ORCA3, a Jasmonate-Responsive Transcriptional Regulator of Plant Primary and Secondary Metabolism

See allHide authors and affiliations

Science  14 Jul 2000:
Vol. 289, Issue 5477, pp. 295-297
DOI: 10.1126/science.289.5477.295


Biosynthesis of many classes of secondary metabolites in plants is induced by the stress hormone jasmonate. The gene for ORCA3, a jasmonate-responsive APETALA2 (AP2)-domain transcription factor fromCatharanthus roseus, was isolated by transferred DNA activation tagging. Orca3 overexpression resulted in enhanced expression of several metabolite biosynthetic genes and, consequently, in increased accumulation of terpenoid indole alkaloids. Regulation of metabolite biosynthetic genes by jasmonate-responsive AP2-domain transcription factors may link plant stress responses to changes in metabolism.

Metabolism comprises coordinate series of coupled enzymatic conversions in living organisms. Secondary metabolites are not vital to cells that produce them, but contribute to the overall fitness of the organism. Functions of these compounds in plants include protection against pests and pathogens. For man, plant secondary metabolites are useful as pharmaceuticals, dyes, fragrances, insecticides, or flavors. Production of secondary metabolites is controlled at the level of expression of the biosynthetic genes by developmental and tissue-specific factors or by external signals (1, 2). Furthermore, accumulation of some metabolites is induced by (methyl)jasmonate [(Me)JA] (3), a plant hormone produced in response to stress.

One group of secondary metabolites are the anthocyanin flower pigments. Expression of genes encoding anthocyanin biosynthetic enzymes is controlled by transcription factors with homology to mammalian MYB and basic helix-loop-helix (bHLH) proteins, respectively [reviewed in (1)]. Knowledge about regulation of other secondary metabolic pathways is more limited.

For the biosynthesis of secondary metabolites, plants must accommodate their primary metabolic pathways. Coordinate regulation between these processes has been observed (4), but the regulatory mechanisms are unknown.

MeJA induces the production of terpenoid indole alkaloid (TIA) secondary metabolites (3). To identify regulators of this metabolic pathway (Fig. 1) in the plant speciesCatharanthus roseus (Madagascar periwinkle), we used a transferred DNA (T-DNA) activation tagging approach (5, 6). Enhancers carried on the T-DNA of a plant transformation vector can deregulate a gene flanking the T-DNA integration site in the plant genome, resulting in a dominant mutation. We used tryptophan decarboxylase (TDC) as a selectable marker (7). TDC catalyzes the first step in TIA biosynthesis starting from tryptophan (Fig. 1). In addition, TDC is able to detoxify 4-methyl-tryptophan (4-mT).Agrobacterium-mediated transformation of C. roseus cell suspensions (8) with the activation tagging construct (Fig. 2A) resulted in roughly 4 × 105 to 5 × 105 stable transformants that were screened for 4-mT resistance at a 0.4 mM concentration. This yielded 180 resistant cell lines screened forTdc expression levels by Northern blot analysis to identify mutant lines with a transcriptional regulator activated by the T-DNA insertion. Among 20 lines with increased Tdc expression (9), six lines also showed increased expression of a second gene involved in TIA production (i.e., Str), suggesting that the tag activated a regulator of multiple genes involved in TIA biosynthesis. From one of these six lines (line 46), 1.6 kilobase pairs (kbp) of plant DNA flanking the T-DNA insertion site were isolated by plasmid rescue (10).

Figure 1

Biosynthesis of TIAs in C. roseus. Solid arrows indicate single enzymatic conversions, whereas dashed arrows indicate multiple enzymatic conversions. Abbreviations of enzymes in the figure are as follows: AS, anthranilate synthase; DXS, D-1-deoxyxylulose 5-phosphate synthase; G10H, geraniol 10-hydroxylase; CPR, cytochrome P450-reductase; TDC, tryptophan decarboxylase; STR, strictosidine synthase; SGD, strictosidine β-D-glucosidase; D4H, desacetoxyvindoline 4-hydroxylase; and DAT, acetyl-CoA:4-O-deacetylvindoline 4-O-acetyltransferase. Genes regulated by ORCA3 are underlined.

Figure 2

(A) Schematic representation of the T-DNA tagging construct. Hpt, hygromycin resistance marker; pUC21,E. coli pUC21 vector; gusA, β-glucuronidase reporter gene; 2B4A1, CaMV 35S promoter enhancer sequences (18) without TATA-box or ATG start codon; LB and RB, left and right T-DNA border sequences. (B) Genomic structure ofOrca3 as isolated by plasmid rescue (10) and I-PCR (14). Boxed region, the transcribed region determined by PCR (13); hatched box, the ORF; black box, the DNA-binding AP2-domain. Positions of the TATA box and ATG start codon are shown. The T-DNA integration site in line 46 is indicated as an inverted triangle. Arrows indicate primers used for PCR (13) and I-PCR (14). The sequence data have been submitted to the DDBJ/EMBL/GenBank databases under accession numbers AJ251249(cDNA sequence) and AJ251250 (genomic sequence).

An open reading frame (ORF) was located ∼600 bp downstream of the T-DNA that predicted a protein of 203 amino acids with a conserved AP2-domain (Fig. 2B). This DNA-binding domain is found in the AP2/EREBP class of plant transcription factors, many of which are involved in the regulation of responses to stress (11). Because of its similarity to ORCA2 (Octadecanoid-derivativeResponsive C atharanthus AP2-domain protein 2) (12), we named the protein encoded by the rescued plant DNA ORCA3. The Orca3 gene contains no introns (13), and the T-DNA integration did not induce any DNA rearrangements (14). Expression ofOrca3 is activated by the T-DNA tag (Fig. 3A, upper row). The overexpressed transcript and the endogenous mRNA have identical sizes, indicating that full-length mRNA and protein are produced, which is consistent with the tagged gene structure (Fig. 2B).

Figure 3

Orca3 overexpression causes induction of expression of primary and secondary metabolic genes and increased accumulation of TIAs. Complementary DNAs of the genes indicated were used as probes on (A) RNA extracted from C. roseus BIX cell line 46 (T) and from a representative control line (C), (B) RNA extracted from ten O3-OX lines and five control cell lines generated by particle bombardment in a genetic background of C. roseuscell line MP183L, and (C) RNA extracted from wild-type MP183L cells exposed for 6 hours to 50 μM MeJA (M) or 0.1 % dimethyl sulfoxide (C). Analysis of a control gene (Rps9) showed equal loading of RNA. (D) Expression levels in (B) were quantified. Mean expression of the control lines is set at 1. Bars indicate means ± standard error. Asterisks above the bars indicate statistically significant differences in expression levels as determined with the nonparametrical Wilcoxon–Mann-Whitney test (P < 0.05). (E) Tryptophan, tryptamine, and TIAs [expressed in μmol per gram dry weight (DW)] were extracted from 8-day-oldC. roseus MP183L cultures (five control lines and five O3-OX lines), either without loganin supply (left panel) or 24 hours after addition of 1 mM loganin to the culture medium (right panel). Bars indicate means ± standard error.

Expression of TIA biosynthetic genes Tdc, Str,Sgd, Cpr, and D4h was increased in theOrca3-tagged line (15) (Fig. 3A). TIA biosynthetic genes G10h and Dat were not induced, suggesting that these genes are not controlled by ORCA3. Genes encoding the alpha subunit of AS (ASα) and DXS, enzymes involved in primary metabolism leading to TIA precursor synthesis, were induced byOrca3 overexpression, whereas two other primary metabolic genes not involved in production of TIA precursors (Ggppsand Ics) were not regulated by ORCA3 (Fig. 3A). Except forSgd, which shows variable expression in control cell lines (Fig. 3B), expression patterns observed in cell lines that were independently transformed by particle bombardment with the rescuedOrca3 ORF (Fig. 3, B and D) or cDNA (9) under control of cauliflower mosaic virus (CaMV) 35Spromoter elements (O3-OX lines) were identical to that of line 46. These results indicate that ORCA3 is a regulator of primary as well as secondary metabolite biosynthetic genes involved in TIA biosynthesis. Although previous studies report differential regulation of genes involved in early (i.e., Tdc and Str) and late (i.e., D4h) steps in vindoline biosynthesis in periwinkle plants (16), genes of both classes are controlled byOrca3 in suspension-cultured cells.

Expression of Orca3, as well as of all other genes used in this study, was induced by MeJA (Fig. 3C). However, not all these genes were induced by Orca3 overexpression. Combinations of different jasmonate-responsive transcription factors, such as ORCA2 (12) and ORCA3, may regulate TIA biosynthesis by selective activation of target genes.

Production of TIAs in untransformed C. roseus cell suspensions is low or absent. TIAs were not detected (17) in the O3-OX lines. Tryptamine and tryptophan synthesis was increased upon Orca3 overexpression (Fig. 3E), showing that the metabolic pathway leading to indole precursors was induced. A possible bottleneck in production of the terpenoid moiety of TIAs is G10H because G10h expression was undetectable (Fig. 3B). To provide an excess of terpenoid precursor, we added loganin extracellularly. Upon loganin feeding, Orca3 overexpression caused significantly increased TIA production (3.2-fold increase compared with controls) (Fig. 3E). O3-OX lines accumulated strictosidine (8.11 ± 0.87 μmol/g dry weight of C. roseus cells), ajmalicine (0.25 ± 0.10 μmol/g dry weight), and an unidentified lochnericine-like TIA (1.21 ± 0.66 area units per gram dry weight). Identical TIA patterns as observed in O3-OX lines after feeding with loganin were detected in untransformed cell suspensions after MeJA treatment (9), demonstrating that the JA-induced conversions of TIAs are regulated by ORCA3.

Electrophoretic mobility shift assays (EMSA) (12) and transient expression assays following particle bombardment (18) showed that ORCA3 functions as a DNA-binding protein which is directly involved in transcriptional activation of Tdc, Str, and Cpr. ORCA3 formed sequence-specific complexes with Tdc, Str, and Cpr promoter fragments (Fig. 4A). Expression from Tdc,Str, and Cpr promoter fragments was activated by coexpressed ORCA3 in C. roseus cells, whereas activation was not observed with unrelated Tcyt elements of T-DNA cytokinin gene promoter (19) or with antisense Orca3expression (Fig. 4B).

Figure 4

Tdc, Str, and Cprpromoter fragments are sequence-specifically bound in vitro andtrans-activated in vivo by ORCA3. (A) EMSAs were performed with the indicated amount of ORCA3 protein expressed inE. coli with Tdc (−99 to +198), Str(−145 to +52), Cpr (−114 to –36), and 4Tcyt(19) promoter fragments as probes. The free probe and the ORCA3 complex are marked by circles and asterisks, respectively. (B) C. roseus cells were transiently cotransformed (18) with Tdc (−99 to +198),Str (−145 to +52), Cpr (−87 to +272), or Tcyt (six copies fused to CaMV 35S −47 to +27 minimal promoter) promoter fragments fused to the gusAreporter gene; overexpression vectors containing the Orca3cDNA fused to the CaMV 35S promoter in sense or antisense orientation; and CaMV 35S promoter-chloramphenicol acetyl transferase (cat) (ratio 1:3:1). B glucuronidase (GUS)/CAT activity ratios were determined (18) and expressed as percentage of empty vector controls. Bars represent means ± standard error (n = 6 replicate experiments, except for antisense where n = 3).

Thus, plants can regulate primary metabolic pathways coordinately with secondary metabolism using a single transcription factor. Because the biosynthesis of many secondary metabolites is induced by jasmonate, the identification of an AP2-domain protein as regulator of several genes involved in JA-responsive metabolism uncovers a control mechanism that may be operative in other stress-responsive plant metabolic pathways as well.

  • * To whom correspondence should be addressed. E-mail: Memelink{at}


View Abstract

Navigate This Article