Genomic and Transcriptomic Investigation of the Physiological Response of the Methylotroph Bacillus methanolicus to 5-Aminovalerate

Abstract

The methylotrophic thermophile Bacillus methanolicus can utilize the non-food substrate methanol as its sole carbon and energy source. Metabolism of L-lysine, in particular its biosynthesis, has been studied to some detail, and methanol-based L-lysine production has been achieved. However, little is known about L-lysine degradation, which may proceed via 5-aminovalerate (5AVA), a non-proteinogenic ω-amino acid with applications in bioplastics. The physiological role of 5AVA and related compounds in the native methylotroph was unknown. Here, we showed that B. methanolicus exhibits low tolerance to 5AVA, but not to related short-chain (C4-C6) amino acids, diamines, and dicarboxylic acids. In order to gain insight into the physiological response of B. methanolicus to 5AVA, transcriptomic analyses by differential RNA-Seq in the presence and absence of 5AVA were performed. Besides genes of the general stress response, RNA levels of genes of histidine biosynthesis, and iron acquisition were increased in the presence of 5AVA, while an Rrf2 family transcriptional regulator gene showed reduced RNA levels. In order to test if mutations can overcome growth inhibition by 5AVA, adaptive laboratory evolution (ALE) was performed and two mutants-AVA6 and AVA10-with higher tolerance to 5AVA were selected. Genome sequencing revealed mutations in genes related to iron homeostasis, including the gene for an iron siderophore-binding protein. Overexpression of this mutant gene in the wild-type (WT) strain MGA3 improved 5AVA tolerance significantly at high Fe²⁺ supplementation. The combined ALE, omics, and genetics approach helped elucidate the physiological response of thermophilic B. methanolicus to 5AVA and will guide future strain development for 5AVA production from methanol.

Keywords: 5-aminovalerate; Bacillus methanolicus; adaptive laboratory evolution; bioplastics; differential RNA-Seq; methylotroph; physiology; whole-genome sequencing.

FIGURE 1

Growth of B. methanolicus in the presence of 5AVA (A), GABA (B), 6ACA (C), L-lysine (D), L-glutamate (E), putrescine (F), cadaverine (G), glutarate (H), and succinate (I). B. methanolicus WT (black), AVA6 (light blue; A), and AVA10 (blue) were cultivated in MVcMY medium supplemented with 0, 10, 50, 100, 200, and 400 mM 5AVA (A) and 10, 50, and 100 mM (B–H) of the other compounds, respectively. The growth rates (empty diamonds, dotted lines) and ΔOD₆₀₀ (full squares, straight lines) are shown as means and standard deviations of triplicate cultivations.

FIGURE 2

Growth of B. methanolicus WT with 5AVA as a carbon/nitrogen source. B. methanolicus WT was cultivated in MVcM medium supplemented with 5AVA as a carbon source (blue circles) and as a nitrogen source (green diamonds) or with methanol as a carbon source and ammonium sulfate as a nitrogen source (black squares). Values and error bars represent means and standard deviations of triplicate cultivations.

FIGURE 3

Differential gene expression analysis of B. methanolicus WT in the presence and absence of 50 mM 5AVA (A). The log₂-transformed M/A plot reveals upregulated (UP; blue triangles) and downregulated (DOWN; red triangles) genes with an M-value > | 1|, which are categorized by function (B). (A) Differential expression is below this threshold and non-differential expressions are indicated by dark gray squares and light gray circles, respectively. Cells were cultivated in MVcMY medium to the mid-exponential growth phase, and RNA was subsequently isolated, reverse transcribed to cDNA, and sequenced in biological triplicates. Differential expression was determined by DESeq2 with an adjusted p-value ≤ 0.01 according to the Wald test (Love et al., 2014).

FIGURE 4

ALE for enhanced 5AVA tolerance (A) and growth of strain AVA10 with different 5AVA concentrations (B). B. methanolicus cells were cultivated in MVcMY medium supplemented with various concentrations of 5AVA. (A) OD₆₀₀ values during the ALE experiment with B. methanolicus WT are plotted for growth without (black squares) or with 5AVA (blue squares; 5AVA concentrations ranging from 50 to 400 mM indicated as blue-shaded area). Passage intervals varied (8–72 h), and every second passage was harvested and stored at -80°C after plating and isolation of a single colony. (B) ALE strain AVA10 was cultivated in MVcMY medium supplemented with 0 mM (black squares), 150 mM (light blue circles), and 300 mM (blue diamonds) 5AVA. The 5AVA concentration in the supernatant was determined by HPLC. Values and error bars represent means and standard deviations of triplicate cultivations.

FIGURE 5

SNPs determined by whole-genome sequencing of B. methanolicus WT, AVA6, and AVA10 (A) and predicted protein structures of three key target genes BMMGA3_RS03040 (B), BMMGA3_RS13980 (C), and BMMGA3_RS14080 (D) comprising respective SNPs in AVA10. Isolated genomic DNA of B. methanolicus strains WT, AVA6, and AVA10 were sequenced, and SNPs in all CDSs of B. methanolicus were detected. (A) Venn diagram for observed SNPs. Gene loci were abbreviated such that the leading “BMMGA3_RS” was cut, leaving a five-digit identifier. Optional tailing numbers in brackets indicate multiple SNPs. Yellow identifiers refer to SNPs in genes coding for ribosomal RNA, and blue and red identifiers highlight SNPs in upregulated and downregulated genes, respectively, according to the differential RNA-Seq analysis. (B–D) 3D protein models were predicted using Phyre2 (Kelley et al., 2015). Selected SNPs resulting in amino acid exchanges are highlighted in red. Intergenic mutations found in AVA6 and AVA10 are described in Table 4.

FIGURE 6

Tolerance of B. methanolicus WT (pBV2xp) (A), XP03040 (B), XP13980 (C), and XP14080 (D) for 5AVA with cofactor titration. B. methanolicus cells were cultivated in MVcMY in the presence (+) and absence (−) of 50 mM 5AVA. The growth rates (empty diamonds) and ΔOD₆₀₀ (full squares) were determined. The cofactors FeSO₄ (0–1,000 μM) and CuSO₄ (0–3,200 μM) were titrated for strains XP13980 (C) and XP14080 (D), respectively, according to the function of proteins coded by their overexpressed genes and ligand prediction with COACH-D. Values and error bars represent means and standard deviations of triplicate cultivations.