Deacetylation of topoisomerase I is an important physiological function of E. coli CobB

Abstract

Escherichia coli topoisomerase I (TopA), a regulator of global and local DNA supercoiling, is modified by Nε-Lysine acetylation. The NAD+-dependent protein deacetylase CobB can reverse both enzymatic and non-enzymatic lysine acetylation modification in E. coli. Here, we show that the absence of CobB in a ΔcobB mutant reduces intracellular TopA catalytic activity and increases negative DNA supercoiling. TopA expression level is elevated as topA transcription responds to the increased negative supercoiling. The slow growth phenotype of the ΔcobB mutant can be partially compensated by further increase of intracellular TopA level via overexpression of recombinant TopA. The relaxation activity of purified TopA is decreased by in vitro non-enzymatic acetyl phosphate mediated lysine acetylation, and the presence of purified CobB protects TopA from inactivation by such non-enzymatic acetylation. The specific activity of TopA expressed from His-tagged fusion construct in the chromosome is inversely proportional to the degree of in vivo lysine acetylation during growth transition and growth arrest. These findings demonstrate that E. coli TopA catalytic activity can be modulated by lysine acetylation-deacetylation, and prevention of TopA inactivation from excess lysine acetylation and consequent increase in negative DNA supercoiling is an important physiological function of the CobB protein deacetylase.

Figure 1.

2-Dimensional chloroquine gel analysis comparing DNA supercoiling in BW25113 (wild-type) and JW1106 (ΔcobB) mutant strains. Plasmids pBAD/Thio and pCobB were extracted from LB cultures at OD₆₀₀ = 0.8. The first dimension electrophoresis was carried out in 0.8% agarose gel with TAE buffer containing 3 μg/ml chloroquine at 3 V/cm for 16 h. The agarose gel was then rotated 90° for electrophoresis in the second dimension with 25 μg/ml chloroquine in the TAE buffer. The schematic diagram shows the expected distribution of positively and negatively supercoiled DNA topoisomers following electrophoresis.

Figure 2.

Relaxation assay of TopA activity in total soluble cell lysates of BW25113 (wild-type) and JW1106 (ΔcobB) and their pCobB transformants. The indicated amount of total soluble proteins from BW25113 (wild-type) and JW1106 (ΔcobB) LB cultures at OD₆₀₀ = 0.8 were incubated with 150 ng of negatively supercoiled plasmid DNA substrate (S) at 37°C for 20 min. DNA were electrophoresed in 0.8% agarose gel with TAE buffer containing 1 μg/ml chloroquine at 3 V/cm for 16 h. Reaction products from partial relaxation of the supercoiled plasmid DNA substrate by increasing amounts of purified recombinant TopA are included to show the effect of increasing TopA catalytic activity on plasmid electrophoretic mobility.

Figure 3.

TopA expression level in BW25113 (wild-type) and JW1106 (ΔcobB) mutant. (A) 20 μg total proteins were resolved by 10% SDS-PAGE, and TopA protein was detected by western blot analysis using monoclonal antibodies against TopA. (B) Quantification of four independent western blot experiments as shown in (A). The level of TopA in the wild-type strain in each experiment was set at 100%, and the corresponding level of TopA in the ΔcobB mutant strain is shown. (C) Quantitative PCR measurements of topA transcript levels from three independent experiments normalized against either idnT or hcaT transcript level as internal reference. The topA transcript level in the wild-type strain was set at 100% for each experiment. Error bars indicate standard deviation of relative topA transcript level in the ΔcobB mutant strain.

Figure 4.

Effect of recombinant TopA overexpression on the slow-growth phenotype of the ΔcobB mutant. (Filled circle) Wild-type BW25113 strain transformed with pBAD/Thio; (unfilled circle) ΔcobB JW1106 strain transformed with pBAD/Thio; (filled square) Wild-type strain transformed with pETOP for TopA overexpression; (unfilled square) ΔcobB mutant strain transformed with pETOP; (filled triangle) wild-type strain transformed with pCobB; (unfilled triangle) ΔcobB mutant strain transformed with pCobB. The LB medium contained no arabinose (A) or 0.0001% arabinose (B).

Figure 5.

Decrease in TopA activity from acetyl phosphate mediated nonenzymatic lysine acetylation. (A) Western blot analysis of nonenzymatically acetylated TopA. Purified E. coli TopA (1 μg) was incubated with 2 and 5 mM acP at 37°C for 4 h. Acetylated TopA was visualized by western blot analysis with anti-acetyllysine antibody. TopA on the membrane was stained with Coomassie blue. (B) Following incubation with and without acetyl phosphate, serial dilutions of TopA (32, 16, 8, 4, 2 ng) were incubated with 150 ng negatively supercoiled plasmid DNA (S) at 37°C for 30 min to compare the relaxation activity.

Figure 6.

CobB deacetylation counters the effect of acetyl phosphate mediated lysine acetylation on TopA relaxation activity. Purified E. coli TopA (1 μg) and 2 mM acP was incubated with or without CobB (0.65:1 molar ratio to TopA) at 37°C for 4 h. (A) Acetylation level of TopA was compared by western blot analysis with anti-acetyllysine antibody. TopA protein was visualized by Coomassie blue staining. (B) Following incubation with acetyl phosphate in the absence or presence of CobB, serial dilutions of TopA (20, 10, 5 ng) were incubated with 150 ng negatively supercoiled plasmid DNA (S) at 37°C for 30 min to assay the TopA relaxation activity.

Figure 7.

Direct interaction between TopA and CobB. Purified recombinant TopA and His-tagged CobB were allowed to interact before pull-down by HisPur Cobalt Agarose Resin. The presence of TopA in the bound proteins eluted from the cobalt resin was visualized by western blot analysis with anti-TopA antibodies (A). The co-eluted His-CobB was stained by Coomassie blue following SDS PAGE (B) or silver on the transfer membrane (C).

Figure 8.

Topoisomerase I specific activity is inversely proportional to degree of lysine acetylation during growth transition and growth arrest. (A) His-TopA expressed from the chromosome in strain YN1434 was purified from buffered and glucose supplemented TB7 cultures at exponential phase (OD₆₀₀ = 0.7), entry into stationary phase (OD₆₀₀ = 1.7) or growth arrest by Ni-affinity column. Following transfer of equal amounts of His-TopA (1 μg) from SDS gel onto nitrocellulose membrane, the His-TopA was stained with MemCode reversible stain and analyzed for level of lysine acetylation with antibodies against acetylated lysine (acK). The signals of His-TopA staining was used to normalize the acK signal for comparison of degree of lysine acetylation. (B) Serial dilutions of the purified His-TopA was assayed for relaxation activity using supercoiled plasmid DNA (S) as substrate.

Figure 9.

Location of lysines in E. coli TopA observed as acetylated lysines in proteomics studies. The lysine residues reported to be acetylated are shown in the crystal structures of the (A) TopA N-terminal domains D1–D4 in covalent complex with cleaved ssDNA substrate (PDB 3PX7); (B) full length TopA with ssDNA non-covalently bound to the C-terminal domains (PDB 4RUL).