A corepressor participates in LexA-independent regulation of error-prone polymerases in Acinetobacter

Abstract

The DNA damage response of the multidrug-resistant pathogen Acinetobacter baumannii, which induces mutagenic UmuD'₂C error-prone polymerases, differs from that of many bacteria. Acinetobacter species lack a LexA repressor, but induce gene transcription after DNA damage. One regulator, UmuDAb, binds to and represses the promoters of the multiple A. baumannii ATCC 17978 umuDC alleles and the divergently transcribed umuDAb and ddrR genes. ddrR is unique to the genus Acinetobacter and of unknown function. 5' RACE (rapid amplification of cDNA ends) PCR mapping of the umuDAb and ddrR transcriptional start sites revealed that their -35 promoter elements overlapped the UmuDAb binding site, suggesting that UmuDAb simultaneously repressed expression of both genes by blocking polymerase access. This coordinated control of ddrR and umuDAb suggested that ddrR might also regulate DNA damage-inducible gene transcription. RNA-sequencing experiments in 17 978 ddrR^- cells showed that ddrR regulated approximately 25 % (n=39) of the mitomycin C-induced regulon, with umuDAb coregulating 17 of these ddrR-regulated genes. Eight genes (the umuDC polymerases, umuDAb and ddrR) were de-repressed in the absence of DNA damage, and nine genes were uninduced in the presence of DNA damage, in both ddrR and umuDAb mutant strains. These data suggest ddrR has multiple roles, both as a co-repressor and as a positive regulator of DNA damage-inducible gene transcription. Additionally, 57 genes were induced by mitomycin C in the ddrR mutant but not in wild-type cells. This regulon contained multiple genes for DNA replication, recombination and repair, transcriptional regulators, RND efflux, and transport. This study uncovered another regulator of the atypical DNA damage response of this genus, to help describe how this pathogen acquires drug resistance through its expression of the error-prone polymerases under DdrR and UmuDAb control.

Keywords: DNA damage; DdrR; LexA; SOS response; UmuDAb; repressor.

Fig. 1.

5' RACE (rapid amplification of cDNA ends) analyses reveal the relationship of the UmuDAb binding region to the ddrR-umuDAb promoters. 5' RACE experiments were conducted on RNA from MMC-treated 17978 and ADP1 cells. Upon mapping the first (+1) mRNA base of each gene, the −35 elements of each gene were observed to overlap either the defined UmuDAb binding site (for 17978) or an inverted repeat that, when mutated, abolished transcription of both genes in ADP1 cells [14]. The dotted box marks a UmuDAb binding site defined in 17978 through gel shift experiments [13], the dashed lines indicate the inverted repeats proposed to be regulatory binding sites in ADP1 [18], and the solid boxes represent −10 and −35 promoter consensus elements suggested by the +1 transcriptional start site(s). Distances between the +1 transcriptional start sites and the coding regions are shown in vertical boxes.

Fig. 2.

ddrR regulates expression of the UmuDAb repressor. RT-qPCR experiments measured umuDAb expression in wild type (WT) strains and ddrR mutant strains MSUcds2730 ( A. baylyi ) and JH1700 ( A. baumannii ), in the absence or presence of DNA damage (MMC, 2 μg ml⁻¹). Mutation of ddrR resulted in the loss of the umuDAb repression that exists in WT cells in the absence of DNA damage. Additionally, expression of umuDAb after DNA damage was significantly induced in the ADP1 ddrR mutant MSUcds2730 (P<0.05 in a t -test). Asterisks indicate statistical significance in a Student’s t -test for P<0.05. The standard error of the mean from technical triplicates of biological triplicates is shown.

Fig. 3.

A DdrR protein probably exerts the regulatory actions of ddrR. RT-qPCR experiments measured the expression of umuDAb in two different ddrR mutants of ADP1 in the absence or presence of DNA damage (MMC, 2 μg ml⁻¹). DR-Stop, a ddrR nonsense (stop codon) mutant, showed the same derepression of umuDAb in the absence of DNA damage as did the null ddrR mutant MSUcds2730. There was a significant induction in umuDAb expression (2^−ΔCT) after DNA damage (P<0.01 in a one-tailed t -test) in each strain, but no significant difference between strains in their induction amount (P>0.05 in a two-tailed t -test comparing 2^−ΔΔCT), or between expression in either the presence or the absence of MMC (P>0.05 in a two-tailed t -test.) The standard error of the mean from technical triplicates of biological triplicates is shown.

Fig. 4.

ddrR is required for repression of multiple error-prone polymerases in 17978. RT-qPCR experiments measured the expression of genes in 17978 WT and ddrR strain JH1700 in the absence or presence of DNA damage (MMC, 2 μg ml⁻¹). Mutation of ddrR resulted in the derepression of all of these genes in the absence of DNA damage. Expression was measured in both un-induced and induced (MMC) conditions of WT or ddrR mutant cells. Genes are indicated by name or abbreviation (uDAb, umuDAb; uD, umuD; uC, umuC), and A1S gene locus number. Each gene was assayed in one RT-qPCR experiment (plate), with error bars indicating the standard error of the mean from technical triplicates of biological triplicates. For every gene, expression in WT cells was significantly (P<0.05 in a one-tailed t -test) increased after induction with MMC. Expression of every gene in the absence of DNA damage was significantly (P<0.05; in a one-tailed t -test) induced in ddrR mutant cells as compared to WT cells.

Fig. 5.

ddrR mutation does not affect the induction of genes that are not regulated by UmuDAb. RT-qPCR experiments measured the expression of genes (ACIAD0724 nrdA, ACIAD0445 gst or ACIAD1436 benA in ADP1; A1S_0408 gst or A1S_1215 benA in 17978) in un-induced (-I) and induced (‘+I’ for inclusion of inducing agent: MMC for nrdA and gst; or benzoate for benA) conditions for WT, ddrR and umuDAb strains of ADP1 and 17978. There was no significant difference in the induction level of any of these genes in the ddrR or umuDAb mutants (P>0.05; in a two-tailed t -test) after MMC or benzoate exposure (for benA gene expression only). nd , benA expression was not examined (not done) in umuDAb mutants of either 17978 or ADP1 by RT-qPCR, although previous experiments have shown that benzoate-mediated induction of benA is unaffected by the umuDAb mutation [18]. Each gene was assayed in one RT-qPCR experiment (plate), with error bars indicating the standard error of the mean from technical triplicates of biological triplicates.

Fig. 6.

Proportion of the DNA damage-inducible genes that are regulated by ddrR and umuDAb. The relationship between 17978 genes induced after MMC in WT cells and ddrR- and umuDAb-dependent genes induced is shown in an area proportional Venn diagram constructed using the BxToolBox at bioinforx.com.