Escherichia coli DNA ligase B may mitigate damage from oxidative stress

Abstract

Escherichia coli encodes two DNA ligases, ligase A, which is essential under normal laboratory growth conditions, and ligase B, which is not. Here we report potential functions of ligase B. We found that across the entire Enterobacteriaceae family, ligase B is highly conserved in both amino acid identity and synteny with genes associated with oxidative stress. Deletion of ligB sensitized E. coli to specific DNA damaging agents and antibiotics resulted in a weak mutator phenotype, and decreased biofilm formation. Overexpression of ligB caused a dramatic extension of lag phase that eventually resumed normal growth. The ligase function of ligase B was not required to mediate the extended lag phase, as overexpression of a ligase-deficient ligB mutant also blocked growth. Overexpression of ligB during logarithmic growth caused an immediate block of cell growth and DNA replication, and death of about half of cells. These data support a potential role for ligase B in the base excision repair pathway or the mismatch repair pathway.

Fig 1. E. coli ligase B.

(A) Ligase B amino acid sequence similarity tree. Ligase B peptide sequences from 46 species of Enterobacteriaceae are shown compared to the outgroup, ligase A. The distance of 0.2 substitutions per site is indicated. (B) The conserved ligB gene neighborhood (S2 Fig), represented by the region in E. coli MG1655.

Fig 2. Effect of ligB deletion on bacterial response to DNA damaging agents.

Survival of the parent strain MG1655 (●) compared to the isogenic ΔligB strain (○). Percent survival was determined by dividing the number of CFU/ml with the indicated treatment by the number of CFU/ml without treatment and these data are shown on a semi-log plot as a function of increasing DNA damaging agent. Error bars denote the standard deviation calculated from experiments done in triplicate (repeated three times with similar results). Comparing the parent and the mutant strain for each specific condition by Student’s T-test, p < 0.05(_*).

Fig 3. Effect of ligB overexpression on E. coli growth.

(A) E. coli MG1655 either without (■) or with (●) plasmid pCA24N JW3622 [48] (pLigB), which encodes ligB under control of the T5 lac promoter, were grown in the absence or presence of IPTG. As a negative control, E. coli MG1655 with pCA24N [48] (pEmpty) (○), which is the parental “empty” vector, was included. Average lag time was quantified using a custom R program (21) and plotted as a function of IPTG concentration. Error bars denote the standard deviation calculated from experiments done in triplicate repeated at least two times. Where error bars are not shown, they were smaller than the symbol. (B) Representative growth curves of MG1655 with overexpression of pLigB and of ligase-deficient pLigB K124A with increasing levels of IPTG induction (lighter color (lower) to darker color (higher)). The experiment was repeated three times each in triplicate with the same results.

Fig 4. Effect of ligB overexpression on DNA replication and viability.

The experiments shown in A,B, and C were performed concurrently. Effect of ligB overexpression on (A) bacterial growth, (B) cell viability (value in parentheses determined using OD₆₀₀ value), and (C) incorporation of [methyl-³H]-thymidine. Lines connecting data points were added merely to aid visualization.

Fig 5. Effect of ligB overexpression during logarithmic phase.

E. coli MG1655 with plasmid pLigB or pEmpty were grown overnight, diluted into fresh growth medium, grown to OD₆₀₀ = ~0.1, and induced with 0.2 mM IPTG. At times 0, 1, and 2 hr after IPTG addition, the OD₆₀₀ of the growing cultures was measured (A) and aliquots of each culture were submitted to fluorescent live/dead staining (B). Representative images are shown from two separate experiments with the same results. n is the total number of cells counted for each experimental condition. (C) Plot of the percentage of living cells for each strain at each time point post IPTG induction.

Fig 6. Effect of ΔligB on growth in the presence of cadmium.

(A) Growth of E. coli MG1655 or the isogenic ΔligB strain was measured at OD₄₅₀ [19] in the presence of the various indicated concentrations of cadmium. Growth without cadmium is in black and increasing cadmium is depicted from light blue (lowest) to dark blue (highest). The experiment shown was repeated three additional times (each in triplicate) with similar results. (B) Averaged growth lag times of E. coli MG1655 (●) or the isogenic ΔligB mutant (○) as a function of cadmium concentration. Lag times were calculated with a custom R program [18]. Error bars denote standard deviation.