Identification of an α-ketoglutarate specific chemoreceptor in Pseudomonas aeruginosa
Chemotaxis to different Krebs cycle intermediates in P. putida KT2440 is mainly mediated by the McpS chemoreceptor that recognizes most of the intermediates (Lacal et al. J. Biol. Chem. 285:23126-36). Responses to this compound class in P. aeruginosa are based on other chemoreceptors that are characterized by an elevated ligand specificity. In Martín-Mora et al. (2016) (put the link) we identify a chemoreceptor, termed McpK, that recognizes and mediates chemotaxis exclusively to α-ketoglutarate. Previous work by the Harwood laboratory has led to the identification of the malate specific chemoreceptor PA2652 (Alvarez-Ortega and Harwood (2007) Appl Environ Microbiol. 73:7793-5). Chemoreceptors McpK and PA2652 are not paralogous since they employ different ligand binding domains for signal sensing.
Quantitative capillary chemotaxis assays of wild type, mcpK mutant and complemented mutant strains of P. aeruginosa PAO1 to different α-ketoglutarate concentrations.
High-throughput based ligand screening to identify a chemoreceptor for metabolizable purine derivatives
The saprophytic soil bacterium Pseudomonas putida KT2440 has 27 chemoreceptors and a major research lines in our laboratory consists in their functional annotation. One of the un-annotated chemoreceptors is the gene product of PP0320. This receptor contains a periplasmic ligand binding domain of the double PDC type of approximately 260 amino acids. To identify which ligands bind to this receptor, its ligand binding domain was produced recombinantly and then submitted to high-throughput thermal shift assays in the absence and presence of compounds from commercially available ligand collections (Krell, 2015). Increases in the thermal stability (as expressed by the Tm value) strongly indicate ligand binding.
This figure shows Tm changes in the presence of 95 different screened compounds. Four compounds, adenine, guanine, xanthine and uric acid, caused significant protein stabilization or destabilization. Interestingly, these four compounds are differently substituted purine derivatives; hence the hypothesis that this receptor may respond to purines.
Isothermaland to test binding of other related compounds. The results from this series of experiments are shown below titration calorimetry experiments were conducted to verify the hits obtained by the thermal shift assays.
This receptor domain was found to bind purine, a compound that does not exist in nature, as well as five different purine derivatives that form part of the purine degradation pathway, which permits growth of the bacterium on these compounds as sole nitrogen source. Binding was highly selective, since no interaction was observed for non-metabolizable purine derivatives (caffeine, theophylline and theobromine), pyrimidine bases, nucleotides or nucleosides. We conclude that this receptor, which we have termed McpH, binds exclusively metabolizable purine derivatives.
Shown below are qualitative capillary chemotaxis assays of the wild type strain, the mcpH mutant as well as the complemented mutant.
The wt strain showed significant taxis towards the McpH ligands identified. However, deletion of the mcpH gene abolished chemotaxis, which was re-established upon complementation.
This is the first report of chemotaxis towards purine derivatives, which also shows that this type of taxis is mediated by a dedicated purine receptor. This study also illustrates the usefulness of the thermal shift based approach, which can not only be used to functionally annotate chemoreceptors, but also other bacterial sensor proteins.
Mechanism for the specific targeting of a CheR methyltransferase to a chemoreceptor
Studies with E. coli that contain a single CheR and CheB show that, apart from binding at the methylation site, some chemoreceptors recognize CheR and CheB at additional high-affinity sites at C-terminal pentapeptides in the chemoreceptor. The opportunistic human pathogen Pseudomonas aeruginosa PAO1 has 4 CheR paralogues and several receptors with C-terminal extensions. As published in García-Fontana (2014) Science Signaling 7, ra34 (full text: click here, pdf version: click here) we found that exclusively CheR2 bound to the McpB chemoreceptor, which contains the GWEEF pentapeptide at its C-terminus.
We show CheR2 was the only paralogue that methylated McpB, and deletion of the pentapeptide abolished both the CheR2-McpB interaction and the methylation of McpB. The cheR2 and mcpB genes are vicinal in the P. aeruginosa genome and form part of the gene cluster that encodes the Che2 chemosensory pathway. We conclude that the CheR-pentapeptide interaction enabled the specific targeting of one CheR methyltransferase to one chemoreceptor. We also found that bacterial CheR proteins form two distinct protein families when clustered according to sequence, those that bind pentapeptide-containing chemoreceptors and those that do not, distinguished by an insertion of three amino acids in the β-subdomain of CheR. Deletion of this insertion in CheR2 prevented its interaction with and methylation of McpB. Because many bacteria contain pentapeptide-containing chemoreceptors and several signaling protein paralogues, we predict that the mechanism described may contribute to the specific assembly of signaling proteins to form distinct, parallel pathways that mediate different responses.
Definition of a novel bacterial sensor domain.
We have published recently the 3D structure of the sensor domain of the McpS chemoreceptor (Pineda et al., 2012) that mediates chemotaxis to various organic acids (Lacal et al., 2010). The McpS sensor domain as shown in this figure corresponds to a novel small molecule sensor domain that is composed of two modules to which each signal molecules bind. In Ortega and Krell (2013) we now report the domain signature of the family of McpS-like sensor domain that we term Helical Bimodular Domain (HBM). This domain was found in Bacteria and Archea and forms part of chemoreceptors and histidine kinases. We showed that the most conserved amino acid are located in close proximity of the two ligand binding sites, indicating that family members may recognize similar ligands. Using this profile we have identified so far 1200 members. The domain profile will be sent to relevant databases like Pfam or Prosite.
Figure legend: Three dimensional structure of the HBM domain. Shown is the 3D structure of the sensor domain of the McpS chemoreceptor. Bound acetate and malate are shown in space-fill mode in blue and yellow, respectively. Most conserved amino acids in a sequence alignment of HBM family members are labelled and shown in ball-and-stick mode.
Multiple signals modulate the activity of the complex sensor kinase TodS
Many sensor kinases have a complex domain arrangement of which the physiological relevance is poorly understood. It was hypothesized that this complexity may permit to perceive different types of stimuli that result in a fine-tuning of the transcriptional response, but experimental proof of this hypothesis is scarce. The 7-domain TodS sensor kinase of Pseudomonas putida DOT-T1E is composed of almost 1000 amino acids (see Figure). It transphosphorylates the TodT response regulator that in turn controls the expression of a toluene degradation pathway
We have shown previously that toluene and other monoaromatic compounds modulate TodS autokinase activity by binding to the PAS1 sensor domain (Lacal et al. (2006) PNAS 103:8191). We have now shown that the oxidative stress agent menadione in addition to toluene modulates TodS activity in a specific manner (Silva-Jiménez et al. (2014) Microb. Biotechnol.). We also show that menadione action involves the modification of cysteine residue 320. This residue is the only conserved cysteine in an alignment of protein family members. The activity of TodS is thus controlled by the presence of monoaromatic compounds and the cellular redox potential, providing support to the above hypothesis.
Multiple effects of prebiotic oligosaccharides on Pseudomonas aeruginosa
We report a study on the effects of inulin and its hydrolysed derivative fructooligosaccharide (FOS) on different features of P. aeruginosa (Daddaoua et al., 2014). We show that FOS reduces bacterial growth, biofilm formation and motility. In co-cultures with eukaryotic cells FOS reduced the secretion of several inflammatory cytokines. Further results are shown that suggest that the FOS mediated effects may be related to a decreased activation of the NF-κB pathway. Our results suggest that FOS may be useful components of antibacterial and anti-inflammatory drug cocktails.
Figure legend: Reduction of biofilm formation in the presence of different FOS concentrations.
Correlation between signal input and output in chemoreceptors based signaling
A common feature of signal transduction systems is the existence of mechanisms for signal detection (input) and the generation of the regulatory response (output). A central question consists in establishing whether the magnitude of input, or affinity of the signal for its sensor protein, determines the magnitude of regulatory output. Previous studies in our laboratory of one- and two-component systems have shown that this is not necessarily the case (Guazzaroni et al. JBC (2007) ; Busch et al. PNAS (2007)).
We have now assessed signal input and output relationships for chemoreceptor based signaling cascades using the PctA and PctB amino acid chemoreceptors of Pseudomonas aeruginosa as models (Reyes-Darias et al. Mol. Microbiol. (2015)). To calculate output, receptor chimeras were constructed fusing the PctA and PctB sensor domain with the signaling domain of the Tar receptor from E. coli. These chimeras were introduced into a chemoreceptor free E. coli mutant and the EC50 values of the response towards different amino acids determined with the FRET assay. We were able to show that for both receptors the magnitude of EC50 values (output) correlates with the affinity of the amino acids for the recombinant ligand binding domains (input) as determined previously (Rico-Jiménez et al. Mol. Microbiol. (2013), see Figure). We are thus able to conclude that, for the receptors studied, the signal affinity determines the magnitude of regulatory output.
Pseudomonads with different lifestyles show specific chemotaxis to Gamma-aminobutyrate (GABA)
After the identification of PctC as the receptor responsible for GABA chemotaxis in P. aeruginosa PAO1 (Rico-Jiménez et al. (2013) Mol. Microbiol.) we report in Reyes Darias et al. (2015) Mol. Microbiol. the identification of the McpG chemoreceptor that binds and mediates chemotaxis exclusively to GABA.
The microcalorimetric titration of the recombinant ligand binding domain of McpG (below) revealed a dissociation constant for GABA of 175 nM, which is amongst the highest ever measured affinities of a signal molecule to a bacterial sensor protein.
We have detected GABA in exudates of tomato roots and have shown that the mutation of the mcpG gene reduces the magnitude of root colonization following chemotaxis through soft agar (below).