A widespread metabolic gene cluster family in metazoans

doi:10.21203/rs.3.rs-4859447/v1

Download PDF

Biological Sciences - Article

A widespread metabolic gene cluster family in metazoans

https://doi.org/10.21203/rs.3.rs-4859447/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Octocorals are unique among metazoans in their prolific production of bioactive terpenoid natural products that rival the chemical diversity of plants and microbes. We recently established that these cnidarians uniformly express terpene cyclases and that their encoding genes often reside within putative biosynthetic gene clusters (BGCs), a feature uncommon in animal genomes. In this work, we report the discovery and characterization of a widespread gene cluster family for the biosynthesis of briarane diterpenoids that number over 700 molecules specific to the Scleralcyonaceans, one of the two octocoral orders. We sequenced five genomes from evolutionarily distinct families of briarane-producing octocorals to complement three publicly available briarane-producing coral genomes, enabling the discovery of a conserved five-gene cluster composed of a terpene cyclase, three cytochrome P450s, and a short-chain dehydrogenase. Using Escherichia coli and Saccharomyces cerevisiae as hosts and homologous briarane biosynthesis genes from eight corals, we reconstituted the biosynthesis of cembrene B γ-lactone, which contains the γ-lactone structural feature distinctive of briarane diterpenoids. The discovery of the genomic basis of briarane biosynthesis establishes that animals, like microbes and plants, employ gene cluster families to produce specialized metabolites. Further, the presence of BGCs in octocoral proves that the formation and maintenance of BGCs over evolutionary time is a more widespread phenomenon in specialized metabolite biosynthesis than previously realized.

Biological sciences/Biochemistry/Enzymes/Multienzyme complexes

Earth and environmental sciences/Ocean sciences/Marine biology

Biological sciences/Biotechnology/Metabolic engineering

Biological sciences/Biological techniques/Sequencing/DNA sequencing

Specialized metabolites, or natural products, are small molecules that while not vital for primary growth, enable a producing organism to survive in an ecological niche by mediating time-, tissue-, or stimulus-dependent functions.¹ Although most of these compounds are known from plants and microbes, they are widespread in all kingdoms of life, including animals, where they too serve diverse purposes including communication, development, and defense.² Examples include volatile sex-pheromones from frogs³ and spiders,⁴ polyketide pigments in parrots⁵ and sea urchins,⁶ sugar-derived hormones in nematodes,⁷ hydroquinones employed by rove beetles to fend off predators,⁸ and peptide toxins used by cone snails to stun their prey.⁹

The overwhelming majority of research on specialized metabolite biosynthesis has been conducted in bacteria and fungi, where genes coding for a biosynthetic pathway are typically arranged in physically co-localized, biosynthetic gene clusters (BGCs).^10,11 In contrast, the few characterized biosynthetic pathways in animals are not encoded in BGCs, but have their encoding genes dispersed throughout the genome. This is currently regarded as the general rule for biosynthetic pathways in non-fungal eukaryotes.¹² However, improved genomic resources in other eukaryotes like plants and algae has revealed the colocalization of non-paralogous genes into BGCs, where they code for specialized metabolic pathways (Fig. 1).^1,13,14

Due to their abundance, eukaryotic BGCs were first recognized in fungi where they are now regarded as the norm in specialized metabolite biosynthesis. A prominent example is the lovastatin (1) BGC, which codes for nine non-homologous genes within a compact genomic region of 40 kilobase pairs (kb) and encodes the complete set of enzymes needed to produce 1, a resistance gene, a transporter, and a transcription factor (Fig. 1).¹⁵ The more recently discovered BGCs in plant genomes show clear differences from fungal BGCs. As generally the case in plant genomes, they show larger intergenic regions and often do not code for the complete biosynthetic pathway. An example is the tomatine (2) BGC, which codes for six steps in the pathway starting from cholesterol.¹⁶ However, the first enzymatic step of the pathway is encoded by a gene residing over 7,000 kb away from this cluster, while two additional genes are coded as a gene pair on a different chromosome (Fig. 1).¹⁷ There are only a few examples of small molecule BGCs in animals that have since been identified. However, their origins link to microbial gene sets acquired via horizontal gene transfer (HGT) as in the case of mycosporine-like amino acids such as gadusol (3) that widely occur in vertebrates¹⁸ (Fig. 1) and a penicillin biosynthesis gene pair from the springtail Folsomia candida that shows high homology to fungal sequences.^19,20 In contrast, the genes involved in the biosynthesis of the hybrid polyketide/ non-ribosomal peptide hormone nemamide (4) from the nematode Caenorhabditis elegans do not show clear sequence homology with microbial genes and thus may have evolved de novo in nematodes.²¹ However, none of these genes are colocalized in a cluster, but dispersed over large genomic regions across four separate chromosomes.

Although uncommon, a few gene clusters in animal genomes that are not the result of horizontal gene transfer are known, such as the Hox ²²and b-globin²³ clusters in humans, as well as a laccase gene cluster recently discovered in rove beetles.²⁴ However, these gene clusters are made up of paralogous genes and have arisen from successive tandem duplications from one ancestor gene, and only the last example is involved in small molecule biosynthesis. Generally, the few characterized biosynthetic pathways in animals, including the extraordinarily well investigated examples of steroid hormones²⁵ or melatonin (5)²⁶ in humans, are not encoded in BGCs, but have their encoding genes dispersed throughout the genome.

It was therefore surprising when we recently identified potential terpenoid BGCs in octocorals, consisting of up to 13 co-localized genes spanning up to 500 kb that did not result from HGT or pure tandem duplications.^27,28 Instead, over evolutionary time, coral gene paralogs from different protein families appear to have been recruited to single genomic scaffolds, forming potential BGCs. We observed related putative gene clusters from Renilla muelleri and Erythropodium caribeorum, corals that belong to different families (Pennatuloidea vs. Erythropodiidae) and differ in their lifestyle and morphology. Yet, both corals are known sources of briarane-type diterpenoids, a Scleralcyonacea-specific structural family of diterpenoids, numbering over 700 representatives and showing a wide array of potent biological properties.²⁹

Here, we test the hypothesis that briarane diterpenoids are encoded by broadly distributed and ancient biosynthetic gene clusters, which is unprecedented across metazoans. We assembled and analyzed high-quality genome sequences from six key briarane producing octocorals from different taxonomic families, revealing a conserved gene cluster family. Using a combination of protein biochemistry and pathway engineering, we reconstituted the biosynthesis of cembrene B γ-lactone, which contains the γ-lactone structural feature distinctive of all briarane diterpenoids. These results provide compelling evidence that corals use BGCs to synthesize their abundant bioactive terpenoids, opening the possibility that other metazoans may have similarly evolved BGCs in their own specialized metabolism.

Chemical Validation of Briarane Production

We previously characterized homologous cembrene B synthases identified from the genomes of R. muelleri and E. caribaeorum.^27,28 Both encoding genes were colocalized with cytochromes P450 genes, which also shared homology with their counterparts on the two distinct genomes. The tantalizing possibility that coral animals evolved gene cluster families for the biosynthesis of structurally related specialized metabolites motivated us to evaluate whether diverse corals known to contain briarane diterpenoids also encode cembrene B synthases in their genomes. As such, we first set out to confirm literature reports of briarane producing octocoral families. Previously, Pennatuloidea, Ellisellidae, Coralliidae, Briareidae, and Erythropodiidae were established as briarane containing octocorals (Fig. 2a).^30–36 As each of these five coral families belong to the Scleralcyonacea order, we further examined the remaining 14 families within this taxonomic order and found that five others reported terpenoid chemistry, yet not with briarane scaffolds (Fig. 2b, SI Table 1) .³⁷ For the remaining nine families, no terpenoid literature was found.

We thus collected coral tissue from representative species across five known briarane producing families, including Renilla koellikeri (family Pennatuloidea), Stylatula elongata (family Pennatuloidea), Dichotella gemmacea (family Ellisellidae), the branching and encrusting morphologies of Briareum asbestinum (family Briareidae), E. caribaeorum (family Erythropodiidae), and Corallium rubrum (family Coralliidae). R. koellikeri collected, by SCUBA in San Diego, is not a reported source for briaranes, but is related to the Atlantic R. muelleri that produces briaranes³⁸ and whose genome encodes cembrene B synthase as part of a putative BGC.^27,28 We thus analyzed the tissue of R. koellikeri and isolated 11-hydroxyptilosarcenone (6) whose chemical structure was confirmed by NMR spectroscopy (NMR supplementary note, SI Table 2, SI Fig. 1). This compound was previously identified from the sea pen Ptilosarcus gurneyi, but not yet from a Renilla sea pansy.^39,40

Limitations in accessible biomass for the other octocorals targeted in this study led us to use molecular networking analyses of crude extracts and semi-pure fractions to confirm briarane production. We used 11-hydroxyptilosarcenone (6) as a standard in the molecular network to identify a cluster containing 54 nodes representing unique masses of briarane compounds (Fig. 2, SI Fig. 2, SI Table 3). This analysis not only expanded the ptilosarcenone network from R. koellikeri (Pennatuloidea) but also established related briaranes in D. gemmacea (Ellisellidae), E. caribeorum (Erythropodiidae), and B. asbestinum (Briareidae). Two nodes corresponding to compounds in these corals had exact masses consistent with literature values: gemmacolide R (7) from D. gemmacea and erythrolide D (8) from E. caribaeorum (Fig. 2c, SI Table 1).^41,42 However, we did not detect briarane nodes from extracts of C. rubrum (Coralliidae) or S. elongata (Pennatuloidea). While there is literature precedent of S. elongata and corals in the family Coralliidae, although not specifically C. rubrum, making briaranes, we included these corals in our study assuming that they should have the genomic capacity to produce briaranes even though the samples we analyzed were devoid at the time of collection.^33,34,43

Genome assembly and syntenic analysis of the briarane gene cluster family in Scleralcyonacea corals

At the start of this study, a few briarane containing coral draft genomes with very low continuity (contig N50) were available: E. caribaeorum (Contig N50 2 kb)²⁸, R. muelleri (N50 70.5 kb)⁴⁴, and Pteroeides caledonicum (N50 4 kb) (SI Table 4 for assembly accessions and statistics).To capture intact BGCs, we proceeded to generate higher quality assemblies of corals across four briarane producing families, obtaining long read sequencing data for R. koellikeri (family Pennatuloidea), S. elongata (family Pennatuloidea), D. gemmacea (family Ellisellidae), the branching and encrusting morphologies of B. asbestinum (family Briareidae), and E. caribaeorum (family Erythropodiidae). We assembled 26 to 78 gigabase pairs (Gb) of sequencing reads into genomes ranging in size from 231 megabase pairs (Mb) (R. koellikeri) to 1,237 Mb (B. asbestinum) (Table 1). To assess the completeness of our assemblies, we estimated genome sizes with K-mer based analyses using Illumina short reads (SI Fig. 3), producing genomes ranging from 144 Mb (R. koellikeri) to 1,155 Mb (B. asbestinum) (Table 1). Illumina short read coverage was too shallow for E. caribaeorum to allow for K-mer based analysis (SI Table 5 for all sequencing data used). Additionally, we included our recently reported chromosomally resolved C. rubrum (family Coralliidae) genome at 475 Mb (20 chromosomes, 951 scaffolds)^45,46 in our analysis to enable BGC exploration across all five known briarane producing coral families. The genome assemblies are now some of the most contiguous for octocorals with contig N50 values above 196 kb (Table 1,Fig. 3a, SI Table 6). The chromosomal resolution assembly of C. rubrum had a scaffold N50 of 16,290 kb, on par with the current best scaffolded octocoral assembly from Xenia sp. with a scaffold N50 of 14,832 kb.⁴⁷ Benchmarking universal single-copy ortholog (BUSCO) completeness scores ranged from 86.4–88% (Table 1). BUSCO scores are highly dependent on the database used for the analysis and the low scores could reflect the use of the metazoan database, which was the closest available. The lower-than-expected completeness is on par with the BUSCO for the Xenia assembly, which was 88.1% complete (Fig. 3a).⁴⁷ We further assembled transcriptomes to guide our genomic analyses by identifying actively transcribed genes and determining intron bounds (SI Table 6).

Using the putative briarane BGC gene sequences from R. muelleri and E. caribeorum as queries, we analyzed the five newly established genomes of D. gemmacea, B. asbestinum, R. koellikeri, C. rubrum, and S. elongata and the public P. caledonicum for homologous genes. In total, the genomes from these eight octocoral species span five of the 20 octocoral families in the order Scleralcyonacea. To determine if non-briarane producing octocoral families also have syntenic BGCs, we also evaluated the publicly available Heliopora coerulea (family Helioporidae) genome.⁴⁸ We identified homologs of the cembrene B terpene cyclase (cbTC) in all eight octocorals shown or suspected to produce briaranes, whereas the H. coerulea genome did not contain a homologous sequence, although it did encode nine other TCs. In the eight octocoral species containing a homologous cbTC gene, seven contained iterations of the briarane BGC (Fig. 3b). Given that the P. caledonicum genome had poor contiguity, we were not surprised to find the cbTC homolog on a 3 kb contig and expect that it too is part of a BGC.

We observed one conserved short-chain dehydrogenase (cbSDH) and up to three cytochrome P450s (cbCYPa, cbCYPb, and cbCYPc) per briarane BGC (Fig. 3b, SI Fig. 4). In the eight octocorals with briarane BGCs, four maintained the full five-gene briarane BGC spanning 22 kb in R. muelleri up to 40 kb in C. rubrum, albeit with different gene organization. Notably, in the cases of S. elongata, E. caribeorum, and B. asbestinum, the orthologous genes were split between two BGC loci (SI Table 7). For S. elongata and B. asbestinum, their respective cbTC and cbCYPc genes reside on one contig, while cbCYPa, cbCYPb, and at least two cbSDHs are on a second contig. In the case of E. caribeorum, cbCYPc resides outside the BGC contig. Also, in the case of B. asbestinum, the orthologous genes have much longer gaps resulting in a significantly larger BGCs spanning ~ 150 kb. S. elongata and B. asbestinum further have multiple copies of the SDH gene at three (85%, 89% and 90% sequence similarity) and two (46% sequence similarity) copies, respectively (SI Fig. 5). The duplications of the SDHs and gene rearrangements suggest lineage specific independent evolution across octocoral species.

Since we identified the same set of homologous cytochromes P450s in a wide range of octocoral species, we conducted a phylogenetic analysis of these sequences in the background of broader cnidarian cytochromes P450s to shed light on their evolutionary origin. Our analysis included CYPs from four non-octocoral cnidarians as a reference set: Hydra vulgaris (24 CYPs), Acropora digitifera (24 CYPs), Aurelia aurita (37 CYPs), and Nematostella vectensis (70 CYPs).⁴⁹ From octocorals we included CYPs from six briarane producing species D. gemmacea, B. asbestinum, R. koellikeri, S. elongata, C. rubrum, and E. caribaeorum, as well as the non-briarane producers Xenia sp. and H. coerulea, yielding 318 octocoral CYP sequences at ~ 40 per genome (SI Table 8). We combined the annotated reference cnidarian CYPs with the annotated octocoral CYPs to build a phylogenetic tree (Fig. 3c, SI Fig. 6). The CYPs from the four reference cnidarians were previously associated with accepted CYP clans and families,⁴⁹ thereby enabling us to examine the evolutionary context of the octocoral BGC CYPs. Notably, some octocoral CYPs fell into the clades associated with the cnidarian CYP clans, suggesting that they fulfill primary, widespread functions generally present in cnidarians. Others, including the BGC-associated CYPs, formed new monophyletic clades that were octocoral specific. The three briarane CYPs, cbCYPa-c, form monophyletic clades consisting of cbCYPa, cbCYPb, and cbCYPc homologs from different species (Fig. 3c). While cbCYPb and cbCYPc share a common ancestor that may suggest related enzymatic functions, cbCYPa is distantly related and instead lies closer to clan 46 with predicted functions associated with steroid biosynthesis (Fig. 3c). Interestingly, H. coerulea, while also from the order Scleralcyonacea has CYPs closely related to the cbCYPb and cbCYPc clades, but these genes are not co-localized hinting at the loss of the briarane BGC in this lineage.

To test whether our phylogenetic approach could be used for genome mining in octocorals with lower quality assemblies, we next included CYP sequences from the poor contiguity genome assembly of P. caledonicum. We found corresponding sequences present in two of the three clades, thereby identifying cbCYPa and cbCYPb orthologs in the genome (Fig. 3d).

	Table 1. Assembly statistics and BUSCO assessment of five octocoral genomes
Organism	Contigs (#)	Assembly size (Mb)	K-mer Based Genome size (Mb)	Contig N50 (kb)	Longest (kb)	BUSCO (%C/ %D)
Briareum asbestinum	22,526	1,254	1,155	332	4,126	87.3/4.6
Dichotella gemmacea	8,171	550	401	196	2,036	87.5/15.6
Renilla koellikeri	557	231	144	4,895	22,942	86.9/2.6
Stylatula elongata	392	360	269	5,139	39,131	88.0/1.0
Erythropodium caribeorum	8,289	300	NA	233	1,576	86.4/0.6
Corallium rubrum⁴⁵	20 chromosomes, 951 scaffolds	545	540	1,600 (scaffold N50 18,521kb)	115,000	88.5/1.2

Biochemical validation of the briarane BGC

To provide support that the briarane BGC is involved in the early stages of briarane biosynthesis, we set out to provide biochemical validation of the clustered biosynthesis genes (SI Table 9). We first tested whether all the homologous cbTC terpene synthases functioned as cembrene B cyclases as we originally established in R. muelleri and E. caribaeorum.^27,28 We expressed each, 6 in total, in Saccharomyces cerevisiae and evaluated their production profiles by GC-MS analysis (Fig. 4, SI Fig. 7). In all cases, we confirmed that GGPP (9) was cyclized into cembrene B (10) as the sole product.

For the remaining enzyme assays, we tested the functions of genes from the R. koellikeri BGC and additional representative homologs identified by synteny. All tested enzyme homologs showed identical reactivity. We evaluated the functions of the three co-localized CYP subtypes in pairwise fashion by co-expression with cognate cbTC genes using previously established methods (Fig. 4).^50–52 We consistently measured the production of 19-hydroxycembrene B (11) in yeast upon co-expression with cbCYPb homologs. Its formation proceeded in a regiospecific manner with no other detectable monooxygenated products (SI Fig. 8). Scale-up fermentation and chromatographic purification provided 11 in pure form, which we characterized by NMR spectroscopic analyses (SI Tables 10 and 11, NMR supplementary note, SI Fig. 9). When we instead co-expressed cbCYPc homologs with cbTC genes, we observed formation of 17-hydroxycembrene (S1) which could be detected by LCMS (SI Fig. 10) and characterized by NMR (SI Tables 11 and 12, NMR supplementary note, SI Fig. 9). Because cbCYPb and cbCYPc were found to oxidize complimentary positions on 10 with respect to the briarane pathway, we tested the cbCYPb product 11 by feeding it to yeast expressing cbCYPc homologs (SI Fig. 11). We clearly observed a single oxidized product that upon purification and combined spectroscopic (SI Tables 10 and 11, NMR supplementary note, SI Fig. 9) and x-ray crystallographic characterization (SI Fig. 12, SI Table 12) was established as (7S)-7,19-dihydroxycembrene B (12). Likewise, when yeast harboring cbCYPb genes were fed S1, we similarly obtained 12, albeit with lower rates of conversion (see Methods). The cooperative and related functions of the cbCYPb and cbCYPc enzymes correlates well with their shared evolution (Fig. 3c). Despite numerous attempts with different substrate and gene pairs, we have yet to observe a function for the third cytochrome P450, cbCYPa.

As a variety of cembrene double bond isomers are produced by corals, we considered the possibility that the described CYP450 enzymes could act non-specifically on other cembrene isomers. We thus co-expressed cbCYPb homologs with terpene synthases that selectively produce either cembrene A or C. However, no oxidized products were produced (SI Fig. 13), showing high specificity of cbCYPb for cembrene B.

We next evaluated the co-localized short-chain dehydrogenase cbSDH homologs for their possible involvement in further transforming the diol 12. We expressed cbSDHs in Escherichia coli to conduct a series of in vitro incubation experiments. Incubation of purified cbSDH enzymes with NADP and 11 produced aldehyde (S2), while incubation with the diol 12 produced cembrene B g-lactone (13), a structure feature diagnostic of all briarane diterpenoids (SI Figs. 14 and 15). Two consecutive dehydrogenation steps were detected by LCMS chromatography via the formation of a m/z 301.2183 product ion ([M + H]⁺, calcd. m/z 301.2163 for C₂₀H₂₉O₂⁺). We scaled-up the in vitro reaction of 12 and NADP with the cbSDH to provide lactone 13, which was purified and characterized spectroscopically. Scale up experiments were conducted with representative characterized genes to unambiguously determine all structures. The structure was supported by NMR (SI Tables 11 and 12, NMR supplementary note, SI Fig. 9) and the diagnostic γ-lactone carbonyl stretch in the IR spectrum (v_max1757 cm^− 1). It is possible that 13 forms from 12 via a five-membered hemiacetal intermediate (SI Fig. 16).

We discovered a broadly conserved BGC family in corals and biochemically validated its function in the early biosynthetic steps toward briarane diterpenoids. The distinctive family of bioactive molecules numbers over 700 characterized members that are exclusively found in corals. Our work established the biosynthetic logic of briarane’s diagnostic γ-lactone core structure by four of the co-clustered gene products, cbTC, cbCYPb, cbCYPc, and cbSDH, confirming the clustered function of the briarane locus. The enzymatic transformations begin with cembrene B (10) formation by cbTC, then proceed with subsequent 19- and 7-hydroxylations carried out by cbCYPb and cbCYPc, respectively. Finally, in a unique mechanism of lactone biosynthesis, cbSDH transforms diol 12 directly into lactone 13 (Fig. 5A), likely via aldehyde formation at the 19-position, hemiacetal formation and a second dehydrogenation to produce the diagnostic lactone (SI Fig. 16). We have not yet characterized the enzyme(s) catalyzing bond formation between carbons 1 and 10.²⁹ A noncanonical class 2 TC acting on lactone 13 could introduce this feature to produce 14 (Fig. 5A).⁵³ Alternatively, a Δ^1,14 epoxide intermediate, potentially installed by a P450, might precede a transannular cyclization reaction to 18 (Fig. 5B).

In animals, previous reports indicate that genes can be found in clusters when they result from successive tandem duplications or HGT. However, such factors are not at play here in the octocorals. Homologs of the briarane TC and CYP genes are ancient, found throughout octocorallia (> 500 Mya). The briarane BGC itself is likely evolutionarily ancient and evolved soon after the division of the octocoral orders Scleralcyonacea and Malacalcyonacea. It is only found in briarane-containing corals from Sclearcyonacea and has similar synteny in distantly related corals such as Renilla koellikeri (Family Pennatoloidea) and Erythropodium caribeorum (Family Erythropodiidae) (Fig. 5C).⁵⁴ A key evolutionary event in the formation of the briarane BGC was the duplication of the related cbCYPb and -c genes early in the evolution of Scleralcyonacea. Supporting this timing, cbCYPb / -c pair is found across the Scleralcyonacea coral families Pennatuloidea, Ellisellidae, Coralliidae, Briareidae, and Erythropodiidae, but not elsewhere in sequenced octocorals. Subsequently, we speculate that cbTC, cbSDH, and cbCYPa were recruited and incorporated adjacent to the duplicated cbCYPb–c gene pair, also before speciation in the Scleralcyonacea, via chromosomal rearrangements to form the BGC — a phenomenon not previously reported in metazoans. Following the formation of the briarane BGC, all or parts of it were lost in a group of octocorals, as no briaranes have been reported from several Scleralcyonacea families (Fig. 5C). Supporting the gene loss scenario, H. coerulea (family Helioporidae), does not harbor a cembrene B synthase in its genome but retains cbCYPb and -c relatives, suggesting that these genes were once part of a briarane BGC. More complex evolutionary scenarios are also observable in sequenced genomes, such as the BGCs from S. elongata and B. asbestinum, which contain additional SDH gene copies. Their appearance coincides with a split of the gene cluster into two parts. The observation of metazoan BGC formation and decay suggests potential mechanisms of chromosomal evolution not previously observed in animals.

In contrast to a well preserved central briarane BGC, genes for specialized, late-stage tailoring processes are not colocalized. The core briarane skeleton is modified in many ways to create the > 700 known briarane derivatives (Fig. 5A). For example, diterpenoids such as 11-hydroxyptilosarcenone (6), gemmacolide R (7), bathyptilone A (15), and erythrolide B (16), found in different coral species, have been modified by diverse enzymatic reactions including oxidation, chlorination, and acylation, by genes found in different chromosomal locations. This may represent a general rule in metazoan BGCs, in which the required core genes are found in the BGC, while those for variable and species-specific tailoring steps are distributed through the genomes.

While the innate functions of coral terpenoids are largely unknown outside of a few examples having feeding deterrent properties,⁵⁵ we suspect that the conservation of their BGCs indicates central, unrealized biological roles pertaining to some of the more common coral terpenoid classes.⁵⁶ With the characterization of a coral terpenoid BGC, future genetic experiments are positioned to address their function once genome editing methods have been established in octocorals as recently accomplished in the hexacoral Acropora millepora.⁵⁷

The stage is set for genome mining and bioengineering studies to delineate biosynthetic cascade reactions to rare, bioactive coral molecules for their biosustainable production and application as therapeutic compounds. The briarane gene cluster family is likely the first of many to be described. We previously reported that some corals contain 70 kb and 500 kb putative BGCs with upward of 13 terpenoid biosynthesis genes.^27,28 With over 4,000 terpenoids characterized from corals (determined by searching the CMNPD database),⁵⁸ gene clustering is likely common in this ancient phylum. Looking ahead, we have yet to see if the active assembly of biosynthetic gene clusters by evolutionary forces is an exceedingly rare event that is limited to octocorals or if this is a more common phenomenon in animals and eukaryotes in general that has been overlooked so far due to limited genomic information in the eukaryotic kingdom.

Acknowledgments

We thank P. Zerofski (Scripps Institution of Oceanography) for collecting R. koellikeri and S. elongata, J. Sprung (Two Little Fishes Inc) for the D. gemmacea tissue, and J. Garrabou (CIIMAR) for the C. rubrum tissue. For her help in species identification, we would like to recognize C. S. McFadden (Harvey Mudd College). For coral images, we thank C. S. McFadden- D. gemmacea, F. Zuberer (CNRS)- C. rubrum, P. Webster @underwaterpat- S. elongata, and J. Simpson (Herbert Wertheim UF Scripps Institute for Biomedical Innovation & Technology)- B. asbestinum and E. caribaeorum. We thank T. Damiani (IOCB Prague) and A. Rodriguez (UC San Diego) for their assistance with feature based molecular networking. Chemical isolation and elucidation of the R. koellikeri briarane diterpene was aided by A. Bogdanov (Scripps Institution of Oceanography). This work was supported by the National Institutes of Health (R01-GM146224 to B.S.M., R35-GM148283 to E.W.S, and K99-GM148783 to P.D.S.), a Margaret A. Davidson Graduate Fellowship to N.E.G. (NERRS NA22NOS4200050), an NSF Graduate Research Fellowship to M.L.M., Tang Genomics Fund to T.P.M,, and Fundaço para a Ciência e a Tecnologia funds (UIDB/04423/2020, UIDP/04423/2020 and 2021.00855.CEECIND) to J.B.L. X-ray diffraction research reported in this publication was supported by the Office of the Director, NIH under award S10-OD030326. NMR data obtained at UC San Diego (the Scripps Institution of Oceanography and Biomolecular NMR Facilities) was assisted by B. Duggan. Additional NMR data was obtained at the University of Utah Health Sciences NMR Core. GCMS data was obtained at the University of Utah Health Sciences Proteomics Core by Q. Pearce.

Author Contributions:

Conceptualization: PDS, IB, EWS, BSM

Data creation: NEG, PDS, MLM, IB

Formal analysis and validation: NEG, PDS, IB, JGG, TA

Funding acquisition: PDS, JBL, TPM, EWS, BSM

Investigation and methodology: NEG, PDS, MLM, IB

Project administration: PDS, IB, EWS, BSM

Resources: JBL, TPM, EWS, BSM

Writing – original draft: NEG, PDS, IB, BSM

Writing – review and editing: All authors

Competing interests:

Authors declare that they have no competing interests.

Data and materials availability:

The feature based molecular network parameters used and raw data are available through the MASSIVE dataset ID: MSV000094792.

Accessions numbers for the genomes generated in this paper are as follows; R. koellikeri (SAMN40621396), B. asbestinum (SAMN40621398), D. gemmacea (SAMN40621399), and S. elongata (SAMN40621397).

Medema, M.H., de Rond, T., and Moore, B.S. (2021). Mining genomes to illuminate the specialized chemistry of life. Nat. Rev. Genet. 22, 553–571. https://doi.org/10.1038/s41576-021-00363-7.
Torres, J.P., and Schmidt, E.W. (2019). The biosynthetic diversity of the animal world. J. Biol. Chem. 294, 17684–17692. https://doi.org/10.1074/jbc.REV119.006130.
Poth, D., Wollenberg, K.C., Vences, M., and Schulz, S. (2012). Volatile Amphibian Pheromones: Macrolides from Mantellid Frogs from Madagascar. Angew. Chem. Int. Ed. 51, 2187–2190. https://doi.org/10.1002/anie.201106592.
Chinta, S.P., Goller, S., Lux, J., Funke, S., Uhl, G., and Schulz, S. (2010). The Sex Pheromone of the Wasp Spider Argiope bruennichi. Angew. Chem. Int. Ed. 49, 2033–2036. https://doi.org/10.1002/anie.200906311.
Masello, J.F., Lubjuhn, T., and Quillfeldt, P. (2008). Is the structural and psittacofulvin-based coloration of wild burrowing parrots Cyanoliseus patagonus condition dependent? J. Avian Biol. 39, 653–662. https://doi.org/10.1111/j.1600-048X.2008.04417.x.
Anderson, H.A., Mathieson, J.W., and Thomson, R.H. (1969). Distribution of spinochrome pigments in echinoids. Comp. Biochem. Physiol. 28, 333–345. https://doi.org/10.1016/0010-406X(69)91347-4.
Butcher, R.A., Fujita, M., Schroeder, F.C., and Clardy, J. (2007). Small-molecule pheromones that control dauer development in Caenorhabditis elegans. Nat. Chem. Biol. 3, 420–422. https://doi.org/10.1038/nchembio.2007.3.
Dettner, K., and Schwinger, G. (1982). Defensive secretions of three oxytelinae rove beetles (Coleoptera: Staphylinidae). J. Chem. Ecol. 8, 1411–1420. https://doi.org/10.1007/BF01403104.
Olivera, B.M., Gray, W.R., Zeikus, R., McIntosh, J.M., Varga, J., Rivier, J., de Santos, V., and Cruz, L.J. (1985). Peptide Neurotoxins from Fish-Hunting Cone Snails. Science 230, 1338–1343. https://doi.org/10.1126/science.4071055.
Chevrette, M.G., Gutiérrez-García, K., Selem-Mojica, N., Aguilar-Martínez, C., Yañez-Olvera, A., Ramos-Aboites, H.E., Hoskisson, P.A., and Barona-Gómez, F. (2020). Evolutionary dynamics of natural product biosynthesis in bacteria. Nat. Prod. Rep. 37, 566–599. https://doi.org/10.1039/C9NP00048H.
Rokas, A., Mead, M.E., Steenwyk, J.L., Raja, H.A., and Oberlies, N.H. (2020). Biosynthetic gene clusters and the evolution of fungal chemodiversity. Nat. Prod. Rep. 37, 868–878. https://doi.org/10.1039/C9NP00045C.
Nützmann, H.-W., Scazzocchio, C., and Osbourn, A. (2018). Metabolic Gene Clusters in Eukaryotes. Annu. Rev. Genet. 52, 159–183. https://doi.org/10.1146/annurev-genet-120417-031237.
Steele, T.S., Brunson, J.K., Maeno, Y., Terada, R., Allen, A.E., Yotsu-Yamashita, M., Chekan, J.R., and Moore, B.S. (2022). Domoic acid biosynthesis in the red alga Chondria armata suggests a complex evolutionary history for toxin production. Proc. Natl. Acad. Sci. 119, e2117407119. https://doi.org/10.1073/pnas.2117407119.
Smit, S.J., and Lichman, B.R. (2022). Plant biosynthetic gene clusters in the context of metabolic evolution. Nat. Prod. Rep. 39, 1465–1482. https://doi.org/10.1039/D2NP00005A.
Ryngajłło, M., Boruta, T., and Bizukojć, M. (2021). Complete genome sequence of lovastatin producer Aspergillus terreus ATCC 20542 and evaluation of genomic diversity among A. terreus strains. Appl. Microbiol. Biotechnol. 105, 1615–1627. https://doi.org/10.1007/s00253-021-11133-0.
Itkin, M., Heinig, U., Tzfadia, O., Bhide, A.J., Shinde, B., Cardenas, P.D., Bocobza, S.E., Unger, T., Malitsky, S., Finkers, R., et al. (2013). Biosynthesis of Antinutritional Alkaloids in Solanaceous Crops Is Mediated by Clustered Genes. Science 341, 175–179. https://doi.org/10.1126/science.1240230.
Nützmann, H.-W., Huang, A., and Osbourn, A. (2016). Plant metabolic clusters – from genetics to genomics. New Phytol. 211, 771–789. https://doi.org/10.1111/nph.13981.
Osborn, A.R., Almabruk, K.H., Holzwarth, G., Asamizu, S., LaDu, J., Kean, K.M., Karplus, P.A., Tanguay, R.L., Bakalinsky, A.T., and Mahmud, T. (2015). De novo synthesis of a sunscreen compound in vertebrates. eLife 4, e05919. https://doi.org/10.7554/eLife.05919.
Roelofs, D., Timmermans, M.J.T.N., Hensbergen, P., van Leeuwen, H., Koopman, J., Faddeeva, A., Suring, W., de Boer, T.E., Mariën, J., Boer, R., et al. (2013). A Functional Isopenicillin N Synthase in an Animal Genome. Mol. Biol. Evol. 30, 541–548. https://doi.org/10.1093/molbev/mss269.
Faddeeva-Vakhrusheva, A., Kraaijeveld, K., Derks, M.F.L., Anvar, S.Y., Agamennone, V., Suring, W., Kampfraath, A.A., Ellers, J., Le Ngoc, G., van Gestel, C.A.M., et al. (2017). Coping with living in the soil: the genome of the parthenogenetic springtail Folsomia candida. BMC Genomics 18, 493. https://doi.org/10.1186/s12864-017-3852-x.
Feng, L., Gordon, M.T., Liu, Y., Basso, K.B., and Butcher, R.A. (2021). Mapping the biosynthetic pathway of a hybrid polyketide-nonribosomal peptide in a metazoan. Nat. Commun. 12, 4912. https://doi.org/10.1038/s41467-021-24682-9.
de Rosa, R., Grenier, J.K., Andreeva, T., Cook, C.E., Adoutte, A., Akam, M., Carroll, S.B., and Balavoine, G. (1999). Hox genes in brachiopods and priapulids and protostome evolution. Nature 399, 772–776. https://doi.org/10.1038/21631.
Storz, J.F., Opazo, J.C., and Hoffmann, F.G. (2013). Gene duplication, genome duplication, and the functional diversification of vertebrate globins. Mol. Phylogenet. Evol. 66, 469–478. https://doi.org/10.1016/j.ympev.2012.07.013.
Kitchen, S.A., Naragon, T.H., Brückner, A., Ladinsky, M.S., Quinodoz, S.A., Badroos, J.M., Viliunas, J.W., Kishi, Y., Wagner, J.M., Miller, D.R., et al. (2024). The genomic and cellular basis of biosynthetic innovation in rove beetles. Cell 0. https://doi.org/10.1016/j.cell.2024.05.012.
Turcu, A.F., Rege, J., Auchus, R.J., and Rainey, W.E. (2020). 11-Oxygenated androgens in health and disease. Nat. Rev. Endocrinol. 16, 284–296. https://doi.org/10.1038/s41574-020-0336-x.
Tan, D.-X., Hardeland, R., Back, K., Manchester, L.C., Alatorre-Jimenez, M.A., and Reiter, R.J. (2016). On the significance of an alternate pathway of melatonin synthesis via 5-methoxytryptamine: comparisons across species. J. Pineal Res. 61, 27–40. https://doi.org/10.1111/jpi.12336.
Burkhardt, I., de Rond, T., Chen, P.Y.-T., and Moore, B.S. (2022). Ancient plant-like terpene biosynthesis in corals. Nat. Chem. Biol. 18, 664–669. https://doi.org/10.1038/s41589-022-01026-2.
Scesa, P.D., Lin, Z., and Schmidt, E.W. (2022). Ancient defensive terpene biosynthetic gene clusters in the soft corals. Nat. Chem. Biol. 18, 659–663. https://doi.org/10.1038/s41589-022-01027-1.
Moon, N.G., and Harned, A.M. (2018). Synthetic explorations of the briarane jungle: progress in developing a synthetic route to a common family of diterpenoid natural products. R. Soc. Open Sci. 5, 172280. https://doi.org/10.1098/rsos.172280.
La, M.-P., Li, J., Li, C., Tang, H., Liu, B.-S., Sun, P., Zhuang, C.-L., Li, T.-J., and Zhang, W. (2014). Briarane Diterpenoids from the Gorgonian Dichotella gemmacea. Mar. Drugs 12, 6178–6189. https://doi.org/10.3390/md12126178.
Barsby, T., and Kubanek, J. (2005). Isolation and Structure Elucidation of Feeding Deterrent Diterpenoids from the Sea Pansy, Renilla reniformis. J. Nat. Prod. 68, 511–516. https://doi.org/10.1021/np049609u.
Fenical, W., and Pawlik, J.R. (1991). Defensive properties of secondary metabolites from the Caribbean gorgonian coral Erythropodium caribaeorum. Mar. Ecol. Prog. Ser. 75, 1–8.
Harvell, C.D., Fenical, W., Roussis, V., Ruesink, J.L., Griggs, C.C., and Greene, C.H. (1993). Local and geographic variation in the defensive chemistry of a West Indian gorgonian coral (Briareum asbestinum). Mar. Ecol. Prog. Ser. 93, 165–173.
Ksebati, M.B., and Schmitz, F.J. (1986). Diterpenes from a Soft Coral, Minabea SP., from Truk Lagoon. Bull. Sociétés Chim. Belg. 95, 835–851. https://doi.org/10.1002/bscb.19860950910.
Wratten, S.J., and Faulkner, D.J. (1979). Some diterpenes from the sea pen stylatula SP. Tetrahedron 35, 1907–1912. https://doi.org/10.1016/0040-4020(79)85047-4.
Yang, J., Zhang, S., Qia, S., Pan, J., Qiu, Y., Tao, S., Yin, H., and Li, Q. (2007). Briarane-type diterpenoids from the China gorgonian coral Subergorgia reticulata. Biochem. Syst. Ecol. 35, 770–773. https://doi.org/10.1016/j.bse.2007.06.003.
MarinLit - A database of the marine natural products literature https://marinlit.rsc.org/.
Keifer, P.A., Rinehart, K.L., and Hooper, I.R. (1986). Renillafoulins, antifouling diterpenes from the sea pansy Renilla reniformis (Octocorallia). J. Org. Chem. 51, 4450–4454. https://doi.org/10.1021/jo00373a020.
Hendrickson, R.L., and Cardellina, J.H. (1986). Structure and stereochemistry of insecticidal diterpenes from the sea penXXX. Tetrahedron 42, 6565–6570. https://doi.org/10.1016/S0040-4020(01)88120-5.
Nurco, D.J., Conklin, D.E., Shapiro, N.S., and Tran, E. (2010). The absolute structure of ptilosarcenone 2.5-hydrate, a diterpenoid briarane from the orange sea pen Ptilosarcus gurneyi (Gray). Acta Crystallogr. Sect. E Struct. Rep. Online 67, o181–o182. https://doi.org/10.1107/S1600536810050142.
Li, C., La, M.-P., Sun, P., Kurtan, T., Mandi, A., Tang, H., Liu, B.-S., Yi, Y.-H., Li, L., and Zhang, W. (2011). Bioactive (3Z,5E)-11,20-Epoxybriara-3,5-dien-7,18-olide Diterpenoids from the South China Sea Gorgonian Dichotella gemmacea. Mar. Drugs 9, 1403–1418. https://doi.org/10.3390/md9081403.
Pordesimo, E.O., Schmitz, F.J., Ciereszko, L.S., Hossain, M.B., and Van der Helm, D. (1991). New briarein diterpenes from the Caribbean gorgonians Erythropodium caribaeorum and Briareum sp. J. Org. Chem. 56, 2344–2357. https://doi.org/10.1021/jo00007a020.
Molina, S.L., Forero, A.M., Ayala, F.I., Puyana, M., Zea, S., Castellanos, L., Muñoz, D., Arboleda, G., Sandoval-Hernández, A.G., and Ramos, F.A. (2020). Metabolic Profiling of the Soft Coral Erythropodium caribaeorum (Alcyonacea: Anthothelidae) from the Colombian Caribbean Reveals Different Chemotypes. Mar. Drugs 18, 4. https://doi.org/10.3390/md18010004.
Jiang, J.B., Quattrini, A.M., Francis, W.R., Ryan, J.F., Rodríguez, E., and McFadden, C.S. (2019). A hybrid de novo assembly of the sea pansy (Renilla muelleri) genome. GigaScience 8, giz026. https://doi.org/10.1093/gigascience/giz026.
Ledoux, J.-B., Gomez-Garrido, J., Cruz, F., Ferreira, F.C., Matos, A., Sarropoulou, X., Ramirez-Calero, S., Aurelle, D., Lopez-Sendino, P., Grayson, N., et al. (2024). Chromosome-level genome assembly and annotation of Corallium rubrum: a Mediterranean coral threatened by overharvesting and climate change. Preprint at bioRxiv, https://doi.org/10.1101/2024.07.13.603384 https://doi.org/10.1101/2024.07.13.603384.
Corominas, M., Marquès-Bonet, T., Arnedo, M.A., Bayés, M., Belmonte, J., Escrivà, H., Fernández, R., Gabaldón, T., Garnatje, T., Germain, J., et al. (2024). The Catalan initiative for the Earth BioGenome Project: contributing local data to global biodiversity genomics. NAR Genomics Bioinforma. 6, lqae075. https://doi.org/10.1093/nargab/lqae075.
Hu, M., Zheng, X., Fan, C.-M., and Zheng, Y. (2020). Lineage dynamics of the endosymbiotic cell type in the soft coral Xenia. Nature 582, 534–538. https://doi.org/10.1038/s41586-020-2385-7.
Ip, J.C.-H., Ho, M.-H., Chan, B.K.K., and Qiu, J.-W. (2023). A draft genome assembly of reef-building octocoral Heliopora coerulea. Sci. Data 10, 381. https://doi.org/10.1038/s41597-023-02291-z.
Pankov, K.V., McArthur, A.G., Gold, D.A., Nelson, D.R., Goldstone, J.V., and Wilson, J.Y. (2021). The cytochrome P450 (CYP) superfamily in cnidarians. Sci. Rep. 11, 9834. https://doi.org/10.1038/s41598-021-88700-y.
DeJong, J.M., Liu, Y., Bollon, A.P., Long, R.M., Jennewein, S., Williams, D., and Croteau, R.B. (2006). Genetic engineering of taxol biosynthetic genes in Saccharomyces cerevisiae. Biotechnol. Bioeng. 93, 212–224. https://doi.org/10.1002/bit.20694.
Pompon, D., Louerat, B., Bronine, A., and Urban, P. (1996). [6] Yeast expression of animal and plant P450s in optimized redox environments. In Methods in Enzymology, E. F. Johnson and M. R. Waterman, eds. (Academic Press), pp. 51–64. https://doi.org/10.1016/S0076-6879(96)72008-6.
Ro, D.-K., Paradise, E.M., Ouellet, M., Fisher, K.J., Newman, K.L., Ndungu, J.M., Ho, K.A., Eachus, R.A., Ham, T.S., Kirby, J., et al. (2006). Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature 440, 940–943. https://doi.org/10.1038/nature04640.
Rudolf, J.D., and Chang, C.-Y. (2020). Terpene synthases in disguise: enzymology, structure, and opportunities of non-canonical terpene synthases. Nat. Prod. Rep. 37, 425–463. https://doi.org/10.1039/C9NP00051H.
McFadden, C.S., Ofwegen, L.P. van, and Quattrini, A.M. (2022). Revisionary systematics of Octocorallia (Cnidaria: Anthozoa) guided by phylogenomics. Bull. Soc. Syst. Biol. 1. https://doi.org/10.18061/bssb.v1i3.8735.
Barsby, T., and Kubanek, J. (2005). Isolation and structure elucidation of feeding deterrent diterpenoids from the sea pansy, Renilla reniformis. J Nat Prod 68, 511–516. https://doi.org/10.1021/np049609u.
He, J., Xin, P., Ma, X., Chu, J., and Wang, G. (2020). Gibberellin Metabolism in Flowering Plants: An Update and Perspectives. Front. Plant Sci. 11. https://doi.org/10.3389/fpls.2020.00532.
Cleves, P.A., Strader, M.E., Bay, L.K., Pringle, J.R., and Matz, M.V. (2018). CRISPR/Cas9-mediated genome editing in a reef-building coral. Proc. Natl. Acad. Sci. 115, 5235–5240. https://doi.org/10.1073/pnas.1722151115.
Lyu, C., Chen, T., Qiang, B., Liu, N., Wang, H., Zhang, L., and Liu, Z. (2021). CMNPD: a comprehensive marine natural products database towards facilitating drug discovery from the ocean. Nucleic Acids Res. 49, D509–D515. https://doi.org/10.1093/nar/gkaa763.

There is NO Competing Interest.

BriaraneTerpenoidSI080324.pdf
Supporting Information

Download PDF

Version 1

posted

You are reading this latest preprint version

A widespread metabolic gene cluster family in metazoans

Status:

Version 1

Abstract

Figures

Introduction

Results

Chemical Validation of Briarane Production

Genome assembly and syntenic analysis of the briarane gene cluster family in Scleralcyonacea corals

Biochemical validation of the briarane BGC

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1