The ORCA-ome as a key to understanding alkaloid biosynthesis in Catharanthus roseus
Hasnain, G.
Citation
Hasnain, G. (2010, June 29). The ORCA-ome as a key to understanding alkaloid biosynthesis in Catharanthus roseus. Retrieved from
https://hdl.handle.net/1887/15735
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/15735
Note: To cite this publication please use the final published version (if applicable).
Chapter 2
Genome-wide determination of ORCA target genes in Catharanthus roseus
Ghulam Hasnain
1,, Muhammad Khurshid
1Jacob Pollier
2, Alain Goossens
2, Robert Verpoorte
1, Johan Memelink
11 Institute of Biology, Leiden University, P.O. Box 9505 2300 RA Leiden, The Netherlands
2 VIB Department of Plant Systems Biology, UGent-VIB Research Building FSVM, Technologiepark 927, 9052 Ghent, Belgium
Abstract
Catharanthus roseus produces an important class of secondary metabolites known as terpenoid indole alkaloids (TIA). The dimeric TIA vincristine and vinblastine are effective anticancer drugs.
The biosynthetic network of TIA involves about around 35 metabolic steps. All biosynthesis genes tested are responsive to methyl jasmonate (MeJA) in cell suspension cultures. Two transcription factors called ORCA2 and ORCA3 have been reported to regulate the MeJA-responsive expression of several biosynthesis genes. We report here comprehensive transcript profiling analysis of C. roseus cell lines overexpressing ORCA2 or ORCA3 in an inducible manner. By using cDNA-amplified fragment-length polymorphism technology the quantitative temporal accumulation patterns of 11,277 transcript tags were determined and analyzed. In total 74 transcript tags were selected because of their differential accumulation. Thirty six transcripts were upregulated in response to overexpression of either or both ORCA2 and ORCA3 while 22 tags and 16 tags were upregulated specifically in ORCA2 overexpression lines and ORCA3 overexpression lines, respectively. Upon blast searching of the NCBI database, the 74 tags nucleotide sequences gave <22% perfect match with the publically available C. roseus entries. Thus, the majority of the tags identified here are previously un-described C. roseus sequence information. The expression patterns for selected transcript tags were validated by northern blot hybridization and were also found responsive to MeJA. Newly identified ORCA- regulated tags may be novel candidates for missing TIA pathway genes.
Introduction
The medicinal plant Catharanthus roseus L. G. Don is a rich source of an important class of plant secondary metabolites known as terpenoid indole alkaloids (Facchini, 2001). Until now 130 different TIA have been identified in C. roseus and most of these compounds have been screened for pharmaceutical activities (van der Heijden et al., 2004). The monomeric alkaloids ajmalicine and serpentine are in clinical use as antihypertensive drug and tranquilizer, respectively. The two dimeric alkaloids vinblastine and vincristine are very effective drugs against different types of cancer. The low production level of these dimeric alkaloids and the complex purification procedure make these compounds quite expensive. During the past few decades considerable effort has been focused on unraveling the complex regulatory, enzymatic and transport processes involved in TIA biosynthesis.
More than 35 metabolic steps are required to synthesize the most important alkaloids in C. roseus and only 12 enzyme-encoding genes have been isolated to date
Synthesis of TIAs in C. roseus involves two plastidic primary metabolic pathways, the tryptophan pathway and the 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway for terpenoid biosynthesis. All TIAs found in the plant kingdom are synthesized from the central precursor strictosidine, which is a fusion product of tryptamine, derived from tryptophan, and secologanin, derived from geraniol (Fig. 2). Several of the enzymatic steps and corresponding genes acting downstream of strictosidine have been identified but several enzymes, genes and even metabolic intermediates remain unknown. Upstream of strictosidine, the pathway leading from geraniol to secologanin has also several unknown enzymatic steps and uncharacterized metabolic intermediates (Loyola-Vargas et al., 2007).
The TIA biosynthetic pathway is under strict developmental and environmental control. The plant hormone (methyl) jasmonate has a positive effect on gene regulation and alkaloid production in C. roseus. MeJA induces most of the known TIA pathway genes (Collu et al., 2001; Memelink et al., 2001) which leads to the accumulation of elevated level of TIAs in C. roseus cell culture (El- Sayed and Verpoorte, 2002). Transcription factors are responsible for coordinating the expression of biosynthetic genes in response to external and internal stimuli. Two AP2/ERF-domain transcription factors called ORCA2 and ORCA3 encoded by MeJA-responsive genes have been isolated (Menke et al., 1999; van der Fits and Memelink, 2000). For the strictosidine synthase (STR) gene it was shown that ORCA proteins control its transcription by binding specifically to a GCC box in the
Figure 1. Biosynthetic pathway for terpenoid indole alkaloids in Catharanthus roseus. Solid arrows indicate single enzymatic conversions, whereas dashed arrows indicate multiple enzymatic conversions. Indicated are enzymes, for which the corresponding genes were cloned. Enzymes encoded by genes previously reported to be regulated by ORCA3 are shown against a gray background. AS: anthranilate synthase, CPR: cytochrome P450 reductase, D4H: desacetoxyvindoline 4-hydroxylase, DAT: acetyl-CoA:4-O-deacetylvindoline 4-O-acetyltransferase, DXS: 1-deoxy-D-xylulose-5-phosphate synthase, DXR: 1-deoxy-D-xylulose-5-phosphate reductoisomerase, MECS: 2-C-methyl-D-erytritol 4-phosphate cytidyltransferase, LAMT: loganic acid methyltransferase, G10H: geraniol 10-hydroxylase, PRX1: Peroxidase 1, SGD:
strictosidine β-D-glucosidase, SLS: secologanin synthase, STR: strictosidine synthase TDC: tryptophan decarboxylase, T16H: tabersonine 16-hydroxylase, 16OMT: 16-hydroxytabersonine-16-O-methyltransferase.
promoter (Menke et al., 1999; Van der Fits and Memelink, 2001). Constitutive over expression of ORCA3 resulted in elevated expression of several genes involved in the production of primary precursors and of TIA (van der Fits and Memelink, 2000). ORCA2 was shown to regulate the STR gene but the effect of ORCA2 on other TIA pathway genes has not been reported. It has been suggested that ORCA2 and ORCA3 regulate different but overlapping sets of genes (Memelink et al., 2001). Studies on the effects of ORCA2 and ORCA3 on TIA pathway genes were based on already known TIA pathway genes. It is clear that these transcription factors regulate several other genes in the TIA pathway which are not characterized yet.
In this chapter we describe studies aimed at the determination of the large majority of possible ORCA target genes. The approach was to overexpress each ORCA in C. roseus cells and determine changes in gene expression. Since constitutive overexpression of ORCAs may have deleterious effect on cell metabolism and can cause the activation of remedial pathways by feedback mechanisms, we used the inducible XVE expression system, which is based on the artificial XVE transcription factor responding to estradiol as an inducer (Zuo et al., 2000). We used cDNA-amplified fragment length polymorphism (cDNA-AFLP) technology to study the genome-wide effect of ORCA2 and ORCA3 overexpression in C. roseus cells. This technology does not require prior genome sequence information and can identify already known gene transcripts as well as unknown transcripts (Vuylsteke et al., 2007). The results show that ORCA2 and ORCA3 regulate distinct but overlapping sets of genes. All known TIA biosynthesis found in the cDNA-AFLP analysis are in the gene set controlled by both ORCA2 and ORCA3, suggesting that this set may be the prime candidate for harboring unknown TIA biosynthesis genes.
Materials and Methods
Cell Cultures, Stable Transformation, and Treatments
Catharanthus roseus cell suspension line MP183L was maintained by weekly 10-fold dilution in 50 mL of Linsmaier and Skoog (LS) medium containing 88 mM sucrose, 2.7 µM 1-NAA and 0.23 µM kinetin and was grown at 25°C in a 16/8 hour light/dark regime at 200 µE m-2 s-1 at 70% relative humidity on a rotary shaker at 120 rpm. For the construction of cell lines with estradiol-responsive transgene expression, the GFP open reading frame (ORF) was excised from 35S-omega-sGFP(S65T) (Chiu et al., 1996) with SalI/PstI and cloned in pBluescript SK+. It was then excised with ApaI/
SpeI and cloned in pER8 (Zuo et al., 2000). The ORCA2 ORF was excised from pBluescript SK+- ORCA2 with SalI/SpeI, and cloned into pER8 digested with XhoI/SpeI (Zuo et al., 2000). The ORCA3 ORF was excised from pIC-20H with XhoI/SpeI and cloned into the binary vector pER8.
For stable transformations of cell line MP183L, plasmids were introduced by particle bombardment (van der Fits and Memelink, 1997). Transgenic cells were selected on solid LS medium containing 50 µg/ml hygromycin-B and individual transgenic calli were converted to cell suspensions. For the induction of transgenes, 10 ml of cell culture 4 days after transfer to fresh medium were transferred to sterile flasks and β-estradiol (Sigma) dissolved in DMSO was added at a concentration of 10 µM. Induced cells were harvested 24 h after estradiol addition. Control cell lines were treated with
DMSO at a final concentration of 0.1% (v/v). Harvested cell samples were frozen in liquid nitrogen and stored at -80 ºC.
RNA extraction and Northern blot analysis
Total RNA was isolated from frozen cells ground in liquid nitrogen by extraction with two volumes of hot phenol buffer (1:1 mixture of phenol and 100 mM LiCl, 10 mM Na-EDTA, 1% sodium dodecyl sulfate (SDS), 100 mM Tris) and one volume of chloroform. After centrifugation, the aqueous phase was re-extracted with one volume of chloroform. RNA was precipitated overnight with LiCl at a final concentration of 2M, washed twice with 70% ethanol, and resuspended in water. Northern blot analyses were performed as described (Memelink, 1994) with the following modifications.
Ten µg RNA samples were subjected to electrophoresis on 1.5% agarose/1% formaldehyde gels, and blotted to GeneScreen nylon membranes (Perkin-Elmer Life Science, Boston, MA). Blots were prehybridized for several hours in 1M NaCl, 10% dextran sulfate (sodium salt, Sigma), 1 % SDS, and 50 µg/ml denatured salmon sperm DNA at 65 ºC before addition of denatured 32P-labeled DNA probes. After overnight hybridization, blots were washed twice at 42 ºC for 30 min with 0.1 x SSPE (saline/sodium/phosphate/EDTA) and 0.5% SDS. Finally, the blots were washed briefly with 0.1 x SSPE at room temperature. Blots were exposed to X-ray films (Fuji, Tokyo, Japan).
cDNA-AFLP Analysis and Data Processing
Two independent cell lines for each construct (GFP, ORCA2 and ORCA3) were selected and treated with estradiol and DMSO. Treatments were done in triplicate and RNA was extracted as described above. Replicate samples were pooled and 6 µg RNA were used for cDNA-AFLP analysis. Sample preparation and cDNA-AFLP based transcript profiling were performed as described (Vuylsteke et al., 2007). For genome-wide transcript profiling, all 128 possible BstYI + 1/MseI + 2 primer combinations were used. The data were processed essentially as described (Vandenabeele et al., 2003). For normalization of data within each primer combination, 25% of the genes with the lowest coefficient of variation value were marked as constitutively expressed. Transcript tags displaying expression values with a coefficient of variation >0.6 were considered as differentially expressed and, after visual inspection, were taken for further analysis. Selected differentially expressed tags were excised from the gels and reamplified with BstYI + 2/MseI + 0 primers. PCR fragments were directly sequenced and tags which did not give good quality sequences were cloned in pGEM-T Easy (Promega, USA) and sequenced. To characterize the tag sequences, they were compared against nucleotide sequences in the NCBI database using the TBLASTX algorithm (Altschul et al., 1997).
Tag sequences were replaced with longer EST or isolated cDNA sequences, when available, to increase the chance of finding significant similarities. The similarity threshold for BLAST searches was defined at 1 x 10-3. However, because of the small size of some tags, in some cases a lower value was accepted when unambiguous matches were observed.
Results
Inducible overexpression of ORCAs
Constitutive overexpression of ORCA transcription factors may have deleterious effects on cell growth due to production of toxic TIA or may block TIA biosynthesis via negative feedback inhibition mechanisms. For these reasons and to maximize the likelihood that primary target genes would be detected, we used the estradiol-inducible XVE system (Zuo et al., 2000) to express the ORCAs in an inducible manner in transgenic C. roseus cells. XVE expression modules containing ORCA2 or ORCA3 were introduced in C. roseus cells via particle bombardment. To generate control lines an XVE expression module carrying GFP was introduced in cells. Independent transgenic lines were screened for elevated levels of expression of introduced genes upon induction with estradiol.
Two ORCA2 overexpressing cell lines, ORCA2-3 and ORCA2-27, and two ORCA3 overexpressing cell lines, ORCA3-NV and ORCA3-29, were selected for further studies (Fig. 2). Two transgenic GFP lines, GFP-2 and GFP-3, were selected as controls. The expression of ORCAs and GFP under control of the XVE system was somewhat leaky with low mRNA levels detected in the absence of inducer (not shown). All selected lines showed high overexpression of respective genes when treated with estradiol and were therefore suitable for transcript and metabolite profiling.
Transcript Profiling
The cDNA-AFLP technology (Breyne et al., 2003) was applied for genome-wide transcript profiling of ORCA overexpressing cell lines. Using 128 Bsty1 + 1/MseI + 2 primer combinations, the quantitative temporal accumulation patterns of 11,277 transcripts tags were determined and analyzed. In total 1874 transcripts were differentially expressed between cell lines and treatments from which those judged most relevant are shown in Table 1. Since ORCAs are positive regulators of the TIA pathway, for identification of the tags we focused on those differentially expressed genes, which were up-regulated in estradiol-treated ORCA2 and ORCA3 lines alone or in both and at lower levels in estradiol-treated GFP lines and in DMSO-treated ORCA lines (Fig. 3). In total 76 differentially expressed transcript tags were isolated and amplified by PCR (hereafter referred to as CR tags) (Table 1, set 8, 10 12). The
Figure 2. ORCA2 and ORCA3 overexpression in C. roseus cell lines under control of the estradiol-inducible XVE expression system.Northern blots containing identical amounts of total RNA from independent transgenic lines were hybridized with the indicated probes. Each cell line was treated with 10 µM estradiol (E) or DMSO (D). GFP-2 and GFP-3 lines contain GFP in the XVE expression module. The ethidium bromide (EtBr) stained gel is shown as a control for RNA loading.
amplified tags were subjected to direct sequencing. Tags, which did not yield a good quality sequence after direct sequencing, were cloned in the pGEM-T Easy cloning vector and subjected
to sequencing. The CR tag sequences were used to search the NCBI database using the TBLASTX algorithm and the first blast hit was used for annotation. Blasting of 17 of the 74 nucleotide sequences of the cDNA-AFLP tags gave a perfect match with a nucleotide sequence entry in the data base. Thus, the majority of the tags identified here are previously undescribed C. roseus sequences. The BLAST
search revealed that 20 of the CR tags displayed no sequence similarity to any entry in the data base.
According to the Functional Catalog of the Munich Information Center for Protein Sequence (http://
mips.gsf.de/projects/funcat), the CR tags can be classified into eight broadly defined functional groups.
The functional group “Metabolism and Energy” in particular is one of the major groups (Fig. 4) and includes, as anticipated, all TIA pathway genes detected in this analysis.
Figure 3. cDNA-AFLP fingerprint of transgenic C. roseus cell lines. Two independent transgenic cell lines for each construct (GFP, ORCA2, and ORCA3) are represented in lanes 1–12. Lanes 1-4 are GFP lines treated with estradiol (2, 4) and DMSO (1, 3), lanes 5-8 are ORCA2 lines treated with estradiol (6, 8) and DMSO (5,7), lanes 9-12 are ORCA3 lines treated with estradiol (10, 12) and DMSO (9, 11) analyzed using four BstYI + 1/MseI + 2 primer combinations (PCs) (labeled PC1–5).
Selective amplification was performed using a [γ-33P]ATP-labeled BstYI primer and fingerprints were visualized using phosphorimager technology. A 10 bp size marker (M) was included on both sides of each primer combination. An examples of a differentially expressed tag is indicated by an arrow.
Transcripts upregulated by both ORCA2 and ORCA3
This data set contains all those transcripts, which are upregulated by both ORCA2 and ORCA3 (Table 2). Out of 76 selected tags 37 fall in this group, which is about 48 % of the total tags (Fig. 5; Table 2). Twelve of the tags had TBLASTX hits to genes encoding enzymes, among which were 4 that were annotated as genes from C. roseus involved in TIA biosynthesis, encoding ASα, TDC, SLS and 10HGO (Fig. 1). The tags CR-37 and CR-60 gave 100% identity TBLASTX hits to the TDC gene while CR-11 and CR-79 gave 100 % amino acid identity hits to ASα and SLS respectively (Table. 2).
The tag CR-36 has 76% amino acid identity to a protein annotated as 10-hydroxy geraniol
Figure 4. Pie chart representation of the distribution (%) of the CR gene tags according to the classification in MIPS functional categories
Table 1. List of differentially expressed tags.
oxidoreductase (10HGO) and probably corresponds to an isoenzyme of 10HGO. The CR tags with hits with enzyme-encoding genes from other plant species might correspond to some of the missing links of the TIA pathway. The CR-75 tag sequence gave a TBLASTX hit to a gene from Sesbania rostrata annotated as aldo/keto oxido-reductase , which is an NADPH dependent oxidoreductase. A closely related enzyme from Papaver somniferum has been described to be involved in morphine biosynthesis (Unterlinner et al., 1999). Another tag, CR-93, gave similarity to polyneuridine aldehyde esterase from Rauvolfia serpentina, which is a key enzyme in the biosynthesis of ajmaline (Dogru et al., 2000), a TIA which does not occur in C. roseus.
The tags CR-61, CR-71, CR-80 and CR-67 gave no blast hit to any gene in the NCBI database while CR-6 and CR-18 showed similarity to unknown proteins. Three tags, CR-46, CR- 50 and CR-74, gave hits to hypothetical proteins. Tag CR-19 gave a hit to a transporter protein.
Tag CR-52 gave a hit to glutathione S-transferase which is involved in flavonoid transport into the vacuole (Zhao and Dixon, 2010). These latter two genes might be involved in the transport of TIA intermediates within C. roseus cells. The tags CR-45 and CR-53 gave hits to cytochrome P450
Figure 5. Relative expression levels and average linkage hierarchical clustering of CR gene tags up-regulated upon overexpression of ORCA2 and ORCA3. Red and green boxes reflect transcriptional activation and repression, respectively, relative to the average expression level of all samples in this group. Gray boxes correspond to missing data points. GFP-2 and GFP-3 are negative control lines containing GFP in the XVE expression module. Each cell line was treated with DMSO (D) or estradiol (E).
enzymes, a large class of multifunctional enzymes involved in plant secondary metabolism.
Northern blot hybridization performed for a number of selected tags to verify the expression pattern found in the cDNA-AFLP analysis revealed that the selected transcripts were upregulated in ORCA2 and ORCA3 cell lines treated with estradiol (Fig. 6). There was no effect of estradiol on the expression of these genes in GFP lines. All selected transcripts were responsive to MeJA, which is consistent with a possible role of the encoded proteins in TIA biosynthesis.
Transcripts upregulated by ORCA2
The total number of transcript tags which were regulated by ORCA2 only was 24 (Fig. 7; Table 3), out of which 5 gave hits to ORCA2 itself (CR-25, CR-28, CR-29, CR-35 and CR-56). Seven tags (CR-4, CR-20, CR-21, CR-30, CR-33, CR-42, and CR-78) did not show similarity to any gene in the database. Six tags showed similarity to enzymes in the TBLASTX search. CR-63 gave a hit with a cytochrome P450 enzyme and therefore may correspond to an enzyme involved in TIA biosynthesis.
The tag CR-65 showed similarity to 2-hydroxyisoflavanone dehydratase from Glycyrrhiza echinata, which is an enzyme involved in flavonoid biosynthesis. The expression of transcripts CR-63 and CR- 65 was verified by Northern blot hybridization and was confirmed to be regulated by ORCA2 but not by ORCA3. It was also found that both CR-63 and CR-65 transcripts accumulated in response to MeJA (Fig. 8). The complete list of tags regulated by ORCA2 is listed in Table 3.
Figure 6. Northern blot analysis with CR tags regulated by both ORCA2 and ORCA3.Identical amounts of total RNA from independent lines were hybridized with the indicated probes. Each cell line was treated with 10 µM estradiol (E) or DMSO (D). GFP transgenic cell lines were also treated with 10 µM MeJA (MJ) for 24 hrs. Wild type (WT) cells were also treated with MeJA and DMSO for 24 hrs. T0 represents a sample taken at time 0. EtBr indicates ethidium bromide staining of the RNA gel prior to blotting.
Table 2. List of CR tags up regulated in both ORCA2 and ORCA3 cell lines upon induction.
Figure 7. Relative expression levels and average linkage hierarchical clustering of CR gene tags specifically upregulated upon overexpression of ORCA2. See legend of Fig. 5 for explanation.
Figure 8. Northern blot analysis with CR tags specifically regulated by ORCA2.See legend of Fig. 6 for explanation.
Transcripts upregulated by ORCA3
Sixteen transcript tags were specifically upregulated by ORCA3 (Fig. 9; Table 4). When subjected to a TBLASTX sequence database search six showed similarity to enzymes while five tags did not show a hit. Two tags, CR-23 and CR-49, seem to correspond to enzymes involved in oxidation/
reduction reactions because they showed similarity with aldo/keto reductase family genes. The tag CR-2 showed 100 % amino acid identity to the ORCA3 gene itself. The tag CR-97 gave a hit to another kind of transcription factor belonging to HD-zip class 3. Two other tags, CR-5 and CR-64, gave hits to cytochrome P450 enzymes. The tags CR-39, CR-94, CR-91 and CR-12 gave hits to a purine transporter, nucleoid DNA binding like protein, ubiquitin activating enzyme and glutamine synthetase respectively. To validate our cDNA-AFLP results we checked the expression of selected tags by Northern blot hybridization. The tag sequences were used as probes and it was found that transcripts CR-5, CR-97, CR-94, CR-39, and CR-59 were upregulated only in ORCA3 cell lines treated with estradiol while transcript CR-64 was also weakly upregulated by ORCA2 but the signal on the RNA gel blot was much lower than after ORCA3 induction (Fig. 10).
Table 3. List of CR tags up regulated only in ORCA2 cell lines upon induction but not in ORCA3 cell lines.
Figure 10. Northern blot analysis with CR tags specifically regulated by ORCA3 See legend of Fig. 6 for explanation.
Figure 9. Relative expression levels and average linkage hierarchical clustering of CR gene tags specifically upregulated upon overexpression of ORCA3. See legend of Fig. 5 for explanation.
Effect of ORCA2 and ORCA3 overexpression on known TIA biosynthesis genes
Certain genes, which were reported to be the targets of ORCA3, were not detected in our cDNA- AFLP profiling, i.e. STR, CPR, and DXS. For STR this is due to the fact that the cDNA lacks the restriction sites used in generation of cDNA-AFLP tags. We used northern blot hybridization to study the expression of all known TIA pathway genes in our inducible ORCA cell lines. The same independent cell lines, which were used in the cDNA-AFLP profiling, were also used in this experiment. Cell lines were treated with DMSO and estradiol. The expression of TDC, STR, ASα, D4H, and SLS were upregulated in both ORCA2 and ORCA3 samples when treated with estradiol while the expression in GFP cell lines was not affected by estradiol (Fig. 11). LAMT which was discovered recently (Murata et al., 2008), was also upregulated in both ORCA2 and ORCA3 cell lines treated with estradiol but the expression was very weak when compared to the MeJA-treated sample. We did not find induced expression of CPR and DXS in our samples, which differs from the results obtained with cell lines constitutively overexpressing ORCA3 (van der Fits and Memelink, 2000). G10H was also not controlled by either ORCA2 or ORCA3. In fact G10H expression was not detectable except in those samples that were treated with MeJA (fig. 11). The expression of the vindoline biosynthesis genes T16H, 16OMT, and DAT was not affected by induction of ORCA2 or ORCA3. In contrast D4H transcript accumulated when ORCA2 or ORCA3 were induced. PRX1, encoding a class III peroxidase gene which is involved in the formation of bisindole alkaloids, was also not regulated by either ORCA2 or ORCA3. Interestingly all the known TIA pathway genes were induced by MeJA in our cell line (Fig. 11). Analysis of a control gene (RPS9) showed equal RNA loading.
Table 4. List of CR tags up regulated only in ORCA3 cell lines upon induction but not in ORCA2 cell lines
Discussion
ORCA2 and ORCA3 are two AP2/ERF-domain transcription factors isolated from C. roseus, which regulate genes involved in the biosynthesis of TIA (Menke et al., 1999; van der Fits and Memelink, 2000). ORCA3 was shown to regulates several genes in the TIA pathway (van der Fits and Memelink, 2000) but the knowledge about ORCA2-regulated genes was more limited. It has been suggested that these two transcription factors may regulate different but overlapping gene sets. To study the impact of these transcription factors on C. roseus metabolism we used the estradiol-inducible XVE over- expression system, because we thought that constitute overexpression of these transcription
factors may be toxic for cells and may lead to the activation of negative feedback mechanisms.
Genome-wide expression profiling methods, such as serial analysis of gene expression or microarray analysis, were not applicable to C. roseus because of the lack of large sequence
Figure 11. Northern blot analysis of known genes acting in the TIA pathway and in precursor pathways. Identical amounts of total RNA from independent lines were hybridized with the indicated probes. Each cell line was treated with 10 µM estradiol (E) or DMSO (D). Wild type (WT) cells were also treated with MeJA and DMSO for 24 hrs. See legend of Fig.1 for most gene names. MAT; 19-hydroxy-O-acetyltransferase, ICS; isochorismate synthase.
Table 5. List of CR tags which have TBLASTN hits to Catharanthus EST data base.
repertoires. In contrast the cDNA-AFLP technology can be used to identify expressed genes in non- model plant species and acquire quantitative expression profiles at the same time (Goossens et al., 2003). For cDNA-AFLP analysis all 128 possible BstYI + 1/MseI + 2 primer combinations were used which potentially can cover 80 % of the total transcripts (Breyne et al., 2003). With this set up we were able to measure the expression levels of 11,277 transcripts. In total 1874 transcripts were differentially expressed with every type of differential expression pattern represented and those judged most relevant are shown in Table 1. Since the ORCA transcription factors are positive regulators, we were interested only in those differentially expressed genes, which were upregulated in ORCA2 and ORCA3 cell lines treated with estradiol. Using this criterion and also by visual observation we selected a total of 76 tags which were upregulated when ORCA2 or ORCA3 were expressed at an elevated level.
All the tags were sequenced and subjected to TBLASTX search using NCBI online blast search tools. About 27 % (20 CR tags) of the tag sequences did not yield any similarity to the NCBI non-redundant gene database. This is consistent with the study of Rischer et al., (2006), in which 37 % of the tags did not yield hits. This may be due to scarcity of gene sequences from either C.
roseus or other species belonging to the Apocynaceae family, which can be used for tag sequence extension in a blast search. Another explanation could be that some of the tags are too short to yield homology and/or contain non-conserved 3`UTR sequences. Therefore isolation of cDNA cloneshas been suggested to be a necessary step in characterizing the cDNA-AFLP derived transcript tags (Breyne et al., 2003).
From the 54 tags that produced hits, 30 tags (55 %) fall in the functional group of “energy and metabolism”, and include, as anticipated, all the known TIA pathway genes detected in this analysis. When the tag sequences were subjected to TBLASTN with the Expressed Sequence Tag (EST) data base of C. roseus, 40 tags gave a perfect match, as listed in Table 4.
Clustering analysis split the selected tags into three main groups:
common set of genes
i. : genes which were up regulated in both ORCA2 and ORCA3 cell lines when induced by estradiol but not in GFP cell lines,
ORCA2-specific genes
ii. : genes which were upregulated only in ORCA2 cell lines when induced by estradiol but not in ORCA3 cell lines and in GFP cell lines,
ORCA3-specific genes
iii. : genes which were up-regulated only in ORCA3 cell lines when induced by estradiol but not in ORCA2 cell lines and in GFP cell lines.
Therefore it can be concluded that ORCA2 and ORCA3 regulate an overlapping set of genes as well as specific sets of genes. An about 50% majority of the tags fall in the overlapping gene set.
The known TIA pathway genes detected in our experimental set up (TDC, ASα, and SLS) fall in this group. Therefore this data set seems to be the most likely to contain missing links in the TIA pathway.
We did not find tags corresponding to STR, DXS, CPR, and D4H, which were found to be regulated by ORCA3 in a previous study (van der Fits and Memelink, 2000). This is because STR does not have the restriction sites for the enzymes weused for cDNA-AFLP analysis and DXS and CPR were not up-regulated by ORCA as it confirmed by northern (Fig. 11). D4H was not detected
on cDNA-AFLP gels but was up-regulated in ORCA2 and ORCA3 cell line when confirmed by northern (Fig 11). Tag CR-36 has 76% amino acid identity to gene bank accession no. AY352047 annotated as C.roseus 10HGO gene. There is no reported biochemical evidence supporting that this enzyme has 10HGO activity. Although we consider CR-36 a likely 10HGO candidate biochemical evidence is needed.
The overlapping gene set contains an additional 8 tags that showed similarity to enzymes.
Tag CR-75 showed homology to chalcone reductase. Therefore the corresponding enzyme may be involved in flavonoid biosynthesis in C. roseus. Alternatively it may catalyze a reaction similar to chalcone reduction in the TIA biosynthesis pathway. This notion is supported by the identification of another close relative of this gene which has was shown to be involved in alkaloid biosynthesis in opium poppy (Unterlinner et al., 1999). We speculate that the CR-75 enzyme is involved in the reduction of cathenamine to ajmalicine, or it may be working as a tetrahydroalstonine synthase.
Both of these steps have been proposed to be carried out by an NADPH-dependent oxidoreductase (Felix et al., 1981; Hemscheidt and Zenk, 1985). Tag CR-93 had a TBLASTX hit to polyneuridine aldehyde esterase, which is involved in ajmaline biosynthesis in Rauvolfia serpentina. The compound ajmaline is structurally related to C. roseus alkaloids but as such has not been reported from C.
roseus. Therefore the CR-93 enzyme is likely to be involved in TIA biosynthesis although we do not have a suggestion for a candidate enzymatic step.
Plants contain a large number of cytochrome P450 enzymes many of which are dedicated to secondary metabolism. Several CYP450s have been described with a role in TIA biosynthesis in C.
roseus, i.e. SLS, G10H and T16H. We picked up several CYP450s in our study; three of which were regulated by both ORCA2 and ORCA3. One of them, CR-79, corresponds to SLS. The regulation of CR-45 and CR-53 transcript tags by both ORCA2 and ORCA3 and in response to MeJA suggests that these CYP450s may have a role in TIA biosynthesis.
Our cDNA-AFLP data contains not only tags corresponding to enzyme-encoding genes but also genes encoding transcription factors, transport proteins and proteins related to protein synthesis.
This means that ORCA transcription factors not only regulate TIA biosynthesis genes but also other genes. We do not know at this point whether these genes are direct or indirect targets of ORCA transcription factors.
The cDNA-AFLP expression data were confirmed for some of the tags by northern blot hybridization. Five tags, CR-45, CR-53, CR-68, CR-69 and CR-90, were up regulated by both ORCA2 and ORCA3 consistent with the cDNA-AFLP results. These transcript tags were also responsive to MeJA, which is to be expected since the ORCAs are responsible for conferring MeJA- responsive gene expression.
The number of ORCA2-specific tags was 23 of which 5 gave a hit to the ORCA2 gene itself. The cDNA-AFLP technique was developed to get a single tag for each gene (Vuylsteke et al., 2007).
Technical reasons to for obtaining more than one tag for the same gene may be incomplete digestion by MseI/BstYI, although we did not observe this in our sequenced tags. The ORCA2-regulated tag CR-63 gives 100% nucleotide identity to a C. roseus CYP450 gene (accession number L19074). The expression of this CYP450 in C. roseus cell lines was reported to be correlated with the induction of
alkaloid biosynthesis (Mangold et al., 1994), which together with its regulation by ORCA2 strongly indicates that this CYP450 is involved in TIA biosynthesis. The tags CR-57 and CR-65 showed similarity to the enzymes aldo/keto reductase and dehydratase respectively. The expression of CR- 63 and CR-65 transcript tags was verified by northern blot hybridization and consistent with the cDNA-AFLP data they were found to be upregulated by ORCA2 alone. In addition their expression was responsive to MeJA which induces all known TIA pathway genes, and therefore is consistent with their possible involvement in TIA biosynthesis.
The ORCA3-specific set of tags was the smallest in number. Two CYP450 tags, CR-5 and CR-64, were upregulated by ORCA3 alone. CR-64 was also picked up by Rischer et al. (2006) in their CDNA-AFLP profiling (tag CRG432). In the gene to metabolite network tag CRG432 correlated to metabolites, suggesting that this CYP450 enzyme may function in the TIA pathway.
Tag CR-5 gave a perfect hit to a partial C.roseus CYP450 EST sequence in the NCBI database.
Two tags, CR-23 and CR-49, showed similarity to the same aldo/keto reductase enzyme as tag CR- 57 A closer analysis of the sequences showed that tags CR-49 and tag CR-57 are identical.CR-39 tag showed similarity to a purine transporter. CR-79 tag showed similarity to a class III HD zip protein.
The expression of all known TIA pathway genes was determined by northern blot hybridization. STR, TDC, SLS, ASsα and D4H genes were upregulated by both ORCA2 and ORCA3.
Van der Fits and Memelink (2000) reported that constitutive overexpression of ORCA3 led to elevated transcript levels of DXS, CPR, STR, ASα, TDC, and D4H. We did not see an effect of overexpression of ORCA3 or ORCA2 on the expression of CPR (Fig. 11), and DXS (not shown)in our inducible system. One explanation for this difference could be the use of a constitutive expression system versus an inducible system. In a recent study with dexamethason dependent inducible overexpression of ORCA3 in a Catharanthus root culture, the expression of CPR was not upregulated (Peebles et al., 2009), which is consistent with our observation, but in that study an effect on DXS expression was observed. In the same study SLS was reported as a target gene for ORCA3, while we observed that both ORCA2 and ORCA3 regulated the SLS gene. The recently discovered gene LAMT (Murata et al., 2008) was also found to be regulated by ORCA2 and ORCA3, but its expression was very low compared with MeJA-treated cells. The secologanin biosynthesis genes LAMT and SLS are regulated by ORCAs but G10H and CPR were not under ORCA control. It is likely that G10H and CPR are regulated by different transcription factors.
The expression of most known TIA biosynthesis genes in response to ORCA overexpression was lower than after MeJA treatment. One reason maybe that the chosen time points are not the optimal ones for each treatment and gene. It is also possible that additional transcription factors are required for optimal expression. Finally it is also possible that the ORCAs need to be activated at the protein level by a MeJA-responsive mechanism for optimal activity. Another interesting observation is that all known genes can be induced in suspension cells with MeJA, even genes that in plants are expressed only in specific cell types. This means that suspension cells may have the capacity to synthesize complex alkaloids under the right conditions, and that it may be possible to force cells to do so with the appropriate set of transcription factors.
Acknowledgments
We thank Maggie Lu (ITRI, Hsinchu, Taiwan) for construction of pER8-ORCA plasmids. Ward de Winter is acknowledged for expert help with cell suspension cultures. G.H. was partially supported by the Institute of Biotechnology and Genetic Engineering NWFP Agricultural University Peshawar, Pakistan and by a van der Leeuw grant from the Netherlands Organization for Scientific Research (NWO) awarded to J.M.
References
Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: 3389-3402
Breyne P, Dreesen R, Cannoot B, Rombaut D, Vandepoele K, Rombauts S, Vanderhaeghen R, Inzé D, Zabeau M (2003) Quantitative cDNA-AFLP analysis for genome-wide expression studies. Mol. Genet. Genomics 269: 173-179
Chiu W-l, Niwa Y, Zeng W, Hirano T, Kobayashi H, Sheen J (1996) Engineered GFP as a vital reporter in plants. Curr. Biol. 6: 325-330
Collu G, Unver N, Peltenburg-Looman AMG, van der Heijden R, Verpoorte R, Memelink J (2001) Geraniol 10-hydroxylase, a cytochrome P450 enzyme involved in terpenoid indole alkaloid biosynthesis. FEBS Lett. 508: 215-220
Dogru E, Warzecha H, Seibel F, Haebel S, Lottspeich F, Stöckigt J (2000) The gene encoding polyneuridine aldehyde esterase of monoterpenoid indole alkaloid biosynthesis in plants is an ortholog of the α/β hydrolase super family. Eur J Biochem. 267: 1397-1406
El-Sayed M, Verpoorte R (2002) Effect of phytohormones on growth and alkaloid accumulation by a Catharanthus roseus cell suspension cultures fed with alkaloid precursors tryptamine and loganin. Plant Cell Tiss. Org. 68: 265-270
Facchini PJ (2001) ALKALOID BIOSYNTHESIS IN PLANTS: Biochemistry, cell biology, Molecular regulation, and metabolic engineering applications. Annu. Rev. Plant Phys.. 52:
29-66
Felix H, Brodelius P, Mosbach K (1981) Enzyme activities of the primary and secondary metabolism of simultaneously permeabilized and immobilized plant cells. Anal. Biochem.
116: 462-470
Goossens A, Häkkinen ST, Laakso I, Seppänen-Laakso T, Biondi S, De Sutter V, Lammertyn F, Nuutila AM, Söderlund H, Zabeau M, Inzé D, Oksman-Caldentey K-M (2003) A functional genomics approach toward the understanding of secondary metabolism in plant cells. Proc. Natl. Acad. Sci. USA 100: 8595-8600
Hemscheidt T, Zenk MH (1985) Partial purification and characterization of a NADPH dependent tetrahydroalstonine synthase from Catharanthus roseus cell suspension cultures. Plant Cell Rep 4: 216-219
Loyola-Vargas V, Galaz-Ávalos R, Kú-Cauich R (2007) Catharanthus biosynthetic enzymes: the road ahead. Phytochemistry Rev. 6: 307-339
Mangold U, Eichel J, Batschauer A, Lanz T, Kaiser T, Spangenberg G, Werckreichhart D, Schroder J (1994) Gene and cDNA for plant cytochrome-P450 proteins (CYP 72 Family) from Catharanthus roseus, and transgenic expression of the gene and a cDNA in tobacco and Arabidopsis thaliana. Plant Sci. 96: 129-136
Memelink J, Kijne J, van der Heijden R, Verpoorte R (2001) Genetic modification of plant
secondary metabolite pathways using transcriptional regulators. In Advances in Biochemical Engineering/Biotechnology, pp 103-125
Memelink J, Swords, K.M.M., Staehelin, L.A., and Hoge, J.H.C. (1994) Southern, Northern and Western blot analysis. in Plant Molecular Biology Manual, S.B. Gelvin, R.A. Schilperoot and D.P.S. Verma, eds. (Dordrecht: Kluwer Academic Publishers): pp. F1-F23
Menke FLH, Champion A, Kijne JW, Memelink J (1999) A novel jasmonate- and elicitor- responsive element in the periwinkle secondary metabolite biosynthetic gene STR interacts with a jasmonate- and elicitor-inducible AP2-domain transcription factor, ORCA2. EMBO J. 18: 4455-4463
Murata J, Roepke J, Gordon H, De Luca V (2008) The leaf epidermome of Catharanthus roseus reveals its biochemical specialization. Plant Cell 20: 524-542
Peebles CAM, Hughes EH, Shanks JV, San KY (2009) Transcriptional response of the terpenoid indole alkaloid pathway to the overexpression of ORCA3 along with jasmonic acid elicitation of Catharanthus roseus hairy roots over time. Metab. Eng. 11: 76-86
Unterlinner B, Lenz R, Toni M (1999) Molecular cloning and functional expression of codeinone reductase: the penultimate enzyme in morphine biosynthesis in the opium poppy Papaver somniferum. Plant J. 18: 465-475
van der Fits L, Memelink J (1997) Comparison of the activities of CaMV 35S and FMV 34S promoter derivatives in Catharanthus roseus cells transiently and stably transformed by particle bombardment. Plant. Mol. Biol. 33: 943-946
van der Fits L, Memelink J (2000) ORCA3, a jasmonate-responsive transcriptional regulator of plant primary and secondary metabolism. Science 289: 295-297
Van der Fits L, Memelink J (2001) The jasmonate-inducible AP2/ERF-domain transcription factor ORCA3 activates gene expression via interaction with a jasmonate-responsive promoter element. Plant J. 25: 43-53
van der Heijden R, Jacobs DI, Snoeijer W, Hallard D, Verpoorte R (2004) The Catharanthus alkaloids: Pharmacognosy and biotechnology. Curr. Med. Chem. 11: 607-628
Vandenabeele S, Van Der Kelen K, Dat J, Gadjev I, Boonefaes T, Morsa S, Rottiers P, Slooten L, Van Montagu M, Zabeau M, Inzĕ D, Van Breusegem F (2003) A comprehensive analysis of hydrogen peroxide-induced gene expression in tobacco. Proc. Natl. Acad. Sci.
USA 100: 16113-16118
Vuylsteke M, Peleman JD, van Eijk MJT (2007) AFLP-based transcript profiling (cDNA-AFLP) for genome-wide expression analysis. Nat. Protoc. 2: 1399-1413
Zhao J, Dixon RA (2010) The ‘ins’ and ‘outs’ of flavonoid transport. Trends Plant Sci. 15: 71-80 Zuo J, Niu Q-W, Chua N-H (2000) An estrogen receptor-based transactivator XVE mediates highly
inducible gene expression in transgenic plants. Plant J. 24: 265-273
