Introduction

When challenged with pathogen invasion, plants activate various defense responses, including activation of signal transduction pathways leading to expression of defense-related genes [such as pathogenesis-related (PR) genes], activation of the hypersensitive response (HR), and synthesis of phytoalexins with antimicrobial activity (Dangl et al. 1996). The major signals involved in plant defense responses are salicylic acid (SA), jasmonic acid (JA), and ethylene (ET; Dong 1998; Reymond and Farmer 1998; Guo and Ecker 2004), which together fine-tune the expression of biotic stress-responsive genes in plants (Hammond-Kosack and Parker 2003). However, the interactions between these signaling networks and the signaling pathways mediated by other phytohormones, such as gibberellin (GA) and auxin, have generally remain unresolved.

GA is a tetracyclic diterpenoid phytohormone important in several aspects of plant development, including seed germination, leaf expansion, stem elongation, flowering, and seed development (Sun and Gubler 2004). During the last decade, much progress has been made to understand the mechanism of GA signaling. It is well known that GAs promote plant growth by inducing the degradation of the DELLA family of nuclear transcription factors. The mechanism of GA-induced disappearance of the growth-restraining DELLA proteins (Jiang et al. 2007) is highly conserved between dicots and monocots (Fleet and Sun 2005). Recently, a role for DELLA proteins has been proposed in the responses of plants to pathogen attack: loss-of-function mutations in DELLA proteins improve the resistance of plants to some pathogens through induction of a salicylic acid-dependent defense pathway (Robert-Seilaniantz et al. 2007; Navarro et al. 2008). The authors indicate that GAs are able to regulate SA biosynthesis during plant responses to pathogens. However, neither the mechanism of GA action on defense responses nor how GA modulates changes in metabolism in response to pathogen attack is known. Recently, Yang et al. (2008) reported that a GA-deactivating enzyme called EUI (Elongated Uppermost Internode) is involved in basal disease resistance against bacterial and fungal pathogens in rice. The Eui gene encodes a P450 that deactivates biologically active GAs through GA 16α,17-epoxidation (Zhu et al. 2006). Thus far, it is unclear whether GA 16α,17-epoxidation is a common GA deactivation reaction in other plant species.

The level of bioactive GA in plants is regulated by the relative rates of its synthesis and deactivation. The important steps in the biosynthetic pathway of bioactive GA are the formation of C19-GA through the oxidation of C20-GA, and the 3β-hydroxylation of C19-GA. The 2-oxoglutarate-dependent dioxygenases, GA 20-oxidase and GA 3-oxidase, catalyze these respective steps in GA biosynthesis (Hedden and Phillips 2000; Olszewski et al. 2002). Another step regulating the endogenous level of GA is the deactivation of the bioactive GA forms GA1 and GA4 and their respective precursors, GA9 and GA20, by 2β-hydroxylation. This step is catalyzed by another 2-oxoglutarate-dependent dioxygenase, GA 2-oxidase (GA2ox), in the cytoplasm (Thomas et al. 1999; Hedden and Phillips 2000; Olszewski et al. 2002). In Arabidopsis, garden pea (Pisum sativum), rice (Oryza sativa), and spinach (Spinacia oleracea), GA2oxs are encoded by small gene families (Schomburg et al. 2003; Lester et al. 1999; Sakamoto et al. 2001; Lee and Zeevaart 2002). Examination of the GA2ox gene from various plants has focused on its role in development, particularly in plant growth (Sakamoto et al. 2001; Lee and Zeevaart 2002; Schomburg et al. 2003; Busov et al. 2003; Ogawa et al. 2003; Frisse et al. 2003; Wang et al. 2004; Lee and Zeevaart 2005). Loss-of-function mutations of PsGA2ox1, AtGA2ox7, and AtGA2ox8 result in a hyperelongated phenotype (Martin et al. 1999; Schomburg et al. 2003). Similarly, in poplar (Populus tremula × Populus alba), overexpression of PtaGA2ox1 causes a stumpy phenotype characterized by extremely short internodes and dark-green leaves with a stiff, leathery texture (Busov et al. 2003). These phenotypes were also observed with overexpression of AtGA2ox6, which is directly regulated by AGL-15, a member of the MADS box transcription factor family (Wang et al. 2004).

Using cDNA microarray analysis, we isolated a putative GA 2-oxidase gene, CaGA2ox1, from pepper plants (Capsicum annuum L. cv. Bukang) challenged with the non-host bacterium Xanthomonas axonopodis pv. glycines 8ra (Xag8ra, soybean pustule pathogen; Hwang et al. 1992). Based on expression analysis of CaGA2ox1, we suggest that CaGA2ox1 in pepper specifically responses to biotic stress. To our knowledge, this is the first report that a GA 2-oxidase, which catalyzes the 2-oxidation of C19 GAs, is involved in plant defense responses.

Materials and methods

Plant materials and pathogen inoculation

Chili pepper (Capsicum annuum L. cv. Bukang) and bell pepper [C. annum L. cv. Early Calwonder (ECW) 30R; bs1/bs1, bs2/bs2, Bs3/Bs3] plants were used in the current study. Peppers were grown in a plant growth room at 24 ± 1 °C using a photoperiod of 16 h light and 8 h dark. Healthy and well-expanded leaves from 6-week-old plants were used for pathogen inoculation and various chemical treatments. The bacterial pathogens used for inoculation were Xanthomonas axonopodis pv. glycines 8ra (Xag8ra), a soy bean pustule pathogen (Hwang et al. 1992), and the pepper bacterial spot pathogens Xanthomonas axonopodis pv. vesicatoria (Xav) race 1 (avrBs3) and race 3 (Kousik and Ritchie 1996). Bacterial infiltration was performed by syringe infiltration of bacterial suspensions (approximately 4 x 108 cfu/ml). Inoculated leaves were harvested at the indicated time points and used as the source of RNA for further analysis.

Chemical treatments

For the various chemical treatments, detached chili pepper leaves were soaked in solutions with dissolved final concentrations of 5 mM salicylic acid, 5 mM ethephon (ET), or 100 μM methyl jasmonic acid (MeJA) in sterile water. All treated leaves were harvested at the indicated time points and frozen in liquid nitrogen until further analyses were conducted.

RNA gel blot analysis

Total RNA was extracted from leaf tissues of pepper plants according to Choi et al. (1996). RNA (20 μg) isolated from leaf tissues treated with pathogens or chemicals was separated by electrophoresis on formaldehyde-containing 1 % agarose gels and transferred onto nylon membranes (Amersham, USA). Hybridizations were performed using the full-length cDNA probe of CaGA2ox1 labeled with 32[P]-dCTP according to the method of Church and Gilbert (1984). Membranes were washed twice with 2× SSC/0.1 % SDS at room temperature for 5 min, once with 1× SSC/0.1 % SDS at room temperature for 15 min, and once with 0.1× SSC/0.1 % SDS at room temperature for 5 min. The membranes were visualized using a BAS-1800II phosphoimager (Fuji Photo Film, Japan). The pepper cDNA control probes in these experiments, CaPR1 and CaPIN-II, were isolated previously (Yi et al. 2004 and Shin et al. 2001).

Plant treatments with GA4+7 and paclobutrazol (PBZ)

Either 50 ng of GA4+7 or PBZ were first dissolved in 0.5 ml of ethanol and then diluted with distilled water to a 50 ml volume. These chemicals were applied to the pot (379-ml plastic pots, Horticulture Nursery Media Low soil mixture; Pu-nong, Korea) of 2-week-old chili pepper plants for a week. The plants were incubated in the growth chamber at 24 °C in conditions as above. Changes in growth were observed and photographed 7 days after treatment. Experiments were performed twice using independently grown pepper seedlings.

RT-PCR and Q-RT-PCR analyses

Material for RNA analysis was ground in liquid nitrogen, and total RNA was isolated using TRI reagent according to the manufacturer’s protocol (MRC, Canada). Reverse transcription was performed using 2 μg of total RNA and SuperScript reverse transcriptase II (Invitrogen). The gene expression levels of various genes were determined by semi-quantitative PCR and quantitative RT-PCR using specific primer sets (Tables S1 and S2). Quantitative PCR was run on a Light cycler system (BIO-RAD) according to the manufacturer’s recommendations with the following conditions: 1 cycle of 15 min at 95 °C, and 40 cycles of 20 s at 95 °C, 20 s at 60 °C, and 20 s at 72 °C. The specificity of the amplifications was verified by melting curve analysis of the PCR products at the end of each experiment. Actin was used as an internal standard.

Subcellular localization

The full-length CaGA2ox1 cDNA without the termination codon was prepared by polymerase chain reaction (PCR) using the N-terminal specific primer (5′-GGGTCTAGAGTCTAATATGGTAGTGG-3′) containing an XbaI site and the C-terminal specific primer (5′-TTTGGATCCCATTGGGCATGGTTCG-3′) containing a BamHI site. The PCR-amplified product was cloned into the pGEM-T easy vector (Promega, Madison, USA), digested with XbaI and BamHI, and purified by agarose gel electrophoresis. The resulting fragment was fused to the coding region of soluble-modified green fluorescent protein (smGFP; David and Vierstra 1996). The CaGA2ox1:smGFP fusion construct was introduced into protoplasts, prepared from chili pepper leaves, by polyethylene glycol-mediated transformation (Park et al. 2001). The fluorescent micrographs of protoplasts were taken using a confocal laser scanning microscope (Carl Zeiss LSM510, Germany) 24 h after transformation. The filter sets used were BP505-530 (excitation 488 nm, emission 505–530 nm) and LP650 (excitation 488 nm, emission 650 nm; Carl Zeiss) to detect GFP and chlorophyll autofluorescence, respectively.

Expression of CaGA2ox1 in E. coli and assay for GA 2-oxidase activity

The coding region of CaGA2ox1 was amplified by PCR with the primers F1, 5′-GTCTAATATGGTAGTGGCAACTC-3′ and R2, 5′-GGTAGCTAAGACTGCTAAGTACTGC-3′. The resulting PCR product was cloned into pGEM T-easy vector (Promega), digested with EcoR1 and Xho1 and then cloned into the corresponding restriction sites of the pET28a vector (Clontech, USA). Either this full length cDNA clone (pET CaGA2ox1) or pET empty vector were transformed into Escherichia coli strain BL21pLysis. Cell lysate was prepared as described by Lee and Zeevaart (2002). The presence of CaGA2ox1 fusion protein in the cell lysate was confirmed by western blot analysis with an anti-His antibody (Invitrogen). Transformation with pGEX-SoGA2ox1 was used as a positive control for the enzyme activity assay (Lee and Zeevaart 2002). Enzyme assays with recombinant CaGA2ox1 were performed with [14C]-labeled GA4. The assays and methods for product identification were conducted as previously described (Lee and Zeevaart 2002; Schomburg et al. 2003).

Results and discussions

Isolation of CaGA2ox1

In order to identify a wide variety of genes induced in pepper plants during resistance to non-host pathogens, cDNA microarray analysis was performed with probes obtained from Xag8ra-infected hot pepper leaves. Xag8ra is not a pathogen of pepper, but does elicit an HR in pepper leaves as well as induce the expression of a number of PR genes (Lee et al. 2004). As a result of the microarray analysis, a gene encoding a key enzyme in gibberellin metabolism, a gibberellin 2-oxidase, designated CaGA2ox1, was isolated. The CaGA2ox1 gene encodes a putative protein that consists of 364 amino acids with a molecular mass of 40 kDa (Fig. 1a). The deduced amino acid sequence of CaGA2ox1 shares 53, 51, and 50 % identity, respectively, with Pisum sativum PsGA2ox2 (Lester et al. 1999), poplar PtaGA2ox1 (Busov et al. 2003), and Arabidopsis thaliana AtGA2ox6 (Wang et al. 2004). Alignment of the amino acid sequence of CaGA2ox1 with those of other GA 2-oxidases confirmed that it belonged to this class of enzymes, as it contains characteristic sequences that are conserved within the GA 2-oxidase class, including the 2-oxoglutarate-binding site (arginine and serine residues) and the iron-binding site (histidine and aspartic acid residues; Thomas et al. 1999). However, the amino acid sequence of the CaGA2ox1 protein shows unique asparagine-rich amino acid sequences (located at positions 188–206) that are different from other GA dioxygenases (Fig. 1a).

Fig. 1
figure 1figure 1

Comparison of amino acid sequences of CaGA2ox1 and other GA2oxs and tissue-specific expression of CaGA2ox1. a Sequence alignment of the deduced amino acid sequence of CaGA2ox1 with four closely related GA2ox genes. The asterisks indicate the three amino acid residues forming the iron-binding site. The closed triangles show the putative 2-oxoglutarate binding site. The boxed region indicates unique asparagine-rich amino acid sequences of CaGA2ox1. b Phylogenetic relationship of CaGA2ox1 with other plant GA2oxs using ClustalW (http://www.ebi.ac.uk/clustalw/). The GenBank, DDBJ, EMBL, or NCBI accession numbers and consensus IDs from the pepper EST database (http://genepool.kribb.re.kr/new) for each protein are: CaGA2ox1 (DQ465393), AtGA2ox1 (CAB41007), AtGA2ox2 (CAB41008), AtGA2ox3 (CAB41009), AtGA2ox4 (AAG51528), AtGA2ox6 (AAG00891), AtGA2ox7 (AAG50945), AtGA2ox8 (CAB79120), LsGA2ox1 (BAB12442), OsGA2ox1 (BAB40934), OsGA2ox2 (BAC16751), OsGA2ox3 (BAC16752), OsGA2ox4 (AAU03107), OsGA2ox5 (BAC10398), OsGA2ox6 (CAE03751), PcGA2ox1 (CAB41036), PsGA2ox1 (AAF08609), PsGA2ox2 (AAD45424), PtaGA2ox1 (AAQ93035) SoGA2ox1 (AAN87571), SoGA2ox2 (AAN87572), and SoGA2ox3 (AAX14674). c Tissue-specific expression of the CaGA2ox1 gene. Total RNA was extracted from leaf (L), stem (S), root (R), at 48 h after 1 mM MgCl2 treatment (B48h), 48 h after Xag 8ra inoculation (X48h), immature fruit (IF), mature green fruit (MGF), breaker (B), red fruit (RF), flower bud (FB), and open flower (OF). RT-PCR analysis was carried out using the primers described in Supplemental Table S1

Full size image

A phylogenetic tree was generated using the deduced amino acid sequences of CaGA2ox1 with the GA2oxs from other plants and other GA2ox homologs, which were the most closely related proteins available at the time these studies were initiated from the pepper EST database (EST ID: cacn18210, CaKS25002C15; http://genepool.kribb.re.kr/new/; Fig. 1b, S1). The GA 2-oxidase family was distributed among three different clades, with class I and II enzymes catabolizing C19-GAs and class III enzymes catalyzing the 2β-hydroxylating reaction of C20-GAs (Lee and Zeevaart 2005). The phylogenetic tree analysis indicates that CaGA2ox1 belongs to the class II enzymes with PsGA2ox2, PaGA2ox, AtGA2ox6, and SoGA2ox2. CaGA2ox1 may be a GA 2-oxygenase that has the conserved regions of GA-dioxygenase activity and generates 2β-hydroxylated C19-GAs, such as the bioactive GA4 and GA1.

In various plants, GA2ox genes are expressed in diverse tissues. To examine the steady-state expression level of CaGA2ox1 in various organs of the hot pepper plant, total RNA was extracted from roots, stems, leaves, fruit, flower bud, and open flower. The level of CaGA2ox1 expression was most abundant in the flower tissues, while low level expression was detected in other tissues (Fig. 1c). In contrast, the other pepper GA2ox homologs, CaKS25002C15, and cacn18210, were more abundant than CaGA2ox1 in all tissues in which they were expressed. The tissue-specific expression pattern of cacn18210 was similar to CaGA2ox1 and was induced by Xag8ra inoculation (Fig. S4).

Expression of CaGA2ox1 gene in HR

To determine the expression pattern of the CaGA2ox1 gene during the non-host plant–pathogen interaction, transcript levels of the CaGA2ox1 gene were monitored over time by RNA gel blot analysis (Figs. 2a, S2). Hot pepper leaves infected with Xag8ra exhibited biphasic CaGA2ox1 transcripts level increases. CaGA2ox1 transcripts were strongly induced within 1 h of pathogen infection, were markedly decreased after 6 h, and then were highly induced again at 12 h. When leaves were infiltrated with the buffer control, rapid accumulation of CaGA2ox1 mRNA was also detected from 1 to 3 h after MgCl2 infiltration, but not induced again until 24 h. This transient upregulation in both treatments partly reflected the effects of mechanical stress caused by infiltration. RT-PCR analysis of wound-treated pepper RNA samples disclosed that this early transient induction of CaGA2ox1 was due to the wounding effect of infiltration (Fig. S2). In contrast, expression of CaPR1, a positive control for inoculation of Xag8ra, was first detected at 9 h after bacterial infection and gradually increased thereafter.

Fig. 2
figure 2

Expression of the CaGA2ox1 gene in pepper plants inoculated with bacterial pathogens. a Chili pepper plants (cv. Bukang) were inoculated with a non-host bacterial pathogen, Xag8ra. b Expression patterns of the CaGA2ox1 gene in cv. ECW30R plants challenged with the bell pepper plant bacterial spot pathogens, Xav race 1 (avrBs3) or race 3. Total RNA was isolated from pepper leaves harvested at the indicated time points

Full size image

In addition to the non-host pathogen resistance response, the expression of CaGA2ox1 in R-gene-mediated disease resistance was studied (Fig. 2b). The bell pepper cultivar ECW-30R, which contains the dominant allele BS3, was used. ECW-30R plants are resistant to Xanthomonas axonopodis pv. vesicatoria race 1 (Xav 1), which possesses the avrBS3 avirulence gene, but are susceptible to the Xav race 3 (Xav 3) strain, which causes bacterial spot disease (Kousik and Ritchie 1996). RT-PCR analysis was carried out using a primer set specific for CaGA2ox1 and CaPR1 (Table S1). As shown in Fig. 2b, strong and early (1.5–3 h after inoculation) induction of CaGA2ox1 transcripts was detected in both compatible and incompatible interactions with Xav. In the incompatible interaction, however, induced CaGA2ox1 mRNA levels were detected until 18 h post-inoculation and then its transcripts were re-induced at 24 h and kept until 72 h; by contrast, early time induction of CaGA2ox1 rapidly disappeared at 12 h in compatible interaction and a later weak induction appeared 72 h after Xav 3 with expression significantly lower than that in the leaves of ECW-30R-infected Xav 1. The expression pattern of CaPR1 was also examined as a positive control for Xav inoculation. Specific induction of CaGa2ox1 during incompatible interactions between pepper and bacterial pathogens (Fig. 2a, b) suggests the possibility that CaGA2ox1 is functional in pathogen-induced defense responses in pepper plants.

Expression of CaGA2ox1 in response to various chemicals

To investigate the expression of CaGA2ox1 in response to endogenous defense-related signals, detached hot pepper leaves were treated with 5 mM SA, 5 mM ethephon, or 100 μM MeJA before isolation of mRNA at each time point. As shown in Fig. 3, the addition of ET and SA dramatically increased CaGA2ox1 gene expression, although the expression pattern was different in response to each treatment. CaGA2ox1 transcripts began to accumulate as early as 6 h after SA treatment, reached their maximum level at 24 h, and continued to be expressed at this level at 48 h after treatment. The ET treatment more effectively induced the early expression of CaGA2ox1. CaGA2ox1 mRNA was strongly expressed within 6 h of ET treatment and diminished by 48 h. On the other hand, the expression of CaGA2ox1 was not affected by MeJA treatment, the defense chemical mediating responses to mechanical wounding (Reymond and Farmer 1998) and which induced the strong expression of the chili pepper proteinase inhibitor within 12 h after MeJA treatment (Fig. 2b; Shin et al. 2001). The CaPR1 gene was detected within 24 h of ET and SA treatments. Recently, it has been proposed that abscisic acid (ABA), which is an important signaling molecule in the abiotic stress responses in plants, is also involved in the regulation of DELLA protein degradation in the plant’s growth response to high salt stress (Achard et al. 2006). The CaGA2ox1 mRNA was not induced with the application of exogenous ABA (Fig. S2). Together, these results indicate that the induction of CaGA2ox1 expression following pathogen inoculation is affected by SA and/or ET accumulation and, moreover, that SA and ethylene indirectly affect GA metabolism.

Fig. 3
figure 3

Expression of the CaGA2ox1 gene in detached leaves treated with chemicals related to various stress responses. Total RNA was extracted from detached pepper leaves treated with buffer, as negative control, 5 mM SA, 5 mM ethephon (ET), or 100 μM MeJA at the indicated time points. RNA (20 μg) was separated by electrophoresis, blotted, and hybridized with a 32P-dCTP-labeled full-length cDNA of CaGA2ox1. The expression of CaPR1 in SA- or ET-treated leaves and CaPIN-II in MeJA-treated leaves was monitored as positive controls. The 25S and 18S rRNA bands in ethidium bromide-stained gels are shown as loading controls. One typical result from three independent experiments was presented

Full size image

CaGA2ox1 protein has a GA 2-oxidase activity and is localized to the cytoplasm

One mechanism regulating the GA biosynthetic pathway may be the subcellular compartmentalization of the pathway. To investigate the cellular distribution of CaGA2ox1, the entire CaGA2ox1 cDNA sequence was fused to a soluble-modified green fluorescent protein (smGFP; David and Vierstra 1996) under the control of the CaMV 35S promoter. The resulting construct was introduced into pepper protoplasts by polyethylene glycol-mediated transformation (Park et al. 2001). Localization of the fusion protein was determined using confocal fluorescence microscopy. The CaGA2ox1:smGFP fusion protein was strongly expressed in pepper protoplasts and was localized to the cytoplasm of the cells (Fig. 4a), indicating that CaGA2ox1 may function in the cytosol.

Fig. 4
figure 4

a Subcellular localization of the CaGA2ox1 gene product. The CaGA2ox1 coding region was fused to the coding region of smGFP and introduced into pepper protoplasts by polyethylene glycol-mediated transformation. Fluorescent micrographs of protoplasts were taken using a confocal laser scanning microscope (Carl Zeiss LSM510, Germany) at 24 h after transformation. b Oxidase activity against GA4 of recombinant CaGA2ox1. The pET:CaGA2ox1 recombinant fusion protein expressed in BL21pLysS cell line was incubated with C14-labeled GA4. The conversion of GA4 to GA34 was confirmed by HPLC

Full size image

To test whether CaGAa2ox1 encodes a gene product with GA 2-oxidase activity, the full-length coding region was expressed in Escherichia coli (Fig. S3). The recombinant fusion protein was present in the soluble portion of the induced cell lysate. As described previously, the crude extract was incubated with C14-labeled GA4, a C19-GA (Lee and Zeevaart 2002; Schomburg et al. 2003). As shown in Fig. 4b, cell lysates expressing CaGA2ox1 fusion protein in E. coli converted C14-GA4 to GA34, indicating that CaGA2ox1 encodes a functional GA 2-oxidase. There are two Arabidopsis genes that sort into the class II clade, GA2ox4 and GA2ox6. GA2ox4 has been shown to act exclusively as a C19-GA 2-oxidase (Jasinski et al. 2005), while GA2ox6 was similarly reported to have activity on GA4 and GA1 (Wang et al. 2004). The activity of CaGA2ox1 on C20-GAs and other C19-GAs was not tested.

Treatment with biologically active GA does not affects CaGA2ox1 expression

Thomas et al. (1999) reported that GA 2-oxidase is involved in maintenance of the concentration of biologically active GA in plant tissues. The effects of GA4+7 and PBZ, an inhibitor of GA biosynthesis (Kitahata et al. 2005; Toh et al. 2008), on the transcript levels of CaGA2ox1 and its homologs were examined using real-time quantitative RT-PCR (Q-RT-PCR; Fig. 5) with the inclusion of the GA biosynthetic genes GA20ox and GA3ox (Gallego-Giraldo et al. 2008) positive controls. CaKS25002C15 and cacn18210 transcript levels were decreased in plants treated with PBZ but were highly increased upon GA4+7 application (Fig. 5b). However, CaGA2ox1 transcripts were not significantly changed under PBZ or GA4+7 treatments. The expression of the GA-biosynthesizing CaGA20ox1 highly increased with PBZ application, but not with GA or buffer treatment. CaGA3ox1 transcript levels were decreased in pepper plants treated with GA4+7, but were increased with PBZ application. Taken together, these results indicated that the expression of CaKS25002C15 and cacn18210 are under feedback regulation by the level of biologically active GA, whereas the expression of CaGA2ox1 is not.

Fig. 5
figure 5

The effects of GA4+7 and PBZ on the level for CaGA2ox1. a Phenotype of pepper treated with buffer, 50 μM GA4+7 or 50 μM PBZ. b The expression of CaGA2ox1, homologous GA2oxs, and GA biosynthesis genes such as GA20 oxidase and GA3 oxidase in response to GA4+7 and PBZ applications to pepper plants. The levels of CaGA2ox1 and GA biosynthesis genes were examined using real-time quantitative RT-PCR. Independent experiments were repeated one more time with similar results

Full size image

Conclusion

GA2ox is the enzyme that catalyzes the deactivation of bioactive GAs, a reaction that is important in the regulation of endogenous GA levels (Hedden and Phillips 2000; Olszewski et al. 2002). Expression studies of the three Arabidopsis GA 2-oxidases revealed that two of these enzymes, AtGA2ox1 and AtGA2ox2, were most abundant in the inflorescence and developing siliques, whereas the third enzyme, AtGA2ox3, could not be detected in any tissue (Thomas et al. 1999). This expression pattern is consistent with a role for GA 2-oxidases in reducing GA levels in seeds to promote dormancy. In pepper, transcripts of CaGA2ox1 were not detectable by RNA gel blot analysis and were only detected by RT-PCR in flower tissue after >38 cycles. However, as shown in Fig. 2a, b, expression of CaGA2ox1 was greatly increased in leaves inoculated with bacterial pathogens. In addition, the transcript level of CaGA2ox1 was not affected by treatment with active GAs, even though it encodes a GA 2-oxidase shown to inactivate GA4 (Fig. 4b). These results indicate that CaGA2ox1 might has a role specifically in disease resistance responses in pepper plants. In summary, through the isolation of the gene CaGA2ox1, which encodes a putative GA oxidizing enzyme, we speculated that two independent, pathogen-activated phytohormonal signaling pathways (SA and ET) regulate plant defense responses through induction of GA metabolism. Further studies in pepper plants silenced for multiple GA 2-oxidase homologs and global gene expressions using a pepper cDNA microarray will be useful in elucidating the roles of GA 2-oxidases. Moreover, directly observing GAs contents during plant defense responses may help to explain the function of GAs and coordinate interactions of GAs with other phytohormones in plant disease resistance responses.