*2.5. 16S rRNA Gene Sequence Analysis of Bacterial Isolates*

Twenty bacterial isolates selected for screening for compounds **1** and **2** production were identified by 16S rRNA gene analysis on the EzBiocloud server [8]. Based on the sequence comparison to reference type strains, the isolates were assigned to the two bacterial phyla (Proteobacteria and Firmicutes). Two isolates (CB1-13 and CB1-18) showed the highest similarity values with *Bacillus hwajinpoensis* SW-72T (99.21%–99.24%) and *Bacillus hemicentroti* JSM 076093T (98.27%). One isolate (CB1-3) shared the highest similarity value with *Pseudovibrio japonicus*WSF2T (99.2%), *Pseudovibrio ascidiaceicola* DSM 16392<sup>T</sup> (98.99%), and *Pseudovibrio denitrificans* DSM 17465<sup>T</sup> (98.91%) from Alphaproteobacteria. The others

were closely related to the species of the genus *Vibrio* from Gammaproteobacteria. The isolates CB1-14, CB2-10, CB2-8, and CB1-5 showed 98.96%–99.58% sequence similarity with *Vibrio hangzhouensis* CN83T. The other isolates CB2-5, CB1-7, and CB2-12 had 98.79%–100% sequence similarity with *Vibrio barjaei* 3062<sup>T</sup> and *Vibrio thalassae* MD16T. Most of isolates (CB1-1, CB1-10, CB1-11, CB2-4, CB2-9, CB2-11, and CB2-13) shared the highest similarity values with *Vibrio mediterranei* CIP 103203T (99.65%) and *Vibrio shilonii* AK1T (99.59%). Three isolates (CB1-6, CB2-1 and CB2-7) showed similarity values less than 97.5% with reference type strains of the species of genus *Vibrio*.

The phylogenetic tree based on the 16S rRNA sequences (1438 bp) clearly showed that *Vibrio* isolates grouped into four clades (Figure 2), three of which included the single type strains, *V. barjaei* or *V. hangzhouensis* or *V. mediterranei*. The fourth clade was at the base of the genus *Vibrio* and did not include any type strains.

**Figure 2.** Bacterial phylogenetic tree on the basis of 16S rRNA gene sequences of isolates recovered from the mucus net of the *C. variopedatus* and closely related sequences of type strains. The tree topology was obtained using the maximum likelihood method based on the Tamura three-parameter model. Bootstrap values above 75% calculated from 1000 re-sampling are shown on the node. The scale bar represents the number of substitutions per site.

Guided by the cutoff value at the species level equal to 98.65% [9] and phylogenetic positions, the isolates CB1-13 and CB1-18 might be identified as *B. hwajinpoensis* and the isolate CB1-3 as *P.* *japonicus*. Among *Vibrio* isolates, the isolates CB2-5, CB1-7, and CB2-12 might be identified as *V. barjaei*, the isolate CB1-5 as *V. hangzhouensis*, and the isolates CB1-1, CB1-10, CB1-11, CB2-4, CB2-9, CB2-11, and CB2-13 as *V. mediterranei*. The clades, containing CB1-14 and CB2-1, might be distinguished as candidates for new species. Thus, one of bacterial strain (CB1-14) presumably producing monanchorins was identified as *Vibrio* sp., closely related to *Vibrio hangzhouensis* CN83T, but probably distinguished from this species.

Therefore, the phylogenetic analysis revealed bacterial diversity in the mucus net of the *C. variopedatus*. The dominant cultured bacteria were members of the genus *Vibrio*, belonging, at least, to three different species.

## *2.6. Multilocus Sequence Analysis of CB1-14*

The phylogenic analysis based on 16S rRNA gene sequences showed the isolate CB1-14 was closely related to *V. hangzhouensis* CN83T, sharing 98.96% identity with this strain. It means that the calculated identity value is within the boundary range proposed for delineating *Vibrio* species [10]. Since the 16S rRNA gene sequence did not help in differentiating closely related bacterial species, the eight-gene MLSA was applied as that currently used for delimitating *Vibrio* species [11,12].

To overcome difficulties in application of universal primers for the MLSA, the draft genome of CB1-14 was obtained and used to retrieve sequences of eight housekeeping genes. Following previously described MLSA scheme [11] and using available genome sequences of nineteen type strains including *V. maritimus* CAIM 1455T, *V. variabilis* CAIM 1457T, *V. mediterranei* NBRC 15635T, and *V. hangzhouensis* CGMCC-1-7062T, the MLSA study was performed. Based on phylogenies generated by ML (Maximum Likelihood), MP (Maximum Parsimony), and NJ (Neighbor Joining) methods (data are not presented) and split tree decomposition analysis (Figure 3), the MLSA placed the isolate CB1-14 into the Mediterranei clade. Within the clade, the isolate CB1-14 formed a separate branch closely related to *V. maritimus* CAIM 1455<sup>T</sup> and *V. variabilis* CAIM 1457T, with a high bootstrap support.

**Figure 3.** Concatenated split network tree based on eight gene loci. The *ftsZ*, *gapA*, *gyrB*, *mreB*, *pyrH*, *recA*, *rpoA*, and *topA* gene sequences from 20 taxa were concatenated including the isolate CB1-14 (bold font). Phylogenetic tree was generated using the SplitsTree4 program.

Thus, the phylogenetic reconstruction showed that the isolate CB1-14 should be recognized as a new species in the genus *Vibrio*. The valid description of this new species in *Vibrio* genus, isolated from an organ of the polychaete, namely from its mucous net, will be done in a special journal.

Marine invertebrates are the oldest animals on Earth, distributed over all the ocean biomes from polar to tropical waters and from shallow to very deep substrates. In the course of their evolution, marine invertebrates have acquired long-term and stable associations with a wide diversity of bacteria, cyanobacteria, archaea, and other groups of microbes, which make up to 60% of the biomass of some these animals and are essential to their survival [13]. There are a number of reports that cultures of microorganisms, isolated from marine sponge [14–19] and ascidian tissues [20], produce secondary metabolites previously isolated from these invertebrates, that indicates their microbial origin [13]. However, up to date, only biosurfactants were identified from polychaete-associated microbial isolates [21]. Production of the 6-*epi*-monanchorin by *Vibrio* sp. CB1-14 isolate is completely unprecedented. Most of the compounds so far isolated from *Vibrio* spp. were proved to be non-ribosomal peptides or their hybrids. Only a few guanidine-containing secondary metabolites were isolated from *Vibrio*, for example siderophore vanchrobactin from *Vibrio anguillarum* [22] as well as Na channel blocker tetrodotoxin and its derivatives from bacteria *V. alginolyticus, V. harveyi, V. fischeri*, and *Vibrio* sp. [23–26].

Our finding that the *Vibrio* sp. CB1-14 isolate, obtained from polychaete food net and which is able to biosynthesize compound **2**, shows that this bacterium (and probably some other close related species) has important unrecognized biosynthetic capabilities, and should be considered as a potential microbial source of monanchorins. From the biotechnological viewpoint, the cultivation of bacteria after optimization of 6-*epi*-monanchorin production could help to solve the recognized supply problem of marine-derived drugs. Identification of this producer also opens prospects of bicyclic guanidine alkaloid biosynthesis.

#### **3. Materials and Methods**

#### *3.1. General*

The 1H-NMR spectra were recorded on a Bruker Avance III-700 spectrometer in CDCl3. Chemical shifts were referenced to the corresponding residual solvent signal (δH 7.26/δC 77.20 for CDCl3). ESI mass spectra (including HRESIMS) were obtained on a Bruker maXis Impact II LC-MS spectrometer by direct infusion in MeOH. MALDI-TOF mass spectra were obtained on a Bruker Ultraflex III TOF/TOF laser desorption spectrometer coupled with delayed extraction using a Smartbeam MALDI 200 laser with α-cyano-hydroxy cinnamic acid as the matrix. HPLC was performed on a Shimadzu instrument with a RID-10A refractive index detector using a YMC-ODS-A (250 × 10 mm) column.
