Taxonomic Positions of a Nyuzenamide-Producer and Its Closely Related Strains

Streptomyces sp. N11-34 is a producer of bicyclic peptides named nyuzenamides A and B. We elucidated its taxonomic position and surveyed its nonribosomal peptide synthetase (NRPS) and polyketide synthase (PKS) gene clusters by whole genome analysis. Streptomyces sp. N11-34 showed 16S rRNA gene sequence similarities of 99.9% and 99.8% to Streptomyces hygroscopicus NBRC 13472T and Streptomyces demainii NRRL B-1478T, respectively. Although these members formed a clade in a phylogenetic tree based on 16S rRNA gene sequences, the clade split into two closely related subclades in multilocus sequence analysis (MLSA). One included Streptomyces sp. N11-34, S. demainii NRRL B-1478T, S. hygroscopicus NBRC 100766, S. hygroscopicus NBRC 16556 and S. hygroscopicus TP-A0867 and the other comprised S. hygroscopicus NBRC 13472T and S. hygroscopicus NBRC 12859. These phylogenetic relationships were supported by phylogenomic analysis. Although Streptomyces sp. N11-34 was classified to S. hygroscopicus at the species level based on MLSA evolutionary distances and DNA–DNA relatedness, these distances and relatedness of members between the two subclades were comparatively far (0.004–0.006) and low (75.4–76.4%), respectively. Streptomyces sp. N11-34 possessed six NRPS, seven PKS and four hybrid PKS/NRPS gene clusters in the genome. Among the seventeen, ten were identified to be biosynthetic gene clusters (BGCs) of nyuzenamide, echoside, coelichelin, geldanamycin, mediomycin, nigericin, azalomycin, spore pigment, alchivemycin and totopotensamide, whereas the remaining seven were orphan in our bioinformatic analysis. All seventeen are conserved in S. hygroscopicus NBRC 100766, S. hygroscopicus NBRC 16556 and S. hygroscopicus TP-A0867. In contrast, S. hygroscopicus NBRC 13472T and S. hygroscopicus NBRC 12859 lacked the BGCs of alchivemycin, totopotensamide, a nonribosomal peptide and a hybrid polyketide/nonribosomal peptide compound. This difference was in a good accordance with the abovementioned phylogenetic relationship. Based on phenotypic differences in addition to phylogenetic relationship, DNA–DNA relatedness and BGCs, strains of S. hygroscopicus should be reclassified to two subspecies: S. hygroscopicus subsp. hygroscopicus and a new subspecies, for which we proposed S. hygroscopicus subsp. sporocinereus subsp. nov. The type strain is NBRC 100766T (=ATCC 43692T = DSM 41460T = INMI 32T = JCM 9093T = NRRL B-16376T = VKM Ac-312T). S. demainii was classified in this subspecies.


Introduction
Nonribosomal peptides and polyketides are the two largest families in the secondary metabolites of actinomycetes. These compounds are structurally diverse and often exhibit pharmaceutically useful biological activities. Half to two thirds of the secondary metabolite-biosynthetic gene clusters (smBGCs) in each actinomycetal genome are nonribosomal peptide synthetase (NRPS), polyketide synthase (PKS) and hybrid PKS/NRPS gene clusters, while each strain, such as that of the genus Streptomyces, harbors dozen of ule and a thiolation (T) domain for carrying the growing polypeptide chain. Sim minimal PKS module consists of an acyltransferase (AT) domain for selecting in acyl-CoAs, a ketosynthase (KS) domain for condensing the new building block w acyl intermediate from the previous module and an acyl carrier protein (ACP) dom carrying the growing polyketide chain. Individual modules are responsible for th poration of either one amino acid or acyl-CoA as a building block into the cha Optional domains may be present in each module, which methylate or epimeriz porated amino acid residues in nonribosomal peptides or reduce a keto group in tide chains. Thus, we can predict backbones of the products based on module n domain organization and the substrates of A and AT domains in each gene clu bioinformatic analysis [3,4]. Hence, PKS and NRPS gene clusters are often investi access the potential of each strain to produce diverse secondary metabolites [5][6][7][8] We recently isolated Streptomyces sp. N11-34 from deep sea water and fou novel compounds designated nyuzenamides A and B from the strain. Nyuzenam bicyclic peptides (Figure 1) with antifungal and cytotoxic activity [9]. Although the pounds seem to be synthesized through an NRPS pathway, the biosynthetic gene (BGC) has not yet been elucidated. In the present study, we investigated the tax position of Streptomyces sp. N11-34 and analyzed NRPS and PKS gene clusters to the nyuzenamide-BGC and reveal hidden potential to produce other compounds quently, we classified Streptomyces sp. N11-34 to S. hygroscopicus.
S. hygroscopicus is known to include strains significant in industrial and bio logical applications. As various bioactive secondary metabolites have been dis from the members, it is expected as a source for searching novel bioactive compo pharmaceutical industries. This species once included four subspecies with valid lished names. However, as they were reclassified to independent species, S. hygro includes no subspecies at present [10]. In the present study, we compared S. hygro N11-34 with its taxonomic neighbors, and consequently revealed that members o groscopicus can be classified into two groups. Thus, we here propose a new subsp S. hygroscopicus.  S. hygroscopicus is known to include strains significant in industrial and biotechnological applications. As various bioactive secondary metabolites have been discovered from the members, it is expected as a source for searching novel bioactive compounds in pharmaceutical industries. This species once included four subspecies with validly published names. However, as they were reclassified to independent species, S. hygroscopicus includes no subspecies at present [10]. In the present study, we compared S. hygroscopicus N11-34 with Microorganisms 2022, 10, 349 3 of 12 its taxonomic neighbors, and consequently revealed that members of S. hygroscopicus can be classified into two groups. Thus, we here propose a new subspecies of S. hygroscopicus.

Materials and Methods
Streptomyces sp. N11-34 was isolated in the previous study [9]. This strain has been deposited to and available from the NBRC Culture Collection as NBRC 113678. EzBio-Cloud [11] was used to search for taxonomic neighbors based on 16S rRNA gene sequences. Multilocus sequence analysis (MLSA) was conducted using DNA sequences of five housekeeping genes-atpD, gyrB, recA, rpoB and trpB-as established in the genus Streptomyces [12]. The accession numbers of gene sequences used for MLSA are listed in Table S1. The phylogenetic trees were reconstructed using ClustalX 2.1 [13]. Genomic DNA of Streptomyces sp. N11-34 for whole genome sequencing was prepared from cultured cells via the method of Saito and Kimura [14]. The whole genome was sequenced by the Kazusa DNA Research Institute using a single-molecule real-time (SMRT) strategy in the same manner of our previous report [7]. The assembled genome sequences were deposited to DDBJ under the accession numbers BNEK01000001-BNEK01000009. Phylogenomic tree was constructed using the TYGS webserver [15]. DNA-DNA relatedness was digitally calculated using whole genome sequences by Formula 2 of the Genome-to-Genome Distance Calculator (GGDC), an in silico method that reliably mimics conventional DNA-DNA hybridization experiments [16]. PKS and NRPS gene clusters in the genomes were surveyed using antiSMASH, which allows the rapid genome-wide identification, annotation and analysis of smBGCs in microbial genomes [4], and then manually analyzed as reported previously [6]. Whole genome sequences used for DNA-DNA relatedness calculation and NRPS and PKS gene cluster analysis are listed in Table 1.
As 16S rRNA gene sequence analysis is known to be low in the resolution, we next conducted MLSA and phylogenomic analysis. MLSA is often used for elucidating phylogenetic relationships with higher resolutions [12], whereas phylogenomic analysis can clarify whole genome sequence-based phylogenies [15]. In the MLSA-based phylogenetic tree, Streptomyces sp. N11-34 formed a clade with members of S. hygroscopicus and S. demainii. However, the topology within the clade was different from that in the 16S rRNA gene sequence-based phylogenetic tree. The clade clearly split into two subclades with the bootstrap values of 100%: one comprises only the type strains of S. hygroscopicus and S. endus whereas the other is composed of the remaining members, including Streptomyces sp. N11-34 and the type strains of S. sporocinereus and S. demainii ( Figure S2). Similarly, the members within the S. hygroscopicus clades split into two in the phylogenomic tree ( Figure 2).  Figure S2). Similarly, the members within the S. hygroscopicus clades split into two in the phylogenomic tree ( Figure 2). DNA-DNA relatedness value of 70% is established as the cut-off for species delineations in bacteria systematics [20]. In the genus Streptomyces, 0.007 in MLSA evolutionary distance is recognized to correspond to the cut-off [12]. Among Streptomyces sp. N11-34 and the phylogenetic neighbors, the DNA-DNA relatedness and MLSA evolutionary distances ranged 75.4-90.1% and 0.000-0.006, respectively ( Table 2). This suggests that these members represent the same species. Hence, Streptomyces sp. N11-34 is classified to S. hygroscopicus. The seven strains have been phylogenetically grouped into two, one is 1-5 and the other is 6-7, as stated above. Within the group 1-5, DNA-DNA relatedness and MLSA evolutionary distances are 85.0-91.8% and 0.000-0.003, respectively. Similarly, these values are 90.1% and 0.000 within the other 6-7, respectively. In contrast, values between the two groups are 75.4-76.4% and 0.004-0.006, respectively. As the threshold for subspecies demarcation is reported to be 79-80% in the DNA-DNA relatedness in bacteria [21], members between groups 1-5 and 6-7 are discriminated at subspecies level.  DNA-DNA relatedness value of 70% is established as the cut-off for species delineations in bacteria systematics [20]. In the genus Streptomyces, 0.007 in MLSA evolutionary distance is recognized to correspond to the cut-off [12]. Among Streptomyces sp. N11-34 and the phylogenetic neighbors, the DNA-DNA relatedness and MLSA evolutionary distances ranged 75.4-90.1% and 0.000-0.006, respectively (Table 2). This suggests that these members represent the same species. Hence, Streptomyces sp. N11-34 is classified to S. hygroscopicus. The seven strains have been phylogenetically grouped into two, one is 1-5 and the other is 6-7, as stated above. Within the group 1-5, DNA-DNA relatedness and MLSA evolutionary distances are 85.0-91.8% and 0.000-0.003, respectively. Similarly, these values are 90.1% and 0.000 within the other 6-7, respectively. In contrast, values between the two groups are 75.4-76.4% and 0.004-0.006, respectively. As the threshold for subspecies demarcation is reported to be 79-80% in the DNA-DNA relatedness in bacteria [21], members between groups 1-5 and 6-7 are discriminated at subspecies level.

NRPS and PKS Gene Clusters of Streptomyces sp. N11-34
Streptomyces sp. N11-34 harbored six NRPS, seven PKS and four hybrid PKS/NRPS gene clusters in the genome, as listed in Table 3. We identified nrps-1 and -2 to be BGCs for echoside and coelichelin, respectively, by bioinformatic analysis. The NRPSs showed high similarities to EchA and SCO0492, responsible for echoside and coelichelin biosyntheses, respectively (Table 4). Although nrps-3 was not a reported gene cluster, we identified it to be the nyuzenamide-BGC because the domain organization well accounts for it. Predicted amino acid residues of the A domains (Thr-X-Val-Gly-Phe-Pro-Leu-Gly-Tyr-Asn) are in a good accordance with those in nyuzenamides (Thr-Hpg-Val-Gly-Hpa-Pro-Leu-Hgy-Htr-Asn). As nrps-4, -5 and -6 were also unreported gene clusters, we predicted their products to be an octapeptide derived from dX-Thr-dX-Val-dX-dVal-dAla-Val, a Thrcontaining molecule and a tripeptide including Gly, respectively, based on the domain organization and predicted substrate of A domains. Among the seven PKS gene clusters in this strain, four type-I PKS (t1pks) and one type-II (t2pks) gene clusters were identified to be BGCs of geldanamycin, mediomycin, nigericin, azalomycin and spore pigment, respectively, which were supported by high sequence similarities of the NRPSs and PKSs to the reported enzymes ( Table 4). The remaining two were not reported gene clusters. Domain organization of t1pks-5 partially resembled that of a butylolactol-BGC. Although the butyrolactol-BGC encodes six PKSs [22], t1pks-5 encodes only five of the six. As the module number of t1pks-5 is eight, the product was predicted to be a compound derived from an octaketide, which is similar to butyrolactol, but the alkyl chain is shorter than that of butyrolactol. Only a single module was present in t1pks-6. As it did not show similarities to BGCs of known compounds, the product could not be predicted. Among the hybrid PKS/NRPS gene clusters in this strain, pks/nrps-1 and -2 were BGCs of alchivemycin [18] and totopotensamide, respectively, whose NRPSs and PKSs correspond to the biosynthetic enzymes (Table 4). In contrast, pks/nrps-3 and -4 were orphan gene clusters. PKSs in pks/nrps-3 were AT-less and resembled to those of leinamycin. However, as the domain organization differed from that of leinamycin-BGC, the product was predicted to be a new macrolactam compound like leinamycin. According to module number and domain organization, pks/nrps-4 was predicted to synthesis octapeptide with a polyketide moiety.

Distributions of the NRPS and PKS Gene Clusters in Streptomyces sp. N11-34 to the Phylogenetically Close Strains
As S. hygroscopicus NBRC 100766, NBRC 16556, TP-A0867, NBRC 12859 and NBRC 13472 T are phylogenetically close to Streptomyces sp. N11-34, as described in the Section 3.1, we examined whether the seventeen PKS and NRPS gene clusters found in Streptomyces sp. N11-34 are present in the genomes of these S. hygroscopicus strains. As summarized in Table 5, all the gene clusters were present in S. hygroscopicus NBRC 100766, NBRC 16556 and TP-A0867, which are closer to Streptomyces sp. N11-34, but S. hygroscopicus NBRC 12859 and NBRC 13472 T , phylogenetically discriminated from Streptomyces sp. N11-34, lacked nrps-6, pks/nrps-1 (avm), -2 (tot) and -3. Members of the group B lack four smBGCs in their genome. The distinctive phenotypic characteristics between the groups A and B are given in the previous reports as follows: spore wall ornamentations are warty or rugose in the group A whereas those are smooth in the group B [25]; although members of the group A utilize D-fructose as a sole carbon source for growth, those of the group B do not [25,26]; and maltose utilization is stronger in members of the group A than of the group B [12]. Taken together, it is considered that members in the group A are a new subspecies of S. hygroscopicus, for which we propose Streptomyces hygroscopicus subsp. sporocinereus subsp. nov.  [25,27]. This subspecies is also discriminated from Streptomyces hygroscopicus subsp. hygroscopicus by the genomic feature shown in Table 5. The genome size ranges from 9.9-10.4 Mb. The type strain is NBRC 100766 T (=ATCC 43692 T = DSM 41460 T = INMI 32 T = JCM 9093 T = NRRL B-16376 T = VKM Ac-312 T ). Accession numbers of the 16S rRNA gene and whole genome sequences in the type strain are AB249933 and BCAN01000001-BCAN01000217, respectively.