Diamide-based screening method for the isolation of improved oxidative stress tolerance phenotypes in Bacillus mutant libraries

Abstract

During their life cycle, bacteria are exposed to a range of different stresses that need to be managed appropriately in order to ensure their growth and viability. This applies not only to bacteria in their natural habitats but also to bacteria employed in biotechnological production processes. Oxidative stress is one of these stresses that may originate either from bacterial metabolism or external factors. In biotechnological settings, it is of critical importance that production strains are resistant to oxidative stresses. Accordingly, this also applies to the major industrial cell factory Bacillus subtilis. In the present study, we, therefore, developed a screen for B. subtilis strains with enhanced oxidative stress tolerance. The results show that our approach is feasible and time-, space-, and resource-efficient. We, therefore, anticipate that it will enhance the development of more robust industrial production strains with improved robustness under conditions of oxidative stress.

Keywords: Bacillus subtilis; diamide; disulfide; library screening; mutagenesis.

Fig 1

LB agar plates with differing diamide concentrations onto which a 100 µL suspension of heat-shocked B. subtilis DB430 ΔlipA carrying pMarB (OD₆₀₀ ≈ 3) was plated. Note that the colony density (varying from fully overgrown to defined single colonies) depended on the diamide concentration (left to right: 0, 100, and 200 µM) and incubation time (40 h in these figures).

Fig 2

Example of the selection of candidate transposon mutants with increased diamide tolerance by microtiter plate screening. Diamide tolerance was measured as the time in hours required to reach exponential growth in LB broth with 2 mM diamide. The solid and dashed red lines indicate the mean and standard deviation, respectively, as measured for the parental strain B. subtilis DB430 ΔlipA that was included as an internal control. Mutants with a growth delay below the lower dashed line were considered potentially diamide-tolerant candidates. The mutant showing the shortest growth delay was subsequently shown to carry a transposon insertion in the pchR gene.

Fig 3

Bar diagram showing the time to reach the exponential phase (OD₆₀₀ > 0.5) for different deletion mutants in microtiter plate in the presence of various concentrations of diamide compared to the parental strain (reference). A reduction in this time compared to the parental strain implies that the phenotype of the respective transposon mutant is related to inactivation of the gene identified by FPNI-PCR and not to unidentified off-target effects.

Fig 4

Growth curves measured for microtiter plate cultures of the parental strain (DB430 ΔlipA, reference) and the ΔpfkA, ΔribT, or ΔbshC deletion strains at various diamide concentrations. Diamide was added during the exponential phase of growth (dashed red line) to test tolerance against higher concentrations compared to the parental strain. In the plot depicting the growth of the ribT mutant at 6 mM diamide, the area shaded in green marks the variation in the individual growth curves.

Fig 5

Measurement of OD₆₀₀ of shake flask cultures of the parental strain (blue and orange) and pfkA or ribT deletion mutants. Optical densities were measured before the addition of diamide (0) and at 0.5, 1, and 4 h after the addition of 2 mM diamide. Bacterial cultures with no added diamide (control group) showed no growth inhibition. Bacterial cultures with 2 mM diamide showed growth phenotypes depending on the strain.

Fig 6

Volcano plots comparing the abundance of proteins of the pfkA deletion strain and the parental strain in the presence of 2 mM diamide added after OD₆₀₀ reached 1.0. The upper and lower panels show differences in protein abundance at the different sampling points (0.5, 1, and 4 h) in the absence or presence of diamide, respectively. Proteins whose abundance was significantly changed are indicated in orange.

Fig 7

Volcano plots comparing the ribT deletion strain with the parental strain in the presence of 2 mM diamide added after OD₆₀₀ reached 1.0. At 4 h for the control group comparison (top right), the majority of proteins whose abundance significantly changed (orange) were those encoded by the rib operon. Overall, little difference was detectable in the proteomes of the two strains except for the diamide-treated group at 4 h after diamide addition, where the ribT deletion strain had resumed growth, while growth of the parental strain was still halted.

Fig 8

Oxidation rates of peptidyl cysteine residues based on differential cysteine labeling measured for control (left) and diamide-treated groups (right) as determined 0.5 h after the introduction of 2 mM diamide or fresh medium, respectively. The data points represent the relative abundance of peptides with oxidized cysteine residues per “bin” showing the degree to which their cysteine residues were oxidized. The oxidation rates are shown for the parental strain (WT) as well as the ΔpfkA and ΔribT mutants.

Fig 9

Volcano plots of the parental strain showing relatively few significant differently abundant proteins (orange) after treatment with diamide after 0.5 h (left) compared to after 1 h (right). The two deletion strains showed a similar response (Fig. S3).