Prof. Christof Holliger, Dr. Julien Maillard, Aamani Rupakula, Géraldine Buttet
Swiss National Science Foundation
October 2011 – September 2016
Prof. Hauke Smidt, University of Wageningen, The Netherlands; Prof. Gabriele Diekert, University of Jena, Germany; Prof. Roland Sigel, University of Zürich, Switzerland; PD Dr. Antonio Pierik, University of Marburg, Germany; Prof. David Leys, University of Manchester, United Kingdom; Prof. Frank Sargent, University of Dundee, United Kingdom; Prof. Dr. Bernhard Kräutler, Universität Innsbruck, Austria.
Environmental pollution with chlorinated compounds such as tetra- (PCE) and trichloroethene (TCE) is widespread and threatens one of our major drinking water resources. Although widespread, this pollution is a quite recent phenomenon and hardly explains the ubiquity of bacteria able to use a chlorinated compound as electron acceptor, the so-called organohalide-respiring bacteria (previously called dehalorespiring bacteria). Despite the fact that it is an apparently widespread respiration process allowing to some bacteria surviving or better living in the absence of oxygen, organohalide respiration is not yet characterized in detail for any of the strains available in pure culture.
We will investigate five main objectives in this project: (I) identify, isolate and characterize the PCE reductive dehalogenase complex (PCE-RDHC), (II) elucidate the physiological function of the individual subunits PceB, PceC, and PceT encoded in the conserved pceABCT gene cluster, (III) develop new strategies allowing the heterologous expression of functional reductive dehalogenases, (IV) study the molecular diversity and the evolution of the reductive dehalogenases identified in newly sequenced genomes, and (V) characterize the functional diversity of the organohalide respiration and related processes in pure cultures as well as in already established dechlorinating enrichment cultures.
Functional screening of OHR regulatory key enzymes
From the genome of Dehalobacter restrictus and other OHR Firmicutes like members of the genus Desulfitobacterium, a family of transcription activators named here RdhK appeared to be important for the transcription of reductive dehalogenase (rdh) genes. While the function of CprK1, the first identified member of the RdhK family, has been well characterized, the specificity of many other RdhK proteins remains to be investigated. In brief, RdhK recognize a specific organohalide which triggers conformational changes, that leads to binding to specific DNA promoters located upstream of rdh gene clusters, and finally activates the transcription. In this part of the work, a method for measuring tripartite interaction between RdhK, the organohalide and the DNA promoter (electrophoretic mobility shift assay, EMSA) was successfully established in our laboratory. Attempts to apply this technique to new RdhK proteins identified in the genome of D. restrictus, however, did not give any clear positive output. Among the 8 different RdhK proteins selected and produced, one (RdhK16) was extensively investigated. Besides EMSA attempts for the tripartite interaction, isothermal titration calorimetry (ITC) was performed to test the interaction of His-tagged and untagged RdhK16 proteins with various organohalide mixtures. Unfortunately, this revealed to be unsuccessful despite positive control experiments with CprK1.
In order to make the functional screening easier, a new strategy was developed in which hybrid proteins are produced by fusing the organohalide-binding N-terminal domain with the DNA-binding C-terminal domain of virtually any two RdhK proteins. As proof of concept two RdhK hybrid proteins were designed by exchanging both domains of two characterized RdhK proteins (CprK1 and CprK4 of D. hafniense) and are currently tested by EMSA using combinations of their respective organohalide substrate and DNA targets.
Comparative proteomic analysis of OHR metabolism
The composition of the respiratory electron transfer chain involved in OHR metabolism is still unresolved. In this project, we aimed at identifying enzymes specifically dedicated to the respiration of tetrachloroethene (PCE). Therefore, we have chosen to apply comparative proteomic analysis with PCE-respiring bacteria such as Dehalobacter restrictus and Desulfitobacterium hafniense.
The quality of proteomic data relies very much on an efficient protein extraction from the biomass under study, especially for the identification of proteins involved in respiratory metabolisms as key redox enzymes are integral membrane proteins and therefore difficult to solubilize.
Several experimental strategies were adopted here. Initially, total cell proteins were dissolved in urea, but the resolution of membrane proteins was not satisfying. Then, attempts to enrich membrane-bound proteins were performed by removing the fraction of soluble proteins. Addition of sodium carbonate clearly helped in removing soluble protein contamination from the membrane fraction, but the detection of low abundant membrane proteins still remained challenging as the dynamic range of detection was biased by a few strongly dominating proteins. Data collected at this step revealed that strong biases also came from the fractionation into soluble and membrane proteins. So finally, it was chosen to run total cell proteins in the proteomic analysis and to focus on significant changes between both strains cultivated in the same medium and also for D. hafniense cultivated with different terminal electron acceptors (PCE vs. sulfite). At this stage, we have faced a new problem of sample reproducibility. Indeed, one of the samples (D. hafniense cultivated on PCE) gave significantly poorer resolution than the other two. Although preliminary conclusions can be drawn from this work, it was not possible to achieve the initial aim in the frame of the present project. One major outcome, however, is the detection for the first time of the PceC protein, with similar intensity as for PceB, the membrane anchor of PceA, suggesting that these three proteins may form a complex in the membrane of OHR bacteria.
Functional characterization of the membrane-bound redox protein PceC
PceC belongs to a family of uncharacterized enzymes (RdhC) encoded in rdh gene clusters of Firmicutes like Dehalobacter spp. and Desulfitobacterium spp., but also in bacteria so far not known to grow by OHR metabolism. While originally considered as possible transcription regulator, RdhC enzymes may represent rather a redox protein involved in electron transfer towards the corresponding reductive dehalogenase (RdhA) in the OHR metabolism.
In the frame of the project, significant progress was made in the production, reconstitution and redox activity of the FMN-binding domain of PceC (PceC-FBD). In brief, sequence analysis and protein domain prediction allowed to define the DNA sequence coding for the FMN-binding domain, which was cloned and expressed in E. coli to produce a recombinant protein. PceC-FBD was produced as inclusion bodies in an aggregated form which therefore needed to be first denatured in urea followed by a reconstitution in the presence of FAD. This protocol was successfully developed using successive dialysis baths to gradually remove urea in the presence of FAD (FMN donor) and a purified flavin-transferase. Reconstituted PceC-FBD was made soluble and fully loaded with FMN. The redox potential of PceC-FBD was then investigated via cyclic voltammetry allowing to characterize the electron-transfer activity of PceC. In addition, variants of PceC-FBD were produced to confirm the covalent bond between FMN and a threonine residue fully conserved in the sequences of RdhC proteins.
Preliminary proteomic data suggested that under standard protein extraction both membrane-bound subunits PceB and PceC are present in similar amount in membranes of D. hafniense, while PceA, the catalytic subunit was detected with a 15-fold excess. The hydrophobic nature of PceB and PceC is likely to explain the strong difference in detection level in comparison to PceA.
Study of the interaction between the molecular chaperone PceT and the signal peptide of PceA
Previous studies have shown that the PceT protein acts as a molecular chaperone for folding of PceA, the key catalytic enzyme in OHR metabolism. Recent progress was made in the frame of the present project to understand and characterize the protein-protein interaction between PceT and the signal peptide of PceA. First, both proteins were produced and purified. When analysed by size-exclusion chromatography, PceT was shown to form a homodimer on its own. A relatively strong binding affinity (KD = 200 nM) was measured by isothermal titration calorimetry (ITC). Moreover, ITC data revealed a 2:1 ratio between PceT and PceA signal peptide which confirmed the dimeric nature of PceT. ITC analysis between PceT and a couple of other signal peptide showed no interaction, revealing its specificity toward PceA. Several attempts were then undertaken to develop an in vivo screening in E. coli for the PceT-PceA interaction in order to identify amino acids in PceT essential for the interaction. So far, however, the strategies chosen failed, most probably due to the transient nature of the interaction.
Competition for tetrachloroethene in a consortium of Sulfurospirillum populations and specificity of their reductive dehalogenase enzymes
The results of our previous studies on PCE dechlorination by a consortium containing Sulfurospirillum have suggested that at least two distinct Sulfurospirillum strains were successively involved in the dechlorination of PCE to trichloroethene (TCE, by strain SL2-PCEc) and further to cis-dichloroethene (cis-DCE, by strain SL2-TCE). While the first strain cannot dechlorinate TCE, the second one is able to dechlorinate both PCE and TCE, as the well-characterized Sulfurospirillum multivorans. Despite the redundancy of PCE dechlorination, a co-culture harbouring both strains could be maintained over the years in our laboratory, raising the question of competition/cooperation of both strains for PCE. Therefore, additional work (initially unplanned in the project proposal) was carried out with the aim to characterize the PceA enzyme of strain SL2-PCEc (called PceATCE), which has the peculiarity of dechlorinating PCE to TCE as end-product exclusively. Sequence comparison of PceATCE with PceA has already shown a relatively high identity level (92%), suggesting that only few changes in the amino acid sequences are responsible for the restricted substrate range of PceATCE. In collaboration with the University of Jena (Germany), this enzyme was purified and characterized. The corrinoid cofactor of the enzyme was extracted and identified as norpseudovitamin B12, as in PceA of S. multivorans. Finally a structure model was developed for PceATCE based on the structure of S. multivorans PceA to identify amino acids likely responsible for the substrate spectrum change.
The growth rate of either one of the SL2 strains was measured by cultivating them separately on various concentrations of PCE, from which the kinetic parameters of both strains were elucidated. The maximal dechlorination rate Vmax (here expressed as mmol Cl-/(L·h)) of strain SL2-PCEc and SL2-TCE were estimated at 1.2 and 2.2, respectively, while the affinity for PCE (KM, in mol/L) were calculated to be at 22 and 98, respectively, explaining the long-term co-existence of both strains in cultures amended with 20 M PCE. Furthermore, competition experiments were performed in which both strains were initially cultivated separately and then mixed together and cultivated in low and high PCE concentrations, thus confirming the concentration dependent growth rate of each of the strains in the co-culture.
Finally, the elucidation of the draft genome of strains SL2-PCEc and SL2-TCE (in collaboration with T. Goris, Jena, Germany) clearly confirmed that the main difference between them resides in the sequence of their reductive dehalogenases, suggesting a recent divergence of a common ancestor.