INFLUENCE OF THE VARIABLE AMINO ACIDS DURING THE DOCKING OF THE DMS

Get Complete Project Material File(s) Now! »

From defensive properties to interesting bioactivities

Thanks to their defensive properties, furanocoumarins exhibit a large range of bioactivities such as photosensitiser, P450-inhibitors, antifungal, antibacterial and antiviral activities, but also antioxidant, anti-inflammatory, anticancer, anticonvulsant, antiproliferative, anti-hyperglycemic, cytoprotective, apoptotic agent or even anti-coagulant activities (Bourgaud et al. 1990; Venugopala et al. 2013; Bruni et al. 2019). Some of these properties have been known for hundreds or thousands of years (Scott et al. 1976), which explains why furanocoumarin-producing plants have been widely used by humans in many domains such as medicine and pharmaceutics (Bourgaud et al. 1990).
For instance, linear furanocoumarins have been historically used against skin disorders, skin depigmentation and other diseases such as psoriasis or vitiligo. Unfortunately, the toxic properties of furanocoumarins also apply to humans and these medical uses have been linked to side effects such as increased skin cancers (Diawara and Trumble 1997; Sarker and Nahar 2004; Melough and Chun 2018). Yet, despite these side effects, furanocoumarin bioactivities are still very interesting for both academic and industrial research, including the medical field. As an example, furanocoumarins are still tested and evaluated for their cytotoxicity against cancer (Chauthe et al. 2015), or for a safe use against diseases such as vitiligo (Zabolinejad et al. 2020).

Defensive properties and spatio-temporal variation within plants

As mentioned previously, in a given plant, furanocoumarins are unequally distributed between plants parts (organs, compartments) and vary over time (constitutive variations and inducibility). In fact, these variations are partly linked to the defensive nature of furanocoumarins.
On the one hand, furanocoumarins fall within the framework of the trade-offs between growth and defence, which means the production and storage of furanocoumarins might negatively impact plant growth. Plants might therefore have evolved strategies to reduce the cost of furanocoumarin production. For instance, plants may produce these toxic compounds only when and where there are needed. In particular, the spatial and temporal variations in the production of furanocoumarins might reflect evolutionary mechanisms that have often been linked to the optimal defence theory – a theory which predicts that plant tissues should be defended according to their contribution to plant fitness and the probability of attack (McKey 1974, 1979; Nitao and Zangerl 1987; Zangerl and Berenbaum 1990; Zangerl and Rutledge 1996; Innocenti et al. 1997a; Zangerl and Nitao 1998). On the other hand, because of the high toxicity of furanocoumarins, plants have evolved strategies to synthesise, store and bring these defensive chemicals into contact with their bioaggressors, without poisoning themselves. Some common strategies to prevent autotoxicity consist in storing defensive compounds at specific locations, in order to isolate them from sensitive tissues. It might therefore explain why – depending on species – furanocoumarins are compartmented in structures such as laticifer cells, trichomes or oil ducts (Reinold and Hahlbrock 1997; Mamoucha et al. 2016; Weryszko-Chmielewska and Chwil 2017). Furanocoumarins can also be stored as inactivated glycosylated forms, in specific cellular compartments, separately from their activating enzymes. As an example, in Psoralea and Coronilla, glycosylated furanocoumarins are stored in the vacuole: upon attack, if the vacuole membrane (tonoplast) is disrupted, glycosylated furanocoumarins are brought into contact with endogenous β-glucosidases that hydrolyse them into activated free-form furanocoumarins (Innocenti et al. 1997b). Lastly, furanocoumarins might also be transported on the surface of the plant, where they represent a lower risk for the plant itself while serving as a barrier against bioaggressors. For instance, furanocoumarins have been found on the surface of the leaves, stems and/or trichomes of several Apiaceae such as parsley (Petroselinum crispum), the rue (Ruta graveolens) or the hogweed (Heracleum sosnowskyi). In these plants, surface furanocoumarins can even represent the major part of the total furanocoumarin content – which might explain many phytophotodermatitis (Zobel and Brown 1988; Reinold and Hahlbrock 1997; Weryszko-Chmielewska and Chwil 2017).
Finally, if furanocoumarins can partially inhibit plant P450s, the P450s of furanocoumarin-producing plants are usually less sensitive to furanocoumarin inhibition than average. This increased resistance might be an adaptative solution that allows furanocoumarin-producing plants to conserve a metabolic activity in presence of furanocoumarins (Gravot et al. 2004).

The different steps of the furanocoumarin biosynthesis pathway

The furanocoumarin pathway derives from the phenylpropanoid pathway. In particular, it starts with the ubiquitous p-coumaroyl-coA, a major branching point that can lead to various compounds such as flavonoids, stilbenes, monolignols or phenolamides. The sequential steps of the main linear furanocoumarin pathway, summarised in Figure 12 are the following:
1) First, p-coumaroyl-coA is converted into 2-4-dihydroxycinnamate through an ortho-hydroxylation catalysed by a dioxygenase. It then spontaneously cyclises into umbelliferone (Bourgaud et al. 2006). The molecular elucidation of this step started in 2008 with some studies done on a 2-oxoglutarate-dependent dioxygenase (2-OGD) involved in the formation of scopoletin in A. thaliana (Kai et al. 2008). Yet, it is only in 2012 that the first dioxygenase involved in the formation of umbelliferone was described. This enzyme, the p-coumaroyl CoA 2’-hydroxylase (C2’H), is a 2-OGD which was identified and characterised in the rue (Ruta graveolens) (Vialart et al. 2012). Similar results have also been obtained in the sweat potato (Ipomea batatas – which does not produce furanocoumarins) with the identification of another C2’H (Matsumoto et al. 2012). More recently, in parsnip (Pastinaca sativa), an α-ketoglutarate-dependent dioxygenase named PsDiox has been described: it catalyses the same reaction than C2’H but shares only 52% identity with the gene from R. graveolens (Roselli et al. 2017).
2) Umbelliferone, which is still a simple coumarin, can subsequently be prenylated to open the way to linear and angular furanocoumarins. This prenylation, catalysed by a prenyltransferase (PT) called umbelliferone dimethylallyltransferases (UDT), is regiospecific: a prenylation on the C6 forms demethylsuberosin (DMS, 6-dimethylallylumbelliferone), which leads to linear isomers (2), while a prenylation on the C8 forms osthenol and paves the way to angular compounds (2’) (Brown and Steck 1973). The first UDT, identified in parsley (Petroselinum crispum), has been named PcPT – for P. crispum PT. It mainly catalyses the C6 prenylation of umbelliferone into DMS but also shows a minor activity at C8 (Karamat et al. 2014). Two years later, two other UDTs were found in parsnip (Pastinaca sativa): PsPT1 and PsPT2. If PsPT1 mainly catalyses the conversion of umbelliferone into DMS (C6), PsPT2 is more efficient at forming osthenol (C8). It should be noted that PsPT2 might have emerged through the duplication and neo-functionalisation of PsPT1 (Munakata et al. 2016), and that PsPT1 is clustered with PsDiox (Roselli et al. 2017). These apiaceous UDTs are all expressed in the plastids, which suggests it is where the furanocoumarin biosynthesis starts. More recently, another UDT specialised in the conversion of umbelliferon into DMS has been found in F. carica (FcPT1) (Munakata et al. 2020).
3) The pathway to linear furanocoumarins continues with the conversion of DMS into (+) marmesin through the cyclisation of the prenyl group on the hydroxyl group. This cyclisation is not spontaneous but catalysed by an enzyme named the marmesin synthase. In Ammi majus, evidences strongly suggest that the marmesin synthase is a cytochrome P450 (Hamerski and Matern 1988a) but, until now, this enzyme has never been characterised and its gene remains unknown.
4) The next step is the conversion of marmesin into psoralen, which is the first linear furanocoumarin that exhibits a strong toxicity. This reaction consists in a P450-mediated cleavage of marmesin that produces psoralen and a molecule of acetone (Hamerski and Matern 1988a). Several genes coding for psoralen synthases have been isolated in apiaceous species. The first one, CYP71AJ1, was isolated and characterised from Ammi majus cell cultures in 2007 – which made it the first known P450 from the furanocoumarin biosynthesis pathway (Larbat et al. 2007). The search for CYP71AJ1 orthologs in other Apiaceae latter led to the identification of two other psoralen synthases: CYP71AJ2 from parsley (P. crispum) and CYP71AJ3 from parsnip (P. sativa) (Larbat et al. 2009).

Evolutionary perspectives and concluding remarks

Furanocoumarins are distributed in distant plant families. Yet, the furanocoumarin biosynthesis pathway seems to be very similar in all producing plants studied so far, which raises the question of the emergence of this pathway. Two alternative hypotheses have been proposed to explain the presence of a similar pathway in distant plant taxa: on the one hand, the pathway may have emerged only once, but it would have then been massively lost in many plant taxa. On the other hand, a similar pathway may have independently emerged several times in different plant taxa (Munakata et al. 2020). The molecular elucidation of the furanocoumarin pathway provided some clues in favour of a multiple origin – even though new data would be welcome to support and strengthen this hypothesis.
The most complete data in favour of this hypothesis results from the work done on the UDTs that open the way to linear or angular furanocoumarins: a gene-family phylogenetic analysis have been conducted on the UDTs from Apiaceae (PcPT1, PsPT1, PsPT2) and Moraceae (FcPT1), and the results strongly suggest that the moraceous UDT evolved from a different ancestor than the three apiaceous UDTs. The UDT activity must thus have independently emerged in Moraceae and Apiaceae (Munakata et al. 2020).
In addition, the hydroxylation of xanthotoxin is catalysed by P450s from the CYP71 family in Apiaceae, but by a P450 from the CYP82 family in Rutaceae. Even though no precise gene-family phylogenetic analyses were conducted on these enzymes, this information also suggests an independent emergence of the enzymes in the Apiaceae and the Rutaceae families.

SRSs, a hot spot for site-directed mutagenesis

The SRSs are highly variable and their amino acid sequences are more diversified than the rest of the P450 primary sequence; the most variable ones being SRS1-3 and SRS6. This fact is consistent with the idea that the SRSs are involved in the substrate specificity: their variations contribute to the diversification of the reactions catalysed by P450s.
As a consequence, many mutations affecting the SRSs significantly impact the substrate specificity, and the SRS have thus become a hotspot for site-directed mutagenesis. Indeed, in many studies, site-directed mutagenesis has been performed on a few residues from the SRSs of various P450s, and resulted in changes in the activity, the substrate range, specificity, regio- and stereoselectivity of the enzyme – even with single amino acid substitutions. A quite common approach consists in modelling a P450 and performing docking experiments to identify key residues that might change its activity. The models are subsequently tested by site-directed mutagenesis and might result in new enzymatic activities and performances, which is of high interest for biotechnology applications and the production of new bioactive compounds. However, even though less common, a few mutations outside the SRSs have sometimes been reported to also affect the activity and specificity of P450s (Gotoh 1992; Urlacher 2006; Baudry et al. 2006; Rupasinghe and Schuler 2006; Li et al. 2008a; Nelson et al. 2008; Seifert et al. 2009; Sirim et al. 2010; Schuler and Rupasinghe 2011; Uno et al. 2011; Vazquez-Albacete et al. 2017; Kuhlman and Bradley 2019).

Atypical P450s and unusual P450-mediated reactions

As mentioned earlier, P450s catalyse a large variety of monooxygenations/hydroxylations. However, if these reactions are the most common, some unusual reactions also have to be mentioned. The most atypical plant P450s are probably from the CYP74 family, that contains the allene oxide synthase. As CYP74A1 and CYP74A2 have been crystallised (Lee et al. 2008; Li et al. 2008b), the CYP74 family is well known. Briefly, CYP74s seem to exist in all plant species, but only in a low number of enzymes. They are localised in the chloroplasts and catalyse an atypical reaction that do not require the activation of a molecular dioxygen: instead, they activate a hydroperoxide that is directly provided by their substrate, which short-circuits the standard catalytic cycle. As should be expected from this atypical reaction, the primary sequence of CYP74s is quite different from the one of typical P450s: for instance, the structures involved in oxygen activation and electron transfer are altered (Werck-Reichhart et al. 2002; Nelson et al. 2008; Schuler and Rupasinghe 2011).
Even if they are not as atypical as the CYP74 family members, many other plant P450s catalyse unusual reactions that are involved in various specialised metabolic pathways. In their review, Mizutani and Sato (2011) provide a nice summary of these reactions and describe their proposed mechanisms. Among them, we can quickly list the formation of methylenedioxy-bridges (CYP719A, CYP80Q1), intramolecular C-O or C-C phenol coupling (CYP80A, CYP80G, CYP719B), N-oxidation (CYP79), sterol desaturation (CYP710A), rearrangement of the carbon skeleton (CYP93C2, CYP88A), C-C bond cleavage (CYP72A1), Baeyer-Villiger oxidation (CYP85A2), or the hydroxylation followed by intramolecular cyclisation performed by the menthofuran synthase (CYP71A32 – Figure 19) (Mizutani and Sato 2011).

The diversification of P450s is at the heart of plant chemical diversity

Not only plants possess a huge number of P450 genes, but their P450s are also quite diversified. Indeed, in plants, a total of 127 P450 families have been identified so far, and a typical angiosperm is expected to possess around 50 P450 families. The diversification of plant P450s mirrors plant biochemical diversity: as the production of new chemicals requires new enzymes, and as oxygen is very useful to build complex molecules, P450s have been recruited in many metabolic networks, in which they contribute to produce new molecules (Nelson 2011; Nelson and Werck-Reichhart 2011). As a result, and even though most of them are still to be characterised, we already know that plant P450s are involved in the production of a large diversity of both primary and specialised metabolites, that are essential for plant growth, development, and adaptation to their environment (Pichersky and Gang 2000; Nelson et al. 2008; Nelson and Werck-Reichhart 2011; Mizutani 2012; Ilc et al. 2018).
Plant P450s can be grouped into 11 clans. The CYP51, CYP74, CYP97, CYP710, CYP711 and CYP746 clans are single-family clans, and most of them seem to fulfil conserved essential functions such as sterol biosynthesis (CYP51, CYP710), oxylipin biosynthesis (CYP74) or carotenoid biosynthesis (CYP97). On the contrary, the CYP71, CYP72, CYP85, CYP86 and CYP727 clans are multiple family clans involved in diverse functions – except maybe for the CYP727 clan that does not have any known function yet. The CYP86 clan is involved in the metabolism of fatty acids, fatty alcohols and alkanes. The CYP85 clan is mainly involved into isoprenoid metabolism. The CYP72 clan metabolises compounds such as fatty acids, isoprenoids, hormones and cytokinins. Last but not least, the CYP71 clan, which contains more than half of all plant P450s, is involved in the metabolism of a great variety of compounds ranging from phenylpropanoids, terpenoids, alkaloids, fatty acids and hormones (Nelson 2004; Nelson et al. 2008; Nelson and Werck-Reichhart 2011).

Kinetic parameters: affinity, catalytic constant and catalytic efficiency

The kinetic parameters that can be measured to characterise an enzyme include the Michaelis constant Km, the catalytic constant Kcat and the catalytic efficiency Kcat/Km. The Km represents the affinity of the substrate toward the enzyme. The Kcat represents the maximum number of molecules of substrate converted per unit of time and per molecule of enzyme. The Kcat/Km is a constant that measures how efficiently an enzyme is able to convert its substrate into product. Determining the kinetic parameters associated to the conversion of DMS into marmesin by CYP76F112 has proved to be a challenging task. Indeed, some preliminary experiments showed that the Km value associated to the conversion of DMS by CYP76F112 was lower than 0.1 μM, which was below the detection limit of the UHPLC-MS I could use. Therefore, I had to develop a specific protocol to overcome this technical limit. This protocol is detailed in the materials and methods section (Chapter VII, B.2.f.3). In brief, because of the need to maintain the incubations at a temperature of 27°C and incubated in presence of DMS and NADPH. All incubations were repeated in triplicates and analysed by UHPLC MS. The peak areas corresponding to the formation of marmesin were mea sured at 315nm. A The incubations were performed at pH=7, with variable temperatures. B. The incubations were performed at 27°C, in buffer solutions of variable pH. agitated, I had to perform the reactions in 2 mL tubes, and was limited to 1 mL of reaction mix at a time. Consequently, for every modality, I performed 5 to 15 incubations of 1 mL, pooled them together and concentrated them in a final volume of 100 μL (for a concentration factor of 50 to 150). The concentrated solutions were then analysed by UHPLC-MS. This adapted protocol allowed me to test DMS concentrations as low as 10 nM. To make sure that the conversion of the substrate did not exceed 50%, even for the smaller concentrations, I also had to dilute the microsomes and to reduce the incubation time – which required many preliminary experiments (not shown).
In addition, in order to calculate the Kcat, I also needed to know the amount of active CYP76F112 added in every incubation mix. For this purpose, I dosed the P450s present in the CYP76F112 microsomal solution by using the differential CO spectrum method. The spectrum I obtained showed a characteristic peak around 450nm, which indicates the presence of functional P450s in the solution (Figure 30). The height of this peak allowed me to calculate the concentration of functional CYP76F112 present in the microsomal solution used for the kinetic assays and therefore the quantity of active enzyme added in every incubation (Figure 30). The details of the calculation can be found in the materials and methods section (Chapter VII, B.2.e).

Table of contents :

CHAPTER I STATE OF ART
I. JOURNEY TO THE CENTRE OF PLANT DEFENCE
A. THROUGH THE LOOKING-GLASS: OPEN YOUR EYES, DON’T BE PLANT BLIND!
B. THE DEFENCE IN OUR PLANTS: FROM MARTIAL STRATEGIES TO THE COST OF WAR
B.1. Defence starts before the establishment of defensive mechanisms
B.1.a. Various enemies imply various defences
B.1.b. How to realise you are under attack?
B.1.c. Signalling pathways to trigger defences
B.2. Various defence mechanisms to respond to various threats
B.2.a. Indirect defences: asking for help?
B.2.b. Direct defences: physical mechanisms to build a reinforced armour
B.2.c. Direct defences: chemical mechanisms to concoct poison pills
B.2.c.1. Defensive proteins
B.2.c.2. Specialised metabolites
B.3. The cost of defence: growing or fighting?
B.3.a. Defence is a resource-consuming process
B.3.b. The multiple theories of trade-off between growth and defence
C. FEED ME KILL ME: AGRONOMICAL INTEREST AND DEFENCE OF FICUS CARICA
C.1. The fig tree, a plant of agronomical interest
C.2. Ficus defence mechanisms
C.2.a. Tough mineralised leaves and specialised tissues
C.2.b. Ficus defensive proteins and metabolites
D. SMELLS LIKE TOXIC FURANOCOUMARINS
D.1. Furanocouma-what?
D.1.a. Furanocoumarins constitute one of the four classes of coumarins
D.1.b. Distribution of furanocoumarins in higher plants
D.1.c. Repartition and variation of furanocoumarins within plants
D.2. Defensive properties of furanocoumarins
D.2.a. Toxic properties and P450 inhibition
D.2.b. Phototoxic properties and phytophotodermatitis
D.2.c. From defensive properties to interesting bioactivities
D.3. Consequences of these defensive properties for the plants themselves
D.3.a. Emergence of linear and angular furanocoumarins
D.3.b. Defensive properties and spatio-temporal variation within plants
D.4. The furanocoumarins in the Moraceae family
D.4.a. Furanocoumarins described in various Ficus species
D.4.b. Repartition of furanocoumarins in Ficus carica
E. THE FURANOCOUMARIN PATHWAY MUST GO ON
E.1. The different steps of the furanocoumarin biosynthesis pathway
E.2. Evolutionary perspectives and concluding remarks
II. ENDLESS P450S MOST BEAUTIFUL
A. INTRODUCTION: ALL YOU NEED IS A P450
B. A BRIEF HISTORY OF P450S: RESEARCHES AND DISCOVERIES OVER TIME
C. ONE CLASSIFICATION TO NAME THEM ALL
D. IF YOU PLEASE DRAW ME A FUNCTIONAL P450: FROM STRUCTURE TO ACTIVITY
D.1. P450s: from the primary to the secondary and tertiary structures
D.1.a. Discovery of the P450 fold
D.1.b. The “P450 fold”: overall architecture
D.1.c. Structurally conserved regions
D.1.d. Substrate Recognition Sites
D.1.d.1. Discovery and description of the SRSs
D.1.d.2. SRSs, a hot spot for site-directed mutagenesis
D.2. Catalytic activity of P450s
D.2.a. P450 redox partner systems
D.2.b. Catalytic cycle of a typical class II P450
D.3. Atypical P450s and unusual P450-mediated reactions
E. HIGHWAY TO PHYTOCHEMISTRY: P450S’ IMPORTANCE IN THE PLANT KINGDOM
E.1. In plants, P450s constitute one of the largest gene superfamily
E.2. The diversification of P450s is at the heart of plant chemical diversity
F. ON THE ORIGIN OF P450 GENES: EVOLUTIVE STORY OF P450S IN LAND PLANTS
F.1. Early emergence of all plant P450 clans
F.2. Subsequent diversification of plant P450 families and subfamilies
F.3. The diversification of P450s families: different functions, different patterns
F.3.a. Plant-specific P450s and the conquest of land
F.3.b. P450s involved in plant specialised metabolism
F.3.b.1. Evolutionary patterns
F.3.b.2. Two examples of lineage-specific evolution
F.6. Summary of plant P450 evolution
III. OBJECTIVE AND APPROACH OF THE PHD
CHAPTER II WHOLE NEW GENES
A. INTRODUCTION AND STRATEGY: IN SEARCH OF THE LOST GENE
B. CANDIDATE GENES AND HOW TO FIND THEM
B.1. APPROACH: WHERE TO SEARCH, WHAT TO SEARCH?
B.2. IN SILICO SCREENING OF THE F. CARICA RNA-SEQ LIBRARY
C. I’LL MAKE AN ENZYME OUT OF YOU: CLONING AND EXPRESSION OF THE P450S
C.1. AMPLIFICATION, CLONING AND SEQUENCING OF THE P450 CANDIDATES
C.2. HETEROLOGOUS EXPRESSION OF THE P450 CANDIDATES
D. CONVERT THIS AND I’LL LOVE YOU: ENZYME ASSAYS AND CHARACTERISATION
D.1. ENZYME ASSAYS AND FUNCTIONAL SCREENING
D.1.a. CYP76F112, a marmesin synthase
D.1.b. CYP82J18, a P450 that hydroxylates auraptene
D.1.c. CYP81BN4, a P450 that hydroxylates cnidilin
D.2. FUNCTIONAL CHARACTERISATION OF CYP76F112, CYP82J18 AND CYP81BN4
D.2.a. CYP76F112: an enzyme with a strong specificity and affinity
D.2.a.1. Substrate specificity
D.2.a.2. Optimal enzymatic conditions
D.2.a.3. Kinetic parameters: affinity, catalytic constant and catalytic efficiency
D.2.b. The cases of CYP82J18 and CYP81BN4
E. DISCUSSION: ANOTHER P450 IN THE PATHWAY
E.1. AN APPROACH THAT PROVED ITSELF EFFECTIVE
E.2. PHYSIOLOGICAL OR PROMISCUOUS ACTIVITIES?
E.2.a. CYP76F112: a physiological marmesin synthase activity?
E.2.b. CYP81BN4 and CYP82J18: promiscuous activities?
E.3. NEW P450 FAMILIES INVOLVED IN THE FURANOCOUMARIN PATHWAY
E.3.a. The CYP76 family: from terpenoids to furanocoumarins?
E.3.a.1. General overview of the CYP76 family
E.3.a.2. Focus on the CYP76F subfamily and the marmesin synthase
E.3.b. The cases of CYP81BN4 and CYP82J18
E.4. CONCLUSION AND FUTURE PERSPECTIVES
CHAPTER III ONCE UPON A P450
A. BACK TO THE PAST: INTRODUCTION AND STRATEGY
A.1. OBJECTIVE AND STRATEGY
A.2. PREREQUISITE: HOW DO GENES EVOLVE AND DIVERSIFY?
A.2.a. Gene duplication
A.2.b. The fate of duplicated genes
A.2.c. All duplicates are not created equal
B. DATA MINING: THE P450S COMING OUT OF THE NITROGEN FIXING CLADE
B.1. APPROACH: FROM FICUS TO THE NITROGEN FIXING CLADE
B.2. CONSTITUTION OF THE DATASETS
C. INFERRING PHYLOGENIES: THE REALM OF THE ELDER GENES
C.1. EXPANSION AND DIVERSIFICATION OF CYP76FS
C.1.a. Gene-family phylogeny of the CYP76Fs in the Nitrogen Fixing Clade
C.1.b. Evolution of CYP76Fs’ SRSs
C.2. EVOLUTIONARY PATTERNS OF CYP81BNS ACROSS THE NITROGEN FIXING CLADE
C.3. CONSERVATION OF THE CYP82J SUBFAMILY ACROSS THE NITROGEN FIXING CLADE
D. DISCUSSION: THE STORY O’ MY P450S
D.1. EVOLUTIVE STORY OF CYP76F112, THE F. CARICA MARMESIN SYNTHASE
D.1.a. Evolution of the CYP76 family and emergence of CYP76F112
D.1.b. Clustering of CYP76Fs
D.1.c. Multiple origin of the furanocoumarin pathway in higher plants
D.1.d. The CYP76Fs: a marmesin synthase activity specific of the Ficus genus?
D.2. EVOLUTIVE STORY OF CYP81BN4
D.3. EVOLUTIVE STORY OF CYP82J18
D.4. CONCLUSION AND FUTURE PERSPECTIVES
CHAPTER IV IN THE ACTIVE SITE OF THE MARMESIN SYNTHASES
A. RISE OF THE MARMESIN SYNTHASES: INTRODUCTION AND STRATEGY
B. THE MOLECULAR SHAPE OF YOUR P450: 3D MODELLING AND DOCKING
B.1. HOMOLOGY MODELLING OF CYP76F111 AND CYP76F
B.1.a. Approach: the homology modelling technique
B.1.b. Building the 3D models of CYP76F111 and CYP76F112
B.2. DOCKING EXPERIMENTS WITHIN CYP76F111 AND CYP76F112
B.2.a. Approach: the docking technique
B.2.b. Docking of the heme within CYP76F111 and CYP76F112
B.2.c. Docking of the DMS within CYP76F111 and CYP76F112
B.2.c.1. Approach: definition of the grid receptor and the flexible residues
B.2.c.2. Docking of the DMS within CYP76F112
B.2.c.3. Docking of the DMS within CYP76F111
C. FINDING KEY AMINO ACIDS INFLUENCING THE DOCKING OF THE DMS
C.1. COMPARISON OF THE DOCKING SITES OF CYP76F111 AND CYP76F112
C.2. INFLUENCE OF THE VARIABLE AMINO ACIDS DURING THE DOCKING OF THE DMS
C.2.a. Preliminary results: identification of 4 key amino acids
C.2.b. Simultaneous modification of the 4 amino acids A, B, C and D
C.2.c. Individual modifications of A, B, C and D
C.2.d. Simultaneous modifications of A and B
D. SITE-DIRECTED MUTAGENESIS: 4 AMINO ACIDS, AND NOTHING ELSE MATTERS?
D.1. THE CHOICE OF THE MUTANTS
D.2. SYNTHESIS AND EXPRESSION OF THE MUTANTS
D.3. FUNCTIONAL CHARACTERISATION: INFLUENCE OF THE 4 AMINO ACIDS
E. GOTTA DOCK THEM ALL: ADDITIONAL DOCKINGS WITH THE F112-LIKE
F. DISCUSSION: A SINGLE AMINO ACID IS MISSING, AND ALL BEGINS ANEW
F.1. CYP76F111 AND CYP76F112: RELIABLE MODELLING AND ACCURATE DOCKING
F.1.a. Overall architecture of CYP76F111 and CYP76F112
F.1.b. Confronting the in-silico dockings with the in-vitro experiments
F.2. THE IMPACT OF THE RESIDUES A, B, C AND D ON THE MARMESIN SYNTHASE ACTIVITY
F.2.a. Key residues from the SRSs
F.2.b. The residues A and B might stabilise the DMS in the active site
F.2.c. The residue D might stabilise the DMS in the active site
F.2.d. The residue C contribute to shape the substrate-docking site
F.2.e. Summary of the influence of the residues A, B, C and D
F.2.f. The residue C: a hotspot position?
F.2.g. The limits of the models and the importance of the access channel
F.3. RECENT AMINO ACIDS FOR A RECENT MARMESIN SYNTHASE ACTIVITY
CHAPTER V THE COST OF FURANOCOUMARINS
A. INTRODUCTION: DO TOMATOES DREAM OF TOXIC FURANOCOUMARINS?
B. BRICK BY BRICK: THE GOLDENBRAID MULTI-GENIC CONSTRUCTIONS
B.1. APPROACH: WHICH PLASMID TO CONSTRUCT, AND HOW?
B.1.a. Overall strategy
B.1.b. Presentation of the GoldenBraid cloning system
B.1.c Using the GoldenBraid technology, or how to construct the desired plasmid
B.2. CONSTRUCTION OF THE PLASMID
C. THE TOMATOES OF EVIL: TOMATO TRANSFORMATION AND REGENERATION
C.1. APPROACH AND CHOICE OF THE NEGATIVE CONTROL
C.2. GENERATION OF THE TRANSGENIC TOMATOES
C.2.a. Tomato transformation and regeneration
C.2.b. Confirmation of the transformations
D. DISCUSSION: GET A BETTER PLASMID, DON’T GIVE UP THE TRANSFORMATIONS
D.1. SUMMARY OF THE PRELIMINARY RESULTS
D.2. LIMITS, HYPOTHESES AND RECOMMENDATIONS FOR A FUTURE CONTINUATION
D.2.a. A matter of size?
D.2.b. Reordering the transcriptional units
D.2.c. Avoiding the repetitive use of identical promoters and terminators
D.2.d. Using inducible instead of 35S promoters
D.2.e. Additional TUs to prevent autotoxicity?
CHAPTER VI GENERAL CONCLUSION AND PERSPECTIVES
A. INTO THE UNKNOWN STEPS OF THE FURANOCOUMARIN PATHWAY
A.1. CYP76F112: A RECENT MARMESIN SYNTHASE THAT OPENS MANY PROSPECTS
A.1.a. The marmesin synthase activity
A.1.b. The furanocoumarin pathway, a case of convergent evolution
A.2. CYP81BN4 AND CYP82J18: PROMISCUOUS AND NON-SPECIES-SPECIFIC ENZYMES?
A.3. NEW PROSPECTS TO PURSUE THE ELUCIDATION OF THE FURANOCOUMARIN PATHWAY
A.3.a. A complete genome for Ficus carica
A.3.b. Finding the ancestral substrate of the Ficus CYP76Fs
A.3.c. Other Ficus CYP76Fs potentially involved in the furanocoumarin pathway
A.3.d. Other P450 families potentially involved in the furanocoumarin pathway
A.3.e. The marmesin synthases in other plant families
A.3.f. Other enzymes families: Ficus methyltransferases and dioxygenases
A.3.g. Application and study of plant biosynthesis pathways
B. TOO MUCH FURANOCOUMARINS WILL COST YOU
CHAPTER VII MATERIALS AND METHODS
A. MATERIALS
A.1. PLANT MATERIAL
A.1.a. Ficus carica
A.1.b. Solanum lycopersicum
A.2. BACTERIAL STRAIN
A.2.a. Escherichia coli MC1022
A.2.b. Escherichia coli ccdB Survival™
A.2.c. Agrobacterium tumefaciens EHA105
A.3. YEAST STRAIN: SACCHAROMYCES CEREVISIÆ WAT21
A.4. VECTORS
A.4.a. pCR™8/GW/TOPO™
A.4.b. pYeDP60 and pYeDP60_GW®
A.4.c. GoldenBraid commercial vectors
A.4.c.1. The pUPD vectors: pUPD, pUPD-35S and pUPD-tNOS
A.4.c.2. The α-level vectors: pDGB1_α1 and pDGB1_α2
A.4.c.3. The Ω-level vectors: pDGB1_Ω1 and pDGB1_Ω2
A.4.d. Recombinant GoldenBraid vectors
A.4.d.1. pDGB1_Ω1 [PsDiox+PsPT1]
A.4.d.2. pUPD-CYP71AJ3
A.4.e. pSoup
A.4.f. plCSL11024 vector
A.5. CULTURE MEDIA
A.5.a. Bacteria culture medium: LB medium and associated antibiotics
A.5.b. Yeast culture media
A.5.c. Tomato in vitro culture media
A.6. BIOINFORMATIC TOOLS
A.6.a. Databases
A.6.a.1. Ficus carica RNA-seq library
A.6.a.2. Public online databases
A.6.b. Software
A.6.b.1. Software used for molecular biology and basic sequence analyses
A.6.b.2. Software used for phylogenetic analyses
A.6.b.3. Software used for modelling and docking
A.6.c. Online tools
B. METHODS
B.1. COMMON MOLECULAR BIOLOGY AND MICROBIOLOGY METHODS
B.1.a. Plant tissue grinding
B.1.b. Extraction and purification of plant RNA
B.1.c. Synthesis of complementary DNA
B.1.d. Extraction of plant genomic DNA
B.1.e. Amplification of DNA fragments by PCR
B.1.e.1. PrimeSTAR® Max DNA Polymerase
B.1.e.2. SapphireAmp® Fast PCR Master Mix 2X
B.1.f. DNA extraction from agarose gel
B.1.g. DNA digestion using restriction enzymes
B.1.h. Cloning techniques
B.1.h.1. Cloning of a PCR-amplified fragment in pCR™8/GW/TOPO™
B.1.h.2. Recombination into the pYeDP60_GW® vector
B.1.i. Preparation of electrocompetent bacteria
B.1.i.1. Preparation of electrocompetent Escherichia coli
B.1.i.2. Preparation of electrocompetent Agrobacterium tumefaciens EHA105
B.1.j. Transformation of competent bacteria
B.1.k. Isolation of plasmid DNA from bacteria
B.1.l. Spectrophotometry quantification
B.1.m. Sequencing
B.1.n. Synthesis of the CYP76F mutants
B.2. METHODS LINKED TO P450 HETEROLOGOUS EXPRESSION, ASSAY AND CHARACTERISATION
B.2.a. Yeast transformation
B.2.a.1. Preparation of competent S. cerevisiae WAT21
B.2.a.2. Transformation of S. cerevisiae WAT21
B.2.b. Isolation of plasmid DNA from yeast
B.2.c. Heterologous expression of P450s in S. cerevisiae
B.2.c.1. Yeast culture and P450 expression
B.2.c.2. Preparation of yeast microsomes
B.2.d. Western-Blotting: confirmation of the presence of the P450s of interest
B.2.d.1. Polyacrylamide gel electrophoresis in denaturing conditions
B.2.d.2. Transfer of the proteins to a polyvinylidene difluoride membrane
B.2.d.3. Immunodetection
B.2.e. Quantification of functional P450s with the differential CO spectrum method
B.2.f. Enzymatic assay and functional characterisation
B.2.f.1. Functional screening
B.2.f.2. Determination of the optimal conditions: temperature and pH
B.2.f.3. Determination of the kinetic parameters
B.3. METABOLIC ANALYSES
B.3.a. Extraction of phenolic compounds from plant grinded sample
B.3.b. UHPLC-MS analyses
B.3.c. Orbitrap-IDX analyses
B.4. THE GOLDENBRAID CLONING TECHNIQUE
B.4.a. Domestication of the genes of interest
B.4.a.1. Domestication of CYP76F112
B.4.a.2. Domestication of the KanaR gene
B.4.b. Cloning into the pUPD vector
B.4.c. Assembly of simple transcriptional units
B.4.d. Repeated assembly of multiple transcriptional units
B.4.d.1. Assembly of two transcriptional units: CYP76F112 and CYP71AJ3
B.4.d.2. Assembly of four transcriptional units: PsDiox, PsPT1, CYP76F112 and CYP71AJ3
B.4.d.3. Assembly of 5 transcriptional units to construct the final 5-TUs plasmid
B.5. TOMATO STABLE TRANSFORMATION AND REGENERATION
B.5.a. Preparation of the tomatoes to be transformed
B.5.a.1. Sterilisation of the tomato seeds
B.5.a.2. Germination of the sterile tomato seeds
B.5.a.3. Preparation of the cotyledons
B.5.b. Preparation of the Agrobacterium suspension to transform the tomatoes
B.5.b.1. Co-transformation of the Agrobacterium
B.5.b.2. Preparation of the Agrobacterium suspension
B.5.c. Transfection of the cotyledon fragments
B.5.d. Regeneration and selection of the transgenic cotyledons
B.5.e. Rooting of the transgenic plantlets
B.5.f. Transfer into the soil and growth of fully developed tomato plants
B.6. BIOINFORMATIC ANALYSES
B.6.a. Identification of P450 candidates from the F. carica RNA-seq library
B.6.b. Phylogenetic analyses
B.6.b.1. Constitution of the CYP76F, CYP81BN and CYP82J initial datasets
B.6.b.2. Sequence alignment
B.6.b.3. Phylogenetic analyses
B.6.c. Modelling and docking analyses
B.6.c.1. Homology modelling of the CYP76Fs
B.6.c.2. Docking experiments
RÉSUMÉ DÉTAILLÉ EN FRANÇAIS
A. ETAT DE L’ART ET OBJECTIF DE LA THESE
A.1. Défense des plantes et furocoumarines
A.2. Les cytochromes P450s
A.3. Objectifs de la thèse
B. IDENTIFICATION DE GENES IMPLIQUES DANS LA VOIE DE BIOSYNTHESE DES FUROCOUMARINES
B.1. APPROCHE, CHOIX DE LA PLANTE MODÈLE ET DES FAMILLES ENZYMATIQUES D’INTÉRÊT
B.2. IDENTIFICATION, CLONAGE ET EXPRESSION HÉTÉROLOGUE DES P450 CANDIDATS
B.3. CRIBLAGE FONCTIONNEL ET CARACTÉRISATION ENZYMATIQUE
B.4. DISCUSSION
C. ANALYSE PHYLOGENETIQUE DE CYP76F112, CYP82J18 ET CYP81BN4
C.1. APPROCHE, CONSTITUTION DES JEUX DE DONNÉES ET CONSTRUCTION DES ARBRES
C.2. CYP76F112 : ANALYSE PHYLOGÉNÉTIQUE ET DISCUSSION
C.3. LES CAS DE CYP81BN4 ET CYP82J18
D. EMERGENCE DE L’ACTIVITE MARMESINE SYNTHASE
D.1. APPROCHE
D.2. MODÉLISATION ET EXPÉRIENCE DE DOCKING MOLÉCULAIRE
D.3. MUTAGENÈSE DIRIGÉE, INFLUENCE DES ACIDES AMINÉS ET DISCUSSION
D.4. PERSPECTIVES ÉVOLUTIVES ET APPARITION DE L’ACTIVITÉ MARMÉSINE SYNTHASE
E. RECONSTITUTION DE LA VOIE DES FUROCOUMARINES DANS LA TOMATE
E.1. APPROCHE GLOBALE
E.2. CONSTRUCTION D’UN PLASMIDE MULTIGÉNIQUE
E.3. GÉNÉRATION DE TOMATES TRANSGÉNIQUES
E.4.DISCUSSION
F. CONCLUSION GENERALE ET PERSPECTIVES
REFERENCES

GET THE COMPLETE PROJECT