NAD+ pool depletion as a signal for the Rex regulon involved in Streptococcus agalactiae virulence

Abstract

In many Gram-positive bacteria, the redox-sensing transcriptional repressor Rex controls central carbon and energy metabolism by sensing the intra cellular balance between the reduced and oxidized forms of nicotinamide adenine dinucleotide; the NADH/NAD+ ratio. Here, we report high-resolution crystal structures and characterization of a Rex ortholog (Gbs1167) in the opportunistic pathogen, Streptococcus agalactiae, also known as group B streptococcus (GBS). We present structures of Rex bound to NAD+ and to a DNA operator which are the first structures of a Rex-family member from a pathogenic bacterium. The structures reveal the molecular basis of DNA binding and the conformation alterations between the free NAD+ complex and DNA-bound form of Rex. Transcriptomic analysis revealed that GBS Rex controls not only central metabolism, but also expression of the monocistronic rex gene as well as virulence gene expression. Rex enhances GBS virulence after disseminated infection in mice. Mechanistically, NAD+ stabilizes Rex as a repressor in the absence of NADH. However, GBS Rex is unique compared to Rex regulators previously characterized because of its sensing mechanism: we show that it primarily responds to NAD+ levels (or growth rate) rather than to the NADH/NAD+ ratio. These results indicate that Rex plays a key role in GBS pathogenicity by modulating virulence factor gene expression and carbon metabolism to harvest nutrients from the host.

Fig 1. Rex regulon and its role in virulence of GBS strain NEM316.

A. Heatmap of a selection of differentially expressed genes in a Δrex mutant compared with wild type. The heatmap indicates log₂ fold change and values that showed a >2-fold change (induced or repressed) are shown. Low to high expression is represented by a change of colour from blue to red, respectively. B. Rex-controlled genes involved in metabolism and virulence. The drawing is based on transcriptome analysis (S2, S3 and S4 Tables). Proteins encoded by genes, that were highly (upper panel) or modestly upregulated (lower panel) in the Δrex mutant versus the WT strain, are indicated. Gua, guanine; ade, adenine; hypox, hypoxanthine; Ado, adenosine; cyt, cytosine. C. Effects of Rex on recovery of GBS from different organs in BALB/c mice following systemic infection; organs were collected 24-hour post-inoculation. Data are pooled from two independent experiments and show medians with interquartile ranges. *P value of <0.05; **P value of <0.01; ***P value of <0.001; Mann-Whitney U tests.

Fig 2. In vitro binding of Rex to an operator sequence.

EMSA assays of ³²P-labeled double stranded DNA fragments containing the predicted ldh Rex operator. The concentration of Rex in the binding reaction is indicated above each lane. A. Sequence of the ldh operator fragment. The axis of symmetry is indicated by the dotted line and nucleotides that were exchanged are shown in bold. B. Complex formation with native ldh operator fragment. C. Complex formation with mutated ldh operator fragment: ldh_m1 (C19→T), ldh_m2 (C19→T, C21→T), ldh_m3 (C9→T, A10→G).

Fig 3. Rex directly binds to the promoters of a selection of upregulated genes.

A. EMSAs with purified Rex protein and promoter DNA fragments from the adhE, rex, ldh, gbs0609, purC, cdnP, menA and gbs0644 (cyl) genes. DNA fragments (30 nM) were incubated with the indicated concentrations (in nM) of Rex protein for 20 min at 30°C. Then, the protein-DNA complexes were resolved by electrophoresis on TBE polyacrylamide gels. As a negative control, an internal fragment of 170 bp from the gbs0455 gene, lacking rex binding site, was used as a DNA competitor (indicated by an asterisk). B. EMSAs were performed with the adhE promoter fragments (30 nM), Rex protein (600 nM), and the indicated concentrations of NADH. No Rex protein was added to the first lane. C. EMSAs were performed with the adhE promoter, as in A, but with different concentrations of NAD⁺. D. EMSAs were performed with the adhE promoter, as in B, but with different NADH and NAD⁺ concentrations.

Fig 4. Structures of Rex alone and in complex with Rex ROP DNA.

Cartoon representations of Rex in complex with NAD⁺ (A and B) and NAD⁺ and DNA (C). The nucleotide binding domain (NBD) that binds NAD⁺ and NADH and the winged helix domain (WHD) that interacts with DNA are indicated. D. Superposition of the three structures. The wing of the WHD is coloured blue and the recognition helix (α-helix 3) that makes sequence-specific DNA contacts is coloured red. E. Schematic representation of protein–DNA interactions in the Rex-ROP complex. Due to the symmetric nature of the complex only half of the 22 bp DNA is shown. Hydrogen bonds are shown as dotted lines and van der Waals contacts as solid lines. Residues that make contacts with the phosphate moieties are indicated. F. Overall structure of the Rex-DNA complex with one Rex subunit coloured grey and the other green. Close up of the Rex-DNA interactions (corresponding to the boxed region). For clarity only the wing and the recognition helix of Rex is shown. Residues that form hydrogen bonds (dotted green lines) to the bases of DNA are labelled and shown as sticks. Water-bridged hydrogen bonds between Ala47 and bases of DNA are shown as dotted yellow lines.

Fig 5. Galactose increases rex expression and decreases the NAD⁺ pool.

WT cells containing a rex promoter-lac fusion on plasmid (Prex-lac plasmid) were grown in M17 supplemented with 0.5%glucose (A) or 0.5% galactose (B). When static cultures reached OD₆₀₀ = 0.5 (indicated by an arrow), they were shifted to aerated condition and grown up to stationary phase. β-galactosidase activities (β-gal.) are expressed in Miller units (M. U.). Data are representative of at least three experiments. Mean values and standard deviations are shown. (C) NAD(H) pool determination. Cells were grown in the presence of glucose or galactose (0.5%) and collected at OD₆₀₀ = 0.5 (T0) and 1 hour after the static/aerobiosis shift (T1h). Data are the mean of three independent experiments and standard deviations are shown.