From Genes to Molecules, Secondary Metabolism in Botrytis cinerea: New Insights into Anamorphic and Teleomorphic Stages

The ascomycete Botrytis cinerea Pers. Fr., classified within the family Sclerotiniaceae, is the agent that causes grey mould disease which infects at least 1400 plant species, including crops of economic importance such as grapes and strawberries. The life cycle of B. cinerea consists of two phases: asexual (anamorph, Botrytis cinerea Pers. Fr.) and sexual (teleomorph, Botryotinia fuckeliana (de Bary) Wetzel). During the XVI International Symposium dedicated to the Botrytis fungus, which was held in Bari in June 2013, the scientific community unanimously decided to assign the most widely used name of the asexual form, Botrytis, to this genus of fungi. However, in the literature, we continue to find articles referring to both morphic stages. In this review, we take stock of the genes and metabolites reported for both morphic forms of B. cinerea between January 2015 and October 2022.


Introduction
The Botrytis genus includes a wide variety of pathogenic fungal species found throughout the world [1,2]. Botrytis species are characterised by grey mycelium and saprotrophic behaviour. They are necrotrophic pathogens responsible for heavy losses among many economically important horticultural and floral crops. Most Botrytis species are specialised pathogens with a narrow host range, generally infecting only one or a few closely related species within a single plant genus [3]. However, Botrytis cinerea is a polyphagous species that infects a wide range of host plants.
Micheli in 1792 characterised the species B. cinerea for the first time, which was later confirmed by Persoon in 1801 [4][5][6]. Since then, a large number of species have been described thanks to progress in molecular genetics and the development of the relevant phylogenetic markers. Today, approximately 30 different Botrytis species have been identified. Seven new species, a hybrid, and a species complex have been isolated and identified over the last decade [7].
B. cinerea is the agent that causes grey mould disease which affects a total of 586 genera of vascular plants, representing over 1400 ornamental and agriculturally important plant species [8,9]. This fungus is a typical necrotroph whose infective cycle includes the killing of plant cells, maceration of plant tissue and then reproduction by forming asexual spores on the decomposing plant material.

From Genes to Molecules
Although the identification of the genes involved in the biosynthetic pathways to botrydial (1) and abscisic acid (ABA) (7) started already two decades ago [26][27][28], the sequencing of the B. cinerea genome provided a complete picture of all genes involved in the biosynthesis of secondary metabolites (SMs) [29,30] and aided research on the genetic determinants of SM production.
Genomic data has revealed that the plant pathogenic fungus B. cinerea appears to have seven STC (Bcstc) genes that encode proteins with the typical two magnesium binding sites required for farnesyl diphosphate (FDP) (14) cyclisation. Four STCs have been functionally characterised to date: Bcstc1 [33], Bcstc5 [34], Bcstc7 [32] and the rare new class of STC as an Open Reading Frame (ORF), the Bcaba3 gene, involved in abscisic acid (ABA) biosynthesis [35].

Botrydial Gene Cluster
As already indicated, Botrydial (1) is a phytotoxic metabolite [13,14] which reproduces the symptoms of the Botrytis infection [15]. This compound is considered to be one of the main toxins of B. cinerea [13,18] and it was one of the first metabolites to be isolated from this fungus.
The botrydial biosynthetic gene cluster has been identified ( Figure 2) [27,33,36]. The cluster consists of seven genes (Bcbot1 to Bcbot7) (for Botrytis cinerea botrydial biosynthesis) coding for a sesquiterpene cyclase (Bcbot2), an acetyltransferase (Bcbot5) and three monooxygenases (Bcbot1, Bcbot3 and Bcbot4), typical of secondary metabolism in filamentous fungi. Subsequently, a transcription factor, Zn(II) 2 Cys 6, (Bcbot6), and a gene encoding a dehydrogenase (Bcbot 7) were reported [37]. The Bcstc1 gene, also named Bcbot2, is responsible for the corresponding cyclisation step of FDP (14) in the biosynthetic route to botrydial (1). Targeted deletion of the Bcbot2 gene therefore stopped production of botrydial (1) and all related probotryane metabolites. Direct evidence for the biochemical function of Bcbot2 came from the demonstration that recombinant BcBot2 protein converted farnesyl diphosphate (FDP) (14) to the parent tricyclic alcohol of presilphiperfolan-8-ol (PSP) or probotryan-9β-ol. The structures and numbering systems for the presilphiperfolanes and botryanes were established before their biosynthetic relationships were known.
The process continues with the involvement of an acyltransferase and three cytochrome P450 monooxygenases encoded by genes Bcbot 1, 3 and 4 [27,36]. The high degree of homology observed between the amino acid sequences obtained from Bcbot5 and acetyltransferases suggests that this gene encodes the enzyme responsible for the introduction of the acetyl group in probotryanes in the final stages of biosynthesis [36]. The gene Bcbot6 encodes the transcription factor Zn(II) 2 Cys 6 , which is responsible for the regulation of the entire toxin gene cluster, while Bcbot7 produces a dehydrogenase that may be involved in the conversion of botrydial (1) into dihydrobotrydial (2) [37] (Figure 2). The identification of Bcbot6 as the main regulator of botryane synthesis is the first step towards a more comprehensive understanding of the whole regulation network of botrydial (1) and relative biosynthesis of its ecological role in the B. cinerea life cycle [37].
2.1.2. ABA Gene Cluster ABA (7) is a plant hormone that plays an important role in many aspects of plant growth and development and in the initiation of adaptive responses to various environmental conditions [38,39]. ABA (7) is mainly produced by plants but is also produced as a secondary metabolite by several species of filamentous fungi such as the genera Botrytis, Penicillium, Cercospora and Rhizoctonia [40]. Studies have shown that the ABA biosynthetic pathway differs between plants and fungi [41]. The understanding of the molecular mechanism driving ABA biosynthesis in fungi remains limited, especially for the steps from FDP (14) to ABA (7) [42].
In 2006, the biosynthetic gene cluster (BcABA), consisting of four putative enzyme genes (Bcaba1-Bcaba4), was identified in B. cinerea. Bcaba1 and Bcaba2 are cytochrome P450s. Bcaba3 shows no homology to functionally characterised enzymes while Bcaba4 is similar to a short-chain dehydrogenase/reductase ( Figure 3A) [28]. Knock-out experiments confirmed that Bcaba1, 2 and 3 are essential for ABA production in B. cinerea, whereas Bcaba4 is involved in the pathway but is not essential [26,28]. It was hypothesised that the cyclisation of FDP (14) requires a sesquiterpene cyclase (STC); however, none of the proteins in the gene cluster exhibited known STC motifs. A study by Izquierdo-Bueno et al. identified an STC gene called Bcaba5, which is co-expressed with, but not located in, the gene cluster [34]. Five STC-coding genes were identified thanks to the complete genome sequencing of the ABA-producing strain ATCC58025. Among them, Bcstc5 exhibits an expression profile coinciding with ABA production. Gene inactivation, complementation and chemical analysis demonstrated that BcStc5/BcAba5 is the key enzyme responsible for the key step of ABA biosynthesis in fungi, located on chromosome 1 in B. cinerea strain B05.10 ( Figure 3B). This gene is involved in the cyclisation of FDP (14) into 2Z,4E-α-ionylidene-ethane ( Figure 4). Hence, an ABA cluster formed by five genes is proposed [34]. Nevertheless, Takino et. al. [35] suggested that the Bcaba3 gene participates in skeletal construction for the formation of ABA (7), identifying this gene as an α-ionylideneethane synthase and thus revealed a three-step reaction mechanism through a series of biotransformation experiments, heterologous expression experiments, and in vitro enzymatic studies. They also showed that ABA (7) could be produced heterologously in an Aspergillus oryzae strain expressing Bcaba1, 2, 3 and 4 [35]. On the other hand, Otto et. al. [43] constructed and characterised an ABA-producing S. cerevisiae strain using the ABA biosynthetic pathway from B. cinerea, expression of the B. cinerea genes Bcaba1, 2, 3 and 4 being sufficient to establish ABA production in the heterologous host [43]. The contradicting results of these two studies raise the question as to whether Bcaba3 and Bcaba5 catalyse the same reaction. Consequently, the gene involved in the cyclisation of FDP (14) for the formation of ABA (7) is still unknown.
In contrast, the most recent studies on ABA (7) have shown that the putative methyltransferase LaeA/LAE1 plays an important role in the regulation of ABA biosynthesis in B. cinerea. The deletion of Bclae1 caused a 95% reduction in ABA yields, accompanied by a decrease in the transcriptional level of the ABA synthesis gene cluster Bcaba1-4. Further RNA-seq analysis indicated that the deletion of Bclae1 also affected the expression level of key enzymes of botcinic acid (BOA) and botryanes (BOT) in secondary metabolism and accompanied clustering regulatory features showing that this gene is a global regulator involved in the biosynthesis of a variety of secondary metabolites in filamentous fungi [44].

Eremophilenol Genes
A new family of cryptic metabolites with a (+)-4-epi-eremophil-9-en-11-ol skeleton was biosynthesised by B. cinerea when sublethal doses of CuSO 4 (5 ppm) were added to the culture media [45,46]. Studies on the in vitro evaluation of the biological role of these metabolites showed their involvement in the self-regulation of asexual spore production and enhanced the production of complex appressoria (infection cushions). This fact indicated for the first time the participation of sesquiterpenoid metabolites in the regulation of infective structures. Moreover, these metabolites possess an enantiomeric carbon skeleton resembling that of phytoalexin capsidiol suggesting that eremophilenols may be effectors that inhibit plant defences or modulate plant immunity to enhance the infection process [46].
PacBio technology, and the resulting update of the Ensembl Fungi (2017) database in the genome sequence, was instrumental in the identification of new possible genes that could be involved in secondary metabolism. As a result, a new STC coding gene (Bcstc7) has been included in the gene list from this plant pathogenic fungus (Table 1). An expression study of B. cinerea genes that encode sesquiterpene cyclases (Bcstc1−5 and Bcstc7) showed that gene Bcstc7 produced a more important level of expression when compared with the other STC-encoding genes. Metabolomic characterisation showed that ∆Bcstc7 was impaired in the production of eremophilenol derivatives while the complemented transformant compl ∆Bcstc7 niaD recovered its capacity to produce this family of compounds, demonstrating that this gene is the key enzyme responsible for the cyclisation of FDP (14) to eremophil-9-en-11-ols ( Figure 4) [32].
The sesquiterpene cyclases STC2, STC3, STC4 and STC6 of the B. cinerea fungus remain uncharacterised to date. Nevertheless, gene Bcstc6 (ID P020710.1) is specific to the T4 strain and absent in many of the other strains studied, including B05.10 [29].

Retinal and Carotenoid Gene Cluster
Diterpene cyclases catalyse the cyclisation of the linear 20-carbon substrate geranylgeranyl diphosphate to produce the diterpene scaffold occurring via a carbocation cascade. This ionisation-dependent reaction is catalysed by class I terpene synthases [47,48].
The diterpene cyclases DTC1, DTC2, DTC3 and PAX1 of the B. cinerea fungus remain uncharacterised to date. However, B. cinerea also contains a Terpene Synthase (TS) encoding gene (PAX1), orthologous to the Penicillium paxilli PaxC gene, involved in the biosynthesis of the indole-diterpene paxillin [29,49]. It is therefore very likely that this gene is involved in the production of a yet to be discovered indole-terpene for this phytopathogen.
Studies carried out by Schumacher et al. [50] on the transcription factor BcLTF1 have shown that its absence affects the expression of secondary metabolism-related genes. The expression patterns of only three key enzyme-encoding genes (Bcnrps2, Bcphs1, Bcstc5) were significantly affected by light. Notably, one of them corresponds to a cluster of lightinduced genes (Bcphs1, Bcphd1, Bccao1) that encode homologues of enzymes involved in the biosynthesis of retinal, the chromophore for opsin, in Fusarium fujikuroi (phytoene synthase, phytoene dehydrogenase, and carotenoid oxygenase). In that publication, the authors proposed a pathway for carotenoid biosynthesis and B. cinerea gene clusters for retinal biosynthesis organised in the same way as for F. fujikuroi ( Figure 5) [50]. There are no reports describing the isolation of carotenoids from B. cinerea. However, β-carotene was isolated from Sclerotinia spp. [51]. This research suggested that the Bcphs1 gene is related to the production of retinal diterpene in B. cinerea.

Botcinic Acid and Botcinins Gene Cluster
PKSs found in fungi, bacteria and plants are large megasynthases related to fatty acid synthases that biosynthesise small molecule polyketides with diverse natural functions and include well known secondary metabolites [52].
The iterative nature of fungal PKSs means that, in most cases, there is only one PKS involved in the synthesis of a particular polyketide. However, some fungal polyketides are known to be assembled by the action of two PKSs [53]. This is the case of botcinic acid and its relatives, botcinins, which require the action of two PKSs, BcBoa6 and BcBoa9 [53]. The expression study of the 20 BcPKS genes predicted from the genome sequence of strain B05.10 was studied at different physiological stages. During infection, only Bcpks6 and Bcpks9 exhibited higher levels of expression than the actin gene. The study of ∆Bcboa6 and ∆Bcboa9 null mutants did not produce botcinic acid or its derivatives, indicating that they act in concert to synthetise botcinic acid. Additionally, these authors proposed a cluster for B. cinerea botcinic acid biosynthesis made up of 17 putative biosynthetic genes (Bcboa1 to Bcboa17) ( Figure 6) [53]. Later studies showed that BcBoa13, a putative Zn 2 Cys 6 transcription factor, is a nuclear protein playing a major positive regulatory role in the expression of other Bcboa1-to-Bcboa12 genes and botcinic acid production [54].

Melanin Gene Cluster
It has been reported that the accumulation of 1,8-dihydroxynaphthalene (DHN) melanin is responsible for the pigmentation of the macroconidia and/or black sclerotia in B. cinerea. Schumacher carried out an interesting study which described the genetic basis and regulation of DHN melanogenesis in B. cinerea [55]. This author identified and functionally characterised the putative melanogenic and regulatory genes. Unlike other DHN melanin-producing fungi, B. cinerea and other Leotiomycetes contain two key (PKS)-encoding enzymes. Bcpks12 and bcpks13 are developmentally regulated and are required for melanogenesis in sclerotia and conidia, respectively ( Figure 7). Regulation of the melanogenic genes involves three pathway-specific transcription factors (TFs), BcSMR1, BcZTF1 and BcZTF2 ( Figure 7). These are clustered with bcpks12 or bcpks13 and other developmental regulators such as light-responsive TFs. Melanogenic genes are dispensable in vegetative mycelia for proper growth and virulence [55,56]. Recently, an exhaustive review about insights from genes studied with mutant analysis in B. cinerea has been published. The biosynthesis of DHN melanin pathways in B. cinerea has been reviewed [57].
Deletion of the Bcpks12 and Bcpks13 genes resulted in albino sclerotia and conidia in B. cinerea, indicating complete melanogenesis disruption [55,56] and studies so far indicate that melanisation does not significantly affect pathogenicity or fungal development [57].

Pyrones, Resorcylic Acids and Resorcinols
Another gene of this family that has been studied is the Bcchs1/Bcpks gene, which was characterised by Jeya et al. [58]. BPKS from B. cinerea is a novel type III polyketide synthase that accepts C4-C18 aliphatic acyl-CoAs and benzoyl-CoA as starters to form pyrones, resorcylic acids and resorcinols by sequential condensation with malonyl-CoA ( Figure 8). This PKS shows the highest catalytic efficiency ever reported for a long chain acyl-CoA ester [58].
The functionality of 14 PKSs in B. cinerea is still unknown, the only thing certain being that they will code for polyketides that are also unknown.

Siderophore Genes
Non-ribosomal peptide synthetases (NRPSs) are multi-modular enzymes, found in fungi and bacteria, which biosynthesise peptides without the aid of ribosomes. Bushley and Turgeon [59] identified genes (NPS) encoding NRPS and NRPS-like proteins in 38 fungal genomes and undertook phylogenomic analyses in order to identify fungal NRPS subfamilies, assess taxonomic distribution, evaluate conservation levels across subfamilies and address evolutionary mechanisms of multi-modular NRPSs. Phylogenomic analysis identified major subfamilies of fungal NRPSs which fall into two main groups: 1) a group of primarily mono/bi-modular enzymes containing the PKS-NRPSs and 2) a group of primarily multi-modular proteins, siderophore synthetases (SID) and Euascomycete-only synthetases (EAS) which appear both restricted to and highly expanded within fungi [59].
All fungal PKS-NRPS hybrids fall into a single, well supported, monophyletic group, which suggests a single origin. In B. cinerea, three PKS-NRPS hybrids, five EAS and four SID were identified. Of these latter enzymes, EAS and SID belong to the NRPS family. NRPS 2,3,7 are possibly involved in the production of secondary metabolites such as ferrichrome siderophores and NRPS 6 coprogene siderophore [59]. However, the production of secondary metabolites of 5 NRPS, and the 2 DMATS of B. cinerea, is still unknown.
The key enzymes related to secondary metabolism in B. cinerea identified to date are summarised in Table 1.
Recently, the genetic and molecular basis of botrydial biosynthesis have been described [36,37]. Genes Bcbot3 and Bcbot4 were deleted by homologous recombination and showed to catalyse regio-and stereospecific hydroxylation at the C-10 and C-4 carbons, respectively, of the probotryane intermediate skeleton (Figure 9) [36].
A detailed study of the ∆bcbot4 null mutant was undertaken in order to discover the metabolic fate of the PSP intermediate biosynthesised by B. cinerea after longer periods of fermentation [67]. This led to the identification of three new presilphiperfolanes (15)(16)(17) and three new cameroonanes (18)(19)(20) (Figure 10 (A)). The rearrangement to cameroonanes was facilitated by the absence of hydroxylation at C-11, whereas functionalisation at this position precludes this rearrangement. This could suggest that the interactions of the C-11 hydroxylated derivatives hinder the stereo-electronic requirements for the migration of the C-11:C-7 sigma bond to C-8 ( Figure 10 (B)) [67].
Another study examined the metabolism of botryane sesquiterpenoids of B. cinerea [68]. The study of metabolites with botryane and presilphiperfolane skeleton of the fungus B. cinerea has shed light on the biosynthesis of this family of sesquiterpenoids and has also led to potentially novel approaches to the selective control of this pathogen. An ecological role of the naturally occurring sesquiterpenoids in terms of their effect on the growth of these plant pathogens has been suggested [68].

Botrydial Applications
A phospholipid second messenger called phosphatidic acid (PA) is involved in the stimulation of plant defence mechanisms. It is produced by either phospholipase D (PLD) or by the concurrent activity of phospholipase C and diacylglycerol kinase (PLC/DGK), two different enzyme mechanisms. Through PLD and PLC/DGK, botrydial (1) causes the production of PA in a matter of minutes [69]. Both PLC and DGK inhibition reduce ROS production sparked by botrydial (1). This shows that PLC/DGK is upstream of ROS production. PLC is encoded in tomato by the multigene family SlPLC1-SlPLC6 and the pseudogene SlPLC7. It has been shown that plants that lack SlPLC2 are less vulnerable to B. cinerea. Additionally, by specifically engineering a microRNA to silence the expression of SlPLC2, it has been possible to investigate the impact of SlPLC2 on botrydial-induced PA generation. SlPLC2-silenced-cell suspensions generate PA levels comparable to those of wild-type cells after botrydial treatments. It is safe to say that PA is a novel element resulting from the responses that botrydial (1) causes in plants [69].
The function of botrydial (1) in the interaction between the phytopathogenic fungus B. cinerea and bacteria associated with plants was examined. Nine types of bacteria found in soil and phyllospheric samples were shown to be susceptible to growth-inhibition caused by B. cinerea. The lack of bacterial inhibition induced by B. cinerea mutants incapable of producing botrydial (1) demonstrated the inhibitory function of botrydial (1). Via taxonomic research, these bacteria were identified as belonging to several Bacillus species (six strains), Pseudomonas yamanorum (two strains) and Erwinia aphidicola (one strain). Bacillus amyloliquefaciens strain MEP 2 18 and WT, and B. cinerea mutants that do not produce botrydial, were inoculated together in soil to show that both microbes exert reciprocal inhibitory effects, B. cinerea's inhibition was dependent on botrydial production. Furthermore, the presence of B. amyloliquefaciens MEP 2 18 in in vitro confrontation assays the affected formation of botrydial (1). In turn, purified botrydial (1) prevented B. amyloliquefaciens MEP 2 18 from producing cyclic lipopeptide (surfactin) and Bacillus strains from developing in vitro. Overall, findings show that B. cinerea has the capacity to suppress potential biocontrol Bacillus genus bacteria due to botrydial (1). It has been suggested to include resistance to botrydial (1) among the criteria determining the choice of biocontrol agents for plant diseases brought on by B. cinerea [70].

New Polyketides from B. cinerea
As described in references [36] and [71], the bcbot4∆ mutant also overproduced a significant number of polyketides which included, in addition to known botcinins, botrylactones and cinbotolide A (9), two new botrylactones (21,22) and two cinbotolides, cinbotolides B (10) and C (23) (Figure 11) [36]. A subsequent detailed study of the polyketides produced by the null mutant bcbot4∆  The cluster of Bcboa genes responsible for botcinins biosynthesis was found to be specifically regulated [54], and it was discovered that this cluster is situated in a subtelomeric genomic region. According to genetic studies, BcBoa13, a putative Zn 2 Cys 6 transcription factor, is a nuclear protein that plays a major positive regulatory role in the expression of Bcboa1 through Bcboa12 genes and botcinic acid production [54]. Interestingly, the structure and regulation of the botcinic acid gene cluster share characteristics with the cluster responsible for the biosynthesis of the other known phytotoxin produced by B. cinerea, the sesquiterpene botrydial (1) [54]. This study demonstrated that the Zn 2 Cys 6 protein Bcboa13 positively controls the clustered Bcboa1-Bcboa12 genes y consequently the synthesis of botcinic acids and other botcinins in B. cinerea [54].

Abscisic Acid Biosynthetic Gene Cluster Studies
Abscisic Acid (ABA) (7) is a well-known hormone produced by plants through the carotenoid's pathway. Surprisingly, this sesquiterpene can also be produced by a small number of fungi including the plant pathogenic species B. cinerea [20]. However, the ABA biosynthetic pathway in fungi differs from the carotenoid pathway described in plants. Hence, in B. cinerea, it has been demonstrated that ABA (7) is obtained from the cyclisation of farnesyl diphosphate (FDP) (14) and subsequent oxidation steps. Inomata et al. therefore proposed a biosynthetic pathway that involves the transformation of FDP (14) to give 2Z,4E,6E-allofarnesene which is then cyclised to form 2Z,4E-α-ionylideneethane [72]. This intermediate is then subjected to several oxidative steps to form ABA (7).
However, most B. cinerea species produce low or even undetectable levels of ABA (7) in vitro. The first gene involved in the biosynthesis of abscisic acid (7) was identified by Siewers et al., using an over-producer isolate of B. cinerea [26]. The Bcaba1 gene was over-expressed in the presence of mevalonic acid in the medium. In addition, mutants deleted in the Bcaba1 gene were impaired in terms of ABA production. The genomic locus includes three other genes (Bcaba2-4) that codify for a P450 monooxygenase, a putative dehydrogenase/reductase, and an unknown protein. However, the neighbouring genomic region does not have a gene that codifies for a sesquiterpene cyclase. Targeted inactivation of the genes proved the involvement of Bcaba2 and Bcaba3 in ABA biosynthesis and suggested a contribution of Bcaba4 [28]. The close linkage of these four genes served as strong evidence for the presence of an abscisic acid gene cluster in B. cinerea.
Questions have arisen around why a plant hormone would be synthetised by a fungus and this is why a negative regulator of disease resistance through the down-regulation of defence response has been considered for ABA (7) [73]. The ABA (7) knock-out mutant proved to be more resistant to B. cinerea, swiftly fortifying the epidermal cell wall and exhibiting a higher induction of some genes involved with defence against B. cinerea. However, the knock-out mutant of the Bcaba1 gene for B. cinerea showed that it is not necessary for fungal virulence [74].
As already mentioned, three abscisic acid biosynthetic studies have been conducted in recent years, but have produced conflicting results. First of all, to identify the STC responsible for the biosynthesis of ABA (7) in fungi, a genomic approach to B. cinerea was taken by Izquierdo-Bueno et al. [34]. Inactivation of the Bcstc5/Bcaba5 gene in the B. cinerea ABA-overproducing strain ATCC58025 abolished ABA production. However, the complemented mutant restored the production of the sesquiterpene, demonstrating that the encoded STC was essential for ABA biosynthesis [34].
However, a more recent study by Takino et al. [35] showed that the four genes of the gene cluster, Bcaba1, 2, 3 and 4, are sufficient to produce ABA (7) in Aspergillus oryzae [35], thereby contradicting the finding of Izquierdo-Bueno et al. [34]. They also studied Bcaba3, a gene of the original gene cluster encoding an enzyme with hitherto unknown functions and no known motifs. In vitro assays showed that BcABA3 can convert FDP (14) to αionylideneethane [35]. Concerning the role of the Bcaba5 gene in the ABA biosynthetic pathway, Takino et al., suggested that this gene could be involved in ABA biosynthesis but is not essential. These studies were reinforced by a heterologous expression of the ABA biosynthetic cluster, Bcaba1-4, in S. cerevisiae, by Otto et al., highlighting the importance of the Bcaba3 gene in the biosynthesis of this metabolite [43]. This work confirmed the finding of Takino et al., i.e., that the four genes in the B. cinerea ABA gene cluster Bcaba1-4 are sufficient to produce ABA (7) and Bcaba5 is not essential. However, it is still possible that the co-expression of Bcaba5 enhances ABA production, as in the case of artemisinic acid biosynthesis [75]. Indeed, Bcaba5 and the monooxygenase BcceP450 are expressed during ABA production in B. cinerea indicating that they could be involved in the pathway [34]. Nonetheless, further analysis is needed to definitively confirm that BcABA5 and monooxygenase BcceP450 are not involved in ABA biosynthesis [34,35,43].
Interestingly, Takino et al. have discovered a brand-new class of sesquiterpene synthase as an Open Reading Frame (ORF) (BCIN_08g03880) called Bcaba3 and its unusual three-step reaction process involving two neutral intermediates called β-farnesene and allofarnesene. Database searches revealed that the homologous enzyme genes are present in more than 100 bacteria and that BcABA3 is not homologous with normal sesquiterpene synthases, indicating that these enzymes belong to a new family of sesquiterpene synthases [35].
Lastly, the function of the four biosynthetic genes Bcaba1-Bcaba4, found in B. cinerea through biotransformation experiments and in vitro enzymatic reactions, was elucidated.
One of them, BcABA2, is notable because it catalyses two rounds of allylic oxidation, the first oxidising the β-face of C4' and the second oxidising the α-face of C1 . The intermediate 30 also undergoes allylic oxidations which are also mediated by enzymes found naturally in A. oryzae. However, when the downstream oxidation enzyme genes, Bcaba2 and Bcaba4, were co-expressed with Bcaba1, this unexpected oxidation was unable to impair the production of 7, as the swift conversion of 30 into 32 stifled the side reaction. The fact that BcABA4 can accommodate diastereomeric molecules 32 and 34, even though 32 converts at a significantly higher rate than 34, makes the alcohol oxidation of 32 catalysed by BcABA4 fascinating [76].

Eremophilenes Isolated from B. cinerea
The high number of key secondary metabolism enzymes encoded in the genome of B. cinerea does not correspond to the total number of chemically characterised metabolites for this fungus. Thus, B. cinerea has a high number of silenced gene clusters (not expressed) under standard laboratory conditions.
A thorough analysis of the metabolites produced by two wild-type strains, B05.10 and UCA992, and the complemented mutant compl ∆Bcstc7 niaD revealed the isolation and structural characterisation of six 11,12,13-tri-nor-eremophilene derivatives (51)(52)(53)(54)(55)(56), in addition to a high number of known eremophilen-11-ol derivatives [77]. A structural characterisation was carried out by means of extensive spectroscopic techniques. The biosynthesis of these compounds has been reported ( Figure 15) and explained by a retroaldol reaction of eremophilenol (A) to give the keto derivative B which, after oxidation at C-7, would yield the corresponding 8-keto-7-hydroxy derivative C, corresponding to compounds 51-53; or by dehydration and oxidative cleavage of C11-C13 carbons which, after reduction, would yield F compounds 54-56 [77]. The structures, occurrences and biosynthesis of 11,12,13-tri-nor-sesquiterpenes have recently been reviewed [78].

Other Metabolites Isolated from B. cinerea
A new cryptic metabolite, botrycinereic acid (57) (Figure 16) was isolated by chemical epigenetic manipulation of B. cinerea strain B05.10 with the histone deacetylase inhibitor SAHA [79]. This compound was also overproduced by the deletion of the stc2 gene encoding an unknown sesquiterpene cyclase. Its structure and absolute configuration were determined by extensive spectroscopic NMR and HRESIMS studies and electronic circular dichroism calculations. Its biosynthesis was studied by feeding 2 H and 13 C isotopically labelled precursors to the B. cinerea ∆stc2 mutant. The results were consistent with a mixed biosynthesis of botrycineric acid (57) as outlined in Figure 17, according to which B. cinerea biosynthesises α-ketoisocaproate and phenylpyruvate via the l-leucine and l-phenylalanine biosynthetic pathways. The reduction of α-ketoisocaproate to the ketoenol derivative and condensation with phenylpyruvate derivative leads to intermediate 58.  Lastly, a high redundancy of phytotoxic compounds contributing to the necrotrophic pathogenesis of B. cinerea has been revealed and reported using multiple knock-out mutants [80]. To comprehensively evaluate the contributions of most of the currently known plant-cell death-inducing proteins (CDIPs) and metabolites for necrotrophic infection, an optimised CRISPR/Cas9 protocol was established. Comparative analysis of mutants confirmed significant roles played by two polygalacturonases (PG1, PG2) and the phytotoxic metabolites botrydial (1) and botcinins for infection but revealed no, or only weak, effects from the deletion of the other CDIPs. This was the first systematic study of the functional redundancy of fungal virulence factors and demonstrates that B. cinerea releases a highly redundant cocktail of proteins leading to the necrotrophic infection of a wide variety of host plants [80].
Moreover, the total synthesis of (+)-cinereain (13) (Figure 1), a fungal cyclotripeptide featuring a complex heterocyclic core isolated from B. cinerea in 1988 by Cutler et al. [25] featuring interesting plant growth regulating properties, has been achieved in a convergent manner [81], confirming the structure of 13 [25].

Botryotinia fuckeliana (de Bary) Wetzel (The Teleomorph Stage)
There are far fewer studies related to the secondary metabolism of Botryotinia fuckeliana, the B. cinerea teleomorph, than to its anamorphic form. The natural products obtained from this microorganism are limited to a few strains, specifically an endophytic strain and two marine strains [82][83][84][85][86]. Metabolic production of the fungi indicates the presence of compounds from terpenoids (sesquiterpenes and diterpenes), dipeptides and some hybrid compound classes. An important characteristic of the metabolic profile discovered is that it is different from the profile found in its anamorphic form. To date, there are no reports on the isolation of sesquiterpenes with a botryane skeleton (1,2), and only botcinin-type polyketides (3)(4)(5)(6) [16,17] but no botrylactones (8) [21] (Figure 1) have been reported.
All metabolites were tested for their cytotoxic activities against four human cancer cell lines (SMMC-7721 human hepatocellular carcinoma cell line, A549 human lung adenocarcinoma cell line, HepG2 hepatocellular carcinoma cell line and MCF-7 human breast adenocarcinoma cell line). Phenochalasin B (67) and [12]-cytochalasin (68) exhibited strong cytotoxicity, significantly inducing apoptosis in the human hepatocellular carcinoma cell line (HepG2).
Later, in 2019, Niu et al., carried out the chemical study of the marine strain B. fuckeliana MCCC 03A00494, isolated from deep-sea water in the Western Pacific Ocean, which produced 71 new (70-140) and eight known metabolites (141-148) (Figures 19 and 20) [85]. All compounds obtained were aphidicolins, tetracyclic diterpenes with a 6/6/5/6 ring system which are potential nuclear DNA replication inhibitors in eukaryotic cells and in some viruses and have, therefore, been thoroughly studied for the treatment of cancer [89,90]. Among the isolated compounds, 12 still need to have their stereochemistry fully defined. Compounds 134-140 are rare norditerpenoids reported in this work for the first time.

Conclusions
This review is a comprehensive account of the genes involved in the biosynthesis of secondary metabolites of the plant pathogenic fungus B. cinerea and of all the metabolites isolated from this fungus from January 2015 to October 2022. In this interval, many metabolites have been isolated and characterised from both morphic forms of the phytopathogenic fungus: approximately forty metabolites were characterised from the anamorphic stage (B. cinerea) and, interestingly, a higher number of metabolites, approximately 110, were isolated from the teleomorph stage (Botryotinia fuckeliana).
The recent genome resequencing of B. cinerea strains, namely B05.10 and T4, has revealed 44 genes encoding key enzymes (KE) involved in the biosynthetic pathways of secondary metabolites of the fungus. These ranged from STCs and DTCs (sesquiterpene and diterpene cyclases), to PKSs and NRPS (polyketide synthases and non-ribosomal peptide synthetases) and some dimethylallyltryptophan synthases (DMATS). Most of these KEencoding genes are co-localised with other enzyme-encoding genes that probably contribute to the biosynthesis of the same SM by modifications of the original skeleton. Taking into account the number of these gene clusters, B. cinerea may be able to produce approximately 40 different families of compounds; however, only a small number of metabolite families has been characterised thus far.
An important factor hampering the identification of fungal SMs is the fact that some of the biosynthesis gene clusters are silent under standard cultivation conditions. In recent years, the field of genome mining has emerged and many secondary metabolites (SMs) have been characterised from fungi sequenced using genomics-guided approaches [98].
In this review, new metabolites which have not been isolated from the anamorphic stage of B. cinerea were isolated from Botryotinia fuckeliana and were characterised. The diterpene cyclases DTC1, DTC2, DTC3 and PAX1 of the B. cinerea fungus remain uncharacterised. However, while no diterpenes have been detected in the broths of B. cinerea, approximately 90 diterpenes, 71 aphidicolins, 8 nor-diterpenes, 1 botry-pimarane and a further 8 diterpenes named botryotins A-H have been reported from a strain of Botrytinia fuckeliana from the Western Pacific Ocean and from some endophytic strains.
Clearly, the production of SMs depends on environmental signals that could be either abiotic or biotic. Both the improvement of culture conditions and a better knowledge of the regulation of secondary metabolism would be helpful in this regard (reviewed in [99,100]).
Lastly, further study of B. cinerea secondary metabolism may provide a useful starting point for the identification of new biological targets to control this crop-devastating fungus. This wealth of knowledge may contribute to the design of selective and rational structurebased fungicides [61].