Stilbenes

We used PRISM 4 and antiSMASH 5 to predict the chemical structures of secondary metabolites encoded within 3759 complete bacterial genomes and 6362 metagenome-assembled genomes (MAGs). All bacterial genomes with an assembly level of ‘Complete’ were downloaded from NCBI Genome, and a set of dereplicated genomes as determined by the Genome Taxonomy Database^{15 (link)} were retained to mitigate the impact of highly similar genomes on our analysis. Similarly, a set of 7902 MAGs^{23 (link)} was obtained from NCBI BioProject (accession PRJNA348753) and the subset of dereplicated genomes was retained. Detected BGCs were matched between PRISM and antiSMASH if their nucleotide sequence overlapped over any range. A small number of PRISM BGC types were mapped to more than one antiSMASH BGC type, including aminoglycosides (‘amglyccycl’ and ‘oligosaccharide’), type I polyketides (‘t1pks’ and ‘transatpks’), and RiPPs (‘bottromycin’, ‘cyanobactin’, ‘glycocin’, ‘head_to_tail’, ‘LAP’, ‘lantipeptide’, ‘lassopeptide’, ‘linaridin’, ‘microviridin’, ‘proteusin’, ‘sactipeptide’, and ‘thiopeptide’). The “hybrid” category encompassed all BGCs assigned any combination of two or more cluster types, i.e., it was not limited to hybrid NRPS-PKS BGCs. The “other” category encompassed aryl polyenes, bacteriocins, butyrolactones, ectoines, furans, homoserine lactones, ladderanes, melanins, N-acyl amino acids, NRPS-independent siderophores, phenazines, phosphoglycolipids, resorcinols, stilbenes, terpenes, and type III polyketides. Producing organism taxonomy was based on genome phylogeny and retrieved from the Genome Taxonomy Database^{15 (link)}.
Cheminformatic metrics, including molecular weight, number of hydrogen bond donors and acceptors, octanol-water partition coefficients, and Bertz topological complexity, were calculated in RDKit. Both platforms occasionally generated very small, non-specific structure predictions (for example, a single unspecified amino acid or a single malonyl unit) that did not provide actionable information about the chemical structure of the encoded product; to remove these from consideration, we applied a molecular weight filter to remove structures under 100 Da output by either platform. To evaluate the internal structural diversity of each set of predicted structures, we computed the distribution of pairwise Tcs for each set⁴⁵ , taking the median pairwise Tc instead of the mean as a summary statistic to ensure robustness against outliers. Structural similarity to known natural products was assessed using the RDKit implementation of the ‘natural product-likeness’ score^{22 (link)}, and by the median Tc between predicted structures and the known secondary metabolite structures deposited in the NP Atlas database^{46 (link)}.

The wild type N2 C. elegans was provided by the Caenorhabditis Genetics Center (CGC). Resveratrol and FUDR (Sigma Aldrich) were used as the test compounds. Resveratrol (3,5,4′-trihydroxy-trans-stilbene) is a type of natural phenol with low aqueous solubility and the molecular weight of 228.24 (Figure 1A). Resveratrol has been shown to extend the lifespan of yeast, fly and C. elegans with clear mechanism of action [30] (link)–[33] (link). FUDR, with the molecular weight of 246.2 (Figure 1B), was soluble in water at the concentration to 50 mg/mL. FUDR was widely used to inhibit the worms to lay eggs in research with the concentration of FUDR from 40 µM to 50 µM [34] , [35] (link).
Resveratrol was dissolved in DMSO to 100 mM as stock solution. In LB medium method, the resveratrol stock solution was diluted into LB liquid medium containing OP50 E. coli bacteria to a final concentration of 100 µM and 1.5 mL of the bacteria solution was applied to each NGM plate (100 mm diameter). In spot dead method, 1.5 mL of 100 µM resveratrol was spotted onto the surface of the NGM plate, then covered with dead bacteria. In NGM live method, the resveratrol stock solution was diluted with NGM (below 65°C after boiled) to the concentration of 100 µM. Then the NGM was poured into petri plates as supporting bed for worms and live OP50 was applied to the surface of NGM plates. In NGM dead method, the plates were made as the NGM live method, except that the food OP50 bacteria was killed by incubating in 65°C for 30 minutes [24] (link). The drug administration of above four methods was summarized in Figure 2. The maintenance of C. elegans in liquid medium was described as previously [23] with slight modification. Briefly, the synchronized N2 adult day 1 worms were cultured in 50 mL centrifuge tubes that contained 35 mL S medium [23] , the concentration of resveratrol in S medium was 100 µM. Dead bacteria were added to S medium as food. FUDR was dissolved in H₂O to 50 mM as stock solution. The procedures of treatment of worms with FUDR were the same with resveratrol, except that the final concentration of FUDR in the five treatment methods was 50 µM.
About 5,000–10,000 adult day 1 wild type worms were transferred to each NGM plate (100 mm diameter). For the liquid growing method, the worms were cultured in several 50 mL centrifuge tubes with each containing about 35 mL S medium [23] . Worms in each method were harvested by using cold M9 buffer at the 10 min, 30 min, 1 hr, 3 hr, 6 hr, 12 hr, day 1, day 2, day 4, day 7, day 14 and day 20 after treated with the compounds [23] , and collected to 15 mL centrifuge tubes. The control group without treatment with compound was also harvested. The tubes were putted into ice for 10 minutes, then spin for 2 minutes at 1,150× g to precipitate the worms. The worm pellets were rinsed three times with cold M9 buffer, air dry, and weighed. The worm pellets were resuspended by using 1 mL methyl ethanol (HPLC grade) (for resveratrol) or H₂O (for FUDR) and sonicated 50 times (200 V, operation 5 seconds every 5 seconds). Then, the worm solution was centrifuged under 12, 000× g for 3 minutes. The supernatant of the worm solution containing resveratrol or FUDR was transferred to a 1.5 mL centrifuge tube.
To test the drug metabolism, the worms treated with 400 µM, 200 µM, 100 µM, 50 µM, 25 µM, and 12.5 µM of resveratrol and FUDR, respectively, for 6 hours under NGM dead method were transferred to NGM plates without resveratrol and FUDR. Then, the worms were harvested at the 10 min, 30 min, 1 hr, 2 hr, 3 hr, 4 hr, 6 hr, 8 hr, 12 hr, and 16 hr time points. Subsequent sample preparation was the same as described above.

A total of 211 compound classes were identified in the positive and negative ion modes. To visualize the compound class diversity, a sunburst plot was conducted (Fig. 4). The most prominently detected classes overall were carboxylic acids and derivatives (mainly due to amino acids, peptides, and analogues), followed by benzene and substituted derivatives, fatty acyls (largely fatty amides), organooxygen compounds (mostly carbohydrates and carbohydrate conjugates), prenol lipids (mostly diterpenoids, retinoids, and sesquiterpenoids), and flavonoids (mostly flavonoid glycosides and hydroxyflavonoids). A large number of features were also classified as stilbenes, the chemical class represented in the ClassyFire chemical ontology that encompasses the characteristic bibenzyls found in Radula spp. Known compounds from liverworts were tentatively annotated and are listed in Table 1.
Fig. 4

Sunburst plot showing an overview on the richness of classified metabolite compounds. Broad compound classes are represented in the center while specific classifications are represented on the exterior. Colours correspond to the assigned classes. Due to readability the names of some classes were removed from the plot. An interactive zoomable plot is available in the supplementary vignettes and on Zenodo

Table 1

Tentatively annotated liverwort specialized metabolites. Full details are found in the Supplementary Information

Compound	Formula	Molar Mass	Ionization	Tentative Feature
Bisabola-1,3,5,7(14),10- pentaene	C15H20	200.32	Positive	FT0671, FT0672
Ar-tenuifolene	C15H20	200.32	Positive	FT0671, FT0672
Eudesma-1,4(15)-11- triene	C15H22	202.23	Positive	FT0692
Myli-4(15)-ene	C15H22	202.33	Positive	FT0692
Cis-calamenene	C15H22	202.33	Positive	FT0692
Cuparene	C15H22	202.33	Positive	FT0692
Xanthorrizol	C15H22O	218.33	Positive	FT0828 - FT0832
2-cuparenol	C15H22O	218.33	Positive	FT0828 - FT0832
Cyclocolorenone	C15H22O	218.33	Positive	FT0828 - FT0832
β-herbertenol	C15H22O	218.33	Positive	FT0828 - FT0832
Trans-Nerolidol	C15H26O	222.37	Positive	FT0861
(E)-farnesol	C15H26O	222.37	Positive	FT0861
3-[2-(3-Methoxyphenyl)ethyl]phenol	C15H16O2	228.29	Positive	FT0923, FT0925
3,4′-Dimethoxybibenzyl	C16H18O2	242.31	Positive	FT1057, FT1059
1,2-Bis(3-methoxyphenyl)ethane	C16H18O2	242.32	Positive	FT1057, FT1059
Lunularic acid	C15H14O4	258.1	Negative	FT0814-FT0820
Radulanin A	C19H20O2	280.37	Positive	FT1451, FT1454, FT1458
2,2-Dimethyl-5-hydroxy- 7-(2-phenylethyl)- chromene*	C19H20O2	280.4	Positive	FT1454, FT1458
4-(3-Methyl-2-butenyl)-5-phenethylbenzene-1,3-diol	C19H22O2	282.38	Positive	FT1480, FT1483, FT1484, FT1487
			Negative	FT1001, FT1008, FT1009, FT1011
4-Prenyldihydropinosylvin	C19H22O2	282.38	Positive	FT1480, FT1483, FT1484, FT1487
			Negative	FT1001, FT1008, FT1009, FT1011
Radulanin A methyl ether	C20H22O2	294.39	Positive	FT1623, FT1624, FT1625, FT1626, FT1627
			Negative	FT1111, FT1112
8-[2-(4-Hydroxyphenyl)ethyl]-3-methyl-2,5-dihydro-1-benzoxepin-6-ol	C19H20O3	296.37	Negative	FT1132, FT1133, FT1135, FT1136, FT1139, FT1140, FT1141, FT1142, FT1143, FT1144, FT1147
5-Methoxy-2-(3-methylbut-2-en-1-yl)-3-(2-phenylethyl)phenol	C20H24O2	296.41	Positive	FT1658, FT1660
			Negative	FT1146, FT1148
4-(3-Methyl-2-Butenyl)-5-(2-Phenylethyl)-3-Methoxyphenol	C20H24O2	296.41	Positive	FT1658, FT1660
			Negative	FT1146, FT1148
2-[(3,3-Dimethyloxiran-2-yl)methyl]-5-(2-phenylethyl)benzene-1,3-diol	C20H24O2	296.41	Positive	FT1658, FT1660
			Negative	FT1146, FT1148
3-Methoxy-5-(2-phenylethyl)-2-prenylphenol	C20H24O2	296.41	Positive	FT1658, FT1660
			Negative	FT1146, FT1148
2-[(3,3-Dimethyloxiran-2-yl)methyl]-5-(2-phenylethyl)benzene-1,3-diol	C19H22O3	298.38	Negative	FT1167, FT1168
Kaempferol 3-methyl-ether	C16H12O6	300.26	Negative	FT1200, FT1201
2,2-Dimethyl-5-hydroxy-7-(2-phenylethyl)-2 H-1-benzopyran-6-carboxylic acid	C20H20O4	324.38	Negative	FT1483, FT1484, FT1485, FT1486, FT1489, FT1491, FT1494, FT1496
Radulanin E	C20H20O4	324.38	Negative	FT1483, FT1484, FT1485, FT1486, FT1489, FT1491, FT1494, FT1496
Radulanin H	C20H20O4	324.4	Positive	FT2017 - FT2020
			Negative	FT1484-1486, FT1489-1494, FT1496

The 3 different T. comosus dried extracts (100 mg each one) were dissolved in 2 ml of the corresponding extraction solvents, namely water (TCI), ethanol 70% (TCT), and ethanol 50% (OpTC). After centrifugation (6,000 × g, 10 min, at 4°C), the supernatants were transferred to HPLC vials and supposed to untargeted phenolic profiling through high-resolution mass spectrometry (HRMS) using a Q-Exactive™ Focus Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Scientific, Waltham, MA, United States) coupled to a Vanquish ultra-high-pressure liquid chromatography (UHPLC) pump, equipped with heated electrospray ionization (HESI)-II probe (Thermo Scientific, United States) (Babotă et al., 2022 (link); Nicolescu et al., 2022 (link)). The post-acquisition data filtering was accomplished using MS-DIAL software (version 4.70), while the annotation was achieved via spectral matching against FoodDB and Phenol-Explorer databases. Overall, the mass accuracy (setting a 5-ppm tolerance for m/z values), isotopic pattern, and spectral matching were used to calculate a total identification score, considering the most common HESI + adducts for the chromatographic conditions adopted, thus reaching a level 2 of confidence in annotation.
Semi-quantitative appreciation of each previously annotated phenolic class was made by analyzing representative pure standard compounds under the same conditions: ferulic acid (phenolic acids), quercetin (flavonols), catechin (flavanols), cyanidin (anthocyanins), luteolin (flavones and other flavonoids), resveratrol (stilbenes), and oleuropein (other remaining phenolics). A linear fitting (R² > 0.99) was built and used for quantification, results being expressed as mg equivalents (Eq.)/g lyophilized extract (n = 3).