Figures and data
Expression of 2-ODD-C23 isoforms induces guanidine accumulation in E. coli. The cDNAs of various 2-ODD-C23 from plants isoforms and Ava5009 from Anabaena variabilis were cloned in pET24 for overexpression of the respective proteins in E. coli. Protein expression was induced over night at 18 °C in LB medium supplemented with 0.2% (w/v) glucose and 10 µM Fe2+. One aliquot of each culture was additionally supplemented with 1 mM homoarginine (HoArg). All cultures reached an OD around 10. For guanidine and pipecolate quantification by LC-MS, aliquots of the cultures were diluted in a 20-fold excess of methanol with 10 µM 13C15N-labelled guanidine as analytical standard. Pipecolate was exclusively detected in the cultures supplemented with homoarginine. Columns are the average ±SD of samples from three independent cultures.
Kinetic analysis of purified recombinant 2-ODD-C23 isoforms. (a)-(c) Specific activities of oxygen consumption at various substrate and inhibitor concentrations were analyzed with a Clark-type electrode. The concentration of the co-substrate 2-oxoglutarate was fixed at 0.5 mM. Only Din11s showed oxygen consumption with both homoarginine (HoArg) or arginine (Arg). The reaction of Din11s and At3g50210 with homoarginine was inhibited in the presence of 1 mM arginine (blue lines) or canavanine (red lines). Specific activities were calculated from the slope of the initial linear reaction rate. KM, and Amax were determined by non-linear regression (see also Supplementary Table S1). Data points represent single measurements, consistent results were obtained with independent enzyme preparations. (d) Reactions catalyzed by 2-ODD-C23 isoforms with subsequent spontaneous decay of the reaction products.
Subcellular localization of 2-ODD-C23 isoforms. Protoplasts were isolated from plants expressing 2-ODD-C23:GFP fusions under control of the constitutive UBQ10-promoter. Spectrally resolved confocal fluorescence images were used to separate GFP (depicted in green) fluorescence from chlorophyll autofluorescence (Chl, depicted in red). BF, brightfield image.
Guanidine and homoarginine contents in leaves of Arabidopsis plants. Plants were grown for two weeks in sterile culture and transferred to fresh plates with 0.2 mM or without (ctrl) homoarginine. Guanidine (a) and homoarginine (b) were extracted from rosettes in 80% (v/v) methanol containing 13C15N-labelled guanidine as analytical standard and quantified by LC-MS. Col-0, wildtype; At3g50210:GFP, At3g50210:GFP overexpressing plants, triple-KO, 2-ODD-C23 triple mutant. The yellow line marks a change in the scale of the Y-axis. Error bars indicate SD from four plant samples. The entire experiment was repeated with consistent results.
Effect of darkness and methyljasmonate (MeJA) on 2-ODD-C23 mutants. (a), Eight-day-old seedlings grown under short-day conditions were placed for 8 days in permanent darkness. Chlorophyll was extracted before and after the dark treatment and quantified spectrometrically. Col-0, wildtype; triple-KO, 2-ODD-C23 triple mutant; Din11s:GFP, At3g50210:GFP, plants overexpressing 2-ODD-C23:GFP fusion proteins. Error bars indicate SD of N = 8 independent plant samples. (b), Three-week-old plants were treated for one week with approximately 50 µM MeJA by applying a stock solution in 50% (v/v) ethanol to the lid of the Petri dishes. Guanidine was extracted from rosettes in 80% (v/v) methanol containing 13C15N-labelled guanidine as analytical standard and quantified by LC-MS. din11-1, single T-DNA insertion mutant of Din11. The yellow line marks a change in the scale of the Y-axis. Error bars indicate SD of 4 independent plant samples. The din11-2 mutant line and the independently generated second triple mutant gave consistent results.
Reduction and oxidation of AASA/P6C by plant enzymes. Recombinant purified enzymes were used to compare the reduction of P6C to pipecolate by pyrroline-5-carboxylate reductase from rice (OsP5CR; a) with the oxidation of AASA by Arabidopsis aldehyde dehydrogenase 7B4 (ALDH7B4; b) at 35 °C with a fixed cofactor concentration of 1 mM. OsP5CR used either NADH or NADPH as electron donor for the conversion of P6C to pipecolate. ALDH7B4 had a much higher specific activity and a lower KM, but efficiently catalyzed the reaction of AASA to aminoadipate only with NAD+ as electron acceptor. Data points are the average ±SD of technical triplicates. Colored lines represent the fit of the experimental data to the Michaelis-Menten equation. The insets show the Lineweaver-Burk plots of the same data. All experiments were repeated with an independent enzyme preparation and showed consistent results.
Phylogeny of the 2-ODD clade C23. Excerpt of the gene tree for Din11 from EnsemblPlants (https://plants.ensembl.org, accessed Feb. 2023). The length of the horizontal blue lines is proportional to the number of amino acid exchanges except the dashed line, which is shortened by a factor of 10. Blue squares indicate speciation nodes, red squares indicate duplication nodes, green squares indicate ambiguous nodes, and open squares indicate species nodes. Grey triangles indicate collapsed branches. The three 2-ODD-C23 paralogues from Arabidopsis thaliana are highlighted by blue lettering.
3′-UTR of Din11. Alignment of RNAseq reads (pink or purple lines) from two independent experiments with the genomic sequence (four-coloured line) of Arabidopsis around the transcription start site of Din11. The 5′-end of the annotated Din11 transcript At3g49620.1/Din11L is hardly covered by RNAseq reads, which are much more frequent several basepairs downstream of the annotated start codon. The graphical representation of the RNAseq reads was modified from the Arabidopsis RNAseq database (Zhang et al., 2020).
Distribution of metabolites in E. coli cultures. (a) E. coli cells expressing Din11s were cultivated over night at 18 °C in medium supplemented with either 5 mM 2-oxoglutarate (2-OG), 5 mM arginine (Arg), or a combination of both. The cultures were diluted with a 20-fold excess of methanol containing 10 µM 13C15N-labelled guanidine as analytical standard. (b), (c) E. coli cells expressing soybean 2-ODD-C23 were grown over night at 18 °C in the presence of 1 mM homoarginine. The cells were harvested by centrifugation and lysed in 20 µl (mg Fw)-1 methanol containing 10 µM labelled guanidine. The supernatant was also mixed with a 20-fold excess of methanol with 10 µM labelled guanidine. Guanidine and pipecolate were quantified by LC-MS. The columns are the average ±SD of samples from three independent cultures. The experiments were repeated with consistent results. (d) and (e) Flame ionization detector traces of samples from the headspace of E. coli cultures overexpressing either ALDH7B4 (negative control), Din11s, or At3g50210 and supplemented with 5 mM Arg or homoarginine (HoArg) as indicated in the figures. One ml of a 1 ppm C2H4 standard or 2 ml of culture headspace were injected into a gas chromatograph (SGI 8610C, SRI Instruments, Los Angeles, CA, United States) equipped with a 3 m Hayesep-D column (60 °C, carrier gas N2). C2H4 was detected with a flame ionization detector (135 °C). The software PeakSimple v4.44 was used to record the chromatograms. The detector output was smoothed by averaging 9 measurements recorded at 5 Hz. Retention times for C2H4 shifted over time but were consistent within each set of measurements.
Biochemical characterization of 2-ODD-C23 isoforms. (a), Lineweaver-Burk plot of the activity data for Din11s from Fig. 2a. Values in brackets are very low activities at low concentrations of homoarginine (HoArg) in presence of 1 mM arginine (Arg) or canavanine (Can) that were excluded from the linear regression. (b), Lineweaver-Burk plot of the activity data for At3g50210 from Fig. 2c. Regression lines intersecting on or very near the Y-axis indicate competitive inhibition. (c) Specific activity of At3g50210 in dependence on the concentration of the co-substrate 2-oxoglutarate was determined at 30 °C. Homoarginine concentration was fixed at 10 mM. The blue line represents the least square fit to the Michaelis-Menten equation with an additional term for substrate inhibition. (d), pH-dependence of the specific activity of At3g50210 in a mixed buffer system with MES, HEPES and CHES. Homoarginine and 2-oxoglutarate (2-OG) concentrations were fixed at 10 mM and 0.5 mM, respectively. The dashed blue line is a polynomic fit with a 4th-order polynomial. Dots represent single measurements and independent enzyme preparations gave consistent results.
2-ODD-C23 mutants in Arabidopsis. (a) Schematic representation of the three genes for 2-ODD-C23 isoforms in Arabidopsis. Thick blue lines represent exons, wide red triangles represent T-DNA insertions, narrow red triangles represent CRISPR/Cas9 target sites. The red lines indicate a fusion event between Din11 and At3g49630, in which the DNA fragment between the two CRISPR/CAs9 target sites was lost. (b) Guanidine content in rosettes of single and triple mutant seedlings cultivated under axenic conditions. The seedlings were cultivated for 9 days under long day conditions on half-strength MS medium supplemented with 2% sucrose. Subsequently, they were transferred for one week to plates supplemented additionally with 0.2 mM homoarginine. Columns represent the average ±SD of N = 4 independent samples. Different letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA with Bonnferroni correction for multiple testing.
Guanidine content in various plants. Black locust (Robinia pseudoaccacia) and Arabidopsis samples were from soil-grown plants. Seeds were surface-sterilised and germinated on filter paper soaked with ddH2O for the indicated times. Guanidine was extracted in 80% (v/v) methanol with 10 µM 13C15N-labelled guanidine as analytical standard and quantified by LC-MS. Columns represent the average ±SD of N = 3 independent samples.
Growth inhibition by homoarginine. (a), (b) Seedlings of WT Arabidopsis plants (Col-0) and 2-ODD-C23 overexpressors or triple mutants were grown for 6 days under long-day conditions on vertical plates containing half-strength MS medium, supplemented with 2% (w/v) sucrose and either arginine (Arg) or homoarginine (HoArg). Columns in (b) represent the average ±SD of N = 40 seedlings. (c) Two-week old seedlings grown on half-strength MS medium with 2% (w/v) sucrose and the indicated concentration of homoarginine. The positioning of the genotypes is identical on each plate. (d) Seven-day-old seedlings grown on vertical plates without or with 1 mM guanidine (Gd). Scale bars in (a), (c) and (d) are 1 cm.
Expression levels of 2-ODD-C23s under stress. RNAseq coverage of the three Arabidopsis 2-ODD-C23 isoforms under various abiotic and biotic stress conditions. The data plots were modified from the Arabidopsis RNAseq database (Zhang et al., 2020). Numbers in brackets are the number of samples per treatment. FPKM, Fragments per kilobase of transcript per million mapped reads.
Characterization of ALDH7B4 (At1g54100). (a) Activity of purified recombinant ALDH7B4 with different concentrations of NAD+ at 35 °C. AASA/P6C concentration was fixed at 1 mM. KM and Amax values were obtained by nonlinear regression to the Michaelis-Menten equation (red line). The inset shows the Lineweaver-Burk plot of the same data. (b) AASA/P6C-dependent reduction of NADP+ by different amounts of ALDH7B4. Concentrations of both NADP+ and AASA/P6C were 1 mM. (c) GSA/P5C-dependent reduction of NAD+ by different amounts of ALDH7B4. NAD+ concentration was 1 mM and DL-GSA/P5C concentration was 2 mM. In b and c, the insets show the initial velocity as a function of the amount of protein in the assay. Data points are the average ±SD of technical triplicates. All experiments were repeated with an independent enzyme preparation.
2-Aminoadipate content after homoarginine feeding. Arabidopsis seedlings were grown for ten days under long-day conditions on half-strength MS medium supplemented with 2% (w/v) sucrose and 5 mM MES-KOH, pH 5.8 before they were transferred for four days to plates additionally containing 0.2 mM homoarginine. Guanidine and 2-aminoadipate were extracted in 80% (v/v) methanol containing 10 µM 13C15N-labelled guanidine as analytical standard and quantified by LC-MS. Columns represent the average ±SD of N = 4 independent samples. n.d., not detected. The experiment was repeated with consistent results.
Kinetic constants of Arabidopsis 2-ODD-C23 enzymes
Arabidopsis T-DNA insertion lines used in this study
