NosP Modulates Cyclic-di-GMP Signaling in Legionella pneumophila

Abstract

Biofilms form when bacteria adhere to a surface and secrete an extracellular polymeric substance. Bacteria embedded within a biofilm benefit from increased resistance to antibiotics, host immune responses, and harsh environmental factors. Nitric oxide (NO) is a signaling molecule that can modulate communal behavior, including biofilm formation, in many bacteria. In many cases, NO-induced biofilm dispersal is accomplished through signal transduction pathways that ultimately lead to a decrease in intracellular cyclic-di-GMP levels. H-NOX (heme nitric oxide/oxygen binding domain) proteins are the best characterized bacterial NO sensors and have been implicated in NO-mediated cyclic-di-GMP signaling, but we have recently discovered a second family of NO-sensitive proteins in bacteria named NosP (NO sensing protein); to date, a clear link between NosP signaling and cyclic-di-GMP metabolism has not been established. Here we present evidence that NosP (Lpg0279) binds to NO and directly affects cyclic-di-GMP production from two-component signaling proteins Lpg0278 and Lpg0277 encoded within the NosP operon. Lpg0278 and Lpg0277 are a histidine kinase and cyclic-di-GMP synthase/phosphodiesterase, respectively, that have already been established as being important in regulating Legionella pneumophila cyclic-di-GMP levels; NosP is thus implicated in regulating cyclic-di-GMP in L. pneumophila.

Figure 1:. NosP operon in Legionella pneumophila.

nosP (Lpg0279) is encoded in an operon adjacent to a histidine kinase, nahK (Lpg0278), followed by a cyclic-di-GMP metabolizing response regulator, narR (Lpg0277). NarR is predicted to have both DGC and PDE activities, as well as a receiver domain that is predicted to be phosphorylated on a conserved aspartate residue.

Figure 2:. NahK is an active kinase and transfers phosphate to NarR.

(A) Autoradiography results with [γ³²P]-ATP to radiolabel phosphorylated proteins. NahK is active and transfers phosphate to NarR and NarR AAL, but not NarR D87N. NarR and NarR AAL are phosphorylated only when NahK is present. (B) Time course of NahK incubated with [γ³²P]-ATP. NahK autophosphorylates rapidly, saturating within 20 minutes of reaction initiation. (C) Time-dependent phosphotransfer from NahK to NarR. NahK was pre-incubated with [γ³²P]-ATP to allow autophosphorylation to occur. NarR was then added and reacted with NahK for the indicated amount of time. Phosphorylated NarR is stable in these conditions for up to 60 minutes.

Figure 3:. Phosphorylation of the NarR receiver domain affects both its DGC and PDE activity.

(A) Time course of NarR incubated with GTP. The amount of cyclic-di-GMP and pGpG produced by NarR is dependent on whether the receiver domain is phosphorylated. Legend: ● NarR phosphorylated, ▲ NarR not phosphorylated. Left: Phosphorylation of NarR causes a decrease in cyclic-di-GMP accumulation. Right: The amount of pGpG produced increases when NarR is phosphorylated. (B) Phosphorylation decreases the DGC activity of NarR AAL. (C) Phosphorylation increases the PDE activity of NarR. Cyclic-di-GMP was used as the substrate to examine the effect of phosphorylation solely on PDE activity. Error bars represent one standard deviation of the mean. *: p < 0.05. **: p < 0.01. p denotes the significance of the corresponding condition vs the far-left condition, where NarR AAL (B) or NarR (C) is phosphorylated.

Figure 4:. NosP binds to NO and modulates NahK activity.

(A) UV-Vis spectroscopy shows that NosP is a hemoprotein that binds to NO. Legend: Fe(III) ─ ─ ─, Fe(II) ───, Fe(II)-NO · · ·. (B) NosP complexes modulate the autophosphorylation of NahK. Fe(II) NosP has the strongest inhibition of NahK autophosphorylation, while Fe(II)-NO NosP has a moderate effect. (C) Fe(II)-NO NosP inhibition of NahK autophosphorylation is independent of NosP concentration. NosP titrated up to 20-fold excess over NahK concentration did not have a significant impact on the inhibition of NahK autophosphorylation. Error bars represent one standard deviation of the mean. *: p < 0.05. **: p < 0.01. p denotes the significance of the corresponding condition vs NahK only.

Figure 5:. NosP and NahK are co-cistronic interacting partners.

(A) RT-PCR results. rt: experimental reverse-transcribed mRNA. gDNA: positive control using L. pneumophila gDNA. nc: negative control, not reverse transcribed to rule out DNA contamination. The successful amplification seen here of the cDNA using nosP forward and nahK reverse primers indicates that nosP and nahK are transcribed on the same mRNA. (B) Quantification of the relative amount of each NosP complex pulled down by GST-NahK. Fe(II) NosP was pulled down the greatest, therefore it has the strongest interaction with NahK. (C) Representative western blotting results for the pulldown assay. NosP complexes do not interact with GST, but they are pulled down by GST-NahK. The amount of NosP pulled down is dependent on which NosP complex is present. Error bars represent one standard deviation of the mean. *: p < 0.05. p denotes the significance of the corresponding condition vs Fe(II) NosP.

Figure 6:. Proposed NosP signaling pathway in Legionella pneumophila.

NosP binds NO, which activates NahK autophosphorylation. NahK transfers phosphate to NarR. Phosphorylated NarR exhibits decreased DGC and increased PDE activity, resulting in a decrease in cyclic-di-GMP.