A cytosolic bifunctional geranyl/farnesyl diphosphate synthase provides MVA-derived GPP for geraniol biosynthesis in rose flowers

Geraniol derived from essential oils of various plant species is widely used in the cosmetic and perfume industries. It is also an essential trait of the pleasant smell of rose flowers. In contrast to other monoterpenes which are produced in plastids via the methyl erythritol phosphate pathway, geraniol biosynthesis in roses relies on cytosolic NUDX1 hydrolase which dephosphorylates geranyl diphosphate (GPP). However, the metabolic origin of cytosolic GPP remains unknown. By feeding Rosa chinensis "Old Blush" flowers with pathway-specific precursors and inhibitors, combined with metabolic profiling and functional characterization of enzymes in vitro and in planta, we show that geraniol is synthesized through the cytosolic mevalonate (MVA) pathway by a bifunctional geranyl/farnesyl diphosphate synthase, RcG/FPPS1, producing both GPP and farnesyl diphosphate (FPP). The downregulation and overexpression of RcG/FPPS1 in rose petals affected not only geraniol and germacrene D emissions but also dihydro-β-ionol, the latter due to metabolic cross talk of RcG/FPPS1-dependent isoprenoid intermediates trafficking from the cytosol to plastids. Phylogenetic analysis together with functional characterization of G/FPPS orthologs revealed that the G/FPPS activity is conserved among Rosaceae species. Site-directed mutagenesis and molecular dynamic simulations enabled to identify two conserved amino acids that evolved from ancestral FPPSs and contribute to GPP/FPP product specificity. Overall, this study elucidates the origin of the cytosolic GPP for NUDX1-dependent geraniol production, provides insights into the emergence of the RcG/FPPS1 GPPS activity from the ancestral FPPSs, and shows that RcG/FPPS1 plays a key role in the biosynthesis of volatile terpenoid compounds in rose flowers.

Duplication and specialization of NUDX1 in Rosaceae led to geraniol production in rose petals

Nudix hydrolases are conserved enzymes ubiquitously present in all kingdoms of life. Recent research revealed that several Nudix hydrolases are involved in terpenoid metabolism in plants. In modern roses, RhNUDX1 is responsible for formation of geraniol, a major compound of rose scent. Nevertheless, this compound is produced by monoterpene synthases in many geraniol-producing plants. As a consequence, this raised the question about the origin of RhNUDX1 function and the NUDX1 gene evolution in Rosaceae, in wild roses or/and during the domestication process. Here, we showed that three distinct clades of NUDX1 emerged in the Rosoidae subfamily (Nudx1-1 to Nudx1-3 clades), and two subclades evolved in the Rosa genus (Nudx1-1a and Nudx1-1b subclades). We also showed that the Nudx1-1b subclade was more ancient than the Nudx1-1a subclade, and that the NUDX1-1a gene emerged by a trans-duplication of the more ancient NUDX1-1b gene. After the transposition, NUDX1-1a was cis-duplicated, leading to a gene dosage effect on the production of geraniol in different species. Furthermore, the NUDX1-1a appearance was accompanied by the evolution of its promoter, most likely from a Copia retrotransposon origin, leading to its petal-specific expression. Thus, our data strongly suggest that the unique function of NUDX1-1a in geraniol formation was evolved naturally in the genus Rosa before domestication.

Functional diversification in the Nudix hydrolase gene family drives sesquiterpene biosynthesis in Rosa × wichurana

Roses use a non-canonical pathway involving a Nudix hydrolase, RhNUDX1, to synthesize their monoterpenes, especially geraniol. Here we report the characterization of another expressed NUDX1 gene from the rose cultivar Rosa x wichurana, RwNUDX1-2. In order to study the function of the RwNUDX1-2 protein, we analyzed the volatile profiles of an F1 progeny generated by crossing R. chinensis cv. 'Old Blush' with R. x wichurana. A correlation test of the volatilomes with gene expression data revealed that RwNUDX1-2 is involved in the biosynthesis of a group of sesquiterpenoids, especially E,E-farnesol, in addition to other sesquiterpenes. In vitro enzyme assays and heterologous in planta functional characterization of the RwNUDX1-2 gene corroborated this result. A quantitative trait locus (QTL) analysis was performed using the data of E,E-farnesol contents in the progeny and a genetic map was constructed based on gene markers. The RwNUDX1-2 gene co-localized with the QTL for E,E-farnesol content, thereby confirming its function in sesquiterpenoid biosynthesis in R. x wichurana. Finally, in order to understand the structural bases for the substrate specificity of rose NUDX proteins, the RhNUDX1 protein was crystallized, and its structure was refined to 1.7 Å. By molecular modeling of different rose NUDX1 protein complexes with their respective substrates, a structural basis for substrate discrimination by rose NUDX1 proteins is proposed.

Linalool and linalool nerolidol synthases in roses, several genes for little scent

Roses are widely appreciated for the appearance of their flowers and for their fragrance. This latter character results from the combination of different odorant molecules among which monoterpenes are often prevalent constituents. In this study, we report the cloning and characterization of three rose monoterpene synthases. In vitro functional characterization of these enzymes showed that one is a (-)-(3R)-linalool synthase whereas the others have a dual (+)-(3S)-linalool nerolidol synthase activity. However, given that the characterized rose cultivars were only able to produce the (-)-(3R)-linalool stereoisomer, the linalool nerolidol synthases are probably not active in planta. Furthermore, these three enzymes were also characterized by a weak expression level as assessed by RT-qPCR and by the low abundance of the corresponding sequences in an EST library. This characteristic is likely to explain why linalool is generally a minor constituent in rose flowers' scents. On this basis, we propose that in roses the monoterpene biosynthesis effort is focused on the production of acyclic monoterpenes derived from geraniol through the recently characterized Nudix biosynthesis pathway, at the expense of conventional monoterpene biosynthesis via terpene synthases such as linalool or linalool nerolidol synthases.

Bornyl-diphosphate synthase from Lavandula angustifolia: A major monoterpene synthase involved in essential oil quality

Lavender essential oils (EOs) of higher quality are produced by a few Lavandula angustifolia cultivars and mainly used in the perfume industry. Undesirable compounds such as camphor and borneol are also synthesized by lavender leading to a depreciated EO. Here, we report the cloning of bornyl diphosphate synthase of lavender (LaBPPS), an enzyme that catalyzes the production of bornyl diphosphate (BPP) and then by-products such as borneol or camphor, from an EST library. Compared to the BPPS of Salvia officinalis, the functional characterization of LaBPPS showed several differences in amino acid sequence, and the distribution of catalyzed products. Molecular modeling of the enzyme's active site suggests that the carbocation intermediates are more stable in LaBPPS than in SoBPPS leading probably to a lower efficiency of LaBPPS to convert GPP into BPP. Quantitative RT-PCR performed from leaves and flowers at different development stages of L. angustifolia samples show a clear correlation between transcript level of LaBPPS and accumulation of borneol/camphor, suggesting that LaBPPS is mainly responsible of in vivo biosynthesis of borneol/camphor in fine lavender. A phylogenetic analysis of terpene synthases (TPS) pointed out the basal position of LaBPPS in the TPSb clade, suggesting that LaBPPS could be an ancestor of others lavender TPSb. Finally, borneol could be one of the first monoterpenes to be synthesized in the Lavandula subgenus. Knowledge gained from these experiments will facilitate future studies to improve the lavender oils through metabolic engineering or plant breeding. Accession numbers: LaBPPS: KM015221.

PLANT VOLATILES. Biosynthesis of monoterpene scent compounds in roses

The scent of roses (Rosa x hybrida) is composed of hundreds of volatile molecules. Monoterpenes represent up to 70% percent of the scent content in some cultivars, such as the Papa Meilland rose. Monoterpene biosynthesis in plants relies on plastid-localized terpene synthases. Combining transcriptomic and genetic approaches, we show that the Nudix hydrolase RhNUDX1, localized in the cytoplasm, is part of a pathway for the biosynthesis of free monoterpene alcohols that contribute to fragrance in roses. The RhNUDX1 protein shows geranyl diphosphate diphosphohydrolase activity in vitro and supports geraniol biosynthesis in planta.

Isolation and functional characterization of a τ-cadinol synthase, a new sesquiterpene synthase from Lavandula angustifolia

In this paper we characterize three sTPSs: a germacrene D (LaGERDS), a (E)-β-caryophyllene (LaCARS) and a τ-cadinol synthase (LaCADS). τ-cadinol synthase is reported here for the first time and its activity was studied in several biological models including transiently or stably transformed tobacco species. Three dimensional structure models of LaCADS and Ocimum basilicum γ-cadinene synthase were built by homology modeling using the template structure of Gossypium arboreum δ-cadinene synthase. The depiction of their active site organization provides evidence of the global influence of the enzymes on the formation of τ-cadinol: instead of a unique amino-acid, the electrostatic properties and solvent accessibility of the whole active site in LaCADS may explain the stabilization of the cadinyl cation intermediate. Quantitative PCR performed from leaves and inflorescences showed two patterns of expression. LaGERDS and LaCARS were mainly expressed during early stages of flower development and, at these stages, transcript levels paralleled the accumulation of the corresponding terpene products (germacrene D and (E)-β-caryophyllene). By contrast, the expression level of LaCADS was constant in leaves and flowers. Phylogenetic analysis provided informative results on potential duplication process leading to sTPS diversification in lavender.

Extracellular localization of the diterpene sclareol in clary sage (Salvia sclarea L., Lamiaceae)

Sclareol is a high-value natural product obtained by solid/liquid extraction of clary sage (Salvia sclarea L.) inflorescences. Because processes of excretion and accumulation of this labdane diterpene are unknown, the aim of this work was to gain knowledge on its sites of accumulation in planta. Samples were collected in natura or during different steps of the industrial process of extraction (steam distillation and solid/liquid extraction). Samples were then analysed with a combination of complementary analytical techniques (gas chromatography coupled to a mass spectrometer, polarized light microscopy, environmental scanning electron microscopy, two-photon fluorescence microscopy, second harmonic generation microscopy). According to the literature, it is hypothesized that sclareol is localized in oil pockets of secretory trichomes. This study demonstrates that this is not the case and that sclareol accumulates in a crystalline epicuticular form, mostly on calyces.

Discovery and functional characterization of two diterpene synthases for sclareol biosynthesis in Salvia sclarea (L.) and their relevance for perfume manufacture

BACKGROUND

Sclareol is a diterpene natural product of high value for the fragrance industry. Its labdane carbon skeleton and its two hydroxyl groups also make it a valued starting material for semisynthesis of numerous commercial substances, including production of Ambrox® and related ambergris substitutes used in the formulation of high end perfumes. Most of the commercially-produced sclareol is derived from cultivated clary sage (Salvia sclarea) and extraction of the plant material. In clary sage, sclareol mainly accumulates in essential oil-producing trichomes that densely cover flower calices. Manool also is a minor diterpene of this species and the main diterpene of related Salvia species.

RESULTS

Based on previous general knowledge of diterpene biosynthesis in angiosperms, and based on mining of our recently published transcriptome database obtained by deep 454-sequencing of cDNA from clary sage calices, we cloned and functionally characterized two new diterpene synthase (diTPS) enzymes for the complete biosynthesis of sclareol in clary sage. A class II diTPS (SsLPPS) produced labda-13-en-8-ol diphosphate as major product from geranylgeranyl diphosphate (GGPP) with some minor quantities of its non-hydroxylated analogue, (9 S, 10 S)-copalyl diphosphate. A class I diTPS (SsSS) then transformed these intermediates into sclareol and manool, respectively. The production of sclareol was reconstructed in vitro by combining the two recombinant diTPS enzymes with the GGPP starting substrate and in vivo by co-expression of the two proteins in yeast (Saccharomyces cerevisiae). Tobacco-based transient expression assays of green fluorescent protein-fusion constructs revealed that both enzymes possess an N-terminal signal sequence that actively targets SsLPPS and SsSS to the chloroplast, a major site of GGPP and diterpene production in plants.

CONCLUSIONS

SsLPPS and SsSS are two monofunctional diTPSs which, together, produce the diterpenoid specialized metabolite sclareol in a two-step process. They represent two of the first characterized hydroxylating diTPSs in angiosperms and generate the dihydroxylated labdane sclareol without requirement for additional enzymatic oxidation by activities such as cytochrome P450 monoxygenases. Yeast-based production of sclareol by co-expresssion of SsLPPS and SsSS was efficient enough to warrant the development and use of such technology for the biotechnological production of scareol and other oxygenated diterpenes.

Production and emission of volatile compounds by petal cells

We localized the tissues and cells that contribute to scent biosynthesis in scented and non-scented Rosa x hybrida cultivars as part of a detailed cytological analysis of the rose petal. Adaxial petal epidermal cells have a typical conical, papillate shape whereas abaxial petal epidermal cells are flat. Using two different techniques, solid/liquid phase extraction and headspace collection of volatiles, we showed that, in roses, both epidermal layers are capable of producing and emitting scent volatiles, despite the different morphologies of the cells of these two tissues. Moreover, OOMT, an enzyme involved in scent molecule biosynthesis, was localized in both epidermal layers. These results are discussed in view of results found in others species such as Antirrhinum majus, where it has been shown that the adaxial epidermis is the preferential site of scent production and emission.

Both the adaxial and abaxial epidermal layers of the rose petal emit volatile scent compounds

The localization and timing of production and emission of scent was studied in different Rosa x hybrida cultivars, focusing on three particular topics. First, it was found that petals represent the major source of scent in R. x hybrida. In heavily scented cultivars, the spectrum and levels of volatiles emitted by the flower broadly correlated with the spectrum and levels of volatiles contained within the petal, throughout petal development. Secondly, analysis of rose cultivars that lacked a detectable scent indicated that the absence of fragrance was due to a reduction in both the biosynthesis and emission of scent volatiles. A cytological study, conducted on scented and non-scented rose cultivars showed that no major difference was visible in the anatomy of the petals either at small magnification in optical sections or in ultrathin sections observed by TEM. In particular, the cuticle of epidermal cells was not thicker in scentless cultivars. Thirdly, using two different techniques, solid/liquid phase extraction and headspace collection of volatiles, we showed that in roses, both epidermal layers are capable of producing and emitting scent volatiles, despite the different morphologies of the cells of these two tissues. Moreover, OOMT, an enzyme involved in scent molecule biosynthesis was localized in both epidermal layers.

Role of petal-specific orcinol O-methyltransferases in the evolution of rose scent

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, predominantly 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.

Progression through meiosis I and meiosis II in Arabidopsis anthers is regulated by an A-type cyclin predominately expressed in prophase I

Meiosis is often described as a special case of cell division since it differs from mitosis in having two nuclear divisions without an intervening S-phase. It will be of great interest to uncover what molecular mechanisms underlie these special features of meiosis. We previously reported that the tardy asynchronous meiosis (tam) mutant of Arabidopsis (Arabidopsis thaliana) is slower in cell cycle progression in male meiosis. Here we report that TAM encodes the A-type cyclin, CYCA1;2. The point mutation in tam replaced a conserved threonine with an isoleucine in the linker region between the alpha4 and alpha5 helices of the first cyclin fold. By studying the dynamics of a CYCA1;2-green fluorescent protein fusion protein under the control of the CYCA1;2 promoter, we found that the fusion protein was most abundant at pachytene, but was undetectable from late prophase I until telophase II. Nonetheless, cell cycle progression in tam was delayed in both pachytene and meiosis II. We conclude either that the CYCA1;2 produced in prophase I indirectly regulates meiosis II progression, or that a very low level of CYCA1;2 directly regulates meiosis II progression. Either of these scenarios is a deviation from the typical mode of action of mitotic cyclins in mitosis and meiosis I, in which each nuclear division is coupled with a peak of expression of mitotic cyclins.

Morphogenesis of maize embryos requires ZmPRPL35-1 encoding a plastid ribosomal protein

In emb (embryo specific) mutants of maize (Zea mays), the two fertilization products have opposite fates: Although the endosperm develops normally, the embryo shows more or less severe aberrations in its development, resulting in nonviable seed. We show here that in mutant emb8516, the development of mutant embryos deviates as soon as the transition stage from that of wild-type siblings. The basic events of pattern formation take place because mutant embryos display an apical-basal polarity and differentiate a protoderm. However, morphogenesis is strongly aberrant. Young mutant embryos are characterized by protuberances at their suspensor-like extremity, leading eventually to structures of irregular shape and variable size. The lack of a scutellum or coleoptile attest to the virtual absence of morphogenesis at the embryo proper-like extremity. Molecular cloning of the mutation was achieved based on cosegregation between the mutant phenotype and the insertion of a MuDR element. The Mu insertion is located in gene ZmPRPL35-1, likely coding for protein L35 of the large subunit of plastid ribosomes. The isolation of a second allele g2422 and the complementation of mutant emb8516 with a genomic clone of ZmPRPL35-1 confirm that a lesion in ZmPRPL35-1 causes the emb phenotype. ZmPRPL35-1 is a low-copy gene present at two loci on chromosome arms 6L and 9L. The gene is constitutively expressed in all major tissues of wild-type maize plants. Lack of expression in emb/emb endosperm shows that endosperm development does not require a functional copy of ZmPRPL35-1 and suggests a link between plastids and embryo-specific signaling events.

ZmEBE genes show a novel, continuous expression pattern in the central cell before fertilization and in specific domains of the resulting endosperm after fertilization

Two novel maize genes expressed specifically in the central cell of the female gametophyte and in two compartments of the endosperm (the basal endosperm transfer layer and the embryo surrounding region) were characterized. The ZmEBE (embryo sac/basal endosperm transfer layer/embryo surrounding region) genes were isolated by a differential display between the upper and the lower half of the kernel at 7 days after pollination (DAP). Sequence analysis revealed ORFs coding for two closely related proteins of 304 amino acids (ZmEBE-1) and 286 amino acids (ZmEBE-2). This size difference was due to differences in the splicing of the two genes. Both protein sequences showed significant similarity to the DUF239 family of Arabidopsis, a group of 22 proteins of unknown function, a small number of which are putative peptidases. ZmEBE genes had a novel cell type-specific expression pattern in the central cell before and the resulting endosperm after fertilization. RT-PCR analysis showed that the expression of both genes started before pollination in the central cell and continued in the kernel up to 20 DAP with a peak at 7 DAP. In situ hybridization revealed that the expression in the kernel was restricted to the basal transfer cell layer and the embryo surrounding region of the endosperm. The expression of ZmEBE-1 was at least 10 times lower than that of ZmEBE-2. Similarly to other genes expressed in the endosperm, ZmEBE-1 expression was subject to a parent-of-origin effect, while no such effect was detected in ZmEBE-2. Sequence analysis of upstream regions revealed a potential cis element of 33 bp repeated 7 times in ZmEBE-1 and ZmEBE-2 between positions -900 and -100. The 1.6 kb ZmEBE-2 upstream sequence containing the seven R7 elements was able to confer expression in the basal endosperm to a Gus reporter gene. These data indicate that ZmEBE is potentially involved in the early development of specialized domains of the endosperm and that this process is possibly already initiated in the central cell, which is at the origin of the endosperm.

The Arabidopsis gene tardy asynchronous meiosis is required for the normal pace and synchrony of cell division during male meiosis

Male meiosis in higher organisms features synchronous cell divisions in a large number of cells. It is not clear how this synchrony is achieved, nor is it known whether the synchrony is linked to the regulation of cell cycle progression. Here, we describe an Arabidopsis mutant, named tardy asynchronous meiosis (tam), that exhibits a phenotype of delayed and asynchronous cell divisions during male meiosis. In Arabidopsis, two nuclear divisions occur before simultaneous cytokinesis yields a tetrad of haploid cells. In tam, cell divisions are delayed, resulting in the formation of abnormal intermediates, most frequently dyad meiotic products, or in rare cases, dyad pollen (two gametophytes within one exine wall). Temperature-shift experiments showed that the percentage of the abnormal intermediates increased at 27 degrees C. Analysis of tam and the tam/quartet1 double mutant showed that most of these abnormal intermediates could continue through the normal rounds of cell divisions and form functional pollen, though at a slower than normal pace. The asynchrony of cell division started at the G2/M transition, with cells entering metaphase at different time points, during both meiosis I and II. In addition, chromosome condensation defects and mis-segregation were sometimes observed in tam. These observations suggest that the TAM protein positively regulates cell cycle progression, perhaps by promoting the G2/M transition. We speculate that there is a signal, perhaps TAM, that couples the normal pace of cell cycle progression with the synchrony of cell division during male meiosis.

Genes normally expressed in the endosperm are expressed at early stages of microspore embryogenesis in maize

Reproduction in flowering plants is characterized by double fertilization and the resulting formation of both the zygotic embryo and the associated endosperm. In many species it is possible to experimentally deviate pollen development towards an embryogenic pathway. This developmental switch, referred to as microspore embryogenesis or androgenesis, leads to the formation of embryos similar to zygotic embryos. In a screen for genes specifically expressed during early androgenesis, two maize genes were isolated by mRNA differential display. Both genes represent new molecular markers expressed at a very young stage of androgenic embryogenesis. When their expression pattern was studied during normal reproductive development, both showed early endosperm-specific expression. Investigation of the cytological features of young androgenic embryos revealed that they present a partially coenocytic organization similar to that of early endosperm. These findings suggest that maize androgenesis may possibly involve both embryogenesis and the establishment of endosperm-like components.

ZmOCL1, an HDGL2 family homeobox gene, is expressed in the outer cell layer throughout maize development

The formation of a morphologically distinct outer cell layer or protoderm is one of the first and probably one of the most important steps in patterning of the plant embryo. Here we report the isolation of ZmOCL1 (OCL for outer cell layer), a member of the HDGL2 (also known as HD-ZIP IV) subclass of plant-specific HD-ZIP homeodomain proteins from maize. ZmOCL1 transcripts are detected very early in embryo development, before a morphologically distinct protoderm is visible, and expression then becomes localised to the protoderm of the embryo as it develops. Subsequently, expression is observed in the L1 cell layer of both the developing primary root and shoot meristems, and is maintained in developing leaves and floral organs. We propose that ZMOCL1 may play a role in the specification of protoderm identity within the embryo, the organisation of the primary root primordium or in the maintenance of the L1 cell layer in the shoot apical meristem. We also show that the expression of ZmOCL1 is different from that of another epidermal marker gene, LTP2 (lipid transfer protein) and, in meristems, is complementary to that of Kn1 (Knotted) which is transcribed only in underlying cell layers.

Isolated maize zygotes mimic in vivo embryonic development and express microinjected genes when cultured in vitro

We established conditions for the regeneration of natural maize zygotes isolated from pollinated plants with the goal of investigating the molecular control of early embryogenesis in higher plants. Viable zygotes were excised from embryo sacs by minimal enzymatic digestion and microdissection. Viable zygotes transferred to coculture with androgenic microspores from barley developed into embryo-like structures in 61% of the cases. No development was observed when zygotes were cultured in the presence of maize anthers undergoing androgenetic embryogenesis. Zygote-derived embryo-like structures regenerated into fertile plants through secondary embryogenesis when transferred to solid medium. The first zygotic division was asymmetrical and bipolar structures similar to pretransitional embryos observed in planta were later produced as observed using light and electron microscopy. Conditions for efficient microinjection of DNA into zygotes were established. Calcofluor and PATAG staining of zygotes showed that cell wall regeneration occurred as early as 20 min after enzymatic isolation and that after 2 hr, each zygote was bordered with cell wall material. Through quantitative microphotometry, DNA synthesis during the first cell cycle of the zygote was shown to occur between isolation and 12 hr of culture. Microinjection of two types of reporter genes (GUS gene and anthocyanin regulatory genes) demonstrates transient expression in plant zygotes. On average, 3.5% of microinjected zygotes showed transgenic expression. Reporter gene expression was observed in zygotes at different time points of their first cell cycle.