This study aims to catalogue and characterize the genes involve in heavy metal ion tolerance and resistance in bacteria. The past 30 years many phenotypical and genetical studies have been conducted to reveal the heavy metal tolerance and resistance systems in bacteria, and the bacterium Cupriavidus metallidurans strain CH34 was the major model organism in this field.The availability of sequence data from the genome project of Cupriavidus metallidurans strain CH34 has now given major additional insights into its heavy metal tolerance and resistance systems.
The genome of C. metallidurans CH34 contains a much higher proportion of genes involved in metal resistance and detoxification than any other of the 64 other genomes tested. For every kind of resistance mechanism, the C. metallidurans genome ranked first for the variety of gene or gene cluster versions. The main illustration of this fact is the family of three component efflux of toxic compounds: most of the genomes contain 0 to 4 versions of this gene family but there are 12 of them in C. metallidurans CH34.
The genome of this bacterium contains also 8 P-type ATPase involved in metal efflux specialized in lead, cadmium, thallium and/or copper efflux, and several others mechanisms involved in metal processing. The genomic analysis combined with the proteomic analysis put also in evidence new proteins (mostly encoded by the large plasmids) that were, up to now, never described in other bacteria.
Understanding gene regulatory networks by sequence motif detection
The reliable reconstruction of genome-wide regulatory networks is one of the major challenges for present and future bioinformatics research and should be seen as the cornerstone for a better understanding of how organisms can cope with different forms of environmental stress. Living cells are highly complex biological systems that must be able to react and adapt to a wide variety of environmental stimuli. Such a response involves the concerted action of many genes that are controlled at the transcriptional level.
Different changes in the environmental conditions such as heat, desiccation, sunlight, nutrient levels, heavy metal concentrations, or radiation, evoke different cellular responses and deploy different genetic circuitries - though these responses and circuitries often contain common mechanisms. In order to deal with these different conditions, organisms have developed complex regulatory interactions with a hierarchical structure. The total sum of all these interactions is conceptualized as a transcriptional regulatory network.
To understand how living cells make use of such networks i.e. to respond to external stimuli it is necessarily to break down these networks into their individual components and reconstruct them bottom-up. In this, the prediction and subsequent analysis of regulatory DNA motifs is indispensable.
Early motif detection tools were based on the hypothesis that genes that either display a synchronous change in expression level or that have a similar functional annotation are co-regulated and thus must share the same transcriptional regulatory mechanisms. However these inference methods are, by assumption, error-prone and alternative methods were sought.
The public availability of full genome sequences for a wide range of prokaryotic species has created the opportunity to compare many different prokaryotic genomes and has resulted in the application of “phylogenetic footprinting” by aligning the upstream sequences of genes and their orthologs. The idea underlying phylogenetic footprinting is that selective pressure causes functional elements to evolve at a slower rate than the non-functional surrounding sequence. Therefore, the most conserved motifs in a collection of homologous regions are excellent candidates to act as functional elements.
Through alignment of the regulatory target sites and the construction of nucleotide position-specific matrices, a consensus sequence (i.e. “motif logo”) can then be proposed. By combining different phylogenetic footprinting outputs with a two-step clustering (involving the alignment of motif models and their subsequent grouping based on alignment scores) an even better prediction can be obtained.
The Microbiology expert group has performed many high-throughput transcriptional analyses in bacterial and eukaryotic systems under diverse conditions and produced a massive bulk of expression data. The computational prediction of regulatory components and the reconstruction of regulatory modules or even entire networks is a logic extension of this experimental research.
Projects & Partners
Sebastien Monchy PreDoc project (SCK•CEN fellowship, ULB, 2002-2007), Sebastien Van Aelst PreDoc project (ULB, 2004-2008), Alfred Cubaka PreDoc project (ULB, 2005-2009), Pieter Monsieurs PostDoc project (SCK•CEN AWM Fellowship, 2008-2009)
In collaboration with: Dr. N. Verbruggen from ULB in Belgium, Dr. R. Wattiez from UMH in Belgium, Dr. N. van der Lelie from BNL in USA, Prof. C.A. Ouzounis from King’s College London, UK, Prof. K. Marchal from KULeuven, Belgium
Contact: Dr. Janssen Paul, Prof. Dr. Mergeay Max