Isolation, cultivation, and characterization

Initial cultures were isolated from field samples collected from Tim’s Branch Creek a second-order stream on the SRS in Aiken, South Carolina, USA (Fig. 1B)¹¹. These samples were collected for a monitoring study of the geochemistry of Tim’s Branch. These samples were plated on tryptic soy agar plates (Fisherbrand™) and placed in a refrigerator at 4 °C. After several days, the plates were observed with clear zones around colonies of what turned out to be the bacterial isolate of interest. After proving to be a Gram Negative (GN) species, the Biolog GN2 kit (Biolog, Hayward, CA, USA) was used for initial identification of the isolated bacterium based on substrate utilization patterns²³. B. subtilis ATCC 49760 was used as a gram positive control.

To isolate the novel strain fresh soil samples were collected from Tims Branch Creek and were initially inoculated onto nutrient agar (Difco, Detroit, MI). Following this the most probable number (MPN) was determined. Cultures were fixed onto glass microscope slides, stained using a gram stain kit (Thermo Fisher Scientific), and observed under an Olympus BX41TF (Tokyo, Japan) microscope.

A further identification of the isolate was confirmed by sequencing of the 16S rRNA gene. DNA was isolated from cultures by placing one isolated colony in 150 µl of QuickExtract Reagent (Epicentre) according to the manufacturer’s instructions. The 16S rRNA gene was PCR amplified (30 cycles: 95C for 45 s, 57C for 45 s, 72C for 120 s) using the 16S-27F (5′-AGA GTT TGA TCM TGG CTC AG; Lane, 1991) and 16S-1492R (5′-GGT TAC CTT GTT ACG ACT T²⁴) primers and ExTaq (Takara) polymerase. Sequencing was performed by dye-terminator sequencing on an ABI Prism 3130XL Genetic Analyzer at the University of Delaware Sequencing and Genotyping Center using the 27F, 515F (5′-GTG YCA GCM GCC GCG GTA A²⁵), 926R (5′-GGACTACHVGGGTWTCTAAT²⁵), and 1492R to obtain bidirectional sequencing of the amplicon.

Resulting sequences were trimmed for quality and to remove primer sequences, then assembled into a 1426 nt contiguous sequence in Geneious (ver 10.2.6). The consensus sequence was analyzed using the RDP Classifier algorithm (ver. 2.1.3; 16S rRNA training set 18²⁶ and further homology-based analysis of the sequence was performed using RDP SeqMatch (release 11, update 5; Cole et al. ²⁷) and blastn (ver. 2.8.1²⁸) against the NCBI nr database (release 242). The sequence and the available near-full length type strain sequences from the genus Cupriavidus and Ralstonia pickettii were aligned to the SILVA (v.138.1) universal SSU rRNA profile using SINA (ver. 1.2.12²⁹). An approximate maximum likelihood tree was constructed with FastTree (ver. 2.1.11³⁰) with a GTR + CAT model and optimized gamma likelihoods. Final formatting of the tree was performed in Iroki (rev2020-06-29-1³¹).