN), E. kotschyanum (KOT), E. pieninicum (PIE), E. 969 pseudorhaeticum (PSE), E. pulchellum (PUL), E. rhaeticum (RHA), E. sylvestre (SYL), E. 970 virgatum (VIR). Sequences correspond to protein phylogeny in Figure 5c. 971 972 CLUSTAL O(1.2.4) multiple sequence alignment 973 974 975 ECE_2g27660 ---------------------------------------MSWWWAGAIGAAKKKLDDDEP 976 HIE -----------------------------------------------MESESG--ILMRR 977 VIR -----------------------------------------------MESENG--ILMRR 978 PUL -----------------------------------------------MESEKG--ILMGR 979 PIE -----------------------------------------------MESESG--ILMRR 980 HUN -----------------------------------------------MESESG--ILMRR 981 ECE_6g7150 -----------------------------------------------MESESG--ILMRR 982 SYL -----------------------------------------------MESESG--ILMRR 983 ER4 -----------------------------------------------MESESG--ILMRR 984 ER3 -----------------------------------------------MESESG--ILMRR 985 ER1_1 -----------------------------------------------MESESG--ILMRR 986 ER1_2 -------------------------------------------------- MAS--SMMRR 987 KOT MNTTIITRSSITTIITSYSLHISYLFSQSTYNNRLSSFHIKYIFPLIPFSMAS --SMMRR 988 CSS -------------------------------------------------- MAS--SMMRR 989 RHA -------------------------------------------------- MAS--SMMRR 990 PSE -------------------------------------------------- MAS--SMMRR 991 ER2 -------------------------------------------------- MAS--SMMRR 992 993 994 ECE_2g27660 TQSYESVALIIGVTGIVGNSLAEILPLSDTPGGPWKVYGVARRPRPSWNADHPIDYIQCD 995 HIE NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 996 VIR NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MYNFISCD 997 PUL NEVDENVALIFGVTGLVGREIVKTLL---TSEPRWKIYGVARNPEINS----MYDFISCD 998 PIE NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 999 HUN NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 1000 ECE_6g7150 NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 1001 SYL NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 1002 ER4 NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 1003 ASP NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 1004 ER1_1 NEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS----MCNFISCD 1005 ER1_2 SEVDENVALIFGVTGLVGRAIVKTLL---TSEPRWKIYGVARKPEINS-MAEMYNFISCD 1006 KOT SEVDENVALIFGVTGLVGREIVKTLL---TSEPRWKIYGVARNPEINS----MYNFISCD 1007 CSS SEVDENVALIFGATGLVGREIVKTLL---TSEPRWKIYGVARKPETNS-MAEMYNFISCD 1008 RHA SEVEENVALIFGVTGLIGREIVKTLL---TSEPRWKIYGVARKPEISS-MAEMYSFISCD 1009 PSE SEVDENVALIFGVTGLIGREIVKTLL---TSEPRWKIYGVARKPEISS-MAEMYSFISCD 1010 ER2 SEVEENVALIFGVTGLIGREIVKTLL---TSEPRWKIYGVARKPEISS-MAEMYSFISCD 1011 .: *.****:*.**::*. :.: * * **:*****.*. . .:*.** 1012 1013 ECE_2g27660 VSNAEDARSKLSPLTDV-THVFYVTWTNRESE-S-ENCEANGSMIR--------NVLQAI 1014 HIE LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1015 VIR LLNASETKQKLSPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLTNALD --------AI 1016 PUL LLNASETKQKLSPLQDIVSHVFWVTWSGESPLDTDECCVQNKTMLTNALD --------AI 1017 PIE LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1018 HUN LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1019 ECE_6g7150 LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1020 SYL LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1021 ER4 LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1022 ASP LLNASETKQKLTPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1023 ER1_1 LLNASETKQKLSPLQDIVSHVFWVTWSGEYPLDSDECCVQNKTMLSNALD --------AI 1024 ER1_2 LLNASETKQKLSPLQDIVSHVFWVTWCGESPLDSDECSVQNKTMLTNALD --------AI 1025 KOT LLNASETKQKLSPLQDIVSHVFWVTWCGESPLDSDECSVQNKTMLTNALDAILPNALDAI 1026 CSS LLNASETKQKLSPLQDIVSHVFWVTWCGESPLDSDECSVQNKTMLTNALD --------AI 1027 RHA LLNASETKQKLSPLQDIVSHVFWVTWCGESPLDTDECSVQNKTMLTNALD --------AI 1028 PSE LLNASETKQKLSPLQDIVSHVFWVTWCGESPLDTDECSVQNKTMLTNALD --------AI 1029 ER2 LLNASETKQKLSPLQDIVSHVFWVTWCGESPLDTDECSVQNKTMLTNALD --------AI 1030 : **.:::.**:** *: :***:*** .. : * . * :*: ** 1031 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 42 1032 ECE_2g27660 VPHAPNLRHICLQTGTKHYVGPFSNLGGGPRHDPPFTEDMPRLQ -IQNFYYTQEDILFEE 1033 HIE LPNAKRLQHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1034 VIR LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1035 PUL LPNSKRLKHFSLQTGMHHYG----EAHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1036 PIE LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1037 HUN LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1038 ECE_6g7150 LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1039 SYL LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1040 ER4 LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1041 ASP LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1042 ER1_1 LPNAKRLKHFSLQTGMHHYS----ESHGEGSSLCYYSEECPRKSSGKNFYYVLEDLLEE- 1043 ER1_2 LPNAKRLKHFSLQTGMKHYLPLVP-SHREGSSLCYYSEESPRQSSRKNFYYVLEDLLEE- 1044 KOT LPNARRLKHFSLQTGMKYYVPLVPPSHREGSSLCYYTEESPRQTSRKNFYYVLEDLLEE - 1045 CSS LPNAKRLKHFSLQTGMKYYVPLVP-SHREGSSLCYYSEESPRQSSRNNFYYVLEDLLEE- 1046 RHA LPNAKRLKHFSLQTGMRYYVPLVP-SHREGSSLCYYTEESPRQSSRKNFYYVLEDLMEE- 1047 PSE LPNAKRLKHFSLQTGMRYYVPLNP-SHREGSSLCYYTEESPRQISRKNFYYVLEDLMEE- 1048 ER2 LPNAKRLKHFSLQTGMRYYVPLNP-SHREGSSLCYYTEESPRQISRKNFYYVLEDLMEE- 1049 :*:: .*:*:.**** ::* ::*: ** :****. **:: * 1050 1051 ECE_2g27660 IKKKESVTWSIHRPNTIFGFSPYSLMNIVGTLCVYAAICKHEGSPLLFPGSKKAWEG -FT 1052 HIE KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1053 VIR KISGTSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1054 PUL KISGNSVVWSVQRPGLLLGSSTRTLNNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1055 PIE KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1056 HUN KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1057 ECE_6g7150 KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1058 SYL KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1059 ER4 KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1060 ASP KISGNSVVWSVQRPGLLLGSSTRTLKNFMGILCVYGAMCKYLNLPFVFGGTRECWEESYI 1061 ER1_1 KISGTSVVWSVQRPGLLLGSSTRTLKNFMGILCLWSNV---------------------- 1062 ER1_2 KISGTSVVWSVQRPGLLLGSSTRTLKNFMGILCLWSNV---------------------- 1063 KOT KLSGNSVVWSVQRPCLLMGSSSRTLYNFMGSLCVYGAMCKYLNLPFVFGGTRECWEESYI 1064 CSS KISGNSVVWSVQRPCLLMGSSSRTLYNFMGSLCVYGAMCKYLNLPFVFGGTRECWEESYI 1065 RHA KISGNSVVWSVQRPCLLMGSSSRTLYNFMGSLCVYGAMCKYLNLPFVFGGTRECWEESYI 1066 PSE KISGNSVVWSVQRPCLLMGSSSRTLYNFMGSLCVYGAMCKYLNLPFVFGGTRECWEESYI 1067 ER2 KISGNSVVWSVQRPCLLMGSSSRTLYNFMGSLCVYGAMCKYLNLPFVFGGTRECWEESYI 1068 . **.**::** ::* * :* *::* **::. : 1069 1070 ECE_2g27660 TASDADLIAEQQIWAAVDPY--AKNEAFNCNNADIFKWKHLWKILAEQFGIEQYGF---- 1071 HIE DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGLGFTWKEIWPDIGRKLGVQVNETTMFD 1072 VIR DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVEFTWKEIWPDIGRKLGVQVNETTMFD 1073 PUL DGSDANLVAEQHIFAATSGKVRNRGEAFNAINGVGFTWKEIWPELGRKLGVQVNETTMFD 1074 PIE DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVGFTWKEIWPGGNLVCKLMKRRCLMKI 1075 HUN DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVGFTWKEIWPGGNLVCKLMKRRCLMKI 1076 ECE_6g7150 DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVGFTWKEIWPDIGRKLGVQVNETTMFD 1077 SYL DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVGFTWKEIWPDIGRKLGVQVNETTMFD 1078 ER4 DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVGFTWKEIWPDIGRKLGVQVNETTMFD 1079 ASP DGSDANLVAEQHIFAATSGKVRNRGEAFNSINGVGFTWKEIWPDIGRKLGVQVNETTMFD 1080 ER1_1 ------------------------------------------------------------ 1081 ER1_2 ------------------------------------------------------------ 1082 KOT HGSDANLVAEQHIFAATSGKVRNRGEAFNAINGDGFTWKEIWPEIGRKFGVQVNETTMFD 1083 CSS HGSDANLVAEQHIFAATSGKVRNRGEAFNAINGDGFTWKEIWPELGRKFGVQVNETTMFD 1084 RHA DGSDANLVAEQHIFAATSGKVRNRGEAFNAINGDAFTWKEIWPELGKKLGVQVNETTMFD 1085 PSE DGSDANLVAEQHIFAATSGKVRNRGEAFNAINGDAFTWKEIWPELGRKLGVQVNETTMFD 1086 ER2 DGSDANLVAEQHIFAATSGKVRNRGEAFNAINGDAFTWKEIWPELGRKLGVQVNETTMFD 1087 1088 1089 ECE_2g27660 ------EEGKNLG--LVEMMKGKERVWEEMVKENQLQEKKLDE--VGVWWFADVILGVEG 1090 HIE ---ENFWYGREM----ERENMCGTRLW--------------------------------- 1091 VIR ---ENFWYGREMGERWERENMCGTRLW--------------------------------- 1092 PUL ---ENFWYGREM--------GERKHVWDEIVVKERLVRTEIEDLA--NWVFLDGLFRCPY 1093 PIE SGMGERWEREN---------MCGTRLW--------------------------------- 1094 HUN SGMGERWEREN---------MCGTRLW--------------------------------- 1095 ECE_6g7150 ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDLA--NWVFLDGLFRCPF 1096 SYL ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDLA--NWVFLDGLFRCPF 1097 ER4 ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDLA--NWVFLDGLFRCPF 1098 ASP ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDLA--NWVFLDGLFRCPF 1099 ER1_1 ------------------------------------------------------------ 1100 ER1_2 ------------------------------------------------------------ 1101 KOT ---ENFWYGREM--------GERKHVWDEIVVKERLVRTEIEDLAKGNWVFLDGLFRMRH 1102 CSS ---ENFWYGREM--------GERKHVWDEIVVKERLVRTEIEDLAKGNWVFLDGLFRMRH 1103 RHA ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDMAKGNWVFLDGLFRMRH 1104 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 43 PSE ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDMAKGNWVFLDGLFRMRH 1105 ER2 ---ENFWYGREM--------GERKHVWDEIVVKEGLVRTEIEDLAKGNWVFLDGLFRMRH 1106 1107 1108 ECE_2g27660 -MIDSMNKSKEHGFLGFRNSNNSFISWIDKYKAFKIVP- 1109 HIE --------------------------------------- 1110 VIR --------------------------------------- 1111 PUL KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1112 PIE --------------------------------------- 1113 HUN --------------------------------------- 1114 ECE_6g7150 KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1115 SYL KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1116 ER4 KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1117 ASP KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMR------- 1118 ER1_1 --------------------------------------- 1119 ER1_2 --------------------------------------- 1120 KOT KLLGKRDKVDRFGFTRKCRTLDSILYWIDVMRDEKLIPL 1121 CSS KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1122 RHA KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1123 PSE KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPL 1124 ER2 KLLGKRDKVDRFGFKRKCRTLDSILYWIDVMRDEKLIPF 1125 1126 1127 1128 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 44 Figure S15. Multiple sequence alignment of 3β-hydroxysteroid dehydrogenases (3βHSD). 1129 Coding sequences were aligned using Clustal Omega. Species included: Arabidopsis thaliana 1130 (AT/At), Erysimum cheiranthoides (Ec/Erche), Erysimum crepidifolium (Ecre), and Digitalis 1131 lanata (Dl). Sequences correspond to gene phylogeny in Figure 6a. 1132 1133 CLUSTAL O(1.2.4) multiple sequence alignment 1134 1135 1136 AT3G51680 ATGCCTGCCC-----------------------------AAGTGATCGCTGAGCAGACCA 31 1137 Dl3BHSD1 ------------------------------------------------------------ 0 1138 Dl3BHSD2 ------------------------------------------------------------ 0 1139 AtSDR5_AT2G47140 ------------------------------------------------------------ 0 1140 Erche07g001535 ------------------------------------------------------------ 0 1141 Ecre_5540 ------------------------------------------------------------ 0 1142 AT2G47150 ------------------------------------------------------------ 0 1143 AT2G47120 ------------------------------------------------------------ 0 1144 Ecre_7213 ------------------------------------------------------------ 0 1145 AtSR3_AT2G471430 ------------------------------------------------------------ 0 1146 AtSDR4_AT3G29250 ATGCCAAAACTCTTGGTTATATTTAACATTATATATCCTTTGCAATCCCTCAAGAAGGAT 60 1147 Ecre_9065 ------------------------------------------------------------ 0 1148 AT3G29260 ------------------------------------------------------------ 0 1149 1150 1151 AT3G51680 CCTTTCACTCCGTCCACG-------ACACCATTATGGAGGAGACGAATACAACTTTATAT 84 1152 Dl3BHSD1 --------------------------------------------------- ATGTCGTCA 9 1153 Dl3BHSD2 ------------------------------------------------------ ATGTCA 6 1154 AtSDR5_AT2G47140 ------------------------------------------------------ ATGTCT 6 1155 Erche07g001535 ------------------------------------------------------ ATGCTT 6 1156 Ecre_5540 ------------------------------------------------------ atgcct 6 1157 AT2G47150 ------------------------------------------------------------ 0 1158 AT2G47120 --------------------------------------------------- ATGTCGGGA 9 1159 Ecre_7213 ------------------------------------------------------ atgtcg 6 1160 AtSR3_AT2G471430 ------------------------------------------------------ ATGTCG 6 1161 AtSDR4_AT3G29250 ACATTCAATTTTTGGTGGTCTTTTCAAAAGTGGAAATACAACAAAACAATGTCGGGACTT 120 1162 Ecre_9065 --------------------------------------------------- atgtccgga 9 1163 AT3G29260 ------------------------------------------------------ ATGTCG 6 1164 1165 1166 AT3G51680 CCTAAGAGGTTGGAAGGAAAAGTAGCCATCATAACCGGAGGCGCACATGGAATAGGCAAA 144 1167 Dl3BHSD1 AAGCCAAGGTTGGAGGGTAAAGTGGCAATCATCACCGGAGCCGCTAGCGGCATCGGCGAG 69 1168 Dl3BHSD2 AAGCAAAGGTTGGAAGGCAAAGTCGCAATCGTCACCGGCGCTGCTAGCGGTATCGGCGAG 66 1169 AtSDR5_AT2G47140 GGAAAAAGATTGGATGGCAAAATCGTAATTATTACTGGCGGAGCAAGCGGGATTGGGGCT 66 1170 Erche07g001535 GGAAACACATTGGATGGCAAAATCGTAATTATCACCGGTGGAGCAAGCGGGATTGGTGCT 66 1171 Ecre_5540 ggaaacatgttggatggcaaaatcgttattatcaccggtggagcaagcgggattggtgct 66 1172 AT2G47150 ---------ATGGATGGCAAAATCGTGATTATAACAGGGGGAGCCAGTGGGACAGGAGCC 51 1173 AT2G47120 CTCAGTAGATTGGAAGGCAAGATCGTAATTATAACAGGCGGAGCCAGCGGGATTGGAGCC 69 1174 Ecre_7213 ggactcagattggatggcaaaatcgtaattataacaggcggagccagcgggattggagcc 66 1175 AtSR3_AT2G471430 GGACTCAGATTGGATGGCAAAATCGCAATAATAACAGGCGGAGCTAGCGGGATTGGAGCC 66 1176 AtSDR4_AT3G29250 AGGCAAGTATTGGATGGCAAGATCGCAATTATAACAGGCGGAGCTAGTGGGATTGGAGCT 180 1177 Ecre_9065 ctcggaagattagatggcaagatcgcaattataacaggtggagctagcgggattggagcg 69 1178 AT3G29260 GGACAAAGATTGGATGGCAAGATCGTAATTATAACAGGCGGGGCTAGTGGGATCGGAGCT 66 1179 * ** ** ** * * ** * ** ** * ** ** * ** 1180 1181 AT3G51680 GCAACCGTCATGTTATTCGCTAGACACGGTGCCACAGTGGTGATTGCTGACGTGGACAAC 204 1182 Dl3BHSD1 GAGACGGCAAGATTGTTCGTGGAGCATGGCGCCTCAGTGGTGGTGGCGGACGTCCAGGAC 129 1183 Dl3BHSD2 GCGGCAGCCAGACTATTTATCGAGCATGGCGCCAGAGTGGTGGTGGCTGACATACAAGAC 126 1184 AtSDR5_AT2G47140 GAGTCCGTTCGATTATTTACCGAACACGGCGCTCGAGTCGTGATCGTTGATGTACAGGAT 126 1185 Erche07g001535 GAGTCCGTCAGGCTATTTACCGACCATGGCGCTCGAGTCGTGATAGTTGACATACAAGAC 126 1186 Ecre_5540 gagtccgtcaggctatttaccgaccatggcgctcgagtcgtcatagttgacatacaagac 126 1187 AT2G47150 GAATCCGCTAGGTTGTTCACTGACCACGGAGCTCAAGTGGTTGTAGTTGACTTACAAGAA 111 1188 AT2G47120 GACGCCGCTAGGCTGTTCACGGACCACGGAGCTAAGGTGGTTATAGTTGACGTACAAGAA 129 1189 Ecre_7213 gaagcggctaggctgttcacggaccacggagctaaggtgatcatcgttgacttacaagaa 126 1190 AtSR3_AT2G471430 GAAGCGGTTAGGCTGTTCACGGACCACGGAGCTAAGGTGGTCATCGTTGACTTTCAAGAA 126 1191 AtSDR4_AT3G29250 GAAGCGGTTAGGTTGTTCACGGACCATGGAGCTAAAGTGGTCATCGTGGACATACAAGAA 240 1192 Ecre_9065 gaagcggctaggttgttcacggatcatggagctaaagtggttatcgttgacgtacaagaa 129 1193 AT3G29260 GAAGCAGCGAGGTTGTTCACGGACCATGGAGCTAAAGTTGTCATCGTTGACTTACAAGAG 126 1194 * * * * ** ** ** ** ** * * * ** * * * 1195 1196 AT3G51680 GTAGCTGGCTCTTCCCTGGCTAAGTCACTCTCATCCCACAAAACCTCCCCGATGGTGGCA 264 1197 Dl3BHSD1 GAATTGGGGCGCCAGGTCGTCGCTTCCGTAAACT ---------CTGACGACAAGATAAGT 180 1198 Dl3BHSD2 GAATTAGCCCAAAACGTCATTTCCTCGATTAA ------------CTCCGATAAGGTCAGC 174 1199 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 45 AtSDR5_AT2G47140 GAGCTCGGTCAAAACGTTGCAGTTTCGATCGG ------------TGAAGACAAAGCGAGT 174 1200 Erche07g001535 GAGCTCGGTCAAAGCGTTGCCGTTTCGGTCGG ------------AGAAGACAAGGCGACT 174 1201 Ecre_5540 gagctcggtcaaaacgttgccgtttcggtcgg ------------agaagacaaggcgact 174 1202 AT2G47150 GAGCAAGGCA-AAACGTCGCCGTTTCAATCGG------------CAAAGACAGAGCAAGT 158 1203 AT2G47120 GAACTAGGCCAAAACGTCGCAGTTTTGATCGG ------------GAAAGACAAAGCTAGT 177 1204 Ecre_7213 gagcttggtcaaaacttagctgtttccatcgg ------------gctagagaaagcaacg 174 1205 AtSR3_AT2G471430 GAACTTGGTCAAAACGTTGCCGTTTCTGTCGG ------------GAAAGACAAAGCAAGT 174 1206 AtSDR4_AT3G29250 GAGCTTGGCCAAAACCTCGCCGTTTCGATAGG ------------GCTAGACAAAGCAAGT 288 1207 Ecre_9065 gagcttgctcaaaacgtcgccgtttcaatcgg ------------gctagacaaagctagt 177 1208 AT3G29260 GAGCTTGGTCAAAACGTCGCCGTTTCGATCGG ------------GCTAGACAAAGCAAGT 174 1209 * * * * * * 1210 1211 AT3G51680 TTCATTAGCTGCGATGTCTCCGTAGAAGCCGACGTGGAAAACCTTGTGAACGTAACCGTT 324 1212 Dl3BHSD1 TACTACCACTGCGACGTCAGAGATGAAAAACAAGTGGCGGCCACCGTCCGCTACGCGGTG 240 1213 Dl3BHSD2 TACTTTCACTGTGACGTGAGAGATGAGAAACAAGTGGCGGATACCGTCAACTACGCCACC 234 1214 AtSDR5_AT2G47140 TACTATCATTGCGATGTCACGAACGAGACGGAAGTTGAAAACGCCGTTAAGTTCACCGTC 234 1215 Erche07g001535 TACTATCATTGCGATGTTACCAACGAGACGGAAGTTGAAAACGCCGTTAAGTTCACCGTC 234 1216 Ecre_5540 tactatcattgcgatgttaccaacgagacggaagttgaaaacgccgttaagttcaccgtc 234 1217 AT2G47150 TTTTACCGTTGTGATGTTACAAACGAGACGGA ---------------------------- 190 1218 AT2G47120 TTTTACCGTTGCGATGTGACAAATGAGACGGAAGTCGAGGACGCAGTTAAGTTCACCGTC 237 1219 Ecre_7213 ttttaccgttgcgatataacaaaagaaacggaggtagagaacgccgttaagttcaccgtc 234 1220 AtSR3_AT2G471430 TTTTACCGTTGCGATGTTACAAACGAAAAGGAGGTAGAAAACGCCGTTAAGTTCACCGTC 234 1221 AtSDR4_AT3G29250 TTTTACCGTTGTAATGTAACCGACGAGACGGATGTCGAGAACGCCGTTAAGTTCACCGTT 348 1222 Ecre_9065 ttttaccgttgcgatataacggatcagatggaggtagagaatgccgttaaattcacagtt 237 1223 AT3G29260 TTTTACCGTTGCGATATAACGGATGAGACGGAAGTTGAGAACGCCGTTAAGTTCACCGTT 234 1224 * ** * * * * 1225 1226 AT3G51680 GCACGGTACGGTAGGCTTGACATTCTATTCAACAACGCGGGAGTTCTCGGAGATCAGAAG 384 1227 Dl3BHSD1 GAGAAATACGGGCGCCTCGACATCATGCTGAGCAACGCCGGAGTCTTCGGGGCCTTGATG 300 1228 Dl3BHSD2 GCAAAATACGGCACTCTCGACATAATGTTTAGCAACGCGGCAGTCCTCGGACCCGATATA 294 1229 AtSDR5_AT2G47140 GAAAAATATGGGAAGCTTGACGTTCTGTTTAGTAACGCCGGCGTAATAGAGCCGTTTGTG 294 1230 Erche07g001535 GAAAAACACGGGAAGCTTGACGTTCTGCTCAGTAACGCTGGCGTTATAGATCCATTTACG 294 1231 Ecre_5540 gaaaaacacgggaagcttgacgttctgttcagtaacgccggcgttatagaaacatttacg 294 1232 AT2G47150 -------------------------------ACCAACCCGGCGTCTTGGAAACACCGGGA 219 1233 AT2G47120 GAAAAACACGGAAAGCTAGACGTTCTGTTTAGCAACGCCGGAGTCTTGGAACCACTGGAA 297 1234 Ecre_7213 gaaaaatacggtaagcttgacgttctgtttagtaacgctggcgttatggaacagccagga 294 1235 AtSR3_AT2G471430 GAAAAGTACGGGAAGCTTGACGTTCTCTTTAGTAACGCCGGCGTTATGGAACAGCCGGGA 294 1236 AtSDR4_AT3G29250 GAAAAACACGGAAAGCTTGACGTTCTGTTTAGTAACGCCGGGGTCTTGGAAGCGTTTGGA 408 1237 Ecre_9065 gaaaaacatggtaagcttgacgttctgtttagtaacgccggcatcatggaacggccggga 297 1238 AT3G29260 GAAAAACACGGAAAGCTTGACGTTCTGTTTAGTAACGCCGGGGTCATGGAACCGCATGGA 294 1239 * * * * * * 1240 1241 AT3G51680 AAACACAAAAGCATATTAGACTTCGACGCGGACGAGTTTGACCACGTGATGCGTGTGAAC 444 1242 Dl3BHSD1 ------ACGAACGTAATCGATCTCGACATGGTTGACTTTGAAAATGTATTGGCGACTAAC 354 1243 Dl3BHSD2 ---------AGCGTATTGGAACTTGACATGGGAAAATTGGACGACACTTTCGCGACGAAT 345 1244 AtSDR5_AT2G47140 ---------AGCATCCTCGACTTAAACCTCAACGAGTTAGACCGAACGATCGCCATTAAC 345 1245 Erche07g001535 ---------AGCATCCTCGACTTAGATCTGGACGAATTCGACCGAGTGGTCGCCGCTAAC 345 1246 Ecre_5540 ---------agcatcctcgacttagatctggacgaattcgaccgagtggtcaccgttaac 345 1247 AT2G47150 ---------AGCATCCTTGATTTGAATCTTGAACGGTTTCACCGAACAATGGCTGTCAAC 270 1248 AT2G47120 ---------AGCTTCCTCGACTTTGACCTCGAACGGTTTGACCGCATAATGGCTGTTAAC 348 1249 Ecre_7213 ---------agcttcctcgacttggatcttgaacactttgaccgaaccatggcggtcaac 345 1250 AtSR3_AT2G471430 ---------AGCTTTCTCGACTTGAATCTGGAACAGTTTGACCGAACCATGGCGGTCAAC 345 1251 AtSDR4_AT3G29250 ---------AGCGTTCTTGATTTGGATCTTGAGGCGTTTGATCGAACGATGGCGGTTAAC 459 1252 Ecre_9065 ---------agcattcttgatttggatcttgagaagtttgaacgaaccatggcggtcaac 348 1253 AT3G29260 ---------AGCATTCTTGATTTGGATCTTGAGGCGTTTGATCGAACGATGGCCGTCAAC 345 1254 * * * * ** * * ** * * ** 1255 1256 AT3G51680 GTACGTGGCGTAGGACTCGGCATGAAACACGGGGCACGCGCTATGATCAAGAGAGGATTC 504 1257 Dl3BHSD1 GTGCGCGGAGTTGCCAACACTATAAAGCACGCGGCACGAGCCATGGTGGAGGGGAAGGTC 414 1258 Dl3BHSD2 GTACGTGGGGTCGCCGCCACGATTAAGCACGCAGGGCGGGCCATGGTGGAGAGGAAGGTA 405 1259 AtSDR5_AT2G47140 CTCCGCGGCACAGCCGCATTCATCAAGCATGCTGCACGTGCCATGGTGGAGAAAGGCATC 405 1260 Erche07g001535 CTCCGTGGCGCAGCCGCATTTATCAAACATGCTGCACGTGCCATGGTCGAGAAAGGCACG 405 1261 Ecre_5540 ctccgtggcgcagccgcatttatcaaatatgctgcacgtgccatggtcgagaaaggcacg 405 1262 AT2G47150 GTTCGTGGAGCGGCTGTGTCTATCAAACACGCCGCACGAGCCATGGTGGAGAAAGGCACA 330 1263 AT2G47120 GTTCGCGGTGCAGCTGCGTTTATCAAACACGCGGCACGAGCCATGGTGGAGAAAGGCACG 408 1264 Ecre_7213 gttcgcggggcggctgcctttatcaaacacgcggcacgtgccatggtggagaagggcacg 405 1265 AtSR3_AT2G471430 GTTCGTGGTGCGGCTGCGTTTATCAAACACGCTGCACGAGCCATGGTGGAGAAAGGCACG 405 1266 AtSDR4_AT3G29250 GTTCGCGGTGCGGCTGCGTTTATCAAACACGCGGCACGTTCGATGGTGGCTAGTGGTACT 519 1267 Ecre_9065 gttcgcggtgcggctgcgtttatcaaacatgtggcgcgtactatggttgctaatgggaca 408 1268 AT3G29260 GTTCGCGGTGCGGCTGCGTTTATAAAACATGCAGCGCGTTCGATGGTGGCTAGTGGAACT 405 1269 * ** ** * ** ** * * * ** * *** * 1270 1271 AT3G51680 AAAGGCTGCATAATCTCCACGGCGAGTGTAGCCGGTGTGATGGGTGGAATGGGACCACAC 564 1272 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 46 Dl3BHSD1 AAGGGGTCCATCATTTGCACCGCCAGCGTGTCGGCGAGCCTTGGAGGCATGGGCCCGCCC 474 1273 Dl3BHSD2 AAGGGGTCCATTATCTGCACTGCCAGCGTGGCGGCGTGCGTTGGGGGGGTGGCCCCGATT 465 1274 AtSDR5_AT2G47140 CGCGGCTCCATCGTTTGCACCACTAGCGTCGCGGCTGAGATCGCTGGCACGGCACCACAC 465 1275 Erche07g001535 CGCGGCTCCATCGTTTGCACGACCAGCGTCGCAGCTGAGGTCGCCGGCACCGGACCACAC 465 1276 Ecre_5540 cgcggctccatcgtttgcacgaccagcgtctcaggtgaggtcgccggcaccggaccacac 465 1277 AT2G47150 CGTGGCTCCATCGTTTGTACGACCAGTGTTACGTCGGAGATCGTGGTC -AGGGACCTCAT 389 1278 AT2G47120 CGTGGCTCCATCGTTTGTACGACGAGCGTCTCGGCGGAGATCGGTG---GGGGGCATCAT 465 1279 Ecre_7213 cgtgggtcaatcgtatgtaccacaagcgtcgcggcggagatcggtggtccgggacctcat 465 1280 AtSR3_AT2G471430 CGTGGGTCAATCGTATGTACGACCAGCGTCGCGTCGGAGATCGGTGGTCCAGGACCTCAC 465 1281 AtSDR4_AT3G29250 CGTGGCTCCATTGTATGTACGACGAGTATTGCGGCAGAGATAGGTGGTCCGGGACCTCAT 579 1282 Ecre_9065 cgtggttcgattgtatgtactactagtatcgcggccgagattggtggtcctgcacctcat 468 1283 AT3G29260 CGTGGCTCCATTGTTTGTACAACGAGTGTGACGGCTGAGATTGGTGGTCCGGGACCTCAC 465 1284 ** * ** * * ** * ** * * * * * * * 1285 1286 AT3G51680 GCTTACACAGCCTCGAAACATGCGATCGTTGGTTTGACCAAGAACGCAGCGTGTGAGCTA 624 1287 Dl3BHSD1 GCTTACACGGCTTCCAAACACGCCGTCCTGGGCCTAGTCAAGGGCGCTTGCGCCGAATTG 534 1288 Dl3BHSD2 GCTTACACCGCTACTAAACATGCTGTGGTGGGGCTGGTGAAGTCGGCTTGCGGCGAGCTT 525 1289 AtSDR5_AT2G47140 GGGTACACGACGTCGAAGCATGGGCTGTTGGGTTTGATCAAATCGGCTTCTGGTGGATTA 525 1290 Erche07g001535 GGGTACACGGCGTCTAAGCATGGGCTATTGGGTTTGATCAAATCAGCTTCTGGTGGATTG 525 1291 Ecre_5540 gggtacacggcggctaagcatgggctagtgggtttgatcaaaacagcttctggtggattg 525 1292 AT2G47150 GAATACACGGCGTCGAAGCATGGGCTCT -------------------------------- 417 1293 AT2G47120 GGTTACACGGCGTCTAAACACGGGCTTGTCGGCCTGATTAGATCGGCCTGTGGTGATTTG 525 1294 Ecre_7213 gcgtacacggcgtcgaagcacgcgcttctcgggctgatgagatcggcatgtggcgggctg 525 1295 AtSR3_AT2G471430 GCGTACACGGCGTCTAAGCACGCTCTTCTCGGGCTGGTTAAATCGGCTTGTGGCGGGCTA 525 1296 AtSDR4_AT3G29250 AGTTACACGGCTTCTAAGCACGCGCTTCTCGGGCTGATCCGATCAGCGTGTGCTGGACTG 639 1297 Ecre_9065 agttacacagcgtctaagcacgcgctcctcgggctgattcgatcagcgtgtggtgggtta 528 1298 AT3G29260 AGTTACACAGCGTCGAAACACGCGCTTCTCGGGTTGGTTCGATCAGCGTGTGGCGGGTTG 525 1299 ***** * * ** ** * * 1300 1301 AT3G51680 GGCAAGTATGGGATTAGGGTTAATTGTATATCACCGTTTGGAGTTGCCACGTCGATGCTG 684 1302 Dl3BHSD1 GGGGTGCACGGGATCCGAGTCAACTCGGTGGCGCCGTACGGTGTGGCGACCCCGATGCCG 594 1303 Dl3BHSD2 GGCGCGCATGGGATTAGGGTTAACTGTGTCTCGCCGTATTTGGTGGCGACGCCGCTCACT 585 1304 AtSDR5_AT2G47140 GGAAAATATGGTATAAGAGTAAACGGCGTTGCTCCATTTGGTGTCGCAACACCGTTAGTT 585 1305 Erche07g001535 GGGAAATACGGAATAAGAGTGAACGGCGTTGCTCCGTATGGGGTCGCCACACCGTTAGTT 585 1306 Ecre_5540 gggaaatacggaataagagtaaacggcgttgctccgtttggggtcgctacaccgttagtt 585 1307 AT2G47150 ------------------------------------------------------------ 417 1308 AT2G47120 GGGAAGTATGGGATTAGAGTCAATGGAGTCGCACCGTACGCGGTGGCAACGCCAATGACT 585 1309 Ecre_7213 gggaagtacgggattagagtcaacggtgttgcaccgtatgcggtggcgacggcaatcaat 585 1310 AtSR3_AT2G471430 GGGAAGTACGGGATTAGAGTCAACGGCGTTGCACCATACGCGGTGGCGACGGCGATAAAT 585 1311 AtSDR4_AT3G29250 GGGCAGTACGGTATTAGAGTCAACGGTGTTGCACCGTATGGGGTTGCCACGGGGATGACT 699 1312 Ecre_9065 gggaagtacgggattagagtcaatggcgttgcaccgtatggggttgcgacggggttgact 588 1313 AT3G29260 GGGAAGTACGGGATTAGAGTCAACGGTGTTGCACCGTATGGAGTAGCAACGGGGTTAACT 585 1314 1315 1316 AT3G51680 GTTAACGCGTGGCGAAAGACGAGTGGTGGTGACGTGGAAGATGATGACGTGGAGGAGATG 744 1317 Dl3BHSD1 TGCAGTGCTTACGGAATGACACCGAGTCAG ---------------------------ATG 627 1318 Dl3BHSD2 TGCAGCGCGTACGAGATGGAGCCGAGTGAA ---------------------------GTA 618 1319 AtSDR5_AT2G47140 TGTAATGGTTTCAAGATGGAACCAAACGTG ---------------------------GTG 618 1320 Erche07g001535 TGTAATGGTTTCAATATAGAGCCAAACGTG ---------------------------GTG 618 1321 Ecre_5540 tgtgataggtacaatatggagccaaacgcg ---------------------------gtg 618 1322 AT2G47150 ------CACGACGAGGAGACGGCGAAGCAG---------------------------ACG 444 1323 AT2G47120 AGCCA---CGACGAGGTAACGGGAAAGCAG---------------------------TTG 615 1324 Ecre_7213 agccg---tgacgaggaaacagtgaagatg---------------------------gtg 615 1325 AtSR3_AT2G471430 AGCCG---TGACGAGGAAACGGTGAGGATG---------------------------GTG 615 1326 AtSDR4_AT3G29250 AGCGCCTACAATGAGGAAGCGGTGAAGATG ---------------------------CTT 732 1327 Ecre_9065 agcca---caatgaggaaacggtgaagttt---------------------------gtt 618 1328 AT3G29260 AGCTA---CAACGAGGAAACTGTGAAGATG---------------------------GTT 615 1329 1330 1331 AT3G51680 GAGGAGTTTGTGAGGAGTTTGGCTAATTTGAAAGGAGAGACATTGAGAGCGAATGATATA 804 1332 Dl3BHSD1 GAGGAGGCCAATAACTCCAGGGCTAACTTGAAGGGGGTGGTTTTGAAGGCTAAGCATGTA 687 1333 Dl3BHSD2 GAAGCCACCATGTGCGCCAACGCTAACTTGAAGGGAGTGGTTTTGAAAGCTATGCATGTT 678 1334 AtSDR5_AT2G47140 GAGCAGAACACGTCAGCTTCGGCGAATCTAAAGGGCATTGTATTGAAAGCTCGTCACGTG 678 1335 Erche07g001535 GAGGCGAACACTTTAGCGAACGGAAATCTAAAGGGCATCATATTGAAAGCTCGCCACGTG 678 1336 Ecre_5540 gaggagagcactctagcgtccggaaatctaaagggcatcatgttgaaagctcgccacgtg 678 1337 AT2G47150 GAGGAGTATTGCGAAGCCAGAGGGATTTTTAAAGGTGTTGTGTTAAAGGCTCGACACGTG 504 1338 AT2G47120 GAGGATTATTTTGACGCCAAGGGAATTCTCAAGGGGATGGTGCTTAAAGCTAGCCACGTT 675 1339 Ecre_7213 gaggactattgcgccgccacggggattctcaagggtgtggtgcttaaggctcgccatgtg 675 1340 AtSR3_AT2G471430 GAGGAATATAGCGCAGCCACGGGGATTCTCAAAGGTGTGGTGCTTAAGGCTCGCCATGTG 675 1341 AtSDR4_AT3G29250 GAAGAATATGGTGAAGCCCTAGGGAATCTCAAAGGTGTGGTGCTTAAAGCTCGCCACATC 792 1342 Ecre_9065 gaagaatattgtgaagccacggggaatctcaaaggtgttgtgcttaaagctcgtcacgtg 678 1343 AT3G29260 GAGGATTACTGTTCAGCCACGGCGATTCTCAAAGGTGTTGTACTTAAAGCTCGACATGTG 675 1344 ** * * * ** ** * * ** * * 1345 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 47 1346 AT3G51680 GCTGAAGCAGCGTTATATTTGGCGAGTGATGAGTCTAAGTATGTGAACGGACATAATCTT 864 1347 Dl3BHSD1 GCTGAGGCGGCTCTCTTCTTGGCTTCCGATGAGTCGGCTTATGTCAGTGGACAAAACTTG 747 1348 Dl3BHSD2 GCTGAGGTGGCTCTGTTCTTGGCTTCCGATGAGTCCGCTTATGTTAGTGGACAAAATGTG 738 1349 AtSDR5_AT2G47140 GCAGAAGCTGCTCTGTTTTTAGCGTCCGATGAGTCGGCTTACGTTAGCGGACAGAACCTG 738 1350 Erche07g001535 GCAGAAGCTGCTTTGTTTTTAGCATCCGATGCGTCGGCTTACGTTAGCGGACAGAACCTG 738 1351 Ecre_5540 gcagaaactgctttgtttttagcatccgatgcgtcggcttacgttagcggacagaacctg 738 1352 AT2G47150 GCAGAAGCTGCTCTGTTTCTCGCTTCGGATGATTCGGTTTATATCAGTGGTCAGAATTTA 564 1353 AT2G47120 GCACAAGTGGCTCTGTTCTTGGCTTCTGATGATTCGGCTTATATAAGTGGTCAGAATTTG 735 1354 Ecre_7213 gcagaggctgctttgttcctggcttcggatgattcagcttacgttagcggccagaatctg 735 1355 AtSR3_AT2G471430 GCAGAGGCGGCTTTGTTTCTGGCTTCGGATGATTCGGCTTACGTTAGCGGTCAGAATCTG 735 1356 AtSDR4_AT3G29250 GCAGAAGCAGCTTTGTTTTTGGCTTCTGATGATTCGGTTTATATTAGCGGTCAGAATCTG 852 1357 Ecre_9065 gcagaagttgctttgtttctggcttctgatgagtcggtttatattagcggtcagaatcta 738 1358 AT3G29260 GCAGATGCAGCTTTGTTTTTGGCTTCTGATGATTCGGTTTATATTAGTGGTCAAAACTTG 735 1359 ** * ** * * * ** **** ** ** * * ** ** ** * 1360 1361 AT3G51680 GTCGTTGACGGTGGTGTTACGACTGCAAGAAACTGTGTTGGTTTGTGA 912 1362 Dl3BHSD1 GCTGTCGACGGCGGCTTCACCGTCGTGCGTTAG --------------- 780 1363 Dl3BHSD2 GCGGTTGACGGCGGGTTCACGGTCGTTAAGGCTCTCGGAGTTTAA --- 783 1364 AtSDR5_AT2G47140 GCTGTTGACGGTGGTTACTCGGTGGTGAAGCCGTAG ------------ 774 1365 Erche07g001535 GCCGTTGACGGGGGTTACTCGGTGGTTAAGCAGTAG ------------ 774 1366 Ecre_5540 gccgttgacgggggttactcggtggttaagcagtag ------------ 774 1367 AT2G47150 GCAGTGGACGGTGGTTTTTGCGTCGTTAAGCCTATTTGA --------- 603 1368 AT2G47120 GCGGTGGATGGTGGTTATACTGTCGTTAAGCCAAGCCGTGATTAA --- 780 1369 Ecre_7213 gctgttgacggtggttttagcatcgttaagcccatgtga --------- 774 1370 AtSR3_AT2G471430 GCTGTTGACGGGGGTTATAGCGTCGTTAAGCCCATTTGA --------- 774 1371 AtSDR4_AT3G29250 GTGGTTGATGGTGGTTTTAGCGTCGTTAAGCTCATGACCACGTGA --- 897 1372 Ecre_9065 gcggtggacggtggttttagcgtcgtgaagctgtcatga --------- 777 1373 AT3G29260 GGGGTTGACGGCGGTTATAGCGTCGTTAAGCTGACATCGAATTAA --- 780 1374 * ** ** ** ** * 1375 1376 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 48 Figure S16. Multiple sequence alignment of 3-ketosteroid isomerases (3KSI). Coding 1377 sequences were aligned using Clustal Omega. Species included: Arabidopsis thaliana (At/AT), 1378 Calotropis gigantea (Cg), Erysimum cheiranthoides (Ec/Erche), and Digitalis lanata (Dl). 1379 Sequences correspond to gene phylogeny in Figure 6b. 1380 1381 CLUSTAL O(1.2.4) multiple sequence alignment 1382 1383 1384 AT1G47290 ATGGTGATGGAAGTTACAGAGACTGA------------GCGATGGTGCGTTGTAACTGGT 1385 Erche04g028310 ATGGAGATGATAGAAGAGAAGAACAGAGGAATAGAAGGCAAAGTAGTAGTCGTGACGGGT 1386 Erche07g017170 ------ATGCATTTGAGTGAGAACGAAGGTGTCGAAGGTAAGACCTTCTTGGTCACTGGA 1387 AT2G33630 ------ATGCATTTGAGTGAGAATGAAGGTGTCGAAGGTAACACCTTCGTGGTCACTGGA 1388 Dl_DN10245 ------atgcatttgagcgagaacgagggaatagagaagaaaaccttcgtggtgactggc 1389 Cg010433 ------atgcatctaagcgaaaacgaagggattgaaaataacagctttgtggtgacgggt 1390 *** * * * ** ** ** 1391 1392 AT1G47290 GGTAGAGGATTCGCTGCAAGACATCTTGTAGAAATGCTCGTACGTTACCAAATGTTCCAC 1393 Erche04g028310 GGGTTGGGTTTAGTAGGCTCCACCGTGTGCCTTGAGCTGCTCC --------------GCC 1394 Erche07g017170 GGTCTTGGCTTCGTCGGTGCTGCCCTCTGCTTTGAGCTCGTTC --------------GTC 1395 AT2G33630 GGTCTTGGCTTCGTCGGTGCAGCTCTCTGCTTAGAGCTCGTTC --------------GTC 1396 Dl_DN10245 gggctgggcttcgtgggctcggccctctgctcggagctctcca --------------gac 1397 Cg010433 ggccatggatttgtaggcgccgccctctgccttgagctagtta--------------gaa 1398 ** ** ** * * * *** 1399 1400 AT1G47290 GTTCGTATCGCCGATTTAGCTCCTGCGATAGTGCTCAATCCTCACGAGGAAACTGGAATT 1401 Erche04g028310 GTGGCGCTCTCCAGGTTCGCTCCTTAGACTGCCGCACCACTTCT -------------TCT 1402 Erche07g017170 GTGGAGCTCGCCATGTCCGCTCTTTCGACCTCCGCCACTCTTCT -------------CCT 1403 AT2G33630 GTGGTGCTCGCCAAGTTCGCTCTTTTGACCTCCGTCACTCTTCT -------------CCT 1404 Dl_DN10245 gaggggcttatcaagttcgggcctttgaccttcgctttgaatct -------------cct 1405 Cg010433 gaggagctcgtgtagtcaaggcttttgatcttcgtacaaattct -------------ccc 1406 * * * * * ** ** 1407 1408 AT1G47290 CTCGGTGAAGCAATTAGATCCGGTAGAGTTCAATACGTCTCCGCTGATCTTCGGAACAAA 1409 Erche04g028310 TGGTCCGACCGTCTCAAAGAATCCGGCGTCCACTGCATAAATGGAGATGTTGTGAGCAGA 1410 Erche07g017170 TGGTCCGATGACCTCAAAAACAGTGGCGTTCGCTGCATCCAAGGTGATGTGACCAAGAAA 1411 AT2G33630 TGGTCCGATGATCTCAAAAACAGTGGCGTTCGCTGCATTCAAGGTGATGTGACTAAGAAA 1412 Dl_DN10245 tggtccgaggatctccggcatcgtggtgttcgttgcatccaaggggatgttgcaaagaaa 1413 Cg010433 tggtcccaagatctcagtacaaaaggcgtccaatttgtccatggagatgtttcacggaag 1414 * * * ** * * * * *** * * 1415 1416 AT1G47290 ACTCAAGTTGTCAAAGGTTTTCAAGGAGCAGAAGTGGTGTTTCATATGGCAGCTCCAGAT 1417 Erche04g028310 GAGGACGTAGAAGAGGCTGTTGAAGGAGTGGACATTGTTATTCATCTTGCTTCGTACGGT 1418 Erche07g017170 GAAGATGTGGATAAGGCTCTAGATGGTGCGGACTGTGTCTTTCATCTTGCTTCCTATGGC 1419 AT2G33630 CAAGATGTGGATAACGCTTTAGATGGAGCAGACTGTGTTTTGCATCTTGCTTCTTATGGT 1420 Dl_DN10245 gaagatgttgaaaaagctttacgaggcgcagattgtgttttccacctcgcttcatatggc 1421 Cg010433 caagatgtggaaaaggcattatggggagcagattgtgtcttccaccttgcctcttatggc 1422 * ** * * * * ** * ** ** * ** * ** * * 1423 1424 AT1G47290 TCATCGATTAACAATC---------ACCAGCTTCAGTACTCAGTTAATGTTCAAGGGACA 1425 Erche04g028310 GGTTCGGGTAAGGAAATGGTTCGGGCTCATCAGATTGAGGAAGTCAACGTAGAAGGGACG 1426 Erche07g017170 ATGTCTGGTAAAGAGATGCTTCAGTTTGGTCGTTGTGACGAGGTTAACATTAACGGCACT 1427 AT2G33630 ATGTCTGGGAAGGAGATGCTTCGGTTTGGTCGTTGTGATGAGGTTAACATCAACGGGACT 1428 Dl_DN10245 atgtctggcaaagaaatgctccagtatggccgcattgatcagataaatataaatggaact 1429 Cg010433 atgtccggaaaagaaatgctccaatttggtcgtgttgatgaggttaatattactggaacc 1430 ** ** * * * * ** * ** ** 1431 1432 AT1G47290 ACAAATGTAATTGATGCTTGTATTGAGGTTGGAGTAAAGAGGCTAATCTATACAAGTTCT 1433 Erche04g028310 CGTAACGTTTTGGAAACATGTGTGAAGAAAGGGATCACCAGGGTAGTGTATTTAAGCACG 1434 Erche07g017170 TGTAACGTCTTGGAGGCTGTGTTCAAACACGAGATCACAAGACTTGTCTATGTCAGCACT 1435 AT2G33630 TGTAATGTCTTGGAAGCTGCGTTTAAACATGAGATTACAAGAATTGTGTATGTTAGCACT 1436 Dl_DN10245 tgccacattctggatgcttgtctggaccatggaatcagcaggcttgtgtatgtgagcaca 1437 Cg010433 tgtcttgttttggacgcttgcgtggatcttggaatcaagagacttgtttatgtaagcaca 1438 * * ** * * * * * * ** * * *** ** * 1439 1440 AT1G47290 CCGAGTGTTGTGTTTGACGGGGTCCATGGTACTTTGAATGCTGATGAATCACTGCCGTAT 1441 Erche04g028310 AAAGATGTCGTTTTCAGTGGGAA---AGAGATTGAGAATGGAGATGAAACTCTACCTTAT 1442 Erche07g017170 TACAATGTTGTCTTTGGTGGTAA---GGAGATTCAAAATGGCAATGAAGGTTTGCCGTAT 1443 AT2G33630 TACAATGTTGTGTTTGGTGGTAA---AGAGATTCTCAATGGTAATGAAGGTTTGCCTTAT 1444 Dl_DN10245 tacaatgttgtatttggtggaaa---ggaaattgtaaatggtaacgagagtttaccttac 1445 Cg010433 tacaacgttgtttttggtggaaa---ggaaatactgaatggcaatgaaagtctgccttat 1446 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 49 ** ** ** ** * * **** * ** * ** ** 1447 1448 AT1G47290 CCACCTAAGCATAATGA------TTCATATTCAGCTACTAAAGCTGAAGGGGAAGCTTTG 1449 Erche04g028310 GTAGCTTCCAATGAATACGTCAGTTCATATGATCGGACAAAATCTATAGCCGAACAGTTG 1450 Erche07g017170 TTCCCTCTTGATGATCATGTCGACGCCTATGGTCGAACTAAATCCATTGCAGAACAGTTA 1451 AT2G33630 TTCCCTCTTGATGATCATGTTGATGCATATAGTCGAACTAAATCGATTGCAGAACAGTTG 1452 Dl_DN10245 tttccgctggatgaccatgtagatccctatggcagcagtaaatctatagctgaacaattg 1453 Cg010433 ttccctatggatgaccatgtcgatgcatatggccgaagtaaagccattgctgaacagtta 1454 * ** * * * *** * *** * * *** ** 1455 1456 AT1G47290 ATTTTGAAAGCGAATGGAA---------------------GAAGTGGACTACTCACTTGT 1457 Erche04g028310 GTTCTAGAGAACAACGGTCGTACGGTTGAGAACGGACATGGAAGTTTTTTATCGACATGT 1458 Erche07g017170 GTCCTCAAGAGTAATGGTCGACCATTTAAGAATGGGGG ---CAAACGGGTGTACACATGT 1459 AT2G33630 GTTCTCAAGAGTAATGGTCGACCTTTTAAGAATGGAGG ---CAAACGGATGTATACATGT 1460 Dl_DN10245 gttctgaagagcaatagcaaacccttcagaaagaaaca ---aggaaaactctacacatgt 1461 Cg010433 gtattgaagagcaatggccgccccctgaagaacaagaatggaaaacgcctctatacatgt 1462 * * * ** * * ** *** 1463 1464 AT1G47290 TGCATACGTCCTAGCAGCATATTTGGTCCTGGTGATAAATTAATGGTTCCATCGCTTGTT 1465 Erche04g028310 GTGATTCGATGCCCGATTGTTTATGGACCTGGTGAAGAGAAGTATCTTAATAGGATAATC 1466 Erche07g017170 GCAATTCGTCCAGCGGCTATATATGGACCCGGCGAAGACAGGCATCTTCCTAGGATCGTT 1467 AT2G33630 GCGATTCGTCCAGCGGCTATATATGGACCTGGCGAAGATAGGCATCTTCCTAGGATTGTT 1468 Dl_DN10245 gcaatccgccccggtgctatatacgggccaggtgaagaaagacacttgcccagaattata 1469 Cg010433 gcagtgcgtcctgctgctatttatggacctggtgaagaaaggcatcttccgagaatatta 1470 * ** * * ** ** ** ** * * * * 1471 1472 AT1G47290 ACTGCTGCCAGGGCTGGGAAATCCAAGTTCATTATAGGTGATGGGAGTAACTTCTATGAT 1473 Erche04g028310 TCTGATGCAAGATTGGGTTTATTTCTCTTCAAAATTGGCGATACAAGCTCAAAAACCGAC 1474 Erche07g017170 AATCTAGCAAAGTTAGGTTTGATGCTCTTCAAGATCGGTGAACCGAGTGTCAAAACAGAC 1475 AT2G33630 ACTCTAACAAAGTTGGGTTTGGCTCTCTTCAAGATCGGTGAACCGAGTGTCAAATCAGAC 1476 Dl_DN10245 aagcttgccaaattaggtcttctgcccttcaagattggcacaaagaatgtgaaagcagac 1477 Cg010433 aaattgatcaaattaggtcttgtgccatttaagattggtgatgccactgcgaaaactgac 1478 * ** ** * ** ** * ** 1479 1480 AT1G47290 TTCACTTATGTTGAAAATGTTGTGCACGCCCATGTCTGTGCTGAGCG ---------AGCT 1481 Erche04g028310 TGGATTTATGTGGATAACATTGTATTCGCACTCATGTTGGCAACTACTGATTTACTCAAT 1482 Erche07g017170 TGGATTTATGTTGAAAACCTTGTCCTCGCAATCATCCTTGCAAGCATGGGACTCTTGGAC 1483 AT2G33630 TGGATTTATGTCGAAAACCTTGTCCTAGCAATCATCTTGGCAAGTATGGGACTCTTGGAT 1484 Dl_DN10245 tgggtttatgtcgataaccttgtgttggccctgatcttagctagcatgggtcttttagac 1485 Cg010433 tgggtctacgtggataaccttgtattagccatattattggccagtatgggccttttagat 1486 * ** ** ** ** **** ** * ** 1487 1488 AT1G47290 CTAGCATCAGGAGGAGAAGTATGTGCAAAAGCTGCTGGCCAGGCTTACTTCATTACCAAC 1489 Erche04g028310 GAACGCT------------------CGAAAGCCTCAGGGAAGGCCTACTTTGTTTCTGAT 1490 Erche07g017170 GACATTCCGGGCAGAGAAGGGCAGCCCGTCGCTGCTGGTCAACCATACTTTGTTTCTGAT 1491 AT2G33630 GACATTCCTGGCAGAGAAGGACAACCTGTTGCTGCTGGTCAACCATATTTTGTCTCTGAC 1492 Dl_DN10245 gacattcctgggagagttggagaaccagttgctgctggtcaaccttatttcatatcagat 1493 Cg010433 gacattcctggacgagaaggacagccagtcgctgctgggcaaccttacttcatatcagat 1494 * ** * ** * * ** ** * * * 1495 1496 AT1G47290 ATGGAGCCAATTAAATTTTGGGAGTTTATGTCACAGCTTCTTGAAGGACTTGGCTATGAG 1497 Erche04g028310 GGTAATCCAATAAATTTCTTCGAATTTCTTCAGCCACTTTTAAAGAATCTTGATTATGAT 1498 Erche07g017170 GGTTCACCGGTGAATACCTTTGAGTTTCTACGTCCCTTACTAAGAAGTTTAGACTATGAC 1499 AT2G33630 GGTTACCCGGTGAATACCTTTGAGTTTCTGCGTCCTTTACTAAAAAGTTTAGACTATGAC 1500 Dl_DN10245 gggtctccagtcaatagttttgagttcctccagcccttactcaagagtctggagtatgat 1501 Cg010433 ggctcgccagttaatagttttgagttcatccgtccactggttaatagcctggattatgac 1502 ** * ** * ** ** * * * * * * ***** 1503 1504 AT1G47290 AGGCCAAGTATAAAGATACCTGCAAGTCTCATGATGCCAATAGCATATCTTGTGGAACTA 1505 Erche04g028310 CTTCCCAAGTTATCCCTATCTATTTCGCTTGCGGTTTTCCAGGGAAAAATATGTGAAACA 1506 Erche07g017170 CTGCCCAAGTTTACAATATCTGTACCTTTTGGGCTGACCTTGGGTAAGATCTTCCAAGGT 1507 AT2G33630 CTGCCCAAGTGTACAATCTCTGTACCTTTTGCGCTGTCCTTGGGTAAGATCTTCCAAGGT 1508 Dl_DN10245 atgccgaatttgtcactggaagtaccacatgctctatttctcggaaaactcttttggggt 1509 Cg010433 ctgccacaaatgtcattagcagttccccatgcacttcttttgggaaaattttgtactgct 1510 ** * * * * * 1511 1512 AT1G47290 GCAT------------ATAAATTACTCGGACCGTATGGGATGAAAGTACCAGTGCTAACA 1513 Erche04g028310 ATATACGCAATGCTATCACCGCTACTAAACCAATGGTGGATTCCACAACCATTGATTCTT 1514 Erche07g017170 TTCTACACTGTGTTATATCCATGGCTATCACAGAGCTGGTTACCACAGCCCCTTATTCTT 1515 AT2G33630 TTCTACACTGTGCTATATCCATGGCTATCAAAGAGCTGGTTACCACAGCCCCTTGTTCTT 1516 Dl_DN10245 ttctactcgcttttgcatccatggctacaccagtggtggattcctccgcccttgatcctt 1517 Cg010433 ctctactctattttgtatccttggttgtgtaaaaagtggcttcctcagccctttatactt 1518 * * ** * ** * * 1519 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 50 1520 AT1G47290 CCTTCTAGGGTTAGGCTACTCTCTTG-CAACAGAACATTTGATTCTTCAAAAGCAAAGGA 1521 Erche04g028310 CCACCTGAAGTTTATAAGGTTGGGGTGACTCATTACTACTCGA -TCGGCAAAATCAAGGA 1522 Erche07g017170 CCTGCTGAAGTCTACAAGGTCGGTGTTACCCATTACTTCTCAT -ATCTCAAAGCCAAGGA 1523 AT2G33630 CCTGCTGAAGTCTACAAGGTCGGTGTTACCCATTATTTCTCAT -ATCTTAAAGCCAAGGA 1524 Dl_DN10245 cctgctgaagtatataaggtgggagttacccattacttctcat -tcttaaaagcaaggaa 1525 Cg010433 cctgctgaagtctacaaggttggtgtaacacattacttttcgt -atatgaaagcaagaga 1526 ** ** ** * ** * * *** * * 1527 1528 AT1G47290 TCGTTTAGGCTATTCTCCTGTTGTCCCACTTCAGGAAGGTATAAAGAGGACAATAGATTC 1529 Erche04g028310 AGAACTTGGGTACGAGCCCAAGACACAACCTGAAGAAGGCATGACCGAAACCATTTCAT - 1530 Erche07g017170 AGAGCTTGGATATGTTCCATTCAAGAGCTCCCAGGAAGGTATGGCTGCAACTATCTCAT - 1531 AT2G33630 GGAGCTTGGATATGTTCCATTCAAGAGCTCCAAGGAAGGTATGGCTGCAACTATCTCAT - 1532 Dl_DN10245 cgaactgggatataccccaatggtcagtcctcaagagggcatgaacgcaacaattgaat - 1533 Cg010433 agagcttgggtacatcccaatagtgagccctcgagagggtatggctgccactatctcgt - 1534 * ** ** ** ** ** ** ** ** * 1535 1536 AT1G47290 ATTCTCACACTTGAAAGCTCAAAATCAACCCAAAACAGAAGTTACTGAAACAATTCAATG 1537 Erche04g028310 --------ATTTTAAAGAAAAGAAGAGGAGAGAGGTTGATGGTCCAAGCATTTACGCATG 1538 Erche07g017170 --------ACTGGCAGGAAAGGAAACGGAGATCTTTAGTCGGTCCAACCATATTCACTTG 1539 AT2G33630 --------ACTGGCAGGAAAGGAAGCGGAGATCTTTAGACGGTCCAACCATGTTCACTTG 1540 Dl_DN10245 --------actggaaggatagaaaaaggagaacaatagatggcccaaaaatatatgaatg 1541 Cg010433 --------attggaaagagaagaagaagagaagtctggacggaccaacaatattcgcatg 1542 * * * * ** * * * * ** 1543 1544 AT1G47290 GAAAA--------AGCAGACTCTCATTGCCATAGTCATTCTGATTACTCTCTATC----- 1545 Erche04g028310 GATTTTCTGTGTGATCGGTTTGCCTAGCATTATATCAGTCGCGTGGCTACCTGACATTGG 1546 Erche07g017170 GATAGCTGTTATACTATCAATGTCAGCTCTTTTTGCTTCTGGATGGTTACCCGACATAGG 1547 AT2G33630 GATAGCTGTTACAATAGGAATGTCTGCTCTTTTTGCTGCCGGATGGTTACCCGACATAGG 1548 Dl_DN10245 gctatttgtgttgactggaatgacgacgctattttgttctgtatacatgcctgatgttgg 1549 Cg010433 gctgttctgtgtgattggaatgttggttgtatttgcagcagcatatctgccagattttgg 1550 * * * * * 1551 1552 AT1G47290 ----ATAACTTTGTTGC-----AACCACCGGATCATCTTCCGTCA--------TAATAAC 1553 Erche04g028310 ACCCATACCATTCTTTCGAGTCATTGCAATGTTTATCTTTAGGTCGATGTTGGTGTTACG 1554 Erche07g017170 ACCTGTGCCTTACTTTAGAGCCATACACCTCTTCTTCTTCCGGAAAATCATGATTGTAAG 1555 AT2G33630 ACCTGTGCCTTTCTTAAGAGCCATACACCTCTTCTTTTTCCGGACAATCACGATTGTAAA 1556 Dl_DN10245 tccctttcccctatttagagcagtgcacctcttctttttccgctctatgtggctgctaca 1557 Cg010433 tcctattcccataatcagggcttttgctcttttcttcctacgctctatgacagcattaag 1558 * * * * * * * ** 1559 1560 AT1G47290 TGCTGTTTCCAAGGTTTTGTTGGTATCATCAATTTTTAT --------------------- 1561 Erche04g028310 GATAGCATCCGGGATTGTGGTGACAACTCACGTGGGTGAAGCGGTGTATGCGTTATGGCT 1562 Erche07g017170 GGCTGTCTTCATCATATCAGTGGGACTACATGTTGGAGAAGCAATCTATGCGTGGTTTCT 1563 AT2G33630 GGCTGTCTTCATTGTAGCAGTGGTACTACATGTTGCAGAAGGAATCTATGCGTGGTTTCT 1564 Dl_DN10245 tatagtctggcttttatctgcagcagcacatattggtgaggctgtttatgcatggaaact 1565 Cg010433 ggtgcttttccttgcgtctgcagcagctcatattggcgaagccttatacgcatggcaact 1566 * * * 1567 1568 AT1G47290 ------------------------------------------------------------ 1569 Erche04g028310 GGCTCGGAGAGTGGACCCAAAGAATGCAAAAGCTTGGTTTTGGCGGACGCTACTTCTTGC 1570 Erche07g017170 GGCTAAACGAGTCGACCCAGGAAATGCAGTGGGGTGGTTTTGGCAAACCGTCGCTCTCGG 1571 AT2G33630 GGCTAAACGAGTCGACCCAGGAAATGCAATGGGGTGGTTCTTGCAAACCAGCGCTCTCGG 1572 Dl_DN10245 tgcaagaaaggtcgatcctgcaaatgcgaaagcctggttttggcaaacgttcgctttggg 1573 Cg010433 tgcaaaaagagtggatcctgcaaatgcaaagggctggttttggcaaactttagcattggg 1574 1575 1576 AT1G47290 --GTTCATCAATGGCATT-----TTACCAGAGAAAATG------AAAGTGTTCGGGAGCA 1577 Erche04g028310 TACTTTCTCGCTTCGATTACTTTTGAAAAGAGCCAAGGAAGTAAAACAATCAACACTAAG 1578 Erche07g017170 GATATTCGGTATGCGGTTTTTGTTAAAGAGAGCCAAAGACCACCAAGTTTAA -------- 1579 AT2G33630 GTTTTTCTCAATGCGGTTTTTATTGAAGAGAGCGAAAGAGCACCAGATTTAA -------- 1580 Dl_DN10245 attcttttcattgcgttttctgttgaagagagccaaaacctag ----------------- 1581 Cg010433 gttcttttccctgcgttttctgttaaagagggctaaagaaaatacttag ----------- 1582 * * ** * * ** * ** 1583 1584 AT1G47290 AGAAGATCGACTAA-------------------------- 1585 Erche04g028310 AGAAGGTCTACTAACCAATTCGGGGTCGTCAATTGCTTGA 1586 Erche07g017170 ---------------------------------------- 1587 AT2G33630 ---------------------------------------- 1588 Dl_DN10245 ---------------------------------------- 1589 Cg010433 ---------------------------------------- 1590 1591 1592 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 51 Figure S17. Multiple sequence alignment of Progesterone 5β-reductases (P5βR). Coding 1593 sequences were aligned using Clustal Omega. Species included: Arabidopsis thaliana (At/AT), 1594 Erysimum cheiranthoides (Ec/Erche), Erysimum crepidifolium (Ecre), Digitalis lanata (Dl), and 1595 Marchantia polymorpha (Mp). Sequences correspond to gene phylogeny in Figure 6c. 1596 1597 CLUSTAL O(1.2.4) multiple sequence alignment 1598 1599 1600 Dl_DN67489 ------------------------------------------------------------ 1601 AT5G58750 AT---------------------------------------GGGGTC----------TGA 1602 Erche06g007150 AT---------------------------------------GGAGTC----------TGA 1603 Erche06g011020 AT---------------------------------------GGGGTC----------TGA 1604 Ecre_P5BR2 AT---------------------------------------GGGGTC----------TGA 1605 Mp0191s0002 ------------------------------------------------------------ 1606 Dl_DN4564 atgtatactgacacaacgacttggtggtacaaaagatc ---cattggcgatattaaacag 1607 DlP5BR2 ATGTATACCGACACAACGACTTGGTGGTACAAAAGATC ---CATTGGCGATATTAAACAG 1608 AT4G24220_VEP1 ---------------ATGAGTTGGTGGTGGGCTGGCGCCATCGGAGC------TGCCAAG 1609 Erche02g027660 ---------------ATGAGTTGGTGGTGGGCTGGCGCCATCGGAGC------TGCCAAG 1610 Ecre_P5BR1 ---------------ATGAGTTGGTGGGGGGCTGGCGCCATCGGAGC------TGCCAAG 1611 Dl_DN477 ------------atgagcagctggtgggcggctggagctaccggcgccgcgaaggagaaa 1612 DlP5BR1 ---------------atgagctggtggtgggctggagcgataggcgc------tgcaaag 1613 1614 1615 Dl_DN67489 ---------atggctgtgaatgaccaaaatcatggatctcttgttgcagccgtctttgga 1616 AT5G58750 AAATGGCAGCTTGATGAGAAGAAACGA--AGTAGACGAGAATGTAGCGTTAATCTTTGGC 1617 Erche06g007150 AAGTGGCATCTTGATGAGACGAAACGA--AGTAGACGAGAACGTCGCGTTGATCTTCGGC 1618 Erche06g011020 AAATGGCATCTTGATGAGAAGAAACGA--AGTAGACGAGAACGTCGCGTTGATCTTCGGC 1619 Ecre_P5BR2 AAATGGCAGCTCGATGAGAAGAAACGA--AGTAGACGAGAACGTCGCGTTGATCTTCGGC 1620 Mp0191s0002 ---------------------------------ATGGGTAGAGCGGCCCTGATCTGCGGT 1621 Dl_DN4564 aaaaatgtcgaaacaaatggtgttgctttaaactacaagagtgtcgccctcatagttggg 1622 DlP5BR2 AAAAATGTCGAAACAAATGGTGTTGCTTTAAACTACAAGAGTGTCGCCCTCATAGTTGGG 1623 AT4G24220_VEP1 AAGAAACTCGACGAAGATGAACCATCACAAAGCTTCGAGAGCGTCGCTCTCATCATCGGC 1624 Erche02g027660 AAGAAACTCGACGACGATGAGCCGACGCAAAGCTACGAGAGCGTCGCTCTCATCATTGGC 1625 Ecre_P5BR1 AAGAAACTCGACGACGATGAGCCGTCGCAAAGCTACGAGAGCGTCGCTCTCATCATCGGC 1626 Dl_DN477 aaaactggagaagataatgaggcaccgtcaaagcaccagagcgtgggcttgatagtcgga 1627 DlP5BR1 aaaaggttggaagaagatgacgcacagccaaagcattcgagcgtggcgttgatagttggg 1628 * * * ** 1629 1630 Dl_DN67489 gttactggccttgtgggaaaggagctcgccggaaaactt ---ctttccaccggt------ 1631 AT5G58750 GTCACCGGGCTTGTGGGTCGAGAGATTGTTAAGACGCTATTAATGTCTAAACCA ------ 1632 Erche06g007150 GTCACGGGACTCGTTGGTCGAGCGATTGTAAAAACGCTACTAACGTCTGAACCC ------ 1633 Erche06g011020 GTCACGGGACTCGTTGGTCGAGAAATTGTAAAAACGCTACTCACGTCGAAACCC ------ 1634 Ecre_P5BR2 GTCACGGGACTCGTTGGTCGAGAGATTGTAAAAACGCTACTAACGTCAAAACCC ------ 1635 Mp0191s0002 GCGACCGGAATTGTGGGCAAATATTTGGTGGACATGCTTTCCAGCCCCGACTGCCCGTCC 1636 Dl_DN4564 gtcaccggcatcgccggaagcggtctggctgaaactctatccatgtccgacacgccaggt 1637 DlP5BR2 GTCACCGGCATCGCCGGAAGCGGTCTGGCTGAAACTCTATCCATGTCCGACACGCCAGGT 1638 AT4G24220_VEP1 GTTACTGGAATCGTCGGAAACAGCTTGGCGGAGATTCTCCCTCTTTCCGACACACCCGGT 1639 Erche02g027660 GTTACAGGAATCGTCGGAAACAGCCTTGCAGAGATTCTCCCTCTTTCCGACACACCCGGA 1640 Ecre_P5BR1 GTTACAGGAATCGTCGGAAACAGCCTGGCGGAGATTCTCCCTCTTTCCGACACACCCGGT 1641 Dl_DN477 gttaccggaattgttgggaatagtctggcggagatcctgccgcgctccgacacccccggc 1642 DlP5BR1 gtaaccggaatcatcggcaacagcctggcggagatcctgccactggccgacacccccggc 1643 * ** ** * ** * * * ** * 1644 1645 Dl_DN67489 ---aaatggaagatctatggaatagcaagaagccggaaggaaagttcgagttctcaatat 1646 AT5G58750 ---GGATGGAGAATCTACGGCGTAGCGCGTAACCCGGAGAT------------------- 1647 Erche06g007150 ---AGATGGAAAATTTACGGCGTGGCGCGTAAGCCGGAGAT------------------- 1648 Erche06g011020 ---AGATGGAAAATCTACGGCGTAGCGCGTAACCCGGAGAT------------------- 1649 Ecre_P5BR2 ---AGATGGAAAATTTACGGCGTAGCGCGTAACCCGGAGAT------------------- 1650 Mp0191s0002 GGACCATGGAAGGTGTACGCGCTAGCTCGAAGATCCCACACTGGTGCCAAAGT ------- 1651 Dl_DN4564 ggaccgtggaaagtctacggggtcgcgcgccgcccttgtccagagtggctagc ------- 1652 DlP5BR2 GGACCGTGGAAAGTCTACGGGGTCGCGCGCCGCCCTTGTCCAGAGTGGCTAGC ------- 1653 AT4G24220_VEP1 GGTCCATGGAAAGTCTACGGCGTCGCTCGTCGTCCTCGTCCTACCTGGAACGC ------- 1654 Erche02g027660 GGTCCCTGGAAAGTCTACGGCGTCGCTCGCCGTCCCCGTCCCAGCTGGAACGC ------- 1655 Ecre_P5BR1 GGTCCCTGGAAAGTCTACGGCGTCGCTCGCCGTCCCCGTCCCAGCTGGAACGC ------- 1656 Dl_DN477 ggcccatggaaggtctacggcgtcgcccgccgcccgcgccccagttggaacga ------- 1657 DlP5BR1 ggtccgtggaaggtatacggcgtcgcccgccgcaccagacccgcctggcatga ------- 1658 **** * ** * * ** * 1659 1660 Dl_DN67489 tttgcccgaactgaaagcgatccaaattaccatttcatttcgtgtgatcttcttaatcca 1661 AT5G58750 --------CAATTCCATGACGAAGATGTACAACTTCATCTCCTGCGATCTGCTTAACGCA 1662 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 52 Erche06g007150 --------CAATTCCAT---------GTGCAATTTCATCTCCTGCGATCTGCTTAACGCA 1663 Erche06g011020 --------CAATTCCATGGCGAAGATGTACAATTTCATCTCCTGCGATCTGCTTAACGCA 1664 Ecre_P5BR2 --------CAATACCATGGCGAAGATGTACAGTTTCATCTCCTGCGATCTGCTTAACGCA 1665 Mp0191s0002 --------TTGCTTCAAGTCCGAGAACTATACGTACATTCAAGCGAATCTGCTGGATAAG 1666 Dl_DN4564 -----------------caaactccacgtcagctacatccaatgtgacatcgggtcaact 1667 DlP5BR2 -----------------CAAACTCCACGTCAGCTACATCCAATGTGACATCGGGTCAACT 1668 AT4G24220_VEP1 -----------------CGATCATCCGATCGATTACATCCAGTGCGATGTCTCCGACGCC 1669 Erche02g027660 -----------------CGATCATCCGATCGATTACATCCAGTGCGATGTCTCCAACGCT 1670 Ecre_P5BR1 -----------------CGATCATCCGATCGATTACATCCAGTGCGATGTCTCCAACGCC 1671 Dl_DN477 -----------------agaaaaccctattaactacatccggtgtgacatatccgacccg 1672 DlP5BR1 -----------------ggataatccgatcaattacgtccagtgcgacatatccgatcca 1673 * * * * * 1674 1675 Dl_DN67489 tgggaaacacaagcaaagctctccccga---tgaatgacataactcatgtattttgggtc 1676 AT5G58750 TCTGAGACTAAACAGAGGTTGTCTCCATTACAAGACATCGTGAGTCACGTGTTTTGGGTC 1677 Erche06g007150 TCTGAGACTAAACAGAAGCTGACTCCATTACAGGACATCGTGAGTCACGTGTTTTGGGTC 1678 Erche06g011020 TCTGAGACTAAACAAAAGCTGTCTCCATTACAGGACATCGTGAGTCACGTGTTTTGGGTC 1679 Ecre_P5BR2 TCTGAGACTAAACAGAAGCTGTCTCCATTACAGGACATCGTGAGTCACGTGTTTTGGGTC 1680 Mp0191s0002 GACGACACCTACGAGAAGCTATCGCCTC---TGAAAGACGTGAGCCATATCTTCTGGGTG 1681 Dl_DN4564 gatgacacgtcagccaaattgtctccac---tttccgacatcactcacattttctacgtg 1682 DlP5BR2 GATGACACGTCAGCCAAATTGTCTCCAC---TTTCCGACATCACTCACATTTTCTACGTG 1683 AT4G24220_VEP1 GAAGACACAAGATCCAAGCTTTCCCCTT---TAACCGATGTCACACATGTCTTCTACGTC 1684 Erche02g027660 GAAGATGCAAGATCCAAGCTTTCCCCTT---TAACCGATGTCACTCACGTCTTCTACGTC 1685 Ecre_P5BR1 GAAGATGCAAGATCCAAGCTTTCCCCTT---TAACCGATGTCACTCACGTCTTCTACGTC 1686 Dl_DN477 gatgatacccgggtcaagctgtcacctc---tcacagatatcactcacgtgttttacgta 1687 DlP5BR1 gatgactcccaagccaagctgtcacctc---tgactgatgttacccacgtgttctacgtt 1688 ** * * * * ** * * ** * ** * ** 1689 1690 Dl_DN67489 acttggtccagccagtttcctctggataacctcgagtgctctgagcagaacagggcaatg 1691 AT5G58750 ACGTGGTCTGGTGAGTTTCCATTGGATACCGACGAATGTTGCGTACAGAACAAGACGATG 1692 Erche06g007150 ACGTGGTCTGGCGAGTATCCATTGGATAGCGACGAGTGCTGTGTCCAGAACAAGACGATG 1693 Erche06g011020 ACGTGGTCTGGCGAGTATCCATTAGATAGCGACGAGTGCTGTGTCCAGAACAAGACGATG 1694 Ecre_P5BR2 ACGTGGTCTGGCGAGTATCCATTGGATACCGACGAGTGCTGTGTCCAGAACAAGACGATG 1695 Mp0191s0002 ACTTGGGTACAGGGCA------AGGACGAGGAAGAGAGCTGTGCGAAGAATGGTCAGATG 1696 Dl_DN4564 tcctggaccggatccg------a------------agattgtgataagaacgcaatcatg 1697 DlP5BR2 TCCTGGACCGGATCCG------A------------AGATTGTGATAAGAACGCAATCATG 1698 AT4G24220_VEP1 ACCTGGACCAATCGTG------AATCGGAAAGTGAAAACTGTGAAGCAAATGGCTCAATG 1699 Erche02g027660 ACCTGGACCAACCGCG------AATCGGAGAGCGAAAACTGCGAGGCTAACGGCTCAATG 1700 Ecre_P5BR1 ACCTGGACCAAGCGCG------AATCGGAGAGCGAAAACTGCGAGGCTAACGGCTCAATG 1701 Dl_DN477 acatgggcaaaacggt------ccaccgaggctgaaaattgcgaagccaatggcaaaatg 1702 DlP5BR1 acctgggctaatcgat------ccaccgaacaagaaaactgtgaagccaatagcaaaatg 1703 * *** * * ** *** 1704 1705 Dl_DN67489 atgtccaacgccctgaacgccatactccccacgtccaaggcgttaaagcatttctccctc 1706 AT5G58750 CTGATGAACGCTTTGGACGCGATTCTCCCAAACGCTAAGAGGTTAAAGCATTTCTCGCTT 1707 Erche06g007150 CTGTCAAACGCTTTGGACGCGATTCTCCCAAACGCTAAGAGGTTAAAGCATTTCTCGCTT 1708 Erche06g011020 CTGACAAACGCTTTGGACGCGATTCTCCCAAACGCTAAGAGGTTAAAGCATTTCTCGCTT 1709 Ecre_P5BR2 CTGACAAACGCTTTGGACGCGATTCTCCCAAACGCTAAGAGGTTAAAGCATTTCTCGCTT 1710 Mp0191s0002 CTCATCAATACTCTGGATGCCGTCGTGGCGAATTCCAGTGCTTTGGAGCACATCGTTCTT 1711 Dl_DN4564 ttcaaaaacatcctcgattcagtcatcccaaatgctccaaatctcaaacacatttccctc 1712 DlP5BR2 TTCAAAAACATCCTCGATTCAGTCATCCCCAATGCTCCAAATCTCAAACACATTTCCCTC 1713 AT4G24220_VEP1 CTCCGTAACGTTCTCCAAGCGATTATCCCATACGCGCCAAATCTCCGACATGTTTGTCTC 1714 Erche02g027660 ATCCGTAACGTTCTCCAAGCGATTGTCCCACACGCGCCTAATCTCCGGCACATTTGTCTC 1715 Ecre_P5BR1 CTCCGTAACGTTCTCCAAGCGATTGTCCCACACGCGCCTAATCTCCGGCACATTTGTCTC 1716 Dl_DN477 atgaagaatgttctcgactctgtaatccctaattgccctaatttgaagcacatttgtttg 1717 DlP5BR1 ttcaggaacgtgcttgatgcagttatccctaattgccccaatttgaagcacatctcattg 1718 * ** * * * * * * * ** * * 1719 1720 Dl_DN67489 cagacgggcaccaagcattacgtctctcttcaacaaggaaatacagatcattcaacgatg 1721 AT5G58750 CAAACGGGGATGAAGCATTATGTGTCTTTGGTGGAAGAGACAATGGCTCGTGGAGAAGGT 1722 Erche06g007150 CAAACGGGGATGCACCATTACAG------------------CGAATCCCATGGAGAAGGT 1723 Erche06g011020 CAAACGGGGATGAAGCATTACGTTTCTCTGGTCGAAGAGACTCTGTCCCGCGGAGAAGGT 1724 Ecre_P5BR2 CAAACGGGGATGAAGCATTACGTGTCTCTGGTCGAAGAGACTCTGTCACACGGAGAAGGT 1725 Mp0191s0002 CAGACGGGATGTAAGCATTATCTTGGACCCTTCGACCAGTTCGGTGG ---AGACGTGAAG 1726 Dl_DN4564 caaacaggaatcaagcactattggggcaacatggttgatgaaatggacaccactaatgta 1727 DlP5BR2 CAAACAGGAATCAAGCACTATTGGGGCAACATGGTTGATGAAATGGACATCACTAATGTA 1728 AT4G24220_VEP1 CAGACAGGGACAAAGCACTACCTTGGCCCTTTCACCAACGTTGA ---------CGGACCT 1729 Erche02g027660 CAGACGGGGACGAAGCACTACGTCGGCCCTTTCAGCAACCTCGG ------CGGCGGACCT 1730 Ecre_P5BR1 CAGACGGGGACGAAGCACTACCTCGGCCCTTTCAGCAACCTCGA ---------CGGACCT 1731 Dl_DN477 cagactgggaggagacattacatgggcttttttgaatcagggat ------g------aga 1732 DlP5BR1 cagactgggaggaagcattacatgggaccatttgaatcctacgg ------gaaaatagaa 1733 ** ** ** ** ** 1734 1735 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 53 Dl_DN67489 aaacatgaaattatcttatttgacgagcattcgccgagggttctaaatgcagagcctaac 1736 AT5G58750 TCG----AGTTTGTATTATTACAG--CGAGGAGTGTCCAAGAAAGAGCTCTGGGAAGAAT 1737 Erche06g007150 TCG----AGTTTGTGTTATTACAG--CGAGGAGTGTCCGAGAAAGAGCTCTGGGAAGAAT 1738 Erche06g011020 TCG----AGTTTGTGTTATTACAG--CGAGGAGTGTCCGAGAAAGGGCTCTGGGAAGAAT 1739 Ecre_P5BR2 TCG----AGTTTGTGTTATTACAG--CGAGGAGTGTCCGAGAAAGGGCTCTGGGAAGAAT 1740 Mp0191s0002 CCTCACGACACCCCATTCCGAGA---GGATGTGCC------CAGACTTCCAGGCCAGAAT 1741 Dl_DN4564 tcacatgattgtccctttaatga---atacatgcc------tagacttagacaacccaat 1742 DlP5BR2 TCACATGATTGTCCCTTTAATGA---ATACATGCC------TAGACTTAGACAACCCAAT 1743 AT4G24220_VEP1 CGTCACGATCCACCGTTCACTGA---GGATATGCC------GAGATTGCAGATCCAGAAT 1744 Erche02g027660 CGCCACGATCCGCCTTTTACTGA---GGATATGCC------GAGATTGCAGATCCAGAAT 1745 Ecre_P5BR1 CGCCACGATCCGCCTTTTACTGA---GGATATGCC------GAGATTGCAGATCCAGAAT 1746 Dl_DN477 tgccacgatgctccgcataccga---ggatttgcc------ccgattggattgtttgaat 1747 DlP5BR1 tcccatgatccaccctacactga---ggatttgcc------caggttgaagtacatgaac 1748 * * * * ** 1749 1750 Dl_DN67489 ttttactacgccctcgaagacttgttacagg ------ataaattgccacttcatcgggtg 1751 AT5G58750 TTCTACTATGTTTTGGAGGACTTGCTGAAGG ------AGAAGATCACTCGTAGTTCCGTT 1752 Erche06g007150 TTCTATTACGTTTTGGAGGATTTGCTGGAGG ------AGAAGATCTCTGGTAATTCGGTT 1753 Erche06g011020 TTTTATTACGTTTTGGAGGATTTGCTGAAGG ------AGAAGATCTCTGGTAATTCGGTT 1754 Ecre_P5BR2 TTCTATTACGTTTTGGAGGATTTGCTGAAGG ------AGAAGATCTCTGGTAGTTCGGTT 1755 Mp0191s0002 TTCTACTACACCTTGGAGGATATTGTGTTCAACCACCTCAAGCAGCACCAAGGTCATCTG 1756 Dl_DN4564 ttctattacaatcttgaagacctactttatgaagcctgcaggacacaaaacggtgctcta 1757 DlP5BR2 TTCTATTCCAATCTTGAAGACCTACTTTATGAAGCCTGCAGGACACAAAACGGTGCTCTA 1758 AT4G24220_VEP1 TTCTATTATACCCAAGAGGATATTCTGTTTGAAGAGATCAAGAAGATAGA ---AACCGTG 1759 Erche02g027660 TTCTATTACACTCAGGAGGATATTCTGTTTGAAGAGATCAAGAAGAAAGA ---AAGTGTG 1760 Ecre_P5BR1 TTCTATTACACTCAGGAGGATATTCTGTTTGAAGAGATCAAGAAGAAAGAAATTAGTGTG 1761 Dl_DN477 ttttactacactttagaggatattctctttgaggaggtggggaagaagga ---aggtttg 1762 DlP5BR1 ttttactatgatttagaggatattatgcttgaggaggtggagaagaagga ---gggtttg 1763 ** ** * ** ** * * * 1764 1765 Dl_DN67489 ggatggtccatccaccgccccggattgataatcggttgttcccatagaacaacttacaat 1766 AT5G58750 GTTTGGTCGGTTCAAAGACCTGGTTTGCTAATGGGAAGCTCTTCAAGAACTCTGTACAAC 1767 Erche06g007150 GTTTGGTCGGTTCAAAGACCTGGTTTGCTATTGGGAAGCTCTACAAGAACTCTGAAAAAT 1768 Erche06g011020 GTTTGGTCGGTCCAAAGACCTGGTTTGCTAATGGGAAGCTCTTCAAGAACTCTGTACAAT 1769 Ecre_P5BR2 GTTTGGTCGGTTCAAAGACCTGGTTTGCTAATGGGAAGCTCTTCAAGAACTCTGTATAAT 1770 Mp0191s0002 ACTTACTCCATTCACAGACCGACCAATATCTTTGGGTTCTCAGCTGGGAATTTGATGAAC 1771 Dl_DN4564 acatggaccgtccatcgccccgcgctaattttcgggttctcgccatgtagtttgatgaac 1772 DlP5BR2 ACATGGACAGTCCATCGCCCCGCACTAATTTTCGGGTTCTCACCATGTAGTTTGATGAAC 1773 AT4G24220_VEP1 ACGTGGTCTATCCACAGACCAAACATGATCTTTGGGTTCTCTCCTTATAGTTTGATGAAC 1774 Erche02g027660 ACGTGGTCTATTCATAGACCAAACACGATCTTTGGATTCTCTCCTTACAGTTTGATGAAC 1775 Ecre_P5BR1 ACGTGGTCTATACATAGGCCAAACACGATCTTTGGATTCTCTCCTTACAGTTTGATGAAC 1776 Dl_DN477 agctggtcagtgcacaggcctggtttaattttcgggttctctccgtatagtattatgaac 1777 DlP5BR1 acttggtcggttcatcgcccagggaatatattcgggttttctccatatagtatgatgaat 1778 * * * ** * ** * * ** ** * ** 1779 1780 Dl_DN67489 tttataggtagtttatgcgtgtatggaactatttgcaagtatctgaatctcccttttgtg 1781 AT5G58750 TTCATGGGAAGTCTTTGTGTTTATGGAGCGATGTGTAAGTATTTGAATCTTCCTTTTGTG 1782 Erche06g007150 TTCATGGGAATTCTGTGTGTTTATGGAGCAATGTGTAAGTATTTGAATCTTCCTTTTGTG 1783 Erche06g011020 TTCATGGGAAGTCTTTGCGTTTATGGAGCAATGTGTAAGTATTTGAATCTCCCCTTTGTG 1784 Ecre_P5BR2 TTCATGGGAAGTCTTTGCGTTTATGGAGCAATGTGTAAGTATTTGAATCTCCCTTTTGTG 1785 Mp0191s0002 TTGTTGGGAACGCTTTCAGTGTATGCAGCCATCTGCAAACACGAGGGACTTCCATTTGTG 1786 Dl_DN4564 atagtcgcgacgctaagtgtttacgctgcaatttgcaagtatgagaacaagccattggtg 1787 DlP5BR2 ATAGTCGCAACGCTAAGTGTTTATGCCGCGATTTGCAAATATGAGAACAAGCCGTTGGTG 1788 AT4G24220_VEP1 ATTGTTGGGACTCTCTGTGTCTATGCAGCGATATGTAAGCATGAAGGGTCTCCGTTGTTG 1789 Erche02g027660 ATTGTTGGGACTCTTTGTGTCTACGCAGCGATCTGCAAGCATGAAGGGTCTCCGTTGTTG 1790 Ecre_P5BR1 ATTGTTGGGACTCTCTGTGTGTACGCAGCGATCTGCAAGCATGAAGGGTCGCCGTTGTTG 1791 Dl_DN477 gtgatggttagcttttgtgtttacgcggctatatgtaagcatgagggtgtgaagttgaaa 1792 DlP5BR1 ttggtgggtaccctttgtgtttatgcagctatttgcaaacacgagggaaaggttttgagg 1793 * * * * * ** ** * * ** ** ** * ** 1794 1795 Dl_DN67489 tttgggggcacaaagcaatgctgggaagaaatgtttgtagatgcttcagatgctcgactg 1796 AT5G58750 TTTGGAGGAACAAGAGAATGTTGGGAAGAGAGTTACATAGATGGCTCTGATTCGAATTTA 1797 Erche06g007150 TTTGGAGGGACAAGAGAGTGTTGGGAAGAGAGTTACATTGATGGCTCTGATGCGAATTTA 1798 Erche06g011020 TTTGGAGGGACACGAGAATGTTGGGAAGAGAGTTACATTGATGGCTCTGATTCGAATTTA 1799 Ecre_P5BR2 TTTGGAGGGACAAGAGAGTGTTGGGAAGAGATTTACATTGATGGGTCTGATGCCAATTTA 1800 Mp0191s0002 TTTCCAGGAAACAAAATCTCTTGGGAGCAGC ---TGTCCGATGCCTCCTCCGCAGATCTG 1801 Dl_DN4564 tataccgggacagaaacgtcgtggaattgtt ---tagtggatgccgtggattcggattta 1802 DlP5BR2 TATACCGGGACACAAACGTCGTGGAATTGTT ---TAGTGGATGCTGTGGATTCGGATTTG 1803 AT4G24220_VEP1 TTTCCTGGGAGCAAGAAAGCTTGGGAAGGGT ---TCATGACGGCTTCTGACGCGGATTTG 1804 Erche02g027660 TTTCCTGGGAGCAAGAAGGCTTGGGAAGGCT ---TCACGACGGCTTCGGACGCGGACCTG 1805 Ecre_P5BR1 TTTCCTGGGAGCAAGAAGGCTTGGGAAGGGT---TCACGACGGCTTCGGACGCGGACCTG 1806 Dl_DN477 tttccgggttgtaaaaccacgtgggatgggt ---attcagattgctcggattcggatctg 1807 DlP5BR1 tttactggttgtaaggctgcgtgggatgggt ---actcggattgctctgatgcggatttg 1808 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 54 * * ** *** * * * 1809 1810 Dl_DN67489 gtggccgaacagcacatttgggctgcaactacatttgaagctgatgcgcaaggagaagca 1811 AT5G58750 GTCGCGGAGCAGCATATATTCGCTGCAACAAGTGGGAAAGTACGCGAGAAAGGGGAAGCT 1812 Erche06g007150 GTCGCGGAACAGCATATATTCGCTGCAACAAGTGGGAAAGTACGCAACCGAGGTGAAGCA 1813 Erche06g011020 GTCGCGGAACAGCATATATTCGCTGCAACAAGTGGGAAAGTACGCAACAGAGGTGAAGCA 1814 Ecre_P5BR2 GTCGCGGAACAGCATATATTCGCTGCAACAAGTGGGAAAGTACGCAACAGAGGTGAAGCA 1815 Mp0191s0002 ATTGCAGAGCAGGAGATATGGGCCGCCACCACTATCGAT ------GCCAAGAACCAAGCT 1816 Dl_DN4564 ttagcggaccacttggtgtggggtgggacaaaccctaat ------gcgaagaatcaagct 1817 DlP5BR2 TTATCAGAGCACTTGGTATGGGGTGCGATCAGCCCAAAT ------GCGAAGAACCAAGCT 1818 AT4G24220_VEP1 ATTGCGGAGCAGCAGATTTGGGCGGCGGTTGATCCGTAT ------GCGAAGAACGAGGCG 1819 Erche02g027660 ATCGCTGAGCAGCAGATTTGGGCTGCGGTTGATCCGTAT ------GCGAAGAACGAGGCG 1820 Ecre_P5BR1 ATCGCTGAGCAGCAGATTTGGGCTGCGGTTGATCCGTAT ------GCTAAGAACGAGGCG 1821 Dl_DN477 attgcggagcaccagatctgggcggcggtggatcctaaa ------gcgaagaacgaggcg 1822 DlP5BR1 atagcggagcatcatatttgggctgcagtggatccttat ------gcaaaaaacgaggcc 1823 * * ** ** * * * * * * ** 1824 1825 Dl_DN67489 tttaatgccattaatggtgattgttatgcgtggaaggacatgtggcgggatattggattt 1826 AT5G58750 TTTAACGCCATTAATGGAGTAGGGTTTACTTGGAAGGAGATTTGGCCGGAAATAGGGAAG 1827 Erche06g007150 TTTAACTCCATTAATGGGGTAGGATTTACTTGGAAGGAGATTTGGCCGGATATCGGGAGG 1828 Erche06g011020 TTTAACGCCATTAATGGGGTAGGGTTTACTTGGAAGGAGATTTGGCCGGATATCGGGAGG 1829 Ecre_P5BR2 TTTAACGCCATTAATGGGGTAGGGTTTACTTGGAAGGAGATTTGGCCGGAGATTGGGAGG 1830 Mp0191s0002 TACAACATCTCCAATGGAGATGTTTTCAAGTGGAAGAAATTGTGGCCATTGCTTGCCGCT 1831 Dl_DN4564 ttcaatatcaacaatggcgacgtttttaaatggaaacacatatggaaagtgttggcggag 1832 DlP5BR2 TTCAATATCAACAATGGCGACGTTTTTAAATGGAAACATATATGGAAAGTGTTGGCGGAG 1833 AT4G24220_VEP1 TTTAACTGCAACAATGCTGATATCTTCAAGTGGAAGCATCTGTGGAAGATTCTAGCTGAG 1834 Erche02g027660 TTTAACTGCAACAACGCGGATATCTTCAAGTGGAAACATCTGTGGAAGATTTTGGCAGAG 1835 Ecre_P5BR1 TTTAACTGCAACAACGCGGATATCTTCAAGTGGAAACATCTGTGGAAGATTCTGGCTGAG 1836 Dl_DN477 tttaatatcagcaatggggatgtgttcaaatggaagcatttctggaaagtgttggctgag 1837 DlP5BR1 tttaatgtgagtaatggagatgtgtttaaatggaagcatttttggaaggtgttggcggag 1838 * ** ** * * * ***** * * *** * * 1839 1840 Dl_DN67489 aagcttggggcggatgtgccaaaagag---gacatgttatgtgaggatttcgtgttttcg 1841 AT5G58750 AAACTTGGTGTACAAGTTAACGAAACA---ACCATGTTTGATGAAGGTTTCTGGTTTGGG 1842 Erche06g007150 AAACTTGGTGTGCAAGTTAATGAAACG---ACGATGTTTGATGAAAATTTCTGGTATGGG 1843 Erche06g011020 AAGCTTGGGGTGCAAGTTACCGAAACG---ACGATGTTTGATGAAGGTTTCTGGTATGGG 1844 Ecre_P5BR2 AAACTTGGGGTGCAAGTTAACGAAACA---ACGATGTTTGATGAAGGTTTCTGGTATGGG 1845 Mp0191s0002 GAGCTTGGAATTGACGCCGCCC------CATAT---CACGGAGAGCCTCTGAACTTGACT 1846 Dl_DN4564 caatttg------aaatcgagtttgtgggttatgaaggcaaggagtttgtgagcttagag 1847 DlP5BR2 CAGCTTCAGCTTGAGATTGAGTTTGTGGGTTATGAAGGTAAGGAGCCGGTGTCTTTGGAG 1848 AT4G24220_VEP1 CAGTTTGGGATTGAGGAG------TATGGATTTGAGGAAGGGAAGAATTTGGGGTTGGTG 1849 Erche02g027660 CAGTTTGGAATTGAGCAG------TATGGATTTGAGGAAGGAAAGAATTTGGGGTTAGTG 1850 Ecre_P5BR1 CAGTTTGGGATTGAGGAG------TATGGATTTGAGGAAGGGAAGAATTTGGGGTTAGTG 1851 Dl_DN477 caatttgaggttgagtat------gcagagtacgaggaaggggagaatctgagcttccaa 1852 DlP5BR1 cagtttggagtagggtgt------ggagagtatgaagaaggggtggatttgaaattgcag 1853 * ** * * 1854 1855 Dl_DN67489 ttggcgatgggggataagggtggtgtttggagggacattgttgagaaagaggggttggtg 1856 AT5G58750 AGAGAGATGGTTGAGAGAAAACATGTGTGGGATGAGATTGTTGTGAAGGAGAAGCTTGTT 1857 Erche06g007150 AGAGAGATGGGAGAGAGAAAACATGTGTGGGACGAGATTGTGGTGAAGGAGGGGCTTGTT 1858 Erche06g011020 AGAGAGATGGGCGAGAGAAAAAATGTATGGGACGAGATTGTGGTGAAGGAGAGGCTTGTT 1859 Ecre_P5BR2 AGAGAGATGGGAGAAAGAAAACATGTGTGGGACGAGATTGTGGTGAAGGAGAGGCTTGTT 1860 Mp0191s0002 CAGAACTTCAGCGGGAAGGATCAAGCATGGGATACGATTGTGAAGGAAAAGGGCTTGGTT 1861 Dl_DN4564 aatctcatgaaggataaggactcggtgtgggacgagatagtgaagaagtacgatcttgtc 1862 DlP5BR2 GGTCTCATGAAGGATAAGGACTCGGTATGGGACGAGATGGTGGAGAAATACGACCTTGTC 1863 AT4G24220_VEP1 GAGATGATGAAAGGGAAAGAGAGAGTGTGGGAGGAGATGGTTAAGGAGAATCAATTGCAA 1864 Erche02g027660 GAGATGATGAAGGGGAAAGAGAGAGTGTGGGAGGAGATGGTGAAGGAGAATCAGTTGCAG 1865 Ecre_P5BR1 GAGATGATGAAGGGGAAAGAGAGAGTGTGGGAGGAAATGGTGAAGGAGAATCAGCTGCAG 1866 Dl_DN477 gatttgatgaaggataaaggtccgatttgggatgaaattgtgggggagaacggattggag 1867 DlP5BR1 gatttaatgaaggggaaggagccggtttgggaggaaatcgtgagggagaatggattgaca 1868 * * * *** ** ** * * * * 1869 1870 Dl_DN67489 gagactgagatggaggatttggctaactgggagtttctggatattcttttccgatgcccc 1871 AT5G58750 AGGACAGAGATTGAGGATTTGGCGAATTGGTATTTCTTGGATGCTCTGTTTAGATGTCCG 1872 Erche06g007150 CGGACAGAGATTGAGGATTTGGCGAATTGGGTTTTCTTGGATGGGCTGTTTAGATGTCCG 1873 Erche06g011020 CGGACAGAGATTGAGGATTTGGCGAATTGGGTTTTCTTGGATGTGCTGTTTAGATGTCCG 1874 Ecre_P5BR2 CGGACAGAGATTGAGGATTTGGCGAATTGGTTTTTCATGGATGTGCTGTTTAGATGTCCG 1875 Mp0191s0002 CCGACCAAATTGAAAGATGTTGGCAATTTTTGGTTCGCGGACCTCGTCCTCAACGTGCCG 1876 Dl_DN4564 ccgacgaaacttagggacatcgcggcgttttggtttgcagatgtggcgtttagcatcgag 1877 DlP5BR2 CCCACGAAACTTAGGGACATAGCGGCATTTTGGTTTGCGGATGTGGCGTTTAGCATCGAG 1878 AT4G24220_VEP1 GAGAAGAAGCTTGAGGAAGTTGGTGTGTGGTGGTTTGCTGATGTTATACTTGGAGTTGAA 1879 Erche02g027660 GAGAAGAAGCTTGATGAAGTTGGTGTGTGGTGGTTTGCGGATGTTATACTTGGAGTTGAA 1880 Ecre_P5BR1 GAGAAGAAGCTTGAGGAAGTTGGTGTGTGGTGGTTTGCGGATGTTATACTTGGAGTTGAA 1881 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 55 Dl_DN477 actaataaattggaagacattgcgaattgggggttcgcggatcttgtacttggctttgaa 1882 DlP5BR1 cctacgaaactgaaggatgtcggaatttggtggtttggtgatgttatacttgggaatgag 1883 * * * ** * * * ** ** * 1884 1885 Dl_DN67489 gcgaagatgttagcaacccgtgacaagattgatcgaatgggattcgagataaggtatcgt 1886 AT5G58750 TTTAAGCTTCTTGGGAAAAGAGAAAAAGTAGATAGGTTTGGGTTTAAGAGGAAATATAGA 1887 Erche06g007150 TTTAAGCTTCTTGGGAAGAGAGATAAAGTAGATAGGTTTGGGTTTAAGAGGAAATGTAGA 1888 Erche06g011020 TTTAAGCTTCTTGGGAAGAGAGATAAAGTGGATAGGTTTGGGTTTAAGAGGAAATATAGA 1889 Ecre_P5BR2 TTTAAGCTTCTTGGGAAGAGAGATAAAGTGGATAAGTTTGGGTTTAAGAGGAAATATAGA 1890 Mp0191s0002 TTCGAGAATGTGAGCTGCATGAACAAGGCCAAGGAGTTCGGATTCCATGGCTTCCGTGAC 1891 Dl_DN4564 ggagcggt---ttcgagcatgaacaagaacaaggagtttgggttcatggggttcagggat 1892 DlP5BR2 GGAGCGGT---TTCGAGCATGAACAAGAACAAGGAGTTTGGGTTCATGGGGTTCAGGGAT 1893 AT4G24220_VEP1 GGAATGAT---TGATAGTATGAACAAGAGTAAGGAATATGGCTTCCTTGGTTTCAGGAAC 1894 Erche02g027660 GGAATGAT---TGATAGTATGAACAAGAGTAAGGAACATGGCTTCCTTGGTTTTAGGAAC 1895 Ecre_P5BR1 GGAATGAT---TGATAGTATGAACAAGAGTAAGGAACATGGCTTCCTTGGTTTTAGGAAC 1896 Dl_DN477 tgtctcct---tgatacgatgaacaagagcaaggagcatggatttaagggatttaggaac 1897 DlP5BR1 tgtttcct---ggatagtatgaacaagagcaaggagcatggctttttgggatttaggaac 1898 * ** * ** ** 1899 1900 Dl_DN67489 gctttggattcgatatcgtattgcattaatttgatgcgagtcgagagatttatcccgtaa 1901 AT5G58750 ACCCTAGATTCGGTTTTGTATTGGATTGATGTGATGAGAGATGAAAAACTCATTCCTTTG 1902 Erche06g007150 ACCCTAGATTCGATTTTGTATTGGATCGATGTGATGAGAGATGAAAAACTCATTCCTTTG 1903 Erche06g011020 ACCCTAGATTCGATTTTGTATTGGATTGATGTGATGAGAGATGAAAAACTCATTCCTTTC 1904 Ecre_P5BR2 ACCCTAGATTCGATTTTATATTGGATTGATGTGATGAGAGATGAAAAACTCATTCCTTTG 1905 Mp0191s0002 ACGGAAAACTCAGTGAAGTCTATCATTCATGAAATGGTCGAAGCTAAAATCATACCGTCA 1906 Dl_DN4564 agcaccaagtcttttatatcttctgttaacaaagtgagatcttacagatttgtcccttga 1907 DlP5BR2 AGCACCAAGTCTTTTATATCTTCTGTTAACAAAGTGAGATCTTATAGATTTGTCCCTTGA 1908 AT4G24220_VEP1 TCCAACAACTCTTTTATCTCTTGGATTGACAAGTACAAGGCGTTCAAGATCGTACCTTGA 1909 Erche02g027660 TCCAACAACTCTTTTATCTCTTGGATTGACAAGTACAAGGCTTTCAAGATCGTGCCTTGA 1910 Ecre_P5BR1 TCCAACAACTCTTTTATCTCTTGGATTGACAAGTACAAGGCTTTCAAGATCGTGCCTTGA 1911 Dl_DN477 tccaagaactcgttcatttattggattgataagatgagatcttacaagattgtgccctaa 1912 DlP5BR1 tccaagaatgcgttcatttcttggattgacaaggcaaaagcttacaagattgttccttga 1913 * * * * * * * * * * ** * 1914 1915 Dl_DN67489 --------------------------------------- 1916 AT5G58750 TAA------------------------------------ 1917 Erche06g007150 TAA------------------------------------ 1918 Erche06g011020 TAA------------------------------------ 1919 Ecre_P5BR2 TAA------------------------------------ 1920 Mp0191s0002 TTCCACGAAAGATATCAGTTGAGACACACAGCTGAATGA 1921 Dl_DN4564 --------------------------------------- 1922 DlP5BR2 --------------------------------------- 1923 AT4G24220_VEP1 --------------------------------------- 1924 Erche02g027660 --------------------------------------- 1925 Ecre_P5BR1 --------------------------------------- 1926 Dl_DN477 --------------------------------------- 1927 DlP5BR1 --------------------------------------- 1928 1929 1930 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 56 Figure S18. Multiple sequence alignment of steroid 5α-reductases (5αR/DET2). Coding 1931 sequences were aligned using Clustal Omega. Species included: Arabidopsis thaliana (At/AT), 1932 Calotropis gigantea (Cg), Erysimum cheiranthoides (Ec/Erche), Digitalis lanata (Dl), and 1933 Marchantia polymorpha (Mp). Sequences correspond to gene phylogeny in Figure 6d. 1934 1935 CLUSTAL O(1.2.4) multiple sequence alignment 1936 1937 1938 M_polymorpha_0170s0018 -----------------ATGCCAACGTTTGAAA----TTCACGACGAGGCGTGGTATCCT 1939 Cg_12956 ATGGTGGATCTGCCTTTTACCAATTCTGCAACAATGGCTGCAGATGAAGCATTCTTCAAG 1940 Erche07g010950 ---------------------------ATGGAAGAGATCGCCGATCAGAGCTTCTTCCGC 1941 AtDET2_AT2G38050 ---------------------------ATGGAAGAAATCGCCGATAAAACCTTCTTCCGA 1942 Cg_5349 ---ATGACACTATCACAGTTTCCTTCCTGGAAAATGATTACAGATCAAGAACTCTTCCAT 1943 Dl2934 ---------------------------------atgaactccgaccaagatttatacaac 1944 ** * * 1945 1946 M_polymorpha_0170s0018 TGGATACTGGGGGGACTCTTTGCACTCTCTCTAGTAACTTACTGGGCTTGTGACCGTATA 1947 Cg_12956 TACTGTCTCATCATTTTCTCACTATTCTCACCAATTTCAAGCCTCT ---------CCTCA 1948 Erche07g010950 TATTGCGTCCTCACCCTAATTTTCTCCGGCCCACCAACCGCCGTCGCTTTAAAATTCCTC 1949 AtDET2_AT2G38050 TACTGTCTCCTCACTCTTATTTTCGCCGGCCCACCAACCGCCGTCCTTCTGAAATTCCTC 1950 Cg_5349 TACAGCCTCATCGTTTTCTTTCTAATCTCACCACCAACTGCAGTCGCCCTTCGCTATATC 1951 Dl2934 tactgtctcctctcaatcctcttcctcacacctcccaccatcatcgctctcctcttcctc 1952 * * * * * * 1953 1954 M_polymorpha_0170s0018 ACGGCTCCCTATGGC---CGTCATGTAAAGCGTGGGTGGGGTCCTGCCTGGGGAGTACGT 1955 Cg_12956 ACCGCTTCCTACGGCAGACATCGTCGTCCAACGACCACCGGAAAAACCATCCCGGCACCA 1956 Erche07g010950 GAAGCTCCTTACGGC---AAACACTTCCGTTCCGGATGGGGTCCCACCGTATCTCCGCCG 1957 AtDET2_AT2G38050 CAAGCTCCTTACGGT---AAACACAACCGTACCGGATGGGGTCCCACCGTATCTCCACCG 1958 Cg_5349 ACAGCTCCTTACGGC---AAGCACAGGCGTTCAGGCTGGGGACCCACCATCCCCGCACCT 1959 Dl2934 accgccccgtacggc---aagcacaaccgccccggatggggccccaccatcccgccgcca 1960 ** * ** ** * ** ** * 1961 1962 M_polymorpha_0170s0018 GAGTGCTGGATTGTGATGGAGAGTCCTGCTCTATGGGCAATGGTTTTGTTCTATTCTATG 1963 Cg_12956 ATAGCATGGTTCCTCTTCGAAAGTCCTACGCTCTGGCTTACAATCCTCCTATTTCCATTA 1964 Erche07g010950 ATTGCTTGGTTCGTCATGGAGAGTCCGACCTTGTGGCTCACGCTCCTTCTCTTCCCCTTT 1965 AtDET2_AT2G38050 ATTGCTTGGTTCGTCATGGAGAGCCCAACCTTGTGGCTCACTCTCCTCCTCTTCCCCTTT 1966 Cg_5349 GTGGCTTGGTTCCTGATGGAGAGTCCAACACTCTGGCTCACCACCCTCCTCTTCCCTTTT 1967 Dl2934 ctcgcctggtttctaatggagagccccacattgtggctcacactcctcctcttcccccac 1968 *** * * * ** ** ** * * *** * * * * * 1969 1970 M_polymorpha_0170s0018 GGTGAGCAAAAGTTGGGTCGTGTACCATTGATTCTACTCAGATTACACCAGGTCCATTAC 1971 Cg_12956 GGCAAAAATCACTCAAATCCAAAATCCATAATCCACATATCGTTCTACCTAATCCACTAT 1972 Erche07g010950 GGTCGTCACTCTCATAATCCTAAATCTCTCCTCCTCTTCTCTCCTTTTCTCCTCCATTAC 1973 AtDET2_AT2G38050 GGTCGTCACGCTCTCAACCCTAAATCTCTACTTCTATTCTCTCCTTATCTCATTCATTAC 1974 Cg_5349 GGTCAAAACAGGACCAATCCAAAAGCTCTTGTCTTGGGATCACTCTACCTCATCCACTAT 1975 Dl2934 ggcagaaacagccacaatccacgcgccttcatcctcatgtcccccttcctcctccactac 1976 ** * * * * * * * ** ** 1977 1978 M_polymorpha_0170s0018 TTCAACAGGGTACTCATCTATCCCATGCG ------------------------------- 1979 Cg_12956 TTCCACCGCGTTCTAATCTATCCGCTCGTTATGTTCCTAATAGTTAACCGCAAAAACCCT 1980 Erche07g010950 TTCCACCGCACCATCATTTACCCTCTTCGCCTCTTCCGCAGCTCCACCTC ---------- 1981 AtDET2_AT2G38050 TTCCACCGCACCATCATTTACCCTCTTCGCCTCTTCCGCAGCTCCTTCCC ---------- 1982 Cg_5349 TTCCACCGCACTATCCTCTACCCAATTAGACTATATTTAAGGAGCTCACC ---------- 1983 Dl2934 ctccaccgtacgttcatttacccactcagactcttcctcaagatccggaa ---------- 1984 ** ** * * * ** ** * 1985 1986 M_polymorpha_0170s0018 -----------CATGAAAGTACGAGGCAAGGGTATGCCAATCATAGTGGCAGCTTGCGCT 1987 Cg_12956 AAAACTAAAACCCCCAAATCCGGCGGACACGATTACCCGATCAAAGAAACAGTTCTCGGA 1988 Erche07g010950 -----------CTCT------GGCAAAAGTGGTTTCCCGATCACTGTCGCCGCCATGAGT 1989 AtDET2_AT2G38050 -----------CGCC------GGTAAAAACGGATTTCCGATCACCATCGCCGCCTTGGCT 1990 Cg_5349 -----------CCGGAAAA---CGCCTAGCGAGTTTCCTGTAAGTATGGCATCGATGGCT 1991 Dl2934 -----------acagaaaatccggaaacccggtttcccggtcagcatggccctgctggct 1992 * ** * * * 1993 1994 M_polymorpha_0170s0018 TTCGCTTTCAACATTCTCAACAGCTATGTCCAAGCTCGGTGGCTGTCCAACTATGGATCA 1995 Cg_12956 ACAATTTACAATCTCATCAACGCTTATCTCCAAGCGAGATCGGCTTCGGAATACACCGAT 1996 Erche07g010950 TTCACCTTTAATCTCATCAACGGTTATATCCAGGCGAGGTGTGTTTCGCACTACAAGAAC 1997 AtDET2_AT2G38050 TTCACCTTTAATCTCCTCAATGGTTATATCCAGGCGAGGTGGGTTTCGCATTACAAGGAT 1998 Cg_5349 TTTGTTTTCAATATCTTGAATGCTTATTTGCAGATAAGGTGGGTCACACACTATGCT --- 1999 Dl2934 tttgggtttaatttgttgaatggttacttgcaagccagatgggtttctgaatacgcc --- 2000 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint 57 * ** * * ** ** * ** * * * * ** 2001 2002 M_polymorpha_0170s0018 TACCCCGAC------TCATGGCTTACCAGTCCGAAGTTTATTCTCGGAGCAACTCTGTTT 2003 Cg_12956 TTCGGTAATGATAATGATAAATTATTCCGTTTACGATTTATGTTCGGAATGGCTATTTTC 2004 Erche07g010950 GACTACGAA---GACACACATTGGTTCTGGTGGCGGTTTATTATTGGTACGGTGGTGTTT 2005 AtDET2_AT2G38050 GACTACGAA---GACGGAAACTGGTTCTGGTGGCGGTTTGTTATCGGTATGGTGGTTTTC 2006 Cg_5349 GTGTACGAA---GATAGTAAGTGGTTCTGGTGGCGGTTCTTTACTGGACTTGCCATTTTC 2007 Dl2934 gatctagac---ggagatcagtggttctggatgcggtttacaagcggcgcggtggttttc 2008 * * * ** ** * ** 2009 2010 M_polymorpha_0170s0018 GGGCTGGGATTTTTGGGCAATTTTTGGAGCGACTCA ------------------------ 2011 Cg_12956 GTTACAGGAATGACGATAAATGTGAGTTCGGATTACGCTTTACTGAGACTGAAGAGCGAT 2012 Erche07g010950 GTAGCCGGTATGTGGATAAATATCACGTCGGACCGGACTCTGGTACGAC ----------- 2013 AtDET2_AT2G38050 ATAACCGGCATGTATATAAATATCACGTCGGACCGGACTTTGGTACGAT ----------- 2014 Cg_5349 GTGGTGGGCATGGCAGTAAACGTAATGTCAGACTATGCCTTGTTGGAGC ----------- 2015 Dl2934 gtcggcggcatgatggcgaatatatggtcggataatttgttgatggggt ----------- 2016 ** * ** * ** 2017 2018 M_polymorpha_0170s0018 ---------------TACCTCTTCTCACTGCGGGCAGACGAAGATGATAGGAGCTACAAG 2019 Cg_12956 CAAACGACGTCGGGAGCGGGAGCAGGAGAAGGAGAAAAGGGAATAAAAAATAGATACAAA 2020 Erche07g010950 -------------------------------TGAAGAAAGAGAATCGGGGAGGTTATGTG 2021 AtDET2_AT2G38050 -------------------------------TGAAGAAAGAGAACCGGGGAGGTTATGTG 2022 Cg_5349 -------------------------------TGAAGAGTAAGGGTGGTGGGGGATACAAA 2023 Dl2934 -------------------------------tgaagaaagagagtggaggagggtatagg 2024 * ** 2025 2026 M_polymorpha_0170s0018 ATTCCGAAGGCCGGTTTGTTTAAGTTCATTACATGTCCGAATTATTTTTCGGAGATGGTA 2027 Cg_12956 ATACCAAAAGGGGGAATGTTCGAATTGGTGAGCTGTCCAAATTATTTTGGGGAAATAATG 2028 Erche07g010950 ATACCGAGAGGAGGCTGGTTCGAGCTTGTAAGCTGTCCGAATTACTTGGGAGAGGCGATA 2029 AtDET2_AT2G38050 ATACCGAGAGGAGGCTGGTTCGAGTTGGTAAGCTGTCCGAATTATTTTGGAGAGGCGATT 2030 Cg_5349 ATTCCCAGAGGTGGGTTGTTTGAGTTGGTTAGTTGCCCAAATTATTTTGGAGAAGTATTG 2031 Dl2934 gtaccaaagggtgggctgttcgagtgggttagcagcccaaattattttggggagatagtg 2032 * ** * * ** *** * * * * ** ***** ** ** * 2033 2034 M_polymorpha_0170s0018 GAATGGTTGGGATGGGCCATCATGACCTGGTCTCCAGCTGGGCTTGCTTTCTTCATTTAC 2035 Cg_12956 GAATTTTTTGGTTGGGCATTGGTGACGTGTCGTTGGGCACCTTTTATATTCTTCTTAAAC 2036 Erche07g010950 GAGTGGTTGGGCTGGGCTGTAATGACTTGGTCTTGGGCCGGGTTTGGTTTTTTTCTGTAC 2037 AtDET2_AT2G38050 GAGTGGTTGGGCTGGGCTGTTATGACTTGGTCTTGGGCCGGTATTGGATTTTTTCTGTAC 2038 Cg_5349 GAGTGGTTCGGATGGTGTTTGGTGAACTGGTCCTGGTCCAGTTTGGGGTTTTTCTTGTTC 2039 Dl2934 gagtggttcgggtgggccgtgatgacttggtcttggtccgggtttgggttcttcatgtgg 2040 ** * ** ** *** * *** ** * * ** ** * 2041 2042 M_polymorpha_0170s0018 ACCATAGCTAATCTGGCCCCTAGAGCTGTAAGCAACCATCAGTGGTACCTGAGCAAGTTC 2043 Cg_12956 ACTTGTAGTATTTTGATCCCTAGGGCTTTTGCACATCACAAATGGTATTTGGATAAGTTT 2044 Erche07g010950 ACGTGTTCCAATTTGTTCCCGCGTGCACGTGCGAGCCACAAGTGGTACATGGACAAGTTC 2045 AtDET2_AT2G38050 ACGTGTTCCAATTTGTTTCCGCGTGCACGTGCGAGTCACAAGTGGTACATTGCCAAGTTC 2046 Cg_5349 ACTTGTGCTAATTTGATCCCTAGGGGTGTCTCAAATCACAAATGGTACCTTGACAAGTTT 2047 Dl2934 acttgtggtaatttagttcccagagcagcggcaactcacaagtggtatttggagaaattt 2048 ** * * * ** * * ** * ***** * ** ** 2049 2050 M_polymorpha_0170s0018 AATGACTAT---CCCAAAGAGCGAAGAATTCTTATACCCTTTGTGTATTGA 2051 Cg_12956 GGAGAAGAATATCCTAAAAATAGAAAAGCTGTTATTCCTTTCGTATGTTAA 2052 Erche07g010950 AAGGATGAGTATCCCAAGACTCGAAAAGCTGTTATTCCTTTTGTGTACTGA 2053 AtDET2_AT2G38050 AAGGAAGAGTATCCCAAGACTCGTAAAGCTGTTATTCCTTTTGTGTACTGA 2054 Cg_5349 GCAGAGGACTACCCTAAGAACAGAAAAGCTGTTATTCCATTTATTTACTGA 2055 Dl2934 ggagaggattatcccaagaacaggaaagctgttattccatttgtatactaa 2056 ** * ** ** * * * * **** ** ** * * * * 2057 2058 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted April 11, 2024. ; https://doi.org/10.1101/2024.04.10.588904doi: bioRxiv preprint