Genomic adaptations enabling Acidithiobacillus distribution across wide-ranging hot spring temperatures and pHs

Abstract

Background: Terrestrial hot spring settings span a broad spectrum of physicochemistries. Physicochemical parameters, such as pH and temperature, are key factors influencing differences in microbial composition across diverse geothermal areas. Nonetheless, analysis of hot spring pools from the Taupo Volcanic Zone (TVZ), New Zealand, revealed that some members of the bacterial genus, Acidithiobacillus, are prevalent across wide ranges of hot spring pHs and temperatures. To determine the genomic attributes of Acidithiobacillus that inhabit such diverse conditions, we assembled the genomes of 19 uncultivated hot spring Acidithiobacillus strains from six geothermal areas and compared these to 37 publicly available Acidithiobacillus genomes from various habitats.

Results: Analysis of 16S rRNA gene amplicons from 138 samples revealed that Acidithiobacillus comprised on average 11.4 ± 16.8% of hot spring prokaryotic communities, with three Acidithiobacillus amplicon sequence variants (ASVs) (TVZ_G1, TVZ_G2, TVZ_G3) accounting for > 90% of Acidithiobacillus in terms of relative abundance, and occurring in 126 out of 138 samples across wide ranges of temperature (17.5-92.9 °C) and pH (1.0-7.5). We recovered 19 environmental genomes belonging to each of these three ASVs, as well as a fourth related group (TVZ_G4). Based on genome average nucleotide identities, the four groups (TVZ_G1-TVZ_G4) constitute distinct species (ANI < 96.5%) of which three are novel Acidithiobacillus species (TVZ_G2-TVZ_G4) and one belongs to Acidithiobacillus caldus (TVZ_G1). All four TVZ Acidithiobacillus groups were found in hot springs with temperatures above the previously known limit for the genus (up to 40 °C higher), likely due to significantly higher proline and GC contents than other Acidithiobacillus species, which are known to increase thermostability. Results also indicate hot spring-associated Acidithiobacillus have undergone genome streamlining, likely due to thermal adaptation. Moreover, our data suggest that Acidithiobacillus prevalence across varied hot spring pHs is supported by distinct strategies, whereby TVZ_G2-TVZ_G4 regulate pH homeostasis mostly through Na⁺/H⁺ antiporters and proton-efflux ATPases, whereas TVZ_G1 mainly relies on amino acid decarboxylases.

Conclusions: This study provides insights into the distribution of Acidithiobacillus species across diverse hot spring physichochemistries and determines genomic features and adaptations that potentially enable Acidithiobacillus species to colonize a broad range of temperatures and pHs in geothermal environments. Video Abstract.

Keywords: Acidithiobacillus; Adaptation; Genome streamlining; Hot spring; Temperature; pH.

Fig. 1

Plots showing the proportion of Acidithiobacillus ASVs observed in microbial communities across the TVZ hot spring sites. a Total relative abundance (%) of prokaryotic communities summed across all samples collected across six geothermal areas in the TVZ. b Boxplots showing total relative abundance of Acidithiobacillus versus other prokaryotes in each sample. Note that there are 12 samples that did not harbor Acidithiobacillus (shown at 0% in the left box plot, and 100% in the right box plot)

Fig. 2

Scatter plots showing relative abundances and prevalence of the four TVZ hot spring Acidithiobacillus ASVs that correspond with recovered genomes of a A. caldus TVZ_G1; b Acidithiobacillus sp. TVZ_G2; c Acidithiobacillus sp. TVZ_G3; and d Acidithiobacillus TVZ_G4. Bubble size represents the relative abundance of each ASV per sample. The number of bubbles represents the prevalence of each Acidithiobacillus ASV

Fig. 3

Plots showing the physicochemical ranges of TVZ Acidithiobacillus and those known for other Acidithiobacillus. Ranges are given for a temperature (with black bars indicating optimal growth temperatures calculated by GrowthPred) and b pH. Ranges shown for subaerial sinters are from the corresponding hot spring waters from which the digitate sinters form. All temperature and pH ranges shown for the TVZ species represent the springs from where they were found, while those for other Acidithiobacillus are from growth experiment characterizations

Fig. 4

A maximum-likelihood concatenated core gene phylogenetic tree with 100 times bootstrapping of TVZ and other Acidithiobacillus species. Bold script indicates the representative MAG of each species (TVZ_G2-4) or subspecies (TVZ_G1) level cluster. Thermithiobacillus tepidarius from another family and Acidiferrobacter thiooxydans from another class are used as outgroups

Fig. 5

Scatter plots of the TVZ and other Acidithiobacillus species. Plots a–c show positive correlations between a GC content (%) and optimal growth temperature (°C) (R = 0.75, p = 3.0 × 10⁻¹¹); b predicted proline contents (%) and optimal growth temperature (°C) (R = 0.92, p < 2.2 × 10⁻¹⁶); and c percentage of paralogous proteins and coding sequences (CDS) (R = 0.63, p = 1.7 × 10⁻⁷). Plots d–f show negative correlations between d GC content (%) and estimated genome size (Mbp) (R = − 0.51, p = 7.0 × 10⁻⁵); e predicted proline content (%) and estimated genome size (Mbp) (R = − 0.81, p = 3.4 × 10⁻¹⁴); and f optimal growth temperature (°C) and estimated genome size (Mbp) (R = − 0.86, p < 2.2 × 10⁻¹⁶). g Shows a positive correlation between minimum generation time (hr) and estimated genome size (Mbp), with reference A. caldus and the Taiwanese Acidithiobacillus (UBA2486) as outliers (R = 0.36, p = 0.008). h Shows a negative correlation between number of non-coding RNA and optimal growth temperature (°C) (R = − 0.45, p = 0.00044). i Shows a positive correlation between percent of non-coding RNA and estimated genome size (Mbp) (R = 0.24, p = 0.08). Optimal growth temperatures were derived from GrowthPred calculation. The nearest (co-clustering) unclassified Acidithiobacillus species (black rimmed-circle) to TVZ_G2 is Acidithiobacillus sp. UBA2486 in all plots. Pearson’s correlation coefficients and t-distribution tables were used to determine the correlation coefficients and the significances, respectively, as described in the methods

Fig. 6

Heatmap showing the number of proteins associated with pH homeostasis in the TVZ and other Acidithiobacillus species. Bold script indicates the representative genome of the TVZ Acidithiobacillus species. Accession IDs of genes analyzed can be found in Table S7