The helicase DinG responds to stress due to DNA double strand breaks

Abstract

Neisseria meningitidis (Nm) is a Gram-negative nasopharyngeal commensal that can cause septicaemia and meningitis. The neisserial DNA damage-inducible protein DinG is a helicase related to the mammalian helicases XPD and FANCJ. These helicases belong to superfamily 2, are ATP dependent and exert 5' → 3' directionality. To better understand the role of DinG in neisserial genome maintenance, the Nm DinG (DinGNm) enzymatic activities were assessed in vitro and phenotypical characterization of a dinG null mutant (NmΔdinG) was performed. Like its homologues, DinGNm possesses 5' → 3' directionality and prefers DNA substrates containing a 5'-overhang. ATPase activity of DinGNm is strictly DNA-dependent and DNA unwinding activity requires nucleoside triphosphate and divalent metal cations. DinGNm directly binds SSBNm with a Kd of 313 nM. Genotoxic stress analysis demonstrated that NmΔdinG was more sensitive to double-strand DNA breaks (DSB) induced by mitomycin C (MMC) than the Nm wildtype, defining the role of neisserial DinG in DSB repair. Notably, when NmΔdinG cells grown under MMC stress assessed by quantitative mass spectrometry, 134 proteins were shown to be differentially abundant (DA) compared to unstressed NmΔdinG cells. Among the DNA replication, repair and recombination proteins affected, polymerase III subunits and recombinational repair proteins RuvA, RuvB, RecB and RecD were significantly down regulated while TopA and SSB were upregulated under stress condition. Most of the other DA proteins detected are involved in metabolic functions. The present study shows that the helicase DinG is probably involved in regulating metabolic pathways as well as in genome maintenance.

Fig 1. Computational analysis of DinG_Nm.

A) Plotting of the similarity of DinG_Nm sequences from PubMLST with sliding window sizes (w) ranging from 4 to 20 aa. Data were generated with plotcon (EMBOSS) and the average similarity for the whole sequence was set to 1. The helicase motive I, Ia, II, III, IV, V, and VI are under laid in orange colour. B) Similarity of the DinG_Nm sequences taken from PubMLST plotted onto the predicted structure of DinG_Nm from MC58 using default Consurf colouring with cysteins shown in black. The helicase motifs are shown as spheres and the rest of the protein is shown as cartoon.

Fig 2. DNA dependent ATPase activity of DinG.

A) Representative chromatograms of ATPase activity of DinG_Nm (upper panel) and the mutant DinG_NmK72A (lower panel). Lanes: 1) 400 nM DinG protein; 2) 90 nM E. coli UvrD; 3–9) 0, 50, 100, 200, 400, 800, and 1600 nM DinG protein. Lane 2–9 contained 200 nM dT₁₀₀ as DNA cofactor. The released phosphate is indicated by the arrows. B) Graph showing ATPase activities of DinG_Nm and DinG_NmK72A in the presence of DNA as cofactor. The standard deviations from three independent experiments are indicated by bars.

Fig 3. DinG_Nm binding to ssDNA.

Representative gel images of DNA binding assays containing 800 nM DinG_Nm (lane 1), or 800 nM DinG_NmK72A (lane 2) and 100 pM homopolymeric nucleotides of given size. Lane 0 contains the control reaction without proteins.

Fig 4. Protein concentration dependent DNA binding.

A) DNA binding assay with DinG_Nm and DinG_NmK72A. The protein concentrations are given in nM. B) Quantitation of gel images from three independent experiments. Values are plotted as a fraction of bound DNA versus the protein concentration.

Fig 5. Titration of the DinG_Nm DNA unwinding activity.

DNA unwinding activity was tested on 1 nM forked DNA substrate (T1+B1 oligo dimer with a 30mer complementary region and 30mer tails) with increasing concentrations of DinG_Nm or DinG_NmK72A. A) A representative gel showing educts and unwinding reaction products, also schematic depicted on the right. Lanes: 1) heat-denatured substrate, 2) no enzyme, 3–8) 50 nM, 100 nM, 200 nM, 400 nM, 800 nM, and 1600 nM DinG_Nm, respectively, 9–14) 50 nM, 100 nM, 200 nM, 400 nM, 800 nM, and 1600 nM DinG_NmK72A, respectively. B) Quantitation of the unwinding activity of DinG_Nm and DinG_NmK72A. The average of three independent experiments and standard deviations are shown.

Fig 6. DinG_Nm unwinding directionality and cofactor dependency.

A) Unwinding activity on modified forked DNA (left: B9+T8-3′-3′, right: T8+B9-5′-5′). Lanes: 1) heat denatured substrate, 2) no protein, 3) 200 nM protein, 4) 400 nM protein. Upper panel: DinG_Nm, lower panel: DinG_Ec. B) Metal dependent unwinding. Labelled forked DNA substrate (T1+B1) was incubated with 400 nM DinG_Nm in the presence of 2 mM ATP and 2 mM of the metal as indicated on top. 1) heat-denatured DNA substrate, 2) reaction without protein, 3) reaction lacking metal ion. C) Nucleotide dependent unwinding. Labelled forked DNA substrate (T1+B1) was incubated with 400 nM DinG_Nm in the presence of 2 mM Mg²⁺ and 2 mM of NTP or dNTP as indicated on top. 1) heat-denatured DNA substrate, 2) reaction without protein, 3) reaction lacking ATP. The products of DNA unwinding are indicated by arrows.

Fig 7. Co-precipitation of DinG_Nm and SSB_Nm.

Precipitation of DinG_Nm alone and in combination with SSB_Nm by ammonium sulphate. An example of a Coomassie blue stained polyacrylamide gel is shown. The open arrow indicates DinG_Nm and the filled arrow indicates SSB_Nm. The supernatant (s) and pellet (p) fractions are shown.

Fig 8. Protein interaction shown by microscale thermophoresis.

A) MST of DinG_Nm with SSB_Nm and with SSB_NmΔ8C. The average and standard deviation of the normalised responses and the fitted curves are shown. B) MST of DinG_NmK72A with SSB_Nm and with SSB_NmΔ8C. No response could be detected and therefore the normalized fluorescence is plotted. The MST results for each protein combination include the data of three independent experiments.

Fig 9. NmΔdinG cells are sensitive to DNA intrastrand crosslinking agents.

A) Survival rate of Nm MC58 wildtype (MC58 wt) and ΔdinG_Nm (MC58ΔdinG) after exposing the cells to the indicated UV fluences. B) Survival rate of Nm MC58 wt and MC58ΔdinG after treating the cells with 10 mM hydrogen peroxide (H₂O₂), 0.5 mM paraquat, 10 nM MMS or 10 ng/ml MMC as described in the Materials and Methods. The survival rate was calculated relative to the untreated wildtype. The results are from at least 3 independent experiments. A p-value < 0.01 is indicated by an asterisk.

Fig 10. PCA of differential abundant proteins in Nm wildtype and NmΔdinG.

The PCA results of the DA proteins from S1 Table are shown in the form of a Variables Factor Map. Each point represents one gene listed in S1 Table, excluding the outlier Opc. The insert shows the vectors weighted by number of DA proteins and with their tails moved to the corresponding head.

Fig 11. Heatmaps for differential abundant proteins in NmΔdinG.

The abundances of proteins belonging to two different groups, A) DNA metabolism, and B) electron transfer, are shown. Gene and protein names are given on the left and the three experiments for the native condition (N) and the MMC stress condition (S) are indicated on top.