Theodor Lange

Isolation and Expression of Three Gibberellin 20-Oxidase cDNA Clones from Arabidopsis

Using degenerate oligonucleotide primers based on a pumpkin (Cucurbita maxima) gibberellin (GA) 20-oxidase sequence, six different fragments of dioxygenase genes were amplified by polymerase chain reaction from Arabidopsis thaliana genomic DNA. One of these was used to isolate two different full-length cDNA clones, At2301 and At2353, from shoots of the GA-deficient Arabidopsis mutant ga1–2. A third, related clone, YAP169, was identified in the Database of Expressed Sequence Tags. The cDNA clones were expressed in Escherichia coli as fusion proteins, each of which oxidized GA12 at C-20 to GA15, GA24, and the C19 compound GA9, a precursor of bioactive GAs; the C20 tricarboxylic acid compound GA25 was formed as a minor product. The expression products also oxidized the 13-hydroxylated substrate GA53, but less effectively than GA12. The three cDNAs hybridized to mRNA species with tissue-specific patterns of accumulation, with At2301 being expressed in stems and inflorescences, At2353 in inflorescences and developing siliques, and YAP169 in siliques only. In the floral shoots of the ga1–2 mutant, transcript levels corresponding to each cDNA decreased dramatically after GA3 application, suggesting that GA biosynthesis may be controlled, at least in part, through down-regulation of the expression of the 20-oxidase genes.

Feed-back regulation of gibberellin biosynthesis and gene expression in Pisum sativum L.

Treatment of tall and dwarf (3β-hydroxylase impaired) genotypes of pea (Pisum sativum L.) with the synthetic, highly active gibberellin (GA), 2,2-dimethyl GA4, reduced the shoot contents of C19-GAs, including GA1, and increased the concentration of the C20-GA, GA19. In shoots of the slender (la crys) mutant, the content of C19-GAs was lower and GA19 content was higher than in those of the tall line. Metabolism of GA19 and GA20 in leaves of a severe (na) GA-deficient dwarf mutant was reduced by GA treatment. The results suggest feedback regulation of the 20-oxidation and 3β-hydroxylation reactions. Feed-back regulation of GA 20-oxidation was studied further using a cloned GA 20-oxidase cDNA from pea. The cDNA, Ps074, was isolated using polymerase chain reaction with degenerate oligonucleotide primers based on pumpkin and Arabidopsis 20-oxidase sequences. After expression of this cDNA clone in Escherichia coli, the product oxidized GA12 to GA15, GA24 and the C19-GA, GA9, which was the major product. The 13-hydroxylated substrate GA53 was similarly oxidized, but less effectively than GA12, giving mainly GA44 with low yields of GA19 and GA20. Ps074 hybridized to polyadenylated RNA from expanding shoots of pea. Amounts of this transcript were less in the slender genotype than in the tall line and were reduced in GA-deficient genotypes by treatment with GA3, suggesting that there is feed-back regulation of GA 20-oxidase gene expression.

Gibberellin Biosynthesis from Gibberellin A12-Aldehyde in Endosperm and Embryos of Marah macrocarpus

Soluble enzyme preparations from embryos and endosperm of Marah macrocarpus (previously Echinocystis macrocarpa) were incubated with [14C4]gibberellin(GA)12-aldehyde,[14C4]GA12, [14C1] GA9, 2,3-didehydro[14C1]GA9, [14C1]GA20, and [17–13C,3H]GA5. Embryo preparations converted GA12-aldehyde, GA12, and GA9 to GA4 and GA7; 2,3-didehydroGA9 to GA7; GA5 to GA3; and GA20 (incompletely) to GA1 and GA60, but not to GA3. Endosperm preparations converted GA12-aldehyde and GA12 to GA15, GA24, GA25, and GA9, but, unlike embryo preparations, not to GA4 or GA7. However, GA4 and GA7 were formed from GA9 and GA7 was formed from 2,3-didehydroGA9. Metabolism of GA5 to GA3 and GA20 to GA1 was low. 2,3-DidehydroGA9 accumulated when GA9 was incubated with a desalted endosperm preparation. A cDNA clone (M3–8), selected from an embryo-derived cDNA library using a DNA fragment generated by reverse transcriptase polymerase chain reaction, was expressed in Escherichia coli. The fusion protein converted GA12 to GA9 (major) and GA25 (minor); GA53 was metabolized less effectively and only to GA44. Thus, the M3–8 protein is functionally similar to GA 20-oxidases from Arabidopsis thaliana, Spinacia oleracea, and Pisum sativum, but different from that from Cucurbita maxima seeds, to which its amino acid sequence is most closely related. mRNA hybridizing to M3–8 accumulated in embryos and endosperm of M. macrocarpus, but was absent in vegetative tissues.

Separation and characterisation of three 2-oxoglutarate-dependent dioxygenases from Cucurbita maxima L. endosperm involved in gibberellin biosynthesis

Three enzymes of the gibberellin (GA) biosynthetic pathway, a 7-oxidase, a 20-oxidase and a 3β-hydroxylase, were partially purified from Cucurbita maxima endosperm by ammonium sulfate precipitation, gel-filtration and anion-exchange chromatography. The enzyme activities, which were assayed by the oxidation of GA12-aldehyde to GA12, of GA12 to GA15 (and GA24) and of GA15 to GA37, respectively, were completely separated from each other. The apparent molecular masses as estimated by gel-filtration high-performance liquid chromatography were 34.5 kDa for the 7-oxidase, 44.5 kDa for the 20-oxidase and 58 kDa for the 3β-hydroxylase. The Michaelis-Menten constants (Km) were 8.6 μM, 0.15μM and 8.7 μM for the respective substrates. All three enzymes had properties typical of 2-oxoglutarate dependent dioxygenases. 2-Oxoglutarate was essential for activity and served as a co-substrate, giving Km values of 6.1 μM, 91 μM and 41 μM with the 7-oxidase, 20-oxidase and 3β-hydroxylase, respectively. Furthermore, 2 oxo[5-14C]glutarate was oxidised stoichiometrically to [14C]succinate when the GA-substrates were oxidised to their respective products, and the 1∶1 ratio was maintained under different oxygen concentrations. Approximately equimolar amounts of 14CO2 were released from 2-oxo[1-14C]glutarate when GA12 was oxidised to GA15/24 by the 20-oxidase. A crude enzyme preparation containing all three enzyme activities (and a 2β-hydroxylase) converted GA12-aldehyde to [18O2]GA4 and [18O5]GA43 under 18O2, showing that all O-atoms incorporated after GA12-aldehyde originate from O2. Accordingly, the reaction rates were near zero under anaerobic conditions, although very low concentrations of O2 sufficed to sustain the reactions. Both Fe2+ and dithiothreitol stimulated the enzyme activities strongly, but if they were added together, catalase was needed to prevent inhibition. The pH dependence showed two opposite trends; the 7-oxidase was most active at pH 6 and below, whereas the other enzymes were maximally active above pH 6.5.

Purification and partial amino-acid sequence of gibberellin 20-oxidase from Cucurbita maxima L. endosperm.

Gibberellin (GA) 20-oxidase was purified to apparent homogeneity from Cucurbita maxima endosperm by fractionated ammonium-sulphate precipitation, gel-filtration chromatography and anion-exchange and hydrophobic-interaction high-performance liquid chromatography (HPLC). Average purification after the last step was 55-fold with 3.9% of the activity recovered. The purest single fraction was enriched 101-fold with 0.2% overall recovery. Apparent relative molecular mass of the enzyme was 45 kDa, as determined by gel-filtration HPLC and sodium dodecyl sulphate-polyacrylamide gel electrophoresis, indicating that GA 20-oxidase is probably a monomeric enzyme. The purified enzyme degraded on two-dimensional gel electrophoresis, giving two protein spots: a major one corresponding to a molecular mass of 30 kDa and a minor one at 45 kDa. The isoelectric point for both was 5.4. The amino-acid sequences of the amino-terminus of the purified enzyme and of two peptides from a tryptic digest were determined. The purified enzyme catalysed the sequential conversion of [14C]GA12 to [14C]GA15, [14C]GA24 and [14C]GA25, showing that carbon atom 20 was oxidised to the corresponding alcohol, aldehyde and carboxylic acid in three consecutive reactions. [14C]Gibberellin A53 was similarly converted to [14C]GA44, [14C]GA19, [14C]GA17 and small amounts of a fourth product, which was preliminarily identified as [14C]GA20, a C19-gibberellin. All GAs except [14C]GA20 were identified by combined gas chromatography-mass spectrometry. The cofactor requirements in the absence of dithiothreitol were essentially as in its presence (Lange et. al, Planta 195, 98–107, 1994), except that ascorbate was essential for enzyme activity and the optimal concentration of catalase was lower.

Biosynthesis of 12α-and 13-hydroxylated gibberellins in a cell-free system from Cucurbita maxima endosperm and the identification of new endogenous gibberellins.

Gibberellin (GA) biosynthesis in cell-free systems from Cucurbita maxima L. endosperm was reinvestigated using incubation conditions different from those employed in previous work. The metabolism of GA12 yielded GA13, GA43 and 12α-hydroxyGA43 as major products, GA4, GA37, GA39, GA46 and four unidentified compounds as minor products. The intermediates GA15, GA24 and GA25 accumulated at low protein concentrations. The structure of the previously uncharacterised 12α-hydroxyGA43 was inferred from its mass spectrum and by its formation from both GA39 and GA43. Gibberellin A39 and 12α-hydroxyGA43 were formed by a soluble 12α-hydroxylase that had not been detected before. Gibberellin A12-aldehyde was metabolised to essentially the same products as GA12 but with less efficiency. A new 13-hydroxylation pathway was found. Gibberellin A53, formed from GA12 by a microsomal oxidase, was converted by soluble 2-oxoglutarate-dependent oxidases to GA1 GA23, GA28, GA44, and putative 2β-hydroxyGA28. Minor products were GA19, GA20, GA38 and three unidentified GAs. Microsomal 13-hydroxylation (the formation of GA53) was suppressed by the cofactors for 2-oxoglutarate-dependent enzymes. Reinvestigation of the endogenous GAs confirmed the significance of the new metabolic products. In addition to the endogenous GAs reported by Blechschmidt et al. (1984, Phytochemistry 23, 553–558), GA1, GA8, GA25, GA28, GA36, GA48 and 12α-hydroxyGA43 were identified by full-scan capillary gas chromatography-mass spectrometry and Kovats retention indices. Thus both the 12α-hydroxylation and the 13-hydroxylation pathways found in the cell-free system operate also in vivo, giving rise to 12α-hydroxyGA43 and GA1 (or GA8), respectively, as their end products. Evidence for endogenous GA20 and GA24 was also obtained but it was less conclusive due to interference.

The partial purification and characterization of a gibberellin C-20 hydroxylase from immature Pisum sativum L. seeds.

A gibberellin (GA) C-20 hydroxylase that catalyses the conversion of GA53 to GA44 was purified from developing pea embryos by ammonium-sulfate precipitation, gel filtration and anion-exchange column chromatography. The purification was about 270-fold and 15% of the enzymic activity was recovered. The relative molecular mass was 44000 by Sephadex G-200 gel filtration. The apparent Michaelis constant was 0.7 μM and the isoelectric point was 5.6–5.9. The enzymic activity was optimal at pH 7.0 2-Oxoglutarate and ascorbate were required for activity. Low concentrations of Fe2+ stimulated the reaction, but externally added Fe2+ was not essential, even in the most purified preparation. Catalase and bovine serum albumin also stimulated. Dithiothreitol preserved the activity during purification but was not needed during incubation. In fact, the simultaneous presence of dithiothreitol and Fe2+ in the incubation mixture was inhibitory to the purified enzyme. The cofactor requirements are typical for those of 2-oxoglutarate-dependent dioxygenases.When the incubation time was long enough, GA53 was converted to both GA44 and GA19. The proportions of these two products remained constant throughout the purification, but this does not necessarily mean that their formations is catalysed by a single enzyme. Sodium dodecyl sulfatepolyacrylamide gel electrophoresis showed that the final preparation contained several proteins. Although the most prominent protein band was located within the range expected for the enzyme on the grounds of its molecular weight, this band did not represent the enzyme, since it separated from the GA C-20 hydroxylase activity on ultrathin-layer isoeletric focusing.

Gibberellin biosynthesis in cell-free extracts from developing Cucurbita maxima embryos and the identification of new endogenous gibberellins

Gibberellin (GA) biosynthetic pathways from GA12-aldehyde, GA12 and GA53 were investigated in cell-free systems from developing embryos of Cucurbita maxima L. Gibberellin A12-aldehyde and GA12 were converted to GA25, putative 12α-hydroxyGA25, GA13 and GA39 as main products. Minor products were GA4, GA34 and, when GA12 was the substrate, putative 12α-hydroxyGA12. The intermediates GA15 and GA24 accumulated at low protein concentrations. The influence of various factors on GA12 metabolism was examined. At low 2-oxoglutarate and ascorbate concentrations, or at acid pH, 3β-hydroxylated products predominated, whereas with increasing 2-oxoglutarate and ascorbate concentrations, or at neutral pH, the yield of 12α-hydroxylated GAs increased. Gibberellin A53 was metabolised mainly to the C20-GAs GA44, GA19, GA17, GA23 and GA28, with the C19-GAs GA20, GA1 and GA8 as minor products. Only C19-GAs were 2β-hydroxylated, which is a main characteristic of the embryo systems. In addition to GA13, GA25, GA39, GA43, GA49, GA58, GA74, 12α-hydroxyGA25 and GA39 3-isovalerate, which were known previously from embryos of C. maxima, GA1, GA4, GA17, GA28, GA37, GA38, GA48, GA85, 12α-hydroxyGA37 and putative 12α-hydroxyGA43 were identified as endogenous components by full-scan capillary gas chromatography-mass spectrometry and Kovats retention indices. Evidence for putative 2β-hydroxyGA28 and GA23 was also obtained but it was less conclusive because of contamination.

The biosynthesis of ent-kaurene in germinating seeds and the function of 2-oxoglutarate in gibberellin biosynthesis

The two topics of this paper deal with different parts of the gibberellin (GA) biosynthetic pathway. The diterpene hydrocarbon ent-kaurene is the first tetracyclic precursor of gibberellins. Its two-step formation from geranylgeranyl pyrophosphate through the action of kaurene synthetase A and B initiates the diversion of the GA pathway from the general terpenoid pathway, and is a likely point for GA biosynthesis regulation. The capacity of a tissue to biosynthesize ent-kaurene is usually measured by the conversion of labelled precursors, such as mevalonate or geranylgeranyl pyrophosphate, to ent-kaurene in cell-free systems. This method works well with immature seeds, which are often highly active for GA biosynthesis, but cell-free systems from germinating seeds and seedlings, which are less active, give erratic and often negative results. We report here a new method of measuring ent-kaurene formation in vivo. In this procedure, ent-kaurene oxidation is inhibited by the addition of a plant growth retardant, thus causing ent-kaurene to accumulate, presumably at the same rate as its normal biosynthesis. The accumulation is sufficient, even in seedlings, to be quantified by the use of isotope dilution and GC-SIM (Groselindemann et al., 1991).