INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Long-day (LD) rosette plants, such as spinach (Spinacia oleracea), grow vegetatively and do not produce a stem when grown under short-day (SD) conditions. Upon transfer to LD conditions, stem elongation and flowering take place. In spinach, LD-induced stem elongation is dependent on gibberellin (GA)-regulated processes. There is considerable evidence that GA biosynthesis increases in rosette plants in LD conditions (Talon and Zeevaart, 1990; Talon et al., 1991). The major endogenous GAs of spinach belong to the early-13-hydroxylation pathway: GA₅₃, GA₄₄, GA₁₉, GA₂₀, GA₁, GA₈, and GA₂₉ (Fig. 1; Talon et al., 1991), of which only GA₁ is active per se (Zeevaart et al., 1993). Several steps of the GA biosynthetic pathway are regulated by photoperiod. For example, the biosynthesis of ent-kaurene, an early step in the GA biosynthetic pathway, is enhanced by LD (Zeevaart and Gage, 1993). The later steps of sequential oxidation and elimination of C-20 of C₂₀-GAs (GA₅₃GA₄₄GA₁₉GA₂₀), catalyzed by GA 20-oxidase, are also increased when spinach plants are transferred from SD to LD conditions (Wu et al., 1996).

View larger version (13K): [in this window] [in a new window]   Figure 1.   The early-13-hydroxylation pathway from GA12 to GA8 of GA biosynthesis and deactivation in spinach. GA53 is converted to GA20 by GA 20-oxidase via GA44 and GA19. GA20 can be converted to GA1 by GA 3-oxidase. GA12, GA53, GA20, and GA1 can be deactivated by 2-hydroxylation to GA110, GA97, GA29, and GA8, respectively.

3-Hydroxylation is the final step to active GAs, converting GA₉ and GA₂₀ to GA₄ and GA₁, respectively. Genes encoding 3-oxidases have been isolated from several species, and their expression and regulation have been studied (Chiang et al., 1995; Lester et al., 1997; Cowling et al., 1998; Yamaguchi et al., 1998; Itoh et al., 1999, 2001; Rebers et al., 1999). GA 2-oxidases introduce a 2-hydroxyl group to biologically active GAs and deactivate C₁₉-GAs, such as GA₁, GA₄, GA₉, and GA₂₀ (Lester et al., 1999; Thomas et al., 1999; Elliott et al., 2001). Both GA 3-oxidase and GA 2-oxidase can, therefore, directly regulate the levels of the active GAs, GA₁ and GA₄, and their immediate precursors, GA₂₀ and GA₉, respectively (for reviews, see Hedden and Phillips, 2000; Yamaguchi and Kamiya, 2000).

As shown in earlier work with spinach, transfer of plants from SD to LD causes an increase in all GAs of the early-13-hydroxylation pathway, with GA₂₀, GA₁, and GA₈ showing the largest increases (Wu et al., 1996). The increase in GA₂₀ can be explained as being attributable to increased expression of the gene encoding GA 20-oxidase, SoGA20ox1. However, to what extent the levels of GA₁ and GA₈ are determined by photoperiodically induced changes in expression of GA 3-oxidase and GA 2-oxidase is not known. In this study, we describe the molecular cloning of a GA 3-oxidase and two GA 2-oxidases from spinach and the regulation of their expressions by photoperiod in comparison with that of GA 20-oxidase. Our results indicate that expression of GA 20-oxidase1 and GA 2-oxidase1 is affected more by the photoperiod than is expression of GA 3-oxidase1 and GA 2-oxidase2.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Cloning of GA 3-Oxidase and GA 2-Oxidase cDNAs

Reverse transcription (RT)-PCR reactions were performed with total RNA as a template isolated from spinach grown under SD conditions. All combinations of degenerate primers were used in the PCR reactions with first-strand cDNA as a template. The combination of degenerate primers JZ418 and JZ432 (Table I) yielded a 521-bp fragment of GA 3-oxidase (data not shown). The sequence of this RT-PCR product was homologous to the GA 3-oxidase genes of Arabidopsis (Chiang et al., 1995) and pea (Pisum sativum; Lester et al., 1997). 5'- and 3'-RACE were performed with sequence-specific primers (JZ445 and JZ444) using the SMART RACE cDNA Amplification Kit (BD Biosciences Clontech, Palo Alto). The coding region of the SoGA3ox1 gene was obtained with primers JZ509 and JZ510 (Table I). The full-length cDNA clone of SoGA3ox1 (1,612 bp; accession no. AF506280) has an open reading frame (ORF) of 1,149 bp, encoding a putative protein of 382 amino acids with a 223-bp 5'-untranslated sequence and a 240-bp 3'-untranslated sequence. The predicted protein molecular mass is 42 kD, with an pI of 8.39 (Fig. 2). Northern-blot analysis showed two sizes of transcripts of the SoGA3ox1 gene in shoot tips around 1.6 and 2.0 kb (data not shown). The longer transcript is immature because it contains one intron of 361 bp. This exon/intron structure is conserved at the same position as in other GA3ox genes (Lester et al., 1997). When compared with other GA 3-oxidases, the deduced amino acid sequence of SoGA3ox1 belongs to the same class of enzymes and contains certain conserved amino acid sequences, such as His-227, Asp-229, and His-284, that may bind iron (Thomas et al., 1999). It shares 33% to 56% identity with the GA 3-oxidases listed in Figure 2.

View larger version (120K): [in this window] [in a new window]   Figure 2.   Alignment of the deduced amino acid sequences for the SoGA3ox1 gene from spinach with other GA 3-oxidases. Conserved regions are boxed in black. Asterisks indicate the putative Fe2+-binding motif at the active site of 2-oxoglutarate-dependent dioxygenases (Thomas et al., 1999). SoGA3ox1, GA 3-oxidase from spinach (GenBank accession no. AF506280); NtGA3ox, GA 3-oxidase from tobacco (Nicotiana tabacum; accession no. BAA89316); StGA3ox, GA 3-oxidase from potato (Solanum tuberosum; accession no. AAK91507); PsGA3ox, GA 3-oxidase from garden pea (GenBank accession no. AAC96015); AtGA3ox, GA 3-oxidase from Arabidopsis (accession no. T51691); CmGA3ox and CmGA23ox, GA 3-oxidases from pumpkin (Cucurbita pepo; accession nos. AAB64347 and CAB92914); and OsGA3ox, GA 3-oxidase from rice (Oryza sativa; accession no. BAB62072).

The primer combination JZ254 and JZ258 yielded a 236-bp fragment of GA 2-oxidase1, and JZ 254 and JZ 342 yielded a 119-bp fragment of GA 2-oxidase2 (data not shown). The sequence of each RT-PCR product was homologous to the GA 2-oxidase genes of runner bean and Arabidopsis (Thomas et al., 1999) and pea (Lester et al., 1999). 5'- and 3'-RACE were performed with sequence-specific primers (JZ282 and JZ298 for GA 2-oxidase1, and JZ375 and JZ574 for GA 2-oxidase2; Table I) using the SMART RACE cDNA amplification kit (BD Biosciences Clontech). The coding region of the SoGA2ox1 gene was obtained with primers JZ300 and JZ301 (Table I). The full-length SoGA2ox1 cDNA clone (1,229 bp, accession no. AF506281) has an ORF of 1,014 bp and encodes a protein of 337 amino acids with a 114-bp 5'-untranslated sequence and a 101-bp 3'-untranslated sequence. The molecular mass of the predicted protein is 38 kD, with an pI of 8.46. The SoGA2ox1 protein shares 40% to 61% identity with the other GA 2-oxidases listed in Figure 3. Northern-blot analysis showed that the size of the SoGA2ox1 transcript is 1.3 kb (data not shown).

View larger version (116K): [in this window] [in a new window]   Figure 3.   Alignment of the deduced amino acid sequences for the SoGA2ox1 and SoGA2ox2 genes from spinach with other GA 2-oxidases. Conserved regions are boxed in black. Asterisks indicate the putative Fe2+-binding motif at the active site of 2-oxoglutarate-dependent dioxygenases (Thomas et al., 1999). SoGA2ox1 and SoGA2ox2, GA 2-oxidases from spinach (accession nos. AF506281 and AF506282); PcGA2ox1, GA 2-oxidase from runner bean (accession no. AT132438); PsGA2ox1 and PsGA2ox2, GA 2-oxidases from garden pea (accession nos. AF100954 and AF100955); AtGA2ox1, AtGA2ox2, and AtGA2ox3, GA 2-oxidases from Arabidopsis (accession nos. AJ132435, AJ132436, and AJ132437); OsGA2ox1, GA 2-oxidase from rice (accession no. BAB40934).

The full-length sequence of the second GA 2-oxidase (1,130 bp), designated SoGA2ox2 (accession no. AF506282), contains an ORF of 990 bp encoding a protein of 329 amino acids (36 kD) with a 113-bp 5'-untranslated sequence, and a 27-bp 3'-untranslated sequence with an pI of 7.1. The coding region of the SoGA2ox2 gene was obtained by PCR with the sequence-specific primers JZ619 and JZ582 (Table I). SoGA2ox2 shares 37% to 55% identity with other GA 2-oxidases listed in Figure 3 and has a low identity (40%) with the SoGA2ox1 protein. We established by dot-blot analysis that there is no cross-reactivity between the SoGA2ox1 and SoGA2ox2 genes (data not shown). Figure 3 shows the conserved regions among various GA 2-oxidases. Alignment with GA 2-oxidases from other species indicates that the two putative GA 2-oxidases from spinach belong to the same class of proteins and contain highly conserved amino acid sequences that may bind iron at the active site.

Heterologous Expression in Escherichia coli

To determine that the SoGA3ox1 cDNA clone encodes a GA 3-oxidase and that the SoGA2ox1 and SoGA2ox2 clones encode GA 2-oxidases, each coding region was heterologously expressed as a fusion protein of the glutathione S-transferase fusion vector in E. coli, strain BL21p Lys21. Soluble protein extracts were used for assays of GA 3-oxidase and GA 2-oxidase activities with several radioactive GAs as substrates. The reaction products were separated by reverse phase HPLC with an on-line radioactivity detector. Retention times (Rts) of the products were compared with those of standard ¹⁴C-labeled GAs for tentative identification. The SoGA3ox1 protein converted radioactive GA₉ and GA₂₀ to products with Rts of 27.8 and 15.4 min, corresponding to those of GA₄ and GA₁, respectively. The product of [¹⁴C₄]GA₂₀ (specific activity 20 µCi µmol¹) incubated with SoGA3ox1 was identified as [¹⁴C₄]GA₁ (Table II), thus demonstrating conclusively that SoGA3ox1 encodes a GA 3-oxidase.

Both SoGA2ox recombinant proteins converted radioactive GA₉ and GA₂₀ to putative GA₅₁ (Rt = 25.8 min) and GA₂₉ (Rt = 10.1 min), respectively. The products of [¹⁴C₄]GA₂₀ and [²H₂]GA₂₀ metabolism were identified as [¹⁴C₄]GA₂₉ and [²H₂]GA₂₉, respectively; [²H₂]GA₁ was converted to [²H₂]GA₈ (Table II), so that both GA₂₀ and GA₁ are substrates for these GA 2-oxidases. In addition to GA₂₉, a second product (Rt = 15.0 min) was formed in incubations of GA₂₀ with SoGA2ox1. This product was tentatively identified as GA₂₉-catabolite, because [¹⁴C₄]GA₂₉ was converted by SoGA2ox1 to a product with the same Rt.

SoGA2ox1, but not SoGA2ox2, also converted GA₅₃ (Rt = 29.3 min) to GA₉₇ (Rt = 12.4 min; Table II) and a second metabolite (Rt = 17.0 min), which was not identified. Thus, it appears from these results that the substrate specificity of the two GA 2-oxidases isolated from spinach is similar to that of other GA 2-oxidases (Thomas et al., 1999; Elliott et al., 2001) in that they all catalyze 2-hydroxylation of C₁₉-GAs, but that SoGA2ox1 is also capable of 2-hydroxylation of the C₂₀-GA, GA₅₃.

Gene Expression in SD and LD Conditions

SoGA2ox1 transcripts were barely detectable on northern blots, so SoGA2ox1 gene expression was measured by the quantitative RT-PCR method (Cowling et al., 1998). For the PCR reaction, 29 cycles were chosen as the point where PCR products of the SoGA2ox1 and SoActin genes were increasing exponentially (data not shown).

To investigate how the photoperiod regulates expression of the GA dioxygenases isolated, spinach plants grown in SD conditions were transferred to LD conditions for 8 d. Figure 4 shows the expression patterns of four GA dioxygenases in several organs. In SD conditions, the levels of the SoGA20ox1 transcripts were low in blades, petioles, and young leaves, but the SoGA20ox1 transcripts accumulated in all organs tested, especially in petioles and tips, when spinach plants were transferred from SD to LD conditions.

View larger version (89K): [in this window] [in a new window]   Figure 4.   Expression patterns of the SoGA20ox1, SoGA3ox1, SoGA2ox1, and SoGA2ox2 genes in various organs of spinach plants grown in SD and after 8 LD. Northern blots were prepared by separating 30 µg of total RNA of each organ through 1.2% (w/v) agarose gels containing formaldehyde, followed by transfer to Hybond-N+ membranes. The blots were separately hybridized to 32P-labeled cDNAs of SoGA20ox1, SoGA3ox1, SoGA2ox2, or SoActin. After RT-PCR of SoGA2ox1 and SoActin, 10 µL of the PCR reaction products were separated through 2.0% (w/v) agarose gel by electrophoresis and hybridized to a 32P-labeled cDNA of SoGA2ox1 or SoActin. The numbers under the blots indicate the relative amount of each transcript after standardization using SoActin as a loading control. The value for each gene transcript in blades of spinach grown in SD conditions was arbitrarily set at 1.0.

The levels of the SoGA3ox1 transcripts of plants in SD conditions were lower in petioles and tips than in leaf blades and young leaves. There was a slight increase in SoGA3ox1 transcript levels in leaf blades and petioles when spinach plants were transferred from SD to LD conditions, but the change in photoperiod had very little effect on transcript level in young leaves and tips (Fig. 4), the organs where growth takes place.

The levels of the SoGA2ox1 transcripts in SD-grown plants were lower in leaf blades than in petioles, young leaves, and tips. But in LD conditions, the levels of the SoGA2ox1 transcripts were lower in young leaves than in other organs. Thus, the levels of the SoGA2ox1 transcripts increased in blades and petioles when spinach plants were transferred from SD to LD conditions. By contrast, the level of SoGA2ox1 transcripts decreased in young leaves and tips. This decrease in the level of SoGA2ox1 transcripts in young leaves and tips in LD conditions is opposite to the increase in the level of SoGA20ox1 transcripts.

The SoGA2ox2 gene was highly expressed in leaf blades and petioles. However, there were no significant differences between transcript levels in SD and LD, so that expression of SoGA2ox2 was not further investigated in subsequent experiments.

Time Course of LD Treatment

Spinach plants were harvested at different time intervals after transfer from SD to LD conditions to determine the time course of changes in gene expression. The levels of the SoGA20ox1, SoGA3ox1, and SoGA2ox1 transcripts were all up-regulated in petioles in a time-dependent manner (Fig. 5A). This enhancement was much stronger in the case of the SoGA20ox1 gene (26×) than of the SoGA3ox1 gene (1.6×) and of the SoGA2ox1 gene (4×). The levels of the SoGA20ox1 transcripts were also increased time-dependently in young leaves and tips (Fig. 5, B and C). LD conditions caused more rapid accumulation of the SoGA20ox1 transcripts in tips than in petioles and young leaves. However, the SoGA3ox1 mRNA abundance in young leaves and tips was similar or only slightly increased by the LD treatment (Fig. 5, B and C). By contrast, the levels of the SoGA2ox1 transcripts were down-regulated in young leaves and tips (Fig. 5, B and C). The levels of the SoGA2ox1 transcripts were more rapidly decreased in young leaves than in tips. This rapid decrease may compensate for the relatively low increase in the levels of the SoGA20ox1 and SoGA3ox1 transcripts in young leaves to maintain the active GAs. The time course experiment demonstrates the different regulation by the photoperiod of the three GA dioxygenases: (a) expression of SoGA20ox1 changed quickly and strongly in petioles and tips, (b) the SoGA3ox1 transcript level was only slightly changed in petioles, and (c) the SoGA2ox1 transcript level increased in petioles but decreased in young leaves and tips.

View larger version (54K): [in this window] [in a new window]   Figure 5.   Time course of SoGA20ox1, SoGA3ox1, and SoGA2ox1 expression after spinach plants were transferred from SD to LD conditions. GA20ox1 transcripts increased with increasing duration of LD treatment in petioles (A), young leaves (B), and tips (C). GA3ox1 transcripts increased with increasing duration of LD treatment in petioles (A), but they did not change in young leaves (B), and tips (C) after exposure to increasing numbers of LD. GA2ox1 transcripts increased with increasing duration of LD treatment in petioles (A), but they decreased in young leaves (B), and tips (C). The numbers under the blots indicate the relative amount of transcript after standardization using SoActin as a loading control. The value of each transcript level in petioles in SD was arbitrarily set at 1.0.

Effect of Light and Darkness on Gene Expression

To investigate the effect of light and darkness on gene expression, spinach plants grown in SD and LD conditions were harvested at the end of the 8-h high-intensity light period (0 h), at the end of 16 h of darkness (16 D), or at the end of 16 h of low-intensity light (16 L). The levels of the SoGA20ox1, SoGA3ox1, and SoGA2ox1 transcripts were determined in petioles, young leaves, and tips (Fig. 6). Compared with the levels in each group at the end of the 8-h high-intensity light period (0 h), additional light (16 L) greatly increased the levels of the SoGA20ox1 transcripts in petioles of plants that were initially in SD conditions (13× compared with SD at 0 h) and also in petioles in LD conditions (4× compared with LD at 0 h; Fig. 6A). The levels of the SoGA20ox1 transcripts in tips were not changed by additional light (Fig. 6C). A long dark treatment (16 D) did not affect the levels of the SoGA20ox1 transcripts in petioles, young leaves, and tips in either SD or LD conditions. This means that the levels of SoGA20ox1 transcript are regulated much more by light than by darkness. Exposure to weak light or dark treatment had little effect on the levels of the SoGA3ox1 and SoGA2ox1 transcripts (Fig. 6).

View larger version (61K): [in this window] [in a new window]   Figure 6.   Effects of light and darkness on the transcript level of SoGA20ox1, SoGA3ox1, and SoGA2ox1 in petioles (A), young leaves (B), and tips (C). Spinach plants were harvested at the end of the 8-h high-intensity light period (0 h) and after transfer from SD or LD conditions to 16 h of darkness (16 D) or to 16 h of weak incandescent light (16 L). The numbers under the blots indicate the relative amount of each transcript after standardization using SoActin as a loading control. The value of each transcript level in petioles at 0 h in SD was arbitrarily set at 1.0.

Expression in Other Organs

To determine other organ-specific expression patterns of the GA dioxygenases, male and female flowers, stems, and shoot tips were harvested from spinach plants that had been exposed to LD conditions for more than 3 weeks (Fig. 7). The shoot tips included the upper 1 cm; stems were the next 1 cm of the elongated stem. The SoGA20ox1 mRNA was more abundant in stems than in shoot tips, and little transcript was detected in flowers. By contrast, SoGA3ox1 and SoGA2ox1 were highly expressed in male and female flowers, stems, and shoot tips.

View larger version (86K): [in this window] [in a new window]   Figure 7.   Expression of SoGA20ox1, SoGA3ox1, and SoGA2ox1 in male flowers (), female flowers (), stems (St), and shoot tips (ShT). Spinach plants were harvested after more than 3 weeks in LD conditions. In this case, the shoot tips included the upper 1 cm of the shoots; stems were the next 1 cm of the elongated stems. The numbers under the blots indicate the relative amount of each transcript after standardization using SoActin as a loading control. The value of the transcript level of each gene in male flowers was arbitrarily set at 1.0.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

We isolated a full-length cDNA clone encoding GA 3-oxidase and two full-length cDNA clones encoding GA 2-oxidase from spinach. Heterologous expression in E. coli showed that the product of the SoGA3ox1 gene is able to catalyze 3-hydroxylation of GA₂₀ to GA₁, whereas the products of the SoGA2ox genes catalyze 2-hydroxylation of GA₂₀ and GA₁, both C₁₉-GAs with a -lactone in the A ring of the GA structure (Fig. 1). In addition, recombinant SoGA2ox1 converts GA₅₃ to GA₉₇, both C₂₀-GAs (Fig. 1). The recombinant proteins of GA 2-oxidases from other species specifically 2-hydroxylate C₁₉-GAs (Thomas et al., 1999; Elliott et al., 2001). So far, few exceptions have been reported to this rule: GA₁₅, a C₂₀-GA, was 2-hydroxylated by PcGA2ox1 and AtGA2ox2 (Thomas et al., 1999); and GA₄₄, also a C₂₀-GA (Fig. 1), was 2-hydroxylated to GA₉₈ by OsGA2ox1 (Sakamoto et al., 2001). Both of these GAs have a -lactone in the A ring. When this lactone ring was opened, GA₁₅ was no longer metabolized by recombinant PcGA2ox1, but it was (although less efficiently) by AtGA2ox2 (Thomas et al., 1999). Spinach produces large amounts of 2-hydroxy-GA₅₃ (=GA₉₇), and smaller amounts of GA₉₈ (= 2-hydroxy-GA₄₄), GA₉₉ (= 2-hydroxy-GA₁₉; Mander et al., 1996), and GA₁₁₀ (= 2-hydroxy-GA₁₂; Owen et al., 1998). Substrate specificity of the 2-oxidases for the possible precursors of these 2-hydroxy-C₂₀-GAs remains to be determined.

Expression of SoGA3ox1 was very little affected by the photoperiod (Figs. 4 and 5), so that it is possible that the conversion of GA₂₀ to GA₁ in LD is limited by the amount of GA₂₀ substrate available and not by the enzyme level. Although expression of SoGA2ox1 was down-regulated by LD, this may be only a minor factor in the maintenance of a high GA₁ level, because SoGA2ox1 transcripts are much less abundant than those of SoGA2ox2, and expression of the latter is not under photoperiodic control (Fig. 4). In a different system, de-etiolation of pea seedlings, GA 3-oxidase was down-regulated by light, whereas GA 2-oxidase was up-regulated. This resulted in decreased GA₁, increased GA₈ levels, and a reduction in growth rate (Reid et al., 2002).

The increase in growth rate of petioles of spinach in LD occurs during the first 16 h of weak supplementary light (Zeevaart, 1971). This may be mainly because of the increase in SoGA20ox1 transcripts during the additional 16 h of light rather than the small increase in SoGA3ox1 and decrease in SoGA2ox1 transcripts (Fig. 6A). The results in Figure 6 show that moving spinach plants from LD to 16 h darkness did not reduce the transcript level of any of the GA dioxygenases studied. On the other hand, Gilmour et al. (1986) reported that enzyme activities for the conversions GA₅₃ to GA₄₄ and GA₁₉ to GA₂₀, both representing GA 20-oxidase activity, declined quickly in plants in darkness. It is possible, therefore, that regulation of GA 20-oxidase by photoperiod is not only at the transcriptional level, but that there may be posttranscriptional control as well. Thus, transcript levels may not necessarily be indicative of enzyme activity. Further work on photoperiodic control of GA biosynthesis in spinach should therefore attempt to compare the protein level and enzymatic activity of GA 20-oxidase.

The low expression of the SoGA20ox1 gene in spinach flowers (Fig. 7) may mean that it has no role in flower development, unlike the LeGA20ox1 and LeGA20ox2 genes, which are highly expressed in tomato (Lycopersicon esculentum) during flower bud and early fruit development (Rebers et al., 1999). Thus, another GA 20-oxidase gene may be involved in flower development in spinach.

Up-regulation of GA biosynthesis by LD starts at the beginning of the pathway with increased ent-kaurene biosynthesis (Zeevaart and Gage, 1993), and in the later stages, it is mainly attributable to the much increased expression of SoGA20ox1. It is surprising that expression of SoGA3ox1, encoding the product that catalyzes the pivotal step of 3-hydroxylation, shows little fluctuation in response to the photoperiod (Fig. 4). Although expression of SoGA2ox1 is down-regulated in tips and young leaves of plants in LD, the overall effect of LD on GA metabolism is that more GA₂₉ (Metzger and Zeevaart, 1980, 1982) and GA₈ (Wu et al., 1996) are produced in plants in LD than in SD. Thus, it appears that in SD, the steps catalyzed by GA 20-oxidase are enzyme limited, whereas GA 3-oxidase and GA 2-oxidase are limited by the amount of substrate available. In LD, a new balance is established between GA₁ synthesis and deactivation, increasing the GA₁ level to such an extent that the threshold value required for stem elongation is exceeded (Zeevaart et al., 1993).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Material and Growing Conditions

Spinach (Spinacia oleracea L. cv Savoy Hybrid 612) plants were grown in a SD growth chamber for approximately 6 or 7 weeks as described previously (Zeevaart and Gage, 1993). SD conditions consisted of 8 h of light from fluorescent tubes and incandescent bulbs of approximately 300 µmol m² s¹ at 23°C, followed by 16 h darkness at 20°C. For LD conditions, the 8-h main light period was followed by 16 h of weak light from incandescent bulbs of 10 µmol m² s¹ at 20°C.

RT-PCR

RT-PCR was used for the cloning of GA 3-oxidase and GA 2-oxidases. First-strand cDNA from total RNA was synthesized by SuperScript II RNase H-minus Reverse Transcriptase (Invitrogen, Carlsbad, CA), using oligo(dT) primers at 42°C for 1 h. The cDNA product was then used in PCR reactions containing Taq polymerase and several pairs of degenerate primers (Table I). For GA 3-oxidase, degenerate primers were synthesized based on conserved regions of GA 3-oxidases from Arabidopsis (Chiang et al., 1995; Yamaguchi et al., 1998) and pea (Lester et al., 1997). For GA 2-oxidase, the degenerate primers were synthesized based on conserved regions of GA 2-oxidases from runner bean and Arabidopsis (Thomas et al., 1999). These primers were used in all combinations in PCR reactions with first-strand cDNA as a template. The PCR amplifications were performed with a DNA Thermal Cycler (RTC-200, MJ Research, Waltham, MA) in a 50-µL reaction containing 1× PCR buffer (Invitrogen), 1.5 mM MgCl₂, 200 µM dNTP, 2.5 pmol of each primer, 1 unit of Taq DNA polymerase (Invitrogen). The reaction mixtures were heated to 94°C for 3 min and then subjected to 30 cycles of 94°C for 1 min, 48°C for 1 min, and 72°C for 45 s. A final extension was performed at 72°C for 10 min. A 5-µL volume of the PCR reaction products was analyzed by 1.2% (w/v) agarose gel electrophoresis, purified with a Wizard PCR purification system (Promega, Madison, WI), and then cloned into the pGEM Teasy vector (Promega). Analysis of the DNA sequences was carried out using the DNASTAR program (DNASTAR, Inc., Madison, WI). Multiple sequence alignments were performed using the Clustal W 1.8 Multiple Sequence Alignment program and printed using BOXSHADE 3.21 (http://www.ch.embnet.org/software/Box_form.html).

RACE and Cloning

5'- and 3'-RACE of spinach GA 3-oxidase cDNA and GA 2-oxidase cDNA were performed using the SMART RACE cDNA Amplification Kit from BD Biosciences Clontech. Poly(A⁺) RNA was isolated from spinach using the PolyATtract mRNA Isolation System IV (Promega) according to the manufacturer's instructions. After RT with primers supplied by BD Biosciences Clontech, the first-strand cDNA was used directly in 5'- and 3'-RACE PCR reactions. Primary PCR amplification reactions were achieved using a high-fidelity enzyme (Pfu Turbo polymerase; Stratagene, La Jolla, CA) and gene-specific primers JZ282 and JZ298 for GA 2-oxidase1, JZ375 and JZ574 for GA 2-oxidase2, and JZ445 and JZ444 for GA 3-oxidase1 to generate the 5'- and 3'-cDNA fragments, respectively. The PCR reaction consisted of the first denaturation for 3 min at 94°C, a series of 30 cycles (1 min at 94°C, 1 min at 54°C or 55°C, and 1 min at 72°C) with a final extension for 5 min at 72°C using a thermal cycler (RTC-200, MJ Research). A 5-µL aliquot of the RT-PCR and RACE reaction solution was analyzed by 1.2% (w/v) agarose gel electrophoresis. PCR products were purified and cloned into the pCR-Script Cam SK(+) cloning vector from Stratagene. These constructs were sequenced.

Northern-Blot Analysis

Leaf blades, petioles, young leaves, tips (Zeevaart and Gage, 1993), and other organs were harvested, frozen immediately in liquid N₂, and stored at 80°C. Northern blots were prepared by electrophoresis of 30 µg of total RNA in the presence of formaldehyde (Sambrook et al., 1989), and RNA was transferred to nitrocellulose. Full-length cDNAs were labeled with [³²P]dCTP by the Random Primers DNA Labeling System (Invitrogen) and the radioactive probes were used to probe northern blots. Hybridization was carried out at 42°C using a 50% (v/v) formamide system (Sambrook et al., 1989). Membranes were washed twice for 10 min in 2× SSC at room temperature and then twice for 10 min in 0.2× SSC with 0.1% (w/v) SDS at 65°C (high stringency). At low stringency, the membrane was washed once for 10 min in 2× SSC at room temperature and then washed again for 10 min in 2× SSC at 65°C.

The relative amounts of mRNA were determined with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). Each blot was first probed with one of the various cDNAs, and then stripped and reprobed with SoActin. The ratio of the GA dioxygenase mRNA to Actin mRNA was used to compare the mRNA levels in various organs and treatments. Each experiment was repeated at least three times with similar results.

Quantitative RT-PCR

The first-strand cDNA synthesis reactions were carried out using 5 µg of each total RNA sample, random hexamer primer, and avian RNase H-minus reverse Transcriptase. A pair of primers (JZ385 and JZ386) was used to amplify a 401-bp fragment of GA 2-oxidase1. JZ389 and JZ390 primers (Table I) were used to amplify a 324-bp fragment of the SoActin gene as a control. Both of these fragments can be distinguished from any contaminant of genomic DNA by size. The RT-PCR products were cloned and sequenced to verify the sequences of the fragments of the SoGA2ox1 and SoActin genes. The cDNA solutions (2.5 µL) were used as a template in a standard 50-µL PCR reaction. Expression of SoGA2ox1 and SoActin was compared at 29 cycles where both PCR reactions were progressing exponentially. Ten microliters of PCR reaction products was separated by electrophoresis, blotted, and hybridized.

Expression of Recombinant GA 3-Oxidase and GA 2-Oxidase Proteins

Coding regions of 2-oxidases were produced by PCR with the following primer set designed from the RACE product: JZ300 and JZ301 for mature GA 2-oxidase1 (PspAI and XhoI sites inserted to facilitate cloning) and JZ582 and JZ619 (BamHI and XhoI sites inserted to facilitate cloning) for mature GA 2-oxidase2. For GA 3-oxidase, JZ509 and JZ510 (PspAI and SalI sites inserted to facilitate cloning) were used to get the coding region. The resulting PCR fragments were cloned into a pCR-Script Cam SK(+) cloning vector and then digested with PspA1/XhoI, BamHI/XhoI, or PspA1/SalI, respectively, and subcloned into the corresponding restriction sites of the pGEX-5X-2 vector (Amersham Biosciences, Piscataway, NJ). These full-length cDNA clones (pGEXGA3ox1, pGEXGA2ox1, or pGEXGA2ox2) were transformed into Escherichia coli, strain BL21pLysS. Fifty milliliters of freshly cultured cells was added to 1 L of Luria-Bertani medium with 100 mg L¹ ampicillin and incubated at 37°C with vigorous shaking. When the optical density at 600 nm reached 0.6, isopropyl--D-thiogalactopyranoside was added to give a final concentration of 3 mM, and the culture was incubated for another 2 h. The cells were harvested and suspended in lysis buffer (100 mM Tris-HCl, pH 7.5, and 10 mg mL¹ lysozyme), and incubated at room temperature for 10 min. Crude cell extracts were obtained by using a Sonifier Cell Disruptor 200 (Branson Ultrasonics, Danbury, CT) carried out for 10 cycles (10 × 10 s) on ice. The lysates were submerged in liquid N₂ for 2 min and then thawed in an ice bath for 15 min (Johnson and Hecht, 1994). The lysates were centrifuged at 13,000 rpm for 30 min, and the supernatant was stored at 80°C until used for enzyme assays.

Enzyme Assays and Product Identification

The assays for GA 3-oxidase activity and GA 2-oxidase activities were performed with approximately 30,000 dpm of ¹⁴C-labeled GAs. [²H₂]GAs were used only for product identification. The reaction mixture (100 µL) contained 100 mM Tris buffer (pH 7.5), 0.5 mM FeSO₄, 5 mM 2-oxoglutarate, 5 mM ascorbate, and 80 µL of bacterial extract. Cofactors were added again after 1 h. The mixture was incubated at 30°C for up to 6 h with gentle shaking. For identification of products, the assays were scaled up to 500 µL. The products were isolated by reverse phase HPLC with an on-line FLO-ONE detector (Radioanalytic, Tampa, FL; Zeevaart and Gage, 1993). For identification of the products by gas chromatography (GC)-mass spectrometry (Zeevaart et al., 1993), the GC was equipped with a DB-5MS capillary column (30-m × 0.32-mm × 0.25-µm film, J&W Scientific, Folsom, CA), which was operated in splitless mode. The oven temperature was kept at 100°C for 1 min after sample injection and then programmed from 100°C to 230°C at 40°C min¹, from 230 to 280°C at 8°C min¹, and finally to 300°C at 20°C min¹. Specific activities of substrates and products were determined by GC-selected ion monitoring of the molecular ion clusters (Zeevaart and Gage, 1993).

Distribution of Materials

Upon written request, all novel materials described in this publication will be made available in limited quantities and in a timely manner for noncommercial research purposes, subject to the requisite permission from any third-party owners of all or parts of the material. Obtaining any permissions will be the responsibility of the requestor.