Crucial Role of the C-Terminal Domain of Hfq Protein in Genomic Instability

Abstract

G-rich DNA repeats that can form G-quadruplex structures are prevalent in bacterial genomes and are frequently associated with regulatory regions of genes involved in virulence, antigenic variation, and antibiotic resistance. These sequences are also inherently mutagenic and can lead to changes affecting cell survival and adaptation. Transcription of the G-quadruplex-forming repeat (G₃T)_n in E. coli, when mRNA comprised the G-rich strand, promotes G-quadruplex formation in DNA and increases rates of deletion of G-quadruplex-forming sequences. The genomic instability of G-quadruplex repeats may be a source of genetic variability that can influence alterations and evolution of bacteria. The DNA chaperone Hfq is involved in the genetic instability of these G-quadruplex sequences. Inactivation of the hfq gene decreases the genetic instability of G-quadruplex, demonstrating that the genomic instability of this regulatory element can be influenced by the E. coli highly pleiotropic Hfq protein, which is involved in small noncoding RNA regulation pathways, and DNA organization and packaging. We have shown previously that the protein binds to and stabilizes these sequences, increasing rates of their genomic instability. Here, we extend this analysis to characterize the role of the C-terminal domain of Hfq protein in interaction with G-quadruplex structures. This allows to better understand the function of this specific region of the Hfq protein in genomic instability.

Keywords: DNA-directed mutagenesis; bacterial chromatin; genomic instability; nucleoid; quadruplex.

Figure 1

Mutation rates for (G₃T)₈ repeats in plasmids pBR325 and pBR235 in MG1655 hfq-Cm^r (reference strain), MG1655 HfqNTR72-Cm^r (∆CTR), and MG1655 hfq::Cm^r (∆hfq) [38]. Mutation rates were determined as described under Materials and Methods. Data for MG1655 hfq-Cm^r and MG1655 NTR72-Cm^r with both pBR325 and pBR235 represent results from two independent Luria–Delbrück fluctuation analyses. Results for plasmids in MG1655 Δhfq::Cm^r represent a single Luria–Delbrück fluctuation analysis. Error bars represent 84% confidence intervals. Numbers represent false discovery rates.

Figure 2

Hfq Binding to (G₃T)₄. (A) Hfq-CTR binding to (G₃T)₄, (G₃T)₄ concentration 0.1 μM, while Hfq-CTR concentration ranged from 0 to 3 μM. (B) Graphic analysis of Hfq-CTR binding to (G₃T)₄ shown in A. (C) controls: Hfq-NTR and wild type Hfq binding to (G₃T)₄.

Figure 3

Synchrotron radiation circular dichroism (SRCD) analysis of the (G₃T)₄ quadruplex complexed to Hfq-CTR. Spectra of (G₃T)₄ in the absence (red) and presence of Hfq-CTR (blue). Hfq alone (green). The spectrum of the complex (blue) is similar to the sum of the (G₃T)₄ and Hfq-CTR spectra (dotted black), differing only in the strength of its amplitudes. This signifies most likely that upon complex formation, an enhancement of already existing structural features in the quadruplex is occurring. It is not clear whether the Hfq-CTR in contact with (G₃T)₄ changes its structure, to form amyloids, which would change the 210–220 nm amplitudes of the spectrum [38].

Figure 4

Model of direct and indirect effects of Hfq on genome instability. Description of the figure is included in the text. Direct effects: Green ovals represent DNA Pol III; blue ovals, RNA polymerase. Hfq is represented by the donut. Indirect effects: MutS is represented by the brown ovals. Red arrows represent increased or decreased levels.