• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plntphysLink to Publisher's site
Plant Physiol. Jan 2006; 140(1): 18–29.
PMCID: PMC1326028

Role of Petal-Specific Orcinol O-Methyltransferases in the Evolution of Rose Scent1

Abstract

Orcinol O-methyltransferase (OOMT) 1 and 2 catalyze the last two steps of the biosynthetic pathway leading to the phenolic methyl ether 3,5-dimethoxytoluene (DMT), the major scent compound of many rose (Rosa x hybrida) varieties. Modern roses are descended from both European and Chinese species, the latter being producers of phenolic methyl ethers but not the former. Here we investigated why phenolic methyl ether production occurs in some but not all rose varieties. In DMT-producing varieties, OOMTs were shown to be localized specifically in the petal, predominanty in the adaxial epidermal cells. In these cells, OOMTs become increasingly associated with membranes during petal development, suggesting that the scent biosynthesis pathway catalyzed by these enzymes may be directly linked to the cells' secretory machinery. OOMT gene sequences were detected in two non-DMT-producing rose species of European origin, but no mRNA transcripts were detected, and these varieties lacked both OOMT protein and enzyme activity. These data indicate that up-regulation of OOMT gene expression may have been a critical step in the evolution of scent production in roses.

The past 10 years have seen rapid progress in flower scent research, the initial breakthrough coming from the pioneering work on (S)-linalool synthase from flowers of Clarkia breweri (Pichersky et al., 1995). The purification of (S)-linalool synthase provided amino acid sequence information that allowed the isolation of the corresponding gene (Dudareva et al., 1996). The characterization of additional genes involved in the biosynthesis of floral volatiles in C. breweri followed, including those encoding enzymes for the synthesis of the phenylpropanoids methyl eugenol and methyl isoeugenol (Wang et al., 1997), benzyl acetate (Dudareva et al., 1998), and methyl salicylate (Ross et al., 1999). More recently, a wealth of information on the biosynthesis of floral volatiles has come from studies of the common garden snapdragon (Antirrhinum majus; Dudareva and Pichersky, 2000; Dudareva et al., 2000, 2003; Tholl et al., 2004). Until very recently, however, C. breweri and snapdragon represented the only plant species in which isolation of enzymes and genes responsible for the formation of scent volatiles in the flower had been accomplished (Dudareva and Pichersky, 2000). Over the past three years, this situation has begun to change, and several investigations have been launched with the aim of understanding the biosynthesis of floral volatiles in other plant models, including Arabidopsis (Arabidopsis thaliana; Aharoni et al., 2003; Chen et al., 2003), petunia (Petunia hybrida; Verdonk et al., 2003, 2005; Boatright et al., 2004), and the rose (Rosa spp.). Genomic approaches have recently been used to analyze the transcriptome of rose petals, leading to the identification of several genes potentially involved in scent production (Channelière et al., 2002; Guterman et al., 2002). Functional characterization of some of these genes led to the identification of a sesquiterpene synthase involved in the production of germacrene D (Guterman et al., 2002), an alcohol acetyltransferase involved in the formation of geranyl acetate (Shalit et al., 2003), and orcinol O-methyltransferases (OOMTs). OOMTs, encoded by two closely related OOMT genes, OOMT1 and OOMT2, catalyze the last two steps of the biosynthetic pathway leading to 3,5-dimethoxytoluene (DMT), the major scent compound in many rose varieties (Lavid et al., 2002; Scalliet et al., 2002). In Rosa chinensis, OOMTs are also able to catalyze the final steps of an alternative pathway leading from phloroglucinol to the trimethylated molecule 1,3,5-trimethoxybenzene (TMB; Scalliet et al., 2002). In addition to OOMTs, TMB biosynthesis involves a phloroglucinol O-methyltransferase, which was characterized recently (Wu et al., 2004).

Nowadays, phenolic methyl ethers are emitted by flowers from most modern rose varieties (Flament et al., 1993). However, the progenitors of modern roses included both European species (e.g. Rosa gallica, Rosa phoenicia, and Rosa moschata) and Chinese species (e.g. R. chinensis and Rosa gigantea), and, interestingly, emission of phenolic methyl ethers was originally restricted to Chinese rose species and was not found in the European species (Nakamura, 1987; Flament et al., 1993; Joichi et al., 2005). In this study we have developed methods to study OOMT expression and activity at the organ, cellular, and subcellular levels in a range of rose varieties. We show that European roses possess OOMT-like sequences in their genomes but do not express OOMT activity in their petals, providing a molecular explication for the absence of methylated phenolic compounds in the scent of these roses.

RESULTS

DMT Production Is Correlated with the Presence of OOMT Protein and Enzyme Activity

To investigate the reasons for the absence of methylated phenolic compound emission by European rose flowers, OOMT activity was measured in petals of two typical European roses, R. gallica and Damask rose (Rosa damascena). R. gallica is a wild European species which, together with two other European species, is a progenitor of Damask, a rose variety used for attar production (Iwata et al., 2000). Both R. gallica and Damask rose possess a typical European scent, rich in monoterpenes and phenylethanol but devoid of phenolic methyl ethers (Flament et al., 1993; Oka et al., 1999). In contrast to R. chinensis cv Old Blush, no significant OOMT activity could be detected when R. gallica and Damask rose cell-free petal extracts were incubated with orcinol and S-adenosyl-l-[methyl-14C]Met (Fig. 1). These data suggest that the European species do not synthesize DMT because they lack active OOMT enzymes.

Figure 1.
Absence of OOMT activity in European roses. TLC analysis of OOMT activity in European and Chinese rose petal extracts (20 μg of total protein), incubated with orcinol (1 mm) in the presence of S-adenosyl-l-[methyl-14C]Met (50 μm). R.g., ...

To investigate this further, a polyclonal antibody was raised against the OOMT1 protein from the rose variety Old Blush, expressed in Escherichia coli as a glutathione S-transferase-fusion protein, as described by Scalliet et al. (2002). After cleavage of the glutathione S-transferase moiety, purified recombinant OOMT1 was used to immunize rabbits. Note that this antibody recognizes both OOMT1 and the closely related OOMT2 protein, but not the more distantly related caffeic acid O-methyltransferase (accession no. AJ439740; data not shown).

We first used the anti-OOMT antibody to determine the abundance of OOMT proteins in different tissues from R. chinensis cv Old Blush, including petals from flowers at different developmental stages (stages 3, 4, and 5, according to Guterman et al., 2002), three other floral organs (sepals, stamens, and pistils) from flowers at stage 3, and young leaves collected from growing branches. Western-blot analysis of these samples showed that OOMTs accumulated specifically in petals (Fig. 2A). OOMT accumulation was developmentally regulated, reaching a maximum in the open flower (stage 4). OOMTs were not detected in other parts of the flower (pistil, stamen, and sepals) or in leaves. This petal-specific pattern of accumulation of OOMTs was consistent with previous observations of OOMT transcript abundance, with the exception that significant levels of OOMT mRNA had also been detected in Old Blush stamens (Scalliet et al., 2002).

Figure 2.
Western-blot analysis of OOMT expression. A, Western-blot analysis of OOMT expression in different organs of R. chinensis cv Old Blush. L, Leaves; Se, sepals; P3, petals at stage 3; P4, petals at stage 4; P5, petals at stage 5; St, stamens; Pi, pistils. ...

Having established the pattern of accumulation of OOMT protein in the TMB-producing variety Old Blush, we used the same antibody to assay for OOMT protein in the European species R. gallica and Damask rose. No signal corresponding to putative OOMT proteins was detected following western-blot analysis of protein extracts from a range of different tissues including petals at stage 4, where the highest levels of protein were detected in the Old Blush samples (Fig. 2B). The absence of phenolic methyl ethers in the scent of the European species R. gallica and Damask rose was therefore correlated with an absence of both OOMT protein and enzyme activity.

OOMTs Immunolocalize Predominantly to the Rose Petal Epidermis

Rose petal epidermal cells are believed to be the major site of scent production in roses (Stubbs and Francis, 1971; Loomis and Croteau, 1973). However, there is no direct evidence that scent biosynthesis enzymes are located in these cells. The experiments described above showed that OOMT proteins were accumulated specifically in petals of R. chinensis cv Old Blush. To determine the tissue localization of phenolic methyl ethers production, OOMT distribution in rose petals was investigated by immunolocalization. Immunolocalization experiments were carried out using the modern Tea rose Rosa x hybrida cv Lady Hillingdon because this rose has been shown to produce large amounts of DMT (Nakamura, 1987; Joichi et al., 2005) and its petals proved to be well suited to microscopy techniques. Cross sections of petals from Lady Hillingdon were incubated with the anti-OOMT antibody and OOMTs were visualized using secondary antibodies coupled to alkaline phosphatase and nitroblue tetrazolium (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP) as substrate. This analysis indicated that OOMTs were localized principally in the conical cells that constitute the upper epidermis of the petal and, to a lesser extent, in the subepidermal cell layer (Fig. 3, A and B). Weaker staining was also detected in the lower epidermis, indicating the presence of a lower level of OOMTs in these cells. No signal was observed when the petal sections were incubated with preimmune serum from the same rabbit (Fig. 3C).

Figure 3.
Tissue and cellular localization of OOMT proteins. A to C, Immunolocalization of OOMTs in Lady Hillingdon petals. A, Cross sections of young petals (stage 3) were incubated with anti-OOMT polyclonal antibody. B, Higher magnification of adaxial epidermis ...

The observation that OOMTs are located predominantly in the petal epidermis supports the hypothesis that this cell layer plays an important role in scent production. Moreover, the identification of an enzyme involved in scent production in these cell types opens up the possibility of investigating scent biosynthesis at the cellular level. As a first step toward this objective, we developed an in vivo imaging strategy based on the use of green fluorescent protein (GFP) fusion proteins to investigate the subcellular localization of OOMTs in rose petal epidermal cells.

Biolistic Transformation of Rose Petal Secretory Cells and Subcellular Localization of OOMT-GFP Fusion Proteins

Remarkable work using plant secretory cell model systems, such as peppermint (Mentha piperita) glandular trichomes, has provided a greatly improved understanding of the metabolism of monoterpenes (Mahmoud and Croteau, 2002). However, because of the dispersed distribution of secretory structures on the organ surface, currently used secretory model systems are not well suited to a combined use of transient expression from gene gun delivered constructs and in vivo imaging techniques. On the other hand, biolistic transformation of petal cells has been limited to approaches such as promoter characterization (Clark and Sims, 1994; Quattrocchio et al., 1999).

The rose petal epidermis provides an ideal target for biolistic delivery of transgene constructs because it consists of a homogenous layer of secretory cells. We found that petal epidermal cells of several rose varieties, including Old Blush and Lady Hillingdon, could be used for this type of experiment. However, petals of the cut flower variety R. x hybrida cv Anna (PEKcougel) gave the best results because flowers can be obtained regularly throughout the year under greenhouse conditions and they have a long vase life of about 9 d due to their very resistant petals. Moreover, petals of this variety emit large amounts of DMT (data not shown).

To provide a framework for the analysis of gene expression in petal epidermal cells, the internal structure of these cells was determined by transmission electron microscopy of tangential cross sections of Anna petals. Figure 3D shows a cross section in the apical region of a petal secretory cell. The cell is surrounded by a thick cell wall, covered by a waxy cuticle, with undulations that are due to the characteristic striated aspect of these cells (Stubbs and Francis, 1971). The dense cytoplasm contains starch-rich plastids and many vacuolar structures. These vacuoles contain variable amounts of electron-opaque material, as previously observed in petal epidermal cells of other flowers (Weston and Pyke, 1999).

Transformed epidermal cells accumulated large amounts of GFP or GFP fusion proteins, leading to an intense fluorescence 24 h (for GFP alone) or 48 h (for GFP fusion proteins) after transformation. Epidermal cells of excised rose petals remained metabolically active and were able to synthesize chimeric proteins over a period of several days. This is consistent with previous studies, which have shown that rose flowers can maintain scent biosynthesis without an external carbohydrate supply (Helsper et al., 1998). Metabolism in excised rose petals presumably utilizes the large stock of starch that accumulates during the early steps of petal development (Fig. 3D).

Following biolistic transformation, the cellular localizations of GFP fusion proteins were studied using confocal laser scanning microscopy. Initially, a number of markers were used to characterize different compartments of the rose petal secretory cell (Fig. 3, E–I). Expression of nontargeted GFP in Anna petal epidermis cells confirmed that these cells are extremely vacuolized, with a central vacuole surrounded by numerous vacuole-like compartments (Fig. 3, E–G). Plastids were visualized using plastid-targeted GFP (Fig. 3H), and the expression of m-GFP5-HDEL (Haseloff et al., 1997) gave the reticulated pattern of GFP localization typical of endoplasmic reticulum (ER) in plant cells (Fig. 3I; Boevink et al., 1998).

To investigate the subcellular localization of OOMT, constructs encoding GFP fused either to the N terminus (GFP-OOMT) or to the C terminus (OOMT-GFP) of OOMT1 from Old Blush were introduced into petal epidermal cells by biolistic transformation. Anna petal cells expressing GFP-OOMT fusion protein showed a fluorescence pattern very similar to that obtained with nontargeted GFP, indicating that GFP-OOMT behaved as a soluble cytosolic protein (Fig. 3J). In contrast, OOMT-GFP fusion protein was restricted to particular regions of the cell, giving rise to intense fluorescent signals (Fig. 3K). The same kind of pattern was obtained when OOMT-GFP was expressed in petal epidermis cells of other rose varieties, such as Lady Hillingdon (Fig. 3L) or Old Blush (data not shown). These intensely fluorescing regions were often localized in the apical region of the cells (Fig. 3L) but were also found sometimes in the basal region (Fig. 3K). Based on the predominantly membrane-associated localization of the endogenous OOMT enzymes at this stage of development (see next paragraph), we propose that the localization of the OOMT-GFP fusion protein is most likely to reflect that of the endogenous protein. This highly localized distribution was of particular interest because OOMTs catalyze the last steps of DMT biosynthesis and these enzymes might therefore be expected to be associated with a putative structure that would mediate DMT secretion.

A Fraction of the OOMT Protein Is Membrane Bound in Rose Petal Cells

We were not able to investigate the subcellular localization of the GFP fusion proteins in rose petals cells using ultracentrifugation experiments because of the limited number of transformed cells obtained using biolistic techniques. However, ultracentrifugation experiments were used to investigate the localization of the endogenous OOMTs. Rose petals were collected at different developmental stages (stages 3, 4, and 5) and cell-free extracts of these petals were subjected to ultracentrifugation (150,000g for 1 h). Supernatants and pellets were analyzed by western blot, using anti-OOMT antibodies (Fig. 4). Rose OOMTs were detected in both the supernatant and in the pellet fractions after centrifugation at 150,000g, with the relative amounts of the soluble and the membrane-bound forms varying depending on the developmental stage. In young petals (stage 3) the majority of OOMT protein was detected in the 150,000g supernatant, a weaker signal being detected in the pellet. The proportion of membrane-bound OOMT increased in petals from flowers at stage 4 and finally reached 100% at stage 5, no soluble OOMT being detected at this stage (Fig. 4A). To test how tightly OOMTs were associated with membranes, microsomes were prepared from petals sampled at stage 4 and incubated in the presence of either 2 m NaCl, 0.1% Triton X-100, 0.1 m Na2CO3, 0.1 m NaOH, or 6.8 m urea (Fig. 4B). OOMTs were released from the membranes by alkaline treatments such as Na2CO3 and NaOH incubations, but not by any of the other reagents, indicating that the microsome-associated OOMTs were firmly bound to membranes. To summarize, native OOMTs in rose petals were detected in both the soluble fraction and tightly bound to microsomes. The membrane-bound fraction was present throughout petal development but increased in relative abundance during maturation. Based on these observations, it is probable that the highly localized expression pattern of the OOMT-GFP protein in rose petal cells was due to its being associated with a specific membrane compartment of the cell.

Figure 4.
Association of OOMT with rose petal microsomes. A, R. x hybrida cv Anna petals were sampled at development stages 3, 4, and 5. Cell-free extracts (C) from these petals were centrifuged at 150,000g for 1 h and supernatants (S) and pellets (P) were analyzed ...

OOMT-Like Genes Are Present, But Are Not Expressed, in European Roses, Which Do Not Produce Phenolic Methyl Ethers

The above experiments provided a description of OOMT expression at the organ and cellular levels in DMT-producing roses and indicated that European species did not emit DMT because they did not express OOMT enzyme activity. In the genus Clarkia, there is evidence that evolution of the ability to emit (S)-linalool involved up-regulation of an (S)-linalool synthase gene that is also present in unscented Clarkia species (Dudareva et al., 1996). We were therefore interested to determine whether the European rose species possessed OOMT gene sequences in their genomes. PCR amplifications were carried out using genomic DNA from a range of rose species and varieties and oligonucleotides based on Old Blush OOMT sequences. In addition to R. gallica and Damask rose, we also selected the wild Chinese species R. gigantea, which, like R. chinensis, is another progenitor of modern roses, for this study. The flowers of R. gigantea emit almost exclusively DMT, like the flowers of the modern Tea rose R. x hybrida cv Lady Hillingdon, which was also included in this study (Joichi et al., 2005). DNA fragments of 1.3 kb could be amplified from the DNA of all the rose varieties tested (Fig. 5). These fragments were cloned and 10 PCR products were sequenced for each variety. For R. chinensis cv Old Blush, all sequenced clones corresponded either to the OOMT1 cDNA (accession no. AJ439741; Scalliet et al., 2002) or to the OOMT2 cDNAs (accession no. AJ439742), and harbored a 220-bp insertion, which possessed all the characteristics of an intron. Similarly, two sequenced clones amplified from R. x hybrida cv Lady Hillingdon corresponded to the previously described OOMT1-related gene, OOMT3 (accession no. AJ439743), with minor differences probably due to allelic variations. Two other genes amplified from Lady Hillingdon were similar, but not identical, to the previously described OOMT4 (accession no. AJ439744; 97% and 97.3% sequence identity, respectively). These sequences also possessed a 220-bp intron. Four different putative OOMT genes were cloned from R. gigantea, all very similar to Old Blush OOMT genes. Interestingly, the genomic DNA fragments amplified from the European species R. gallica and Damask rose also showed high similarity to R. chinensis OOMT genes. Three different types of amplified genomic fragments were cloned from Damask rose, two of which were extremely similar to R. chinensis OOMT genes (98.9% and 97.5% sequence identity with OOMT1, respectively), with the third being more distantly related (92.5% sequence identity). Two different types of amplified genomic fragments were cloned from R. gallica, both very similar to R. chinensis OOMT genes (98.9% and 97.6% sequence identity with OOMT1, respectively). Taken together, these data showed that European roses, such as R. gallica and Damask rose, possess sequences that are highly similar to previously characterized OOMT genes despite the fact that no OOMT activity could be detected in their petals. The deduced protein sequences were very similar to R. chinensis OOMTs (Fig. 6), and analysis of the alignment of these putative rose OOMTs with R. chinensis OOMTs indicated that these predicted proteins contain all the conserved sequences necessary for OMT activity. In phenolic methyl ether-producing rose varieties, OOMT genes have been shown to be strongly expressed in petals (Lavid et al., 2002; Scalliet et al., 2002). However, when reverse transcription (RT)-PCR was used to screen for OOMT transcripts in a range of floral organs (sepals, petals, stamens, pistils) and in young leaves of R. gallica and Damask rose, no OOMT expression could be detected in any of the tested tissues, despite repeated tests using the same OOMT-specific primers as were used to amplify the genomic sequences (Fig. 7).

Figure 5.
Presence of OOMT-like genes in European roses. Agarose gel analysis of the products of PCR reactions using OOMT-specific primers and genomic DNA from different rose varieties as a template. R.c., R. chinensis cv Old Blush; R.d., Damask rose; R.g., R. ...
Figure 6.
Alignment of proteins deduced from OOMT genes. The complete amino acid sequence of OOMT1 from Old Blush (OB OOMT1, accession no. AJ439741) is shown, and only the variant amino ...
Figure 7.
RT-PCR analysis of OOMT gene expression in different organs of Damask rose, R. gallica, and R. chinensis cv Old Blush. Shown is agarose gel analysis of the RT-PCR reaction products using OOMT-specific primers (top section) and RNA extracted from leaves ...

To determine whether the OOMT homologs from R. gallica potentially encode functional proteins, the OOMT2A and OOMT2B sequences from this species (see Fig. 6) were expressed in tobacco (Nicotiana tabacum) using Agrobacterium-mediated transient transformation (Batoko et al., 2000). As controls, the plasmids encoding GFP fused either to the N terminus or C terminus of OOMT1 from Old Blush (the OOMT-GFP and GFP-OOMT constructs in Fig. 3, J–L) were also transformed into tobacco leaves (Fig. 8A). Control extract, from a leaf expressing GFP alone, exhibited a background level of OOMT activity, indicating that tobacco leaves possess OMTs, which are able to efficiently methylate orcinol to produce 3-methoxy 5-hydroxytoluene (MHT). However, OOMT activity was significantly higher in leaf sectors expressing GFP-OOMT or OOMT-GFP, indicating that the OOMT moieties of both fusion proteins were properly folded and active. As no cDNA was available, the OOMT2A and OOMT2B genes of R. gallica were cloned into the pK7FWG2 vector (Karimi et al., 2002) and transformed into tobacco. Western-blot analysis of infiltrated leaves showed that the 200-bp intron of both OOMT2A and OOMT2B genes was properly spliced and that the OOMT2A-GFP and OOMT2B-GFP fusion proteins were produced as efficiently as the OOMT-GFP control protein (Fig. 8B). The OOMT activity in cell-free extracts of tobacco leaves sectors expressing OOMT2A-GFP and OOMT2B-GFP fusion proteins was higher than in the control extract (leaf sector expressing GFP only), indicating that OOMT2A and OOMT2B from R. gallica possess OOMT activity, although the R. gallica enzymes did not methylate orcinol as efficiently as OOMT1 from Old Blush (Fig. 8C).

Figure 8.
Characterization of OOMT-GFP fusion proteins expressed in tobacco. A, TLC analysis of OOMT activity in tobacco leaves transiently expressing GFP and GFP fusion proteins. Tobacco leaf sectors expressing GFP, or GFP fusion proteins, were excised 48 h after ...

Together, these experiments showed that, although European roses such as R. gallica and Damask rose possess OOMT homologs encoding potentially active proteins, these genes are not expressed in flower tissues and this would therefore explain the absence of DMT production in these species.

DISCUSSION

OOMTs Accumulate Specifically in Petals and Predominantly in the Conical Cells of the Adaxial Epidermis

In most roses, petals are the major site of scent biosynthesis, although in some cases volatile compounds are also produced in stamen and sepals (Dobson et al., 1990). In previous studies, transcripts of rose OOMT genes were detected in petals and, to a lesser extent, in stamens, but were not detected in other parts of the flower or in leaves (Lavid et al., 2002; Scalliet et al., 2002). Using a polyclonal antibody raised against recombinant OOMT1, we show here that OOMTs accumulate specifically in petals. No OOMT protein was detected in stamens, despite the presence of transcripts in this organ, indicating that OOMT expression is repressed in stamens by a posttranscriptional mechanism. In petals, the accumulation of OOMT proteins followed the previously described pattern of transcript accumulation, and enzyme levels reached a maximum in the freshly open flower (stage 4), when scent biosynthesis is most active (Lavid et al., 2002).

The adaxial epidermis of rose petals is composed of specialized, cone-shaped secretory cells, and it has been suggested, based on cytological and biochemical analyses, that the epidermis functions as a layer of secretory cells responsible for the biosynthesis and emission of scent compounds (Stubbs and Francis, 1971; Loomis and Croteau, 1973). Immunolocalization experiments (Fig. 3, A–C) showed that the majority of the OOMT protein was concentrated in the adaxial epidermis in petals of R. x hybrida cv Lady Hillingdon, providing strong support for this hypothesis. A lower level of OOMT protein also accumulated in the abaxial epidermis, indicating a probable differential role for the two petal epidermal layers, the adaxial epidermis being the major site for scent production. A similar localization pattern has been described for another enzyme involved in flower scent biosynthesis, S-adenosyl-l-Met:benzoic acid carboxyl methyltransferase. S-adenosyl-l-Met:benzoic acid carboxyl methyltransferase protein was detected predominantly in the conical cells of the inner epidermal layer and, to a lesser extent, in the cells of the outer epidermis of snapdragon flower petal lobes, providing strong support for the commonly held belief that petal scent biosynthesis occurs in the epidermis (Kolosova et al., 2001).

Taken together with previous studies, these data indicate that the cone-shaped cells covering the petals of many flowers play a major role in attracting pollinators by using both visual and olfactory cues. From a visual point of view, the conical shape of the epidermal cells enhances light absorption by the floral pigments and, thus, the intensity of their color (Gorton and Vogelmann, 1996). This study and that of Kolosova et al. (2001) indicate that the cone-shaped epidermal cells also play a major role in the production of olfactory cues, harboring high concentrations of enzymes involved in scent biosynthesis pathways.

Petal Epidermal Cells as a Secretory Cell Model

To date, little is known about the intracellular biosynthesis of volatile compounds in secretory cells and the trafficking of these compounds from their sites of synthesis to their site(s) of emission (Kolosova et al., 2001). Peppermint glandular trichomes have been successfully used as secretory cell model system to investigate the metabolism of monoterpenes (Mahmoud and Croteau, 2002), and peppermint trichomes have been well characterized at the structural level (Turner et al., 2000). The immunolocalization of several monoterpene biosynthetic enzymes (Turner and Croteau, 2004) showed that monoterpene biosynthesis may involve the trafficking of intermediate products between different cell compartments; however, the cellular events leading to the biosynthesis and secretion of monoterpenes remain poorly understood. In recent years, the use of GFP as an in vivo reporter has emerged as an extremely valuable tool in plant cell biology (Hawes et al., 2001). However, due to the lack of a suitable model system, plant secretory cells have not been investigated using this technology. Rose petal epidermal cells present a distinct advantage for this type of approach, compared with the dispersed glandular trichomes from peppermint, for example, because they constitute a homogenous layer, covering the whole surface of the petal and are therefore particularly well suited for biolistic delivery of gene constructs. Moreover, characterization of the internal structure of the rose petal secretory cell, using GFP-fusion proteins targeted to different cellular compartments, indicated a similar general structure to more commonly studied cellular models such as tobacco or onion (Allium cepa) epidermis cells (Boevink et al., 1998; Scott et al., 1999; Brandizzi et al., 2004), providing the possibility of comparative studies of secretory and nonsecretory cell types. Biolistic transformation can be applied to the petals from other plant species (Clark and Sims, 1994; Quattrocchio et al., 1999), and the results presented in this study therefore underline the potential of petal epidermal cells as a secretory cell model system.

Association of Rose Petal OOMTs with Cell Membranes

There is considerable evidence supporting a role for the ER in the biosynthesis of secondary metabolites. For example, immunolocalization experiments have shown that cytochrome P450 limonene 6-hydroxylases was bound to smooth ER in spearmint (Mentha spicata; Turner and Croteau, 2004). In alfalfa (Medicago sativa), biosynthesis of the phytoalexin medicarpin involves the hydroxylation of liquiritigenin by the ER-associated cytochrome P450 isoflavone synthase and the subsequent methylation of one hydroxyl group by isoflavone O-methyltransferase (IOMT). After elicitation of the isoflavonoid pathway, IOMT localizes to the ER, where it probably interacts with isoflavone synthase to ensure efficient metabolic channeling and hence optimize production of medicarpin (Liu and Dixon, 2001).

In rose petals, both biochemical fractionation and confocal microscopy studies indicated that a significant fraction of the OOMT protein was strongly bound to membranes throughout organ development, with the proportion of membrane-bound OOMT increasing as the petal matured (Fig. 4). When an OOMT-GFP fusion protein was expressed in these cells, fluorescence was associated with compact structures, which were often, but not always, localized at the apical pole of the cells. In contrast to the situation observed for alfalfa IOMT after elicitation (Liu and Dixon, 2001), the fluorescence pattern observed after expression of OOMT-GFP fusion protein was very different from the pattern obtained with ER-targeted GFP-HDEL (Fig. 3I), indicating that the OOMTs were not associated with the ER. The localization of OOMT-GFP to compact cellular structures appears to involve the N terminus of the protein, as the N-terminal GFP-OOMT fusion protein was not concentrated in the same manner but was dispersed freely in the cytosol. Expression of GFP-OOMT and OOMT-GFP in tobacco, using Agrobacterium-mediated transient expression, showed that both fusion proteins retained OOMT activity (Fig. 8A), indicating that their OOMT moieties were properly folded and suggesting that the fluorescence pattern observed in rose petals did not correspond to artefactual aggregates, but rather to a physiologically relevant localization of the OOMT protein. It is unlikely that the OOMTs were targeted to organelles such as vacuoles because they lack any predicted signal peptide, according to programs such as TargetP (Emmanuelsson et al., 2000). Furthermore, membrane-bound and soluble forms of OOMTs exhibited the same electrophoretic mobility, so it is unlikely that they possess a cleaved signal peptide. This also suggests that it is unlikely that membrane association involves the attachment of a lipid moiety to the protein as observed for chalcone isomerase (Burbulis and Winkel-Shirley, 1999). One possibility is that OOMTs were bound to the cytosolic side of the membranes of vacuolar compartments. Indeed, the structures revealed by OOMT-GFP fluorescence could be related to storage vacuoles, which have been shown to be distinct from the central lytic vacuole (Paris et al., 1996). Further subcellular markers (Di Sansebastiano et al., 1998) will be used to address this question.

It is intriguing that the proportion of membrane-associated OOMT increased during rose petal maturation, and this may represent a mechanism for the regulation of enzyme activity. For this reason, we were interested in determining whether OOMT1 and OOMT2 showed the same pattern of membrane association during development. Anti-OOMT antibodies recognized both OOMT1 and OOMT2 and did not allow discrimination between these two proteins. However, biochemical analysis, using the different substrate specificities of OOMT1 and OOMT2 to distinguish between them (Lavid et al., 2002; Scalliet et al., 2002), indicated that both soluble and membrane-bound OOMT fractions exhibited the same intermediate substrate specificity, suggesting that neither OOMT1 nor OOMT2 associated preferentially with membranes (data not shown). Future work will be aimed at investigating the physiological significance of the association of OOMTs with these membrane structures. In particular, it will be important to identify the cell structures that OOMTs are bound to, for example, using immunogold staining and electron microscopy techniques, such as those used successfully for other scent-related enzymes (Kolosova et al., 2001).

Why Don't All Roses Produce DMT?

Most modern rose varieties produce DMT as a component of their scent, having inherited this trait from their Chinese progenitors (Nakamura, 1987; Flament et al., 1993; Joichi et al., 2005). However, while DMT production is almost universal in modern rose varieties, this is not true for older European varieties, which do not produce phenolic methyl ethers (Flament et al., 1993). Investigating the reasons for this absence in two European roses, R. gallica and Damask rose, we found that neither OOMT activity nor OOMT protein could be detected in the floral organs of these varieties. The genomes of both R. gallica and Damask rose contain OOMT homologous sequences, but these genes are not expressed at detectable levels in either floral organs or leaves. There are many other examples of related plant species exhibiting very different scent characteristics (Dudareva et al., 1996; Gang et al., 2002; Guterman et al., 2002), and in some cases this has also been shown to be a result of changes in gene expression. For example, the scented C. breweri is believed to have evolved recently from the nonscented Clarkia concinna (Dudareva et al., 1996). In C. breweri, the gene encoding linalool synthase (Lis) is highly expressed in stigma and in the epidermal cells of petals, as well as in stamens, whereas in the nonscented C. concinna, Lis is expressed only in the stigma and at a relatively low level (Dudareva et al., 1996). Thus, it appears that C. breweri has evolved its ability to emit large amounts of S-linalool simply by increasing the expression level of Lis in the epidermal cells of the petals.

An interesting question is the following: Did the common ancestor of the European and Chinese rose species produce phenolic methyl ethers, with this ability being subsequently lost by European species, or did Chinese roses acquire the ability to carry out this metabolism after their separation from European species? Phylogenetic analysis of the genus Rosa shows that the Chinese roses, grouped under the Indicae section, may have evolved recently within this genus (Wissemann and Ritz, 2005), supporting the idea that new functions may have evolved in this group. For example, Chinese roses may have evolved their ability to produce phenolic methyl ethers as a result of a combination of increased OOMT gene expression in petals and evolution of more catalytically active enzymes. In this hypothesis, the up-regulation of OOMT gene expression may have been a critical step in the evolution of scent production in roses; however, a more detailed analysis of the OOMT gene family will be needed to definitively address the question of the origin of the biosynthesis of phenolic methyl ethers in the genus Rosa.

MATERIALS AND METHODS

Plant Materials

Two main types of roses were used for this study: on one hand, the Chinese roses Rosa chinensis cv Old Blush and Rosa gigantea, and the modern varieties Rosa x hybrida cv Lady Hillingdon and R. x hybrida cv Anna (PEKcougel), which produce phenolic methyl ethers; and on the other hand, the European species Rosa gallica officinalis and Damask rose (Rosa damascena), which do not produce these scent compounds. Old Blush, R. gallica officinalis, and Damask rose were grown in open soil conditions at the Ecole Normale Supérieure de Lyon. R. gigantea and Lady Hillingdon were from the Lyon Botanical Garden. Anna flowers were obtained from the local producer Hortirose and grown under greenhouse conditions. Flower development was divided into six stages, according to Guterman et al. (2002). At stage 1, flower buds are closed. At stage 2, petals start to emerge from the sepals. Stages 3 and 4 are characterized by petal elongation. At stage 5 and 6, petals unroll to reach full size.

Chemicals and Radiochemicals

MHT was prepared according to Scalliet et al. (2002). S-adenosyl-l-[methyl-14C]Met (55 mCi/mmol) was from Amersham. All other chemicals and reagents were from Sigma.

Preparation of Cell-Free Extracts and Microsomes

Rose petals or tobacco (Nicotiana tabacum) leaves were homogenized in buffer A (0.1 m Tris, pH 7.5, containing 20% glycerol [v/v], 5 mm MgCl2, 10 mm NaF, 14 mm 2-mercaptoethanol, and 1% phenylmethylsulfonyl fluoride) with 1% (w/w) polyvinylpolypyrrolidone, using 4 mL of buffer A per gram fresh weight. The homogenate was filtered through glass wool and centrifuged at 5,000g for 15 min at 4°C. The supernatant was used for enzyme assays. For microsome preparation, the same 5,000g supernatant was centrifuged at 150,000g for 1 h at 4°C. The resultant pellet was washed twice with buffer A, and then resuspended in the same buffer.

Measurement of Enzyme Activity

Cell-free rose petal or tobacco leaf extracts were incubated in a final volume of 50 μL with 50 μm S-adenosyl-l-[methyl-14C]Met and 1 mm of orcinol in buffer A, and the incorporated radioactivity was measured by liquid scintillation. Reaction products were analyzed by thin-layer chromatography (TLC) on silica gel (Merck) with chloroform as the solvent, using a Storm 860 phosphoimager (Molecular Dynamics). Enzyme reaction products were identified by comigrating with standards.

Protein Purification and Production of Antibodies

OOMT1 coding regions were cloned into pGEX-4T1. Recombinant OOMT1 protein was expressed, purified, and recovered by cleavage with thrombin as by Scalliet et al. (2002). Polyclonal antibodies, generated in rabbits against the purified recombinant OOMT1, were prepared by Eurogentec.

SDS-PAGE and Western-Blot Analysis

Based on the petal homogenate volume, equivalent amounts of soluble and microsomal proteins were resolved on 10% Tris-Gly gels using a Mini Protean II gel apparatus (Bio-Rad) and transferred to nitrocellulose membranes using a Bio-Rad Trans-Blot apparatus. The membranes were blocked with 2% (w/v) bovine serum albumin (BSA) in Tris-buffered saline (TBS) and probed with a primary antibody against recombinant OOMT1 at a dilution of 1:1,000 in 2% (w/v) BSA in TBS containing 1% (v/v) Tween 20. Goat anti-rabbit IgG horseradish peroxidase conjugate was used as the secondary antibody at a dilution of 1:5,000 in TBS containing 1% (v/v) Tween 20, with visualization using a chemiluminescence assay kit (ECL; Amersham Biosciences).

Plant Vectors

For expression of the N-terminal GFP fusion protein (GFP-OOMT), OOMT coding sequences were amplified by PCR using the upstream primer 5′-AAAAAGCAGGCTATGGAAAGGCTAAACAGCTTTAGACA-3′ and the downstream primer 5′-AGAAAGCTGGGTCAGGATAAACCTCAATGAGAGACCTTAA-3′. For expression of the C-terminal GFP fusion protein (OOMT-GFP), OOMT coding sequences were amplified by PCR using the upstream primer 5′-AAAAAGCAGGCTCCATGGAAAGGCTAAACAGCTTTAGACA-3′ and the downstream primer 5′-AGAAAGCTGGGTTCAAGGATAAACCTCAATGAGAGACC-3′. Amplified DNA fragments were introduced into pDONR 201 (Invitrogen) and then into pK7WGF2 (GFP-OOMT) and pK7FWG2 (OOMT-GFP; Karimi et al., 2002) using GATEWAY cloning technology (Invitrogen), according to the supplier's instructions. Alternatively, for a higher expression efficiency of the OOMT-GFP fusion protein in rose petals, OOMT1 coding sequence was cloned into the NcoI site of the vector pCATS-GFP, which contains a codon-optimized GFP S65C mutant gene (Heim et al., 1995) under the control of the cauliflower mosaic virus 35S promoter and an additional translation enhancing element of Tobacco etch virus. OOMT1 coding sequence was amplified by PCR using the upstream primer 5′-GCGTCATGATAAGGCTAAACAGCTTTAAACACCTTAAC-3′ and the downstream primer 5′-CGCTCATGATAGGATAAACCTCAATGAGAGACCTTAAA-3′; the amplified DNA fragment was cleaved with BspHI and cloned into the NcoI site of the pCATS-GFP vector.

Transient Expression of GFP Fusions

For biolistic transformation, plasmid DNA (5 μg) was mixed with 50 μL of an aqueous suspension containing 7.5 mg of 1.0 μm gold particles (Bio-Rad). The gold DNA suspension was dispersed by vortexing in the presence of 1.25 m CaCl2 and 17 mm spermidine and kept on ice for 10 min. The DNA-coated gold particles then were collected by brief centrifugation, washed, resuspended in ethanol, and spread onto carrier discs for biolistic bombardment using the particle delivery system 1000/He (Bio-Rad). Young (stage 3) rose petals were excised and placed on moist filter paper in petri dishes. The gold particles were fired at 1,100 p.s.i., and bombarded petals were maintained in the dark at 22°C before examination. For Agrobacterium-mediated transient expression, each expression vector was introduced into Agrobacterium tumefaciens strain C58 (pMP90) by electroporation. Tobacco SR1 (cv Petit Havana) leaves were infiltrated with A. tumefaciens cultures (OD600 0.1) according to Batoko et al. (2000). Disks were punched from tobacco leaves 48 h after Agrobacterium infiltration and cell-free extracts were assayed for OOMT activity as described above.

Microscopy Techniques

Rose petals were examined 12 to 48 h after biolistic transformation using a Zeiss LSM 510 confocal imaging system attached to a Zeiss Axioplan 2 microscope (Carl Zeiss). GFP was visualized with a ×40 water immersion apochromat objective (Zeiss) by excitation with the 488 line of a krypton/argon laser and use of a BP 505-550 emission filter. Serial optical sections were obtained at 1-μm intervals, and projections of optical sections were accomplished using LSM image-processing software (Zeiss). For electron microscopy observations, petals were fixed in 1.5% glutaraldehyde in 0.1 m sodium cacodylate buffer (pH 7.2) and post fixed in 1% OsO4. Petals were then embedded in Spurr's resin. Paradermal ultrathin sections were prepared with an RMC MT 6000 ultramicrotome (RMC). The sections were mounted on 100 mesh grids coated with formvar film, stained with 1% uranyl acetate and 0.3% lead citrate, and observed using a Hitachi H-800 transmission electron microscope. For immunolocalization, rose petals were fixed overnight at 4°C in 3.7% formaldehyde, 5% acetic, 50% ethanol. Post fixation, dehydration, clearing, and wax embedding were performed according to the protocol of Jackson (1991). Petal cross sections (10 μm thickness) were first incubated for 1 h at room temperature in TBS containing 0.3% (v/v) Tween 20 and 2% (w/v) BSA, followed directly by incubation either with anti-OOMT1 polyclonal antibodies (1:300 dilution in the same buffer) or with the same dilution of preimmune serum. Goat anti-rabbit IgG alkaline phosphatase conjugate was used as the secondary antibody at a dilution of 1:1,000 in TBS containing 0.3% Tween 20, with visualization using NBT/BCIP.

Cloning of OOMT-Like Genes

Rose genomic DNA was prepared according to Delichère et al. (1999). OOMT genes were amplified by PCR using the upstream primer 5′-ATGGAAAGGCTAAACAGCTTTAGACACCTTAAC-3′ and the downstream primer 5′-TCAAGGATAAACCTCAATGAGAGACCTTAAACC-3′. PCR amplification was carried out for 30 cycles of denaturation at 94°C for 30 s, annealing at 52°C for 30 s, and extension at 72°C for 2 min with a final extension of 5 min, in a GeneAmp PCR system 9700 cycler (Perkin Elmer). Amplified DNA fragments were cloned into pGEM-T Easy (Promega) and the inserts sequenced. Multiple sequence alignments were constructed using ClustalW (Thompson et al., 1994). RT-PCR analyses of OOMT genes expression were performed as described by Channelière et al. (2002), using ExTaq (Takara) and the OOMT-specific primers as above. Glyceraldehyde-3-P dehydrogenase (accession no. BI977235) was amplified as a control using the upstream primer 5′-ATCCATTCATCACCACCGACTACA-3′ and the downstream primer 5′-GCATCCTTACTTGGGGCAGAGA-3′.

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers AJ786302 to AJ786316.

Acknowledgments

We would like to thank Charles Broizat (Hortirose), Alexis Lacroix, Isabelle Desbouchages, Armand Guillermin (Ecole Normale Supérieure de Lyon), and Christophe Ferry (Jardin Botanique de la Ville de Lyon, Parc de la Tête d'Or) for help with plant material. We thank Mohammed Bendahmane (Ecole Normale Supérieure de Lyon) and Guido Jach (Max-Planck Institut für Züchtungsforschung, Cologne, Germany) for the pCATS-GFP vector. We thank Danièle Marty-Mazars (Université de Bourgogne, Dijon), Mickaël Michel, Fabienne Simian-Lermé (Ecole Normale Supérieure de Lyon), and Isabelle Anselme-Bertrand (Centre de Microscopie Electronique Stéphanois) for help with microscopy techniques. We thank Anne-Marie Thierry for assistance with antibody preparation, Hervé Leyral and Claudia Bardoux (Ecole Normale Supérieure de Lyon) for technical assistance, and Nadine Paris (Centre National de la Recherche Scientifique-Unité Mixte de Recherche 6037, Université de Rouen) for helpful discussions.

Notes

1This work was supported by the Région Rhône-Alpes (France), the Institut National de la Recherche Agronomique, and the Centre National de la Recherche Scientifique.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Philippe Hugueney (philippe.hugueney@ens-lyon.fr).

Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.105.070961.

References

  • Aharoni A, Giri AP, Deuerlein S, Griepink F, de Kogel WJ, Verstappen FWA, Verhoeven HA, Jongsma MA, Schwab W, Bouwmeester HJ (2003) Terpenoid metabolism in wild-type and transgenic Arabidopsis plants. Plant Cell 15: 2866–2884 [PMC free article] [PubMed]
  • Batoko H, Zheng HQ, Hawes C, Moore I (2000) A Rab1 GTPase is required for transport between the endoplasmic reticulum and Golgi apparatus and for normal Golgi movement in plants. Plant Cell 12: 2201–2217 [PMC free article] [PubMed]
  • Boatright J, Negre F, Chen X, Kish CM, Wood B, Peel G, Orlova I, Gang D, Rhodes D, Dudareva N (2004) Understanding in vivo benzenoid metabolism in petunia petal tissue. Plant Physiol 135: 1993–2011 [PMC free article] [PubMed]
  • Boevink P, Oparka K, Santa Cruz S, Martin B, Betteridge A, Hawes C (1998) Stacks on tracks: the plant Golgi apparatus traffics on an actin/ER network. Plant J 15: 441–447 [PubMed]
  • Brandizzi F, Irons SL, Johansen J, Kotzer A, Neumann U (2004) GFP is the way to glow: bioimaging of the plant endomembrane system. J Microsc 214: 138–158 [PubMed]
  • Burbulis IE, Winkel-Shirley B (1999) Interactions among enzymes of the Arabidopsis flavonoid biosynthetic pathway. Proc Natl Acad Sci USA 96: 12929–12934 [PMC free article] [PubMed]
  • Channelière S, Rivière S, Scalliet G, Szecsi J, Jullien F, Dolle C, Vergne P, Dumas C, Bendahmane M, Hugueney P, et al (2002) Analysis of gene expression in rose petals using expressed sequence tags. FEBS Lett 515: 35–38 [PubMed]
  • Chen F, Tholl D, D'Auria JC, Farooq A, Pichersky E, Gershenzon J (2003) Biosynthesis and emission of terpenoid volatiles from Arabidopsis flowers. Plant Cell 15: 481–494 [PMC free article] [PubMed]
  • Clark KR, Sims T (1994) The S-ribonuclease gene of Petunia hybrida is expressed in nonstylar tissue, including immature anthers. Plant Physiol 106: 25–36 [PMC free article] [PubMed]
  • Delichère C, Veuskens J, Hernould M, Barbacar N, Mouras A, Negrutiu I, Moneger F (1999) Sly1, the first active gene cloned from a plant y chromosome, encodes a wd-repeat protein. EMBO J 18: 4169–4179 [PMC free article] [PubMed]
  • Di Sansebastiano GP, Paris N, Marc-Martin S, Neuhaus JM (1998) Specific accumulation of GFP in a non-acidic vacuolar compartment via a C-terminal propeptide-mediated sorting pathway. Plant J 15: 449–457 [PubMed]
  • Dobson HEM, Bergström G, Groth I (1990) Differences in fragrance chemistry between flower parts of Rosa rugosa Thunb. (Rosaceae). Isr J Bot 39: 143–156
  • Dudareva N, Cseke L, Blanc VM, Pichersky E (1996) Evolution of floral scent in Clarkia: nove1 patterns of S-linalool synthase gene expression in the C. breweri flower. Plant Cell 8: 1137–1148 [PMC free article] [PubMed]
  • Dudareva N, D'Auria JC, Nam KH, Raguso RA, Pichersky E (1998) Acetyl CoA: benzylalcohol acetyltransferase, an enzyme involved in floral scent production in Clarkia breweri. Plant J 14: 297–304 [PubMed]
  • Dudareva N, Martin D, Kish CM, Kolosova N, Gorenstein N, Faldt J, Miller B, Bohlmann J (2003) (E)-beta-ocimene and myrcene synthase genes of floral scent biosynthesis in snapdragon: function and expression of three terpene synthase genes of a new terpene synthase subfamily. Plant Cell 15: 1227–1241 [PMC free article] [PubMed]
  • Dudareva N, Murfitt LM, Mann CJ, Gorenstein N, Kolosova N, Kish CM, Bonham C, Wood K (2000) Developmental regulation of methyl benzoate biosynthesis and emission in snapdragon flowers. Plant Cell 12: 949–961 [PMC free article] [PubMed]
  • Dudareva N, Pichersky E (2000) Biochemical and molecular genetic aspects of floral scent. Plant Physiol 122: 627–633 [PMC free article] [PubMed]
  • Emmanuelsson O, Nielsen H, Brunak S, von Heijne G (2000) Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J Mol Biol 300: 1005–1016 [PubMed]
  • Flament I, Debonneville C, Furrer A (1993) Volatile constituents of roses: characterization of cultivars based on the headspace analysis of living flower emissions. In R Teranishi, RG Buttery, H Sugisawa, eds, Bioactive Volatile Compounds from Plants. American Chemical Society, Washington DC, pp 269–281
  • Gang DR, Lavid N, Zubieta C, Chen F, Beuerle T, Lewinsohn E, Noel JP, Pichersky E (2002) Characterization of phenylpropene O-methyltransferases from sweet basil: facile change of substrate specificity and convergent evolution within a plant O-methyltransferase family. Plant Cell 14: 505–519 [PMC free article] [PubMed]
  • Gorton HL, Vogelmann TC (1996) Effects of epidermal cell shape and pigmentation on optical properties of Antirrhinum petals at visible and ultraviolet wavelengths. Plant Physiol 112: 879–888 [PMC free article] [PubMed]
  • Guterman I, Shalit M, Menda N, Piestun D, Dafny-Yelin M, Shalev G, Bar E, Davydov O, Ovadis M, Emanuel M, et al (2002) Rose scent: genomics approach to discovering novel floral fragrance-related genes. Plant Cell 14: 2325–2338 [PMC free article] [PubMed]
  • Haseloff J, Siemering KR, Prasher DC, Hodge S (1997) Removal of a cryptic intron and subcellular localization of green fluorescent protein are required to mark transgenic Arabidopsis brightly. Proc Natl Acad Sci USA 94: 2122–2127 [PMC free article] [PubMed]
  • Hawes C, Saint-Jore CM, Brandizzi F, Zheng H, Andreeva AV, Boevink P (2001) Cytoplasmic illuminations: in planta targeting of fluorescent proteins to cellular organelles. Protoplasma 215: 77–88 [PubMed]
  • Heim R, Cubitt AB, Tsien RY (1995) Improved green fluorescence. Nature 373: 663–664 [PubMed]
  • Helsper JPFG, Davies JA, Bouwmeester HJ, Krol AF, van Kampen MH (1998) Circadian rhythmicity in emission of volatile compounds by flowers of Rosa hybrida L. cv. Honesty. Planta 207: 88–95
  • Iwata H, Kato T, Ohno S (2000) Triparental origin of Damask roses. Gene 259: 53–59 [PubMed]
  • Jackson DP (1991) In situ hybridisation in plants. In SJ Gurr, MJ McPherson, DJ Bowles, eds, Molecular Plant Pathology: A Practical Approach. Oxford University Press, Oxford, pp 163–174
  • Joichi A, Yomogida K, Awano K, Ueda Y (2005) Volatile components of tea-scented modern roses and ancient Chinese roses. Flavour Fragr J 20: 152–157
  • Karimi M, Inzé D, Depicker A (2002) GATEWAY vectors for Agrobacterium-mediated plant transformation. Trends Plant Sci 7: 193–195 [PubMed]
  • Kolosova N, Sherman D, Karlson D, Dudareva N (2001) Cellular and subcellular localization of S-adenosyl-L-methionine:benzoic acid carboxyl methyltransferase, the enzyme responsible for biosynthesis of the volatile ester methylbenzoate in snapdragon flowers. Plant Physiol 126: 956–964 [PMC free article] [PubMed]
  • Lavid N, Wang J, Shalit M, Guterman I, Bar E, Beuerle T, Menda N, Shafir S, Zamir D, Adam Z, et al (2002) O-methyltransferases involved in the biosynthesis of volatile phenolic derivatives in rose petals. Plant Physiol 129: 1899–1907 [PMC free article] [PubMed]
  • Liu CJ, Dixon RA (2001) Elicitor-induced association of isoflavone O-methyltransferase with endomembranes prevents the formation and 7-O-methylation of daidzein during isoflavonoid phytoalexin biosynthesis. Plant Cell 13: 2643–2658 [PMC free article] [PubMed]
  • Loomis WD, Croteau RB (1973) Biochemistry and physiology of lower terpenoids. Recent Adv Phytochem 6: 147–185
  • Mahmoud SS, Croteau RB (2002) Strategies for transgenic manipulation of monoterpene biosynthesis in plants. Trends Plant Sci 7: 366–373 [PubMed]
  • Nakamura S (1987) Scent and component analysis of the hybrid tea rose. Perfum Flavor 12: 43–45
  • Oka N, Ohishi H, Hatano T, Hornberger M, Sakata K, Watanabe N (1999) Aroma evolution during flower opening in Rosa damascena Mill. Z Naturforsch C 54: 889–895
  • Paris N, Stanley CM, Jones RL, Rogers JC (1996) Plant cells contain two functionally distinct vacuolar compartments. Cell 85: 563–572 [PubMed]
  • Pichersky E, Lewinsohn E, Croteau R (1995) Purification and characterization of S-linalool synthase, an enzyme involved in the production of floral scent in Clarkia breweri. Arch Biochem Biophys 316: 803–807 [PubMed]
  • Quattrocchio F, Wing J, van der Woude K, Souer E, de Vetten N, Mol J, Koes R (1999) Molecular analysis of the anthocyanin2 gene of petunia and its role in the evolution of flower color. Plant Cell 11: 1433–1444 [PMC free article] [PubMed]
  • Ross JR, Nam KH, D'Auria JC, Pichersky E (1999) S-adenosyl-L-methionine:salicylic acid carboxyl methyltransferase, an enzyme involved in floral scent production and plant defense, represents a new class of plant methyltransferases. Arch Biochem Biophys 367: 9–16 [PubMed]
  • Scalliet G, Journot N, Jullien F, Baudino S, Magnard JL, Channelière S, Vergne P, Dumas C, Bendahmane M, Cock JM, et al (2002) Biosynthesis of the major scent components 3,5-dimethoxytoluene and 1,3,5-trimethoxybenzene by novel rose O-methyltransferases. FEBS Lett 523: 113–118 [PubMed]
  • Scott A, Wyatt S, Tsou PL, Robertson D, Allen NS (1999) Model system for plant cell biology: GFP imaging in living onion epidermal cells. Biotechniques 26: 1128–1132 [PubMed]
  • Shalit M, Guterman I, Volpin H, Bar E, Tamari T, Menda N, Adam Z, Zamir D, Vainstein A, Weiss D, et al (2003) Volatile ester formation in roses: identification of an acetyl-coenzyme A. Geraniol/Citronellol acetyltransferase in developing rose petals. Plant Physiol 131: 1868–1876 [PMC free article] [PubMed]
  • Stubbs JM, Francis MJO (1971) Electron microscopal studies of rose petal cells during flower maturation. Planta Med 20: 211–218 [PubMed]
  • Tholl D, Kish CM, Orlova I, Sherman D, Gershenzo J, Pichersky E, Dudareva N (2004) Formation of monoterpenes in Antirrhinum majus and Clarkia breweri flowers involves heterodimeric geranyl diphosphate synthases. Plant Cell 16: 977–992 [PMC free article] [PubMed]
  • Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673–4680 [PMC free article] [PubMed]
  • Turner GW, Croteau RB (2004) Organization of monoterpene biosynthesis in Mentha: immunocytochemical localizations of geranyl diphosphate synthase, limonene-6 hydroxylase, isopiperitenol dehydrogenase, and pulegone reductase. Plant Physiol 136: 4215–4227 [PMC free article] [PubMed]
  • Turner GW, Gershenzon J, Croteau RB (2000) Development of peltate glandular trichomes of peppermint. Plant Physiol 124: 665–679 [PMC free article] [PubMed]
  • Verdonk JC, Haring MA, van Tunen AJ, Schuurink RC (2005) ODORANT1 regulates fragrance biosynthesis in petunia flowers. Plant Cell 17: 1612–1624 [PMC free article] [PubMed]
  • Verdonk JC, Ric de Vos CH, Verhoeven HA, Haring MA, van Tunen AJ, Schuurink RC (2003) Regulation of floral scent production in petunia revealed by targeted metabolomics. Phytochemistry 62: 997–1008 [PubMed]
  • Wang J, Dudareva N, Bhakta S, Raguso RA, Pichersky E (1997) Floral scent production in Clarkia breweri (Onagraceae). II. Localization and developmental modulation of the enzyme S-adenosyl-L-methionine:(iso)eugenol O-methyltransferase and phenylpropanoid emission. Plant Physiol 114: 213–221 [PMC free article] [PubMed]
  • Weston EL, Pyke KA (1999) Developmental ultrastructure of cells and plastids in the petals of wallflower (Erysimum cheiri). Ann Bot (Lond) 84: 763–769
  • Wissemann V, Ritz CM (2005) The genus Rosa (Rosoideae, Rosaceae) revisited: molecular analysis of nrITS-1 and atpB-rbcL intergenic spacer (IGS) versus conventional taxonomy. Bot J Linn Soc 147: 275–290
  • Wu S, Watanabe N, Mita S, Dohra H, Ueda Y, Shibuya M, Ebizuka Y (2004) The key role of phloroglucinol O-methyltransferase in the biosynthesis of Rosa chinensis volatile 1,3,5-trimethoxybenzene. Plant Physiol 135: 1–8 [PMC free article] [PubMed]

Articles from Plant Physiology are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links