Adaptive laboratory evolution of a thermophile toward a reduced growth temperature optimum

Abstract

Thermophily is an ancient trait among microorganisms. The molecular principles to sustain high temperatures, however, are often described as adaptations, somewhat implying that they evolved from a non-thermophilic background and that thermophiles, i.e., organisms with growth temperature optima (T_OPT) above 45°C, evolved from mesophilic organisms (T_OPT 25-45°C). On the contrary, it has also been argued that LUCA, the last universal common ancestor of Bacteria and Archaea, may have been a thermophile, and mesophily is the derived trait. In this study, we took an experimental approach toward the evolution of a mesophile from a thermophile. We selected the acetogenic bacterium T. kivui (T_OPT 66°C) since acetogenesis is considered ancient physiology and cultivated it at suboptimal low temperatures. We found that the lowest possible growth temperature (T_MIN) under the chosen conditions was 39°C. The bacterium was subsequently subjected to adaptive laboratory evolution (ALE) by serial transfer at 45°C. Interestingly, after 67 transfers (approximately 180 generations), the adapted strain Adpt45_67 did not grow better at 45°C, but a shift in the T_OPT to 60°C was observed. Growth at 45°C was accompanied by a change in the morphology as shorter, thicker cells were observed that partially occurred in chains. While the proportion of short-chain fatty acids increased at 50°C vs. 66°C in both strains, Adpt45_67 also showed a significantly increased proportion of plasmalogens. The genome analysis revealed 67 SNPs compared to the type strain, among these mutations in transcriptional regulators and in the cAMP binding protein. Ultimately, the molecular basis of the adaptation of T. kivui to a lower T_OPT remains to be elucidated. The observed change in phenotype is the first experimental step toward the evolution of thermophiles growing at colder temperatures and toward a better understanding of the cold adaptation of thermophiles on early Earth.

Keywords: Thermoanaerobacter kivui; acetogens; adaptive laboratory evolution; cold adaptation; origin of life; thermophiles.

Figure 1

Serial passaging modified after Strauss et al. (2019). Five subpopulations of T. kivui were incubated for 24 h in a complex medium containing 25 mM glucose as a carbon source. The subpopulation with the highest increase in OD₆₀₀ was used to generate five new subpopulations in a fresh medium.

Figure 2

Previously reported temperature ranges and optimal growth temperatures for different Thermoanaerobacter species. Temperature range, gray; T_OPT, red. Data from Onyenwoke and Wiegel (2015).

Figure 3

Growth of T. kivui strains (A) DSM2030 (type strain) and (B) Adpt45_67 at different temperatures, and (C) the corresponding growth rates. Cells were grown in 50 ml complex medium in 100 ml serum bottles with 25 mM glucose as a carbon source. Red triangles, 66°C; black open circles, 66°C; gray squares, 55°C; black squares, 50°C; gray triangles, 45°C; turquoise inverted triangles, 40°C. The error indicated shows the standard deviation from four biological replicates. Significant statistical differences (***p < 0.0005, two-way ANOVA with Tukey's test) between the type strain and the adapted strains at each temperature condition are shown.

Figure 4

Growth of T. kivui strains (A) DSM2030 (type strain) and Adpt45_67 in the defined medium at 66°C and 60°C, and (B) the corresponding growth rates. Red, type strain at 66°C; white, Type strain at 60°C; gray, Adpt45_67 at 66°C; black, Adpt45_67 at 60°C. Cells were grown in 50 ml of medium in 100 ml serum bottles with 25 mM glucose as a carbon source. The error indicated shows the standard deviation from four biological replicates. Significant statistical differences (**p < 0.005, two-way ANOVA with Tukey's test) between the type strain and the adapted strains at each temperature condition are shown.

Figure 5

Morphology of T. kivui strains as observed by phase-contrast microscopy. T. kivui strains were grown in 50 ml of complex medium in 100 ml serum bottles with 25 mM glucose as a carbon source. (A) DSM2030 (type strain), 20 h at 66°C (T_OPT), (B) DSM2030 after five transfers at 45°C, 24 h at 45°C, (C) Adpt45_67, 24 h at 45°C, and (D) Adpt45_67, 24 h at 60°C (T_OPT).

Figure 6

Effect of growth temperature on the total fatty acid composition of T. kivui. T. kivui strains DSM2030 (A) and Adpt45_67 (B) were grown in a complex medium with 25 mM glucose at optimal (66°C and 60°C) and suboptimal (50°C) temperatures to an OD₆₀₀ of ~1. Left panels: red, iso-branched; gray, anteiso-unbranched; black, anteiso-branched. Middle panels: turquoise, fatty acids; light gray, plasmalogens; dark blue, fatty aldehydes. Right panels: yellow, long-chain >C₁₆; light blue, short-chain <C₁₆.