**3. Materials and Methods**

#### *3.1. Sample Collection and Treatment*

Soil samples were collected from Enggano Island (5◦22 57.0792 S, 102◦13 28.2792 E), Indonesia, in December 2015 (Figure 1B). Marine samples were collected from marine sediments from Bali Island (8◦43 5.5 S, 115◦10 7.8 E), Indonesia, in May 2014 (Figure 1C), and Lombok Island West Nusa Tenggara (8◦24 17.133 S, 116◦15 57.228 E), Indonesia, in May 2017 (Figure 1D). Soil and sediment samples were taken aseptically from 10 cm depth of soil samples and the center of sediment in mangrove and tidal area. Soil and sediment samples were transferred into sterile 50 mL conical tubes and placed on ice and then stored at 4 ◦C until further treatment.

#### *3.2. Isolation of Actinomycetes*

Isolation and enumeration of actinomycetes were done using a serial dilution of Humic Acid-Vitamin (HV) medium [48] and/or NBRC No. 802 Medium [49] by using the direct method [50], the dry heat method [51], and the phenol method [51]. In the direct method, an air-dried soil sample or marine sediment was ground in a mortar and heated in a hot-air oven at 110 ◦C for 30 min. One gram of the heated samples was transferred to 10 mL of sterile water and mixed for 2 min, then diluted with sterile water to 10−1, 10<sup>−</sup>2, and 10−<sup>3</sup> times. In total, 200 μL of each dilution was inoculated on isolation medium

agar of HV [48] or NBRC No. 802 Medium [49] with or without the addition of 1% NaCl. The inoculated plates were incubated for 2–4 weeks at 28 ◦C. The colonies showing the *Streptomyces* morphological characteristics were selected and streaked on fresh plates of the modified *Streptomyces* International Project 2 (ISP2 =ˆ YM) agar [52]. The cultures were resuspended in sterile 0.9% (*w*/*v*) saline supplemented with 15% (*v*/*v*) glycerol and stored at −80 ◦C. This dry-heat method [51] was used to isolate heat-tolerant actinomycetes spores. In the dry-heat method, the soil or sediment samples were incubated at 100 ◦C for 40 min and then cooled to 28 ◦C in a desiccator. The samples were distributed on HV medium agar plates with a spatula tip and incubated at 28 ◦C for 2–3 weeks. The phenol method was used to select for spores, which survive in the presence of phenol. In total, 1 mL of 10−<sup>1</sup> dilution of one gram of oven-dried soil or marine sample was transferred to 9 mL of sterile 5 mM-phosphate buffer (pH 7.0) containing phenol at a final concentration of 1.5%. The sample was then heated and diluted in serial dilution (10−1, 10−2, 10−3). Next, 100 or 200 μL of each dilution was spread over the surface of HV medium agar plates and incubated for 2–4 weeks at 28 ◦C.

### *3.3. Antimicrobial Bioassays*

The preliminary screening of actinomycetal strains for antimicrobial activity was performed using the agar plug diffusion method (see Supplementary for test plate preparation). Gram-positive (*B. subtilis* ATCC6051, *M. luteus*, and *S. carnosus* TM300) and Gram-negative bacteria (*E. coli* K12 W3110 and *P. fluorescens*) were chosen as test organisms. The isolates were spread evenly over the agar plate surface of soya flour mannitol medium (MS) (mannitol 20 g, soy flour (full fat) 20 g, agar 16 g in 1 L of distilled water) [80] and incubated for 10 days at 28 ◦C. Agar discs of the 10 days inoculum were cut aseptically with a cork borer (9 mm diameter) and placed on the bioassay test plate. Bioassays to determine optimal cultivation conditions in the liquid culture were examined using a disc diffusion assay against the test Gram-positive (*B. subtilis* ATCC6051, *M. luteus*, and *S. carnosus* TM300) and Gram-negative bacteria (*E. coli* K12 W3110 and *P. fluorescens*). In total, 10 μL methanolic extract obtained from liquid cultures of the actinomycetal strains was pipetted on a filter disc (6 mm) and then placed on the respective test plates. In addition, 5 μL kanamycin (50 μg/mL) was used as positive control and 10 μl methanol as a negative control.

The bioassay plates were incubated overnight at 37 ◦C for *B. subtilis*, *E. coli*, and *S. carnosus* and at 28 ◦C for *M. luteus* and *P. fluorescens* to allow for the test organisms' growth. The antimicrobial activity of the isolates was assessed by measuring the diameter of the inhibition zone (mm) around the agar plug or the discs. All bioassay tests were carried out as three independent biological replicates.

#### *3.4. Isolation of Genomic DNA and 16S rDNA Phylogenetic Analysis*

For isolation of genomic DNA, the producer strains were grown for two days in 50 mL of R5 medium at 30 ◦C [81]. The genomic DNA was extracted and purified with the Nucleospin® Tissue kit from Macherey-Nagel (catalog number 740952.50) following the standard protocol from the manufacturer. The DNA was applied as a PCR template for 16S rRNA gene amplification using polymerase chain reaction (PCR). Primers used for PCR were 27Fbac (5 -AGAGTTTGATCMTGGCTCAG-3 ) and 1492Runi (5 -TACGGTTACCTTAC GACTT-3 ). The PCR amplicons were subcloned into the cloning vector pDrive (Qiagen) using basic DNA manipulation procedures as previously described by Sambrook et al. [122]. The respective 16S rDNA fragments were sequenced at MWG Eurofins (Ebersberg, Germany) with primers 27Fbac. The 16S rDNA sequence data were analyzed using the EzTaxon database (https://www.ezbiocloud.net, accessed on 28 May 2018).

#### *3.5. Phylogenomic and Genome Mining Analysis*

For phylogenomic and genome mining studies, full-genome sequence data have been obtained as reported previously [57–59]. Genomic DNA was isolated to construct a 10–20 kb paired-end library for sequencing by Macrogen (Seoul, South Korea) with the Pacific

Biosciences RS II technology (Pacbio). The genome was assembled using Hierarchical Genome Assembly (HGAP) V.3. and annotated with Prokka version 1.12b and the NCBI Prokaryotic Genome Annotation Pipeline (PGAP). The phylogenomic analysis of the nine selected strains was carried out with the Type (Strain) Genome Server (TYGS), a free bioinformatics tool (https://tygs.dsmz.de/, accessed on 13 December 2019) for wholegenome-based taxonomic analysis [63]. The identification of potential biosynthesis gene clusters (BGCs) was accomplished by analyzing the genome sequences with antiSMASH version 5.0 [44]. The antiSMASH results were further analyzed using the BiG-SCAPE platform [78] to cluster the predicted BGCs into gene cluster families (GCFs) based on their sequences and Pfam protein family similarities [79]. BiG-SCAPE was conducted on global mode with default parameters [78], with the exception of the raw distance cutoff and the "–mix"parameter. Raw distance cutoff was set to 0.4 to ensure that even clusters with a pairwise distance higher than 0.3 (the default) were included in the output. The resulting network of BiG-SCAPE was visualized with Cytoscape version 3.7.2 [82].
