INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

The biosynthesis of the gibberellins (GAs) has been recently reviewed (MacMillan, 1997). In all systems studied, the pathway has been shown to proceed from the cyclic diterpene ent-kaurene to GA₁₂ aldehyde then to GA₁₂. Depending on the sequence of hydroxylation at the 3- and 13-positions, parallel pathways branch from GA₁₂ to the C₁₉-GAs, the number of these branch pathways varying from species to species. For maize (Zea mays L.) we previously demonstrated (see Fig. 1) the presence of the early 13-hydroxylation branch pathway, a pathway originating from GA₁₂ and leading to the hydroxylated C₁₉-GAs, GA₁, GA₃, and GA₅ (Kobayashi et al., 1996 and refs. therein; Spray et al., 1996). As shown in Figure 1, the steps from GA₁₂ to bioactive GA₁, GA₃, and GA₅, the early 13-hydroxylation branch pathway, have been established by feeding studies using labeled substrates; the immediate metabolites were identified by full-scan gas chromatography-mass spectrometry (GC-MS) and Kovats retention index (KRI) data (Fujioka et al., 1990; Kobayashi et al., 1996). All members of this branch pathway are native to maize (Fujioka et al., 1988a, 1988b).

View larger version (37K): [in this window] [in a new window]   Figure 1.   Maize branch pathways from GA12: right vertical row, the early 13-hydroxylation branch pathway; left vertical row, the presumptive non-early 3,13-hydroxylation branch pathway. All of the GAs are endogenous to maize except 2,3-dehydroGA9, shown in brackets. , Steps established in this paper; , steps previously established; - - , steps tested, not observed.

There is indirect evidence for the presence of a second pathway from GA₁₂, the non-early 3,13-hydroxylation branch pathway. The pathway originates from GA₁₂ and leads via GA₉ to the 3-hydroxylated C₁₉-GAs GA₄, and GA₇ (see Fig. 1). While the pathway has been shown to be present in a number of plant species (for review, see MacMillan, 1997), its presence in maize is based solely on the identification from maize of the five members GA₁₅, GA₂₄, GA₉, GA₄, and GA₇. Moreover, in vivo feeding studies have provided no evidence for the metabolism of GA₁₂ to GA₁₅ (Kobayashi et al., 1996), GA₉ to GA₄ (Davis et al., 1998), and GA₄ to GA₇ (Kobayashi et al., 1993).

In the present study, we describe the metabolism of [17-¹⁴C]GA₁₅, [17-¹⁴C]GA₂₄, [17-¹⁴C]GA₂₅, and [17-¹⁴C]GA₇ in seedlings of tall, dwarf-1 (d1), and dwarf-5 (d5) maize. Given the previous demonstration of the sequence GA₂₀  GA₅ (2, 3-dehydro-GA₂₀)  GA₃ in maize (Fujioka et al., 1990), the possible existence of a parallel sequence of GA₉  2,3-dehydro-GA₉  GA₇ was tested by feeding 2,3-dehydro-[17-¹⁴C]GA₉, a GA not reported to be present in maize (Fujioka et al., 1988b). The data obtained, together with the previous results from the metabolism of [17-¹³C,³H]GA₉ and [17-¹³C,³H]GA₄, are discussed in terms of the biosynthesis of GAs in maize.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Plant Material

Normal (tall), dwarf-1 (d1), and dwarf-5 (d5) maize (Zea mays L.) seedlings came from seed stocks of known genotype (Spray et al., 1996). The seeds were pre-soaked in water for 12 h and planted in vermiculite:soil (1:1). The seedlings were then grown in the greenhouse at the University of California, Los Angeles. Three- to four-week-old seedlings (three- to four-leaf stage) were used for feeds.

Labeled Substrates

[17-¹⁴C]GA₁₅ (2.07 TBq mol¹), [17-¹⁴C]GA₂₄ (2.07 TBq mol¹), and [17-¹⁴C]GA₇ (2.07 TBq mol¹) were purchased from Prof. L.N. Mander (Australian National University, Canberra). [17-¹⁴C]GA₂₅ (2.07 TBq mol¹) was prepared from [17-¹⁴C]GA₂₄ (300 kBq; a gift from Prof. L.N. Mander) with cell lysates (3.5 mL) from Escherichia coli NM522 containing clone E5 by methods detailed by Lange (1997) and purified as described by Lange and Graebe (1993). 2,3-Dehydro-[17-¹⁴C]GA₉ (1.75 TBq mol¹) was prepared as described in MacMillan et al. (1997).

Treatment, Purification, and Analysis

Each of the five labeled GAs, [17-¹⁴C]GA₁₅, [17-¹⁴C]GA₂₄, [17-¹⁴C]GA₂₅, [17-¹⁴C]GA₇, and 2,3-dehydro-[17-¹⁴C]GA₉, was dissolved in 90 µL of ethanol:water (1:1). Two microliters of the [17-¹⁴C]GA₁₅ solution (1,570 Bq; 250 ng) were individually injected into the coleoptile nodes of three sets of 10 plants (normal, d1, and d5). Similar injections were made with [17-¹⁴C]GA₂₄ (1,490 Bq; 250 ng), [17-¹⁴C]GA₂₅ (1,420 Bq; 250 ng), and [17-¹⁴C]GA₇ (1,550 Bq; 250 ng). One set of 10 d5 seedlings was used for the 2,3-dehydro-[17-¹⁴C]GA₉ injections (485 Bq; 88 ng).

The seedlings were incubated in the greenhouse for 24 h, harvested as sets of 10, frozen with dry ice, and stored at 80°C. Each set of frozen seedlings was homogenized, extracted, and solvent-fractionated to give an acidic ethyl acetate-soluble (AE) fraction. Each fraction was concentrated and purified using Bond Elut (Varian, Harbor City, CA) columns and two steps of HPLC (Davis et al., 1998). All samples were methylated and the GAs in each sample were identifed by full-scan GC-MS and KRI (Gaskin and MacMillan, 1991; Spray et al., 1996).

Isotopic Dilution

To determine whether 2,3-dehydro-GA₉ is endogenous to maize, [17-¹⁴C]2,3-dehydro-GA₉ (1.75 TBq mol¹) was used in an isotopic dilution experiment. Fifteen nanograms (3 Bq) was dissolved in 100 µL of 50% (v/v) aqueous ethanol and added to a homogenate from 50 normal maize seedlings (200 g fresh weight). The homogenate was extracted immediately and solvent fractionated to give an AE fraction. The fraction was processed for the determination of isotopic dilution using the isotope dilution fit program described by Croker et al. (1994).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Metabolism
[17-¹⁴C]GA₁₅

The recovered [¹⁴C]labeled metabolites GA₄₄, GA₁₉, GA₂₀, GA₁₁₃, and GA₁₅-15,16-ene (artifact?) are shown in Table I, and are based on identification by the full-scan GC-MS and KRI data presented in Table II. The step from GA₁₅ to GA₄₄ (Fig. 1) is a direct 13-hydroxylation that is new for maize. The observed 13-hydroxylation of GA₁₅ to GA₄₄ in maize (Fig. 1) has also been reported in a cell-free preparation from germinating barley (Grosslindemann et al., 1992). In addition, the opened lactone of GA₁₅ is metabolized to GA₄₄ from in vitro studies using seeds of pea (Kamiya and Graebe, 1983) and bean (Takahashi et al., 1986). The steps GA₄₄  GA₁₉ and GA₁₉  GA₂₀ have been previously demonstrated in maize seedlings (Kobayashi et al., 1996). The step from GA₁₅ to GA₁₁₃ (Fig. 2) is a direct 12-hydroxylation, which is new for maize and for higher plants. GA₁₁₃ has not been found to occur naturally in maize but has been recently isolated from the seeds and shoots of the Japanese radish (Nakayama et al., 1998). The relatively high levels of endogenous GA₄₄ and GA₁₉ present in the normal and d1 seedlings compared with the d5 seedlings (Fujioka et al., 1988a) may create feedback inhibition and thus account for the absence of the labeled metabolite GA₁₉, in the normal and d1 seedlings, in contrast to the recovery of [¹⁴C]GA₁₉ from d5 seedlings.

View larger version (30K): [in this window] [in a new window]   Figure 2.   Structures of metabolites not shown in Figure 1.

[17-¹⁴C]GA₂₄

The recovered [¹⁴C]labeled metabolites, GA₁₉, GA₂₀, and GA₁₇ are shown in Table III, and are based on identification by the full-scan GC-MS and KRI data presented in Table IV. The step from GA₂₄ to GA₁₉ (Fig. 1) is a direct 13-hydroxylation and is new for maize seedlings. The step from GA₁₉ to GA₂₀ has been previously established for maize (Kobayashi et al., 1996) with no evidence for the conversion of GA₁₉ to GA₁₇. However, the conversion of GA₁₉ to GA₁₇ has been demonstrated using GA 20-oxidases from spinach (Wu et al., 1996) and pumpkin (Lange et al., 1994), which have been cloned and expressed in E. coli.

[17-¹⁴C]GA₂₅

The recovered [¹⁴C]labeled metabolites GA₁₇, GA₂₅ 16,17-H₂-17-OH, and HO-GA₂₅ (hydroxyl position not determined) are shown in Table V, based on identification by the full-scan GC-MS and KRI data presented in Table VI. The metabolism of GA₂₅ to GA₁₇ (Fig. 2) is a result of direct 13-hydroxylation. This step is new for plants.

[17-¹⁴C]GA₇

The [¹⁴C]labeled metabolites GA₃₀, GA₃, isoGA₃, and GA₇-diene-diacid (trace amounts) are shown in Table VII, and are based on identification by the full-scan GC-MS and KRI data shown in Table VIII. However, in each case, most of the radioactivity was recovered in fractions (Table VII) that contained products not analyzable by GC-MS. The products are presumed to be conjugates.

2,3-Dehydro-[17-¹⁴C]GA₉

The recovered [¹⁴C]labeled metabolites, GA₅, GA₇ (trace amounts), 2,3 dehydro-GA₁₀ (artifact), GA₃₁, and GA₆₂ are shown in Table IX, based on identification by the full-scan GC-MS and KRI data shown in Table X. Four of the metabolites are formed by hydroxylation at C-1 (GA₆₂, Fig. 2), at C-3 (GA₇, Fig. 1), at C-12 (GA₃₁, Fig. 2), and at C-13 (GA₅, Fig. 1). 2,3-Dehydro-GA₁₀ (Fig. 2) is the product of hydration of the 16,17-double bond and this step may be non-enzymatic. The metabolism of 2,3-dehydro-[17-²H₂]GA₉ to [²H₂]GA₇ has been previously reported from cell-free systems from seeds of wild cucumber and apple (Albone et al., 1990). The metabolism of 2,3-dehydro-GA₉ to GA₆₂, to GA₃₁, and to GA₅ are the first examples of these conversions in plants.

Isotopic Dilution of 2,3-Dehydro-GA₉

In view of the observed conversion of 2,3-dehydro-GA₉ to GA₇, we investigated the possible natural occurrence of 2,3-dehydro-GA₉ in maize. Thus, we determined the level of isotopic dilution of 2,3-dehydro-[17-¹⁴C]GA₉ added to a homogenate of normal maize seedlings. No dilution of label was observed based on a full-scan GC-MS analysis of the recovered 2,3-dehydro-[17-¹⁴C]GA₉ (data not shown), thus indicating that 2,3-dehydro-GA₉ is not endogenous to maize.

General

The structures of the substrates and metabolites presented in this report are shown in Figures 1 and 2, with the exception of the HO-GA₂₅ metabolite for which the hydroxylation site was not determined. In maize, the 13-hydroxylation of GA₁₅ to GA₄₄, GA₂₄ to GA₁₉, GA₉ to GA₂₀, and GA₄ to GA₁ results in the formation of a grid connecting members of the (presumptive) non-early 3,13-hydroxylation pathway to members of the early 13-hydroxylation pathway (Fig. 1). The two steps, GA₁₅  GA₄₄ and GA₂₄  GA₁₉, represent the first demonstration of in vivo crossovers between C₂₀-GAs. A similar grid connecting the two branch pathways has been demonstrated from in vitro studies from a number of plant species (Kamiya and Graebe, 1983; Takahashi et al., 1986; Grosslindemann et al., 1992). The 13-hydroxylation of GA₁₅, GA₂₄, GA₉, and GA₄ in maize may reside in a single 13-hydroxylase with low substrate specificity or with the presence of separate substrate-specific enzymes. The failure to detect the sequence GA₁₂  GA₁₅  GA₂₄  GA₉  GA₄  GA₇ could be because the K_m values for these substrates are much lower for the 13-hydroxylase(s) than for the 20-oxidase(s).

The two labeled metabolites GA₁₅-15,16-ene and GA₇-diene-diacid were probably formed by the non-enzymatic rearrangement of a double bond. Additionally, 2,3-dehydro-GA₁₀ was probably formed as a result of a non-enzymatic hydration of the 16,17-double bond in the substrate 2,3-dehydro-GA₉.

Based on the previous demonstration that GA₅ is an intermediate between GA₂₀ and GA₃ in maize shoots (Fujioka et al., 1990; Spray et al., 1996), we examined the possibility that 2,3-dehydro-GA₉ is an intermediate between GA₉ and GA₇. Our results show that 2,3-dehydro-GA₉ is predominantly 13-hydroxylated to GA₅, 12-hydroxylated to GA₃₁, and 1-hydroxylated to GA₆₂, and converted into GA₇ in trace amounts. However, isotope dilution studies gave no evidence for the natural occurrence of 2,3-dehydro-GA₉ in maize shoots (data not shown). The metabolic origin of GA₁₅, GA₂₄, GA₉, GA₄, and GA₇ in maize remains unresolved.