Genome-based classification of the Streptomyces violaceusniger clade and description of Streptomyces sabulosicollis sp. nov. from an Indonesian sand dune

A polyphasic study was designed to determine the taxonomic provenance of a strain, isolate PRKS01-29T, recovered from an Indonesian sand dune and provisionally assigned to the Streptomyces violaceusniger clade. Genomic, genotypic and phenotypic data confirmed this classification. The isolate formed an extensively branched substrate mycelium which carried aerial hyphae that differentiated into spiral chains of rugose ornamented spores, contained LL-as the wall diaminopimelic acid, MK-9 (H6, H8) as predominant isoprenologues, phosphatidylethanolamine as the diagnostic phospholipid and major proportions of saturated, iso- and anteiso- fatty acids. Whole-genome sequences generated for the isolate and Streptomyces albiflaviniger DSM 41598T and Streptomyces javensis DSM 41764T were compared with phylogenetically closely related strains, the isolate formed a branch within the S. violaceusniger clade in the resultant phylogenomic tree. Whole-genome sequences data showed that isolate PRKS01-29T was most closely related to the S. albiflaviniger strain but was distinguished from the latter and from other members of the clade using combinations of phenotypic properties and average nucleotide identity and digital DNA:DNA hybridization scores. Consequently, it is proposed that isolate PRKS01-29T (= CCMM B1303T = ICEBB-02T = NCIMB 15210T) should be classified in the genus Streptomyces as Streptomyces sabulosicollis sp. nov. It is also clear that streptomycetes which produce spiral chains of rugose ornamented spores form a well-defined monophyletic clade in the Streptomyces phylogenomic tree., the taxonomic status of which requires further study. The genome of the type strain of S. sabulosicollis contains biosynthetic gene clusters predicted to produce new natural products.

Strains classified in the S. violaceusniger clade have an impressive track record as a source of new antibiotics (DeBoer et al. 1970;Chen et al. 2003;Cheng et al. 2010;Xie et al. 2019), antiparasitic metabolites (Sun et al. 2002), antitumour compounds (Lam et al. 1990;Wang et al. 2013), enzymes (Rabe et al. 2017) and immunosuppressants (Vezina et al. 1975) and biocontrol agents (Clermont et al. 2010;Palaniyandi et al. 2016;Sarwar et al. 2019) hence the continued interest in them for genome mining and natural product discovery. Members of this taxon are gifted in the sense of Baltz (2017) as they have large genomes ([ 8 Mbp) rich in biosynthetic gene clusters (BGCs) predicted to encode for specialised metabolites (Baranasic et al. 2013;Horn et al. 2014;Komaki et al. 2018). Prospecting for Streptomyces diversity also shows that sampling strains from unexplored, including extreme habitats, raises the probability of finding new compounds (Nicault et al. 2020) and that streptomycete genomes are a prolific source of novel BGCs (Vicente et al. 2018;Martinet et al. 2020).
The present study was designed to classify a putative new member of the S. violaceusniger clade based on genomic, genotypic and phenotypic data and to gain an insight into its potential as a source of new specialised metabolites. The resultant datasets showed that the isolate represents a novel species, named Streptomyces sabulisicollis sp. nov. Associated phylogenomic data clarified the internal taxonomic structure of the S. violaceusniger clade and relationships to its closest phylogenetic neighbours.

Materials and methods
Isolation, maintenance and cultivation Isolate PRKS01-29 T was isolated from an arid, nonsaline soil sample (pH 5.8., organic matter content 0.06%) collected just below the surface of a sand dune in the Parangkusumo Region (8°1 0 7 513 00 S/ 110°19 0 11.04 00 E) of Yogyakarta Province, Java, Indonesia following incubation on Actinomycete Isolation Agar (HiMedia, Einhausen, Germany), pH 7.3, supplemented with cycloheximide (50 lg/mL), nalidixic acid (25 lg/mL) and nystatin (25 lg/mL) and incubated for 7 days at 45°C, as described previously (Kusuma et al. 2020 Shirling and Gottlieb 1966) and as mixtures of hyphal fragments and spores in 20%, v/v glycerol at -20°C and -80°C. The type strains of S. albiflaviniger and S. iranensis were obtained from the Leibniz Institute DSMZ German Collection of Microorganisms and Cell Cultures GmbH, Braunschweig, Germany and the remaining reference strains were from the personal collection of Professor Michael Goodfellow, Newcastle University, Newcastle-upon-Tyne, United Kingdom. Biomass for the chemotaxonomic studies carried out on the isolate was prepared in 1L Erlenmeyer flasks containing 250 mL of sterile ISP 2 broth (Shirling and Gottlieb 1966), the flasks were shaken at 180 rpm for 14 days at 28°C and the resultant biomass harvested by centrifugation at 4000 rpm for 10 min, washed twice in sterile distilled water and freeze dried for 3 days.

Acquisition of chemotaxonomic, cultural and morphological properties
The isolate was examined for chemotaxonomic, cultural and morphological properties of value in Streptomyces systematics (Kämpfer 2012;van der Aart et al. 2019). Gram-stain (Hucher's modification, Society for American Bacteriology 1957) and micromorphological features were recorded following growth on ISP 3 agar for 7 days at 28°C. Growth from the ISP 2 preparation was examined for sporechain arrangement and spore-surface ornamentation using a scanning electron microscope (Tescan Vega 3, LMU instrument) and the procedure described by O'Donnell et al. (1993). The ability of the test and associated marker strains to grow at different temperatures, pH regimes and in the presence of various concentrations of sodium chloride was carried out in triplicate, as mentioned by Kusuma et al. (2020). Standard chromatographic methods were used to detect the isomers of diaminopimelic acid (A 2 pm) (Staneck and Roberts 1974), whole-organism sugars (Lechevalier and Lechevalier 1970) and for menaquinones and polar lipids by applying the integrated procedure of Minnikin et al. (1984), using appropriate controls. Cellular fatty acids were extracted from freeze dried cells of the isolate and fatty acid methyl esters (FAMES) prepared following saponification and methylation using the procedure described by Miller (1982), as modified by Kuykendall et al. (1988). The FAMES were separated by gas chromatography (Agilent 68,908 instrument), the resulted peaks automatically integrated and the fatty acid names and properties determined using the standard Microbial Identification (MIDI) system, version 4.5 and the ACTIN 6 database (Sasser 1990). The growth and cultural characteristics of the isolate and reference strains were determined on tryptone yeast extract, yeast extract-malt extract, oatmeal, inorganic saltsstarch, glycerol-asparagine, peptone-yeast extractiron and tyrosine agar plates (ISP media 1-7; Shirling and Gottlieb 1966) for 21 days at 28°C., aerial spore mass and substrate mycelial colours and those of diffusible pigments were recorded using colour charts (Kelly 1958).

Whole genome sequencing
Genomic DNA was extracted from wet biomass of single colonies of the isolate, S. albiflaviniger DSM 41598 T and S. javensis DSM 41764 T , grown on ISP 2 agar for 7 days at 28°C, following the protocol provided by MicrobesNG (Birmingham, UK) (http:// www.microbesng.uk) and sequenced on an Miseq instrument (Illumina, San Diego, USA). The quality of the extracted DNA preparations and the sequencing of genomic DNA libraries was achieved, as described by Kusuma et al. (2020). The libraries were sequenced following the 2 9 250-bp paired-end protocol (Mi-crobesNG, Birmingham, UK). Reads under 200 bp were discarded and contigs assembled using SPAdes software version 3.1.1 (Bankevich et al. 2012). The draft genome assemblies of the strains were annotated using the RAST-SEED web server (Aziz et al. 2008;Overbeek et al. 2014) with default options and are available from GenBank database.

Phylogeny
An almost complete 16S rRNA gene sequence (1454 nucleotides [nt]) (GenBank accession number MK503616) was taken directly from the draft genome of the isolate using the ContEst16S tool from the EZBioCloud webserver (https://www.ezbiocloud.net/ tools/contest16s) (Lee et al. 2017); this had been compared with the associated 16S rRNA gene sequence generated using Sanger method. The gene sequence was aligned with corresponding sequences of the most closely related type strains of Streptomyces species retrieved from the EzBiocloud webserver ) using MUSCLE software (Edgar 2004). Pairwise sequence similarities were determined using the single-gene tree option from the Genome-to-Genome Distance Calculator (GGDC) webserver (Meier-Kolthoff et al. 2013a, b). Phylogenetic trees were inferred using the maximum-likelihood (ML., Felsenstein 1981), maximum-parsimony (MP., Fitch 1971) and neighbour-joining (NJ., Saitou and Nei 1987) algorithms. A ML tree was inferred from alignments with RAxML (Stamatakis 2014) using rapid bootstrapping with the auto Maximum-Relative-Error (MRE) criterion (Pattengale et al. 2010) and a MP tree was constructed from the alignments with the Tree Analysis New Technology (TNT) program (Goloboff et al. 2008) using 1000 bootstraps together with tree-bisection-and-reconnection branch swapping and ten random sequence replicates. The sequences were checked for computational bias using the X2 test from PAUP*(Phylogenetic Analysis Using Parsimony) (Swofford 2002).The trees were evaluated using bootstrap analyses based on 1000 replicates (Felsenstein 1985) from the MEGA X software package (Kumar et al. 2018) and the two-parameter model of Jukes and Cantor (1969) then rooted with the 16S rRNA gene sequence from Streptomyces albus subsp. albus NRRL B-1811 T (GenBank accession number JX486031.1), the type strain of the type species of the genus Streptomyces.

Comparison of genomes
The draft genome sequences generated for isolate PRKS01-29 T , S. albiflaviniger DSM 41598 T and S. javensis DSM 41764 T were compared with corresponding sequences of type strains of species classified in the S. violaceusniger 16S rRNA gene clade. The ML phylogenomic tree inferred using the codon tree option in the PATRIC webserver (Wattam et al. 2017), which was based on aligned amino acids and nucleotides derived from 453 single copy genes in the genome dataset matched against the PATRIC PGFams database (http://www.patricbrc.org), was generated using the RAxML algorithm (Stamatakis 2006). The genome sequences of isolate PRKS01-29 T and the S. albiflaviniger and S. javensis strains were compared with one another and with those of S. antimycoticus NRRL B-24289 T , S. himastatinicus ATCC 53653 T , S. hygroscopicus subsp. hygroscopicus NBRC 16556 T , S. iranensis DSM 41954 T , S. malaysiensis DSM 4137 T , S. melanosporofaciens DSM 40318 T , S. milbemycinicus NRRL 5739 T , S. rapamycinicus NRRL 5491 T , S. rhizosphaericus NRRL-24304 T , S. sparsogenes DSM 40356 T and S. violaceusniger DSM 40503 T . Average nucleotide identity (orthoANI., Lee et al. 2016) and digital DNA-DNA hybridisation (dDDH., Meier-Kolthoff et al. 2013a) values were determined between the isolate and members of the S. violaceusniger clade using the ANI calculator from the EzBioCloud (https://www.ezbiocloud.net/tools/ ani) and the GGDC webserver (http://ggdc.dsmz.de/ ggdc), respectively. The presence of natural product-BGCs in the genome of the strains were detected using the antiSMASH 5.0 platform (Blin et al. 2019) with default option available at https://antismash. secondarymetabolites.org.

Phenotypic tests
Isolate PRKS01-29 T and the type strains of its closest phylogenomic neighbours were examined for phenotypic properties that distinguish between species classified in the S. violaceusniger 16S rRNA gene clade (Sembiring et al. 2000;Goodfellow et al. 2007;Goodfellow 2008, 2010;Hamedi et al. 2010;Zhou et al. 2017). Biochemical, degradation and physiological properties were acquired using media and methods described by Williams et al. (1983) and enzyme profiles with API-ZYM strips (BioMériux, France). All of the tests were carried out in duplicate using a standard inoculum equivalent to 5.0 on the McFarland scale (Murray et al. 1999).

Screening for bioactivity
The isolate was screened for antimicrobial activity against a panel of wild type microorganisms (primary screens) and Bacillus subtilis reporter strains (secondary screens) using a standard plug assay (Fiedler 2004). Plugs of isolate PRKS01-29 T were taken from ISP 2 and ISP 3 agar (Shirling and Gottlieb 1966) and from MMM and from 410 agar (Goodfellow and Fiedler2010) plates incubated for 14 days at 28°C and added to cultures of wild type strains of Bacillus subtilis, Candida albicans, Escherichia coli, Micrococcus luteus, Pseudomonas aeruginosa and Staphylococcus aureus., all of the strains were obtained from Public Health Laboratory Service, Freeman Hospital, Newcastle-upon-Tyne, United Kingdom. The wild type strains were prepared by inoculating 500 lL of overnight cultures grown at 37°C in 25 mL Luria Bertani (LB) broth (Sigma Aldrich, UK) to an optical density (OD) of 0.6 and the resultant preparations diluted to give an OD value of 0.0125 by mixing 100 mL of the LB media with the same proportion of nutrient agar (Sigma Aldrich, UK)., each of the resultant preparations was carefully mixed, poured into the square Petri dishes containing the agar plugs of the isolate and the plate incubated overnight at 37°C. The incubated plates were observed for the presence and sizes (in millimetres) of inhibition zones around the agar plugs. In the secondary assays, agar plugs were added to overnight cultures of six B. subtilis reporter strains grown as described above., the reporter strains were designed to detect modes of action of antimicrobial compound(s) produced by the isolate, as shown in Table S1. Overnight cultures of the strains were grown at 37°C in Luria Bertani broth then mixed with a similar volume of nutrient agar (Sigma-Aldrich, UK) to give an optical density reading of 0.0125. The resultant preparations were examined for the presence of blue halos around the circumference of inhibition zones, the latter are formed when bioactive compound(s) produced by the isolate cleave X-gal in the agar media to 5-bromo-4-chloro-3-hydroxy indole (blue compound) and galactose.

Results and discussion
The chemotaxonomic, colonial and morphological properties of the isolate showed that it was a bona fide member of the S. violaceusniger clade (Sembiring et al. 2000;Goodfellow et al. 2007;Goodfellow 2008, 2010;Hamedi et al. 2010;Nguyen and Kim 2015;Zhou et al. 2017). The organism was found to be aerobic, Gram-stain positive, formed an extensively branched substrate mycelium and aerial hyphae that differentiated into spiral chains of rugose ornamented spores (Fig S1), produced a dark grey to black aerial spore mass and a grey yellow substrate mycelium on oatmeal agar (Fig S2), contained LL-A2pm as the diamino acid of the peptidoglycan, MK-9 (H 6 ) (58.4%) and MK-9 (H 8 ) (41.6%) as the predominant isoprenologues, galactose, glucose, mannose and ribose as whole cell sugars and gave a polar lipid profile consisting of diphosphatidylglycerol, two phosphatidylglycerols, phosphatidylinositol, two phosphatidylinositol mannosides and two unknown phospholipids (Fig S3).
The phylogenetic tree (Fig. 1) based on 16S rRNA gene sequences shows that the isolate forms a clade in the Streptomyces gene tree together with the type strains of S. albiflaviniger, S. javensis and S. violaceusniger. It is most closely related to S. javensis NBRC 100777 T and S. violaceusniger NBRC 13459 T sharing a similarity with these strains of 99.4%, a value which corresponds to 9 nucleotide (nt) differences., the corresponding values with S. albiflaviniger NRRL B-1356 T are 99.3% (10 nt differences in 1414  Labeda et al. (2012) who found that streptomycetes producing spores with rugose or rough surfaces belonged to six highly related clades. The phylogenomic tree (Fig. 2) shows that the isolate forms a distinct branch at the periphery of a subclade that encompasses the type strains of S. albiflaviniger, S. iranensis, S. javensis, S. rapamycinicus and S. rhizosphaericus. The S. malaysiensis strain form a distinct lineage between this and a sister subclade composed of the type strains of S. antimycoticus, S. melanosporofaciens and S. violaceusniger. The two remaining members of the S. violaceusniger clade, S. himastatinicus ATCC 58653 T and S. hygroscopicus subspecies hygroscopicus NBRC 16556 T form single membered lineages. The close phylogenomic relationships between the type strains of S. milbemycinicus and S. sporogenes and S. violaceusniger clade is in agreement with the earlier study by Nouioui et al. (2018).
The recommended thresholds used to distinguish between closely related prokaryotic species based on ANI and dDDH similarities are 95 to 96% (Richter and Rosselló-Móra 2009;Chun et al. 2018) and 70% (Meier-Kolthoff 2013a; Chun et al. 2018), respectively. Table 2 shows that on this basis the isolate can be separated from the type strains of its closest phylogenomic neighbours, as shown in Fig. 2. It is most closely related to S. albiflaviniger DSM 41598 T based on a dDDH similarity of 53.9% and an ANI value of 93.5% though this latter value is shared with S. javensis DSM 41764 T and S. iranensis HM 35 T . Identical results were obtained for the duplicated cultures in all of the phenotypic tests. It is also encouraging that the results of the biochemical, degradative and tolerance tests are in agreement with those from earlier analyses on the reference strains that were performed under the same experimental procedures (Al-Tai et al. 1999;Sembiring et al. 2000;Saintpierre et al. 2003;Goodfellow et al. 2007;Kumar and Goodfellow 2008;Hamedi et al. 2010;Zhou et al. 2017). Table 3 shows that the isolate can be separated from the type strains of all of its closest phylogenomic neighbours using a combination of phenotypic properties. It can, for instance, be distinguished from S. albiflaviniger DSM 14548 T , its closest neighbour, as it is positive for esterase (C4), a-glucosidase and lipase (C14), casein, Tween 20 and uric acid, hydrolyses allantoin and grows in the presence of 7% w/v NaCl. In contrast, the S. albiflaviniger strain, unlike the isolate, hydrolyses arbutin. Additional combinations of phenotypic properties distinguish the isolate from the remaining reference strains and also the latter from one another.
The aerial spore mass and substrate mycelial colours produced by the respective reference strains on the ISP media are in agreement with those from earlier analyses (Al-Tai et al. 1999;Goodfellow et al. 2007;Kumar and Goodfellow 2008;Hamedi et al. 2010). Table S2 shows that the isolate and its closest phylogenomic neighbours grew well on nearly all of the ISP media forming a grey-yellowish substrate mycelium bearing a grey aerial spore mass that became moist and black on prolonged incubation on ISP 3 agar, as is the case with the type strains of S. antimycoticus (Kumar and Goodfellow 2008; and Tamura 2020a), S. griseiniger ), S. hygroscopicus (Labeda and Lyons 1991) and S. yatensis (Saintpierre et al. 2003). The isolate and the S. albiflaviniger can be distinguished by their ability to produce diffusible pigments, for instance, only the reference strain produced diffusible pigments on ISP media 3 and 7.
The isolate showed activity in the primary and secondary screens. Growth of the S. aureus strain was inhibited when the isolate was grown on ISP 2, ISP 3, MMM and 410 agar media. Similarly, it inhibited the B. subtilis, C. albicans and M. luteus strains following cultivation on all of the nutrient formulations, apart from medium 410. In contrast, it did not show any activity against the E. coli strain though it did inhibit the growth of the P. aeruginosa strain when grown on ISP 3 and MMM agar. In the secondary screens, the isolate formed blue halos around inhibition zones against B. subtilis reporter strains YpuA ER , YvqI ER , Yjax ER and DinB CH indicating its ability to inhibit cell envelope, DNA, fatty acid and RNA synthesis, respectively. It also inhibited the growth of the other reporter strains, YvgS ER and YheH, without forming blue halos thereby suggesting an ability to produce bioactive compound(s) with unknown modes of action.
Biosynthetic potential of isolate PRKS01-29 T and members of the S. violaceusniger clade The isolate and the type strains of species classified in the S. violaceusniger clade have large genomes (10.1-12.7 Mb) predicted to encode for chemically diverse specialised metabolites. The genome mining studies showed that all of the strains are genetically equipped with bioclusters predicted to encode for 'core secondary' metabolites, such as albaflavenone/geosmin, ectoines, hopenes, melanin and spore pigments, results in good agreement with those of Ward and Allenby (2018). In contrast, most of the bioclusters predicted to encode for druggable molecules, notably antibiotics, were discontinuously distributed in the genomes of the strains with many being strain specific, as has been found in recent studies on streptomycetes (Vicente et al. 2018;Martinet et al. 2020).

Conclusion
It can be concluded from the phylogenetic trees and associated colonial and morphological data that isolate PRKS01-29 T belongs to the S. violaceusniger clade (Sembiring et al. 2000;Goodfellow et al. 2007;Goodfellow 2008, 2010). In addition, the whole genome sequence data show that it belongs to a wellsupported monophyletic clade which includes the type strains of S. albiflaviniger, S. iranensis, S. javensis, S. rapamycinicus and S. rhizosphaericus. It can be distinguished from all of these strains by a broad range of phenotypic properties and by low ANI and dDDH values. It is, therefore, proposed that isolate PRKS01-29 T represents a novel species within the genus Streptomyces for which the name Streptomyces sabulosicollis sp. nov. is proposed.
Description of Streptomyces sabulosicollis sp. nov.
In the case of the genus Streptomyces genomebased classifications have revealed the presence of well-defined species-groups (Labeda et al. 2012(Labeda et al. , 2017Nouioui et al. 2018), the recognition of later heterotypic synonyms of established species (Komaki and Tamura 2020a, b;Madhaiyan et al. 2020) within and outwith the S. violaceusniger phylogenetic clade (Sembiring et al. 2000;Goodfellow et al. 2007;Goodfellow 2008, 2010) and the delineation of the genera Embleya and Yinghuangia for species previously included in the genus (Nouioui et al. 2018). Such developments can be expected to continue and in this respect, it is evident from this study that streptomycetes which form rugose-ornamented spores, spiral spore chains and characteristic colonial properties on oatmeal agar belong to a distinct phylogenomic clade the taxonomic status of which merits further investigation.

Data availability statements
The 16S rRNA gene and whole genome sequences of strain PRKS01-29 T that support the findings of this study have been deposited in GenBank database with the accession numbers MK503616 and JAEEAP000000000.1, respectively. In turn, corresponding accession numbers for the whole genome sequences of Streptomyces albiflaviniger DSM 42598 T and Streptomyces javensis DSM 41764 T are JAEEAR000000000.1 and JAEEAQ000000000.1, respectively. All the whole genome sequences described in this paper is version 1. Biotechnology, Sumbawa University of Technology, Indonesia, Dr. Wasu Pathom-aree and Nilita Mukjang from the Department of Biology, Chiang Mai University, Thailand for their valuable assistance in the deposition of the whole genome sequences of the strains in the NCBI genome database. IN is grateful to Newcastle University for a postdoctoral fellowship and MG for an Emeritus Fellowship from the Leverhulme Trust. We also thank Professor Hans-Peter Klenk (Newcastle University, United Kingdom) for his help in initial stages of the study.
Authors contribution MG and ABK designed the study and prepared the manuscript. ABK helped to collect the soil sample, characterized the strain under the supervision of IN and MG and deposited it in the culture collections. ABK and IN were responsible for the genome sequencing, annotation and the genome analyses. All of the authors approved the final version of the manuscript.
Funding This research was funded by The Indonesian Endowment Fund for Education (LPDP), Ministry of Finance, Indonesia through PhD scholarship scheme awarded to the first author with the grant number 20160822028928.

Declarations
Conflict of interest The authors declare that they do not have any conflicts of interest.
Ethical approval This article does not include any work with human participants and/or animals performed by one of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.