Temperature-gradient incubation isolates multiple competitive species from a single environmental sample

Abstract

High-throughput sequencing has allowed culture-independent investigation into a wide variety of microbiomes, but sequencing studies still require axenic culture experiments to determine ecological roles, confirm functional predictions and identify useful compounds and pathways. We have developed a new method for culturing and isolating multiple microbial species with overlapping ecological niches from a single environmental sample, using temperature-gradient incubation. This method was more effective than standard serial dilution-to-extinction at isolating methanotrophic bacteria. It also highlighted discrepancies between culture-dependent and -independent techniques; 16S rRNA gene amplicon sequencing of the same sample did not accurately reflect cultivatable strains using this method. We propose that temperature-gradient incubation could be used to separate out and study previously 'unculturable' strains, which co-exist in both natural and artificial environments.

Keywords: 16S rRNA; Culture; Environmental sample; Isolation; Methanotroph; Temperature gradient.

Fig. 1.

(a) Sampling location within New Zealand. The sampling site is indicated with a red triangle; (b) photograph of the sampling site, taken from [34].

Fig. 2.

Distribution of 16S rRNA sequences by phyla from the water (GDS1) and sediment (GDS2) samples.

Fig. 3.

Graph of the number of 16S rRNA amplicon sequences assigned to each methanotrophic genus from the two sample sites. The bar colour indicates whether methanotroph strains from that genus were also cultivated from that site: green, successful cultivation: pink, unsuccessful cultivation; grey, cultivation not attempted. The number of strains cultivated is given in Table 1.

Fig. 4.

Molecular phylogenetic analysis of 16S rRNA gene sequences from isolated methanotrophs and closely related strains, with growth temperature ranges for each. The temperature of the original sample (37 °C) is highlighted with a yellow line. Purple names and bars indicate strains isolated in this study; blue names and bars indicate strains identified through Illumina sequencing of DNA extracted from the Golden Springs site; black names and bars indicate closely related strains not identified through sequencing. a, no T _min published; b, no T _max published; c, only growth temperature published. The evolutionary history was inferred using the aximum-likelihood method based on the Jukes–Cantor model [55]. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. The scale bar represents nucleotide substitutions per site. Evolutionary analyses were conducted in mega7 [53].