Computational prediction of conserved operons and phylogenetic footprinting of transcription regulatory elements in the metal-reducing bacterial family Geobacteraceae

Integrating genome sequence and structural data for statistical learning to predict transcription factor binding sites

We report an approach to predict DNA specificity of the tetracycline repressor (TetR) family transcription regulators (TFRs). First, a genome sequence-based method was streamlined with quantitative P-values defined to filter out reliable predictions. Then, a framework was introduced to incorporate structural data and to train a statistical energy function to score the pairing between TFR and TFR binding site (TFBS) based on sequences. The predictions benchmarked against experiments, TFBSs for 29 out of 30 TFRs were correctly predicted by either the genome sequence-based or the statistical energy-based method. Using P-values or Z-scores as indicators, we estimate that 59.6% of TFRs are covered with relatively reliable predictions by at least one of the two methods, while only 28.7% are covered by the genome sequence-based method alone. Our approach predicts a large number of new TFBs which cannot be correctly retrieved from public databases such as FootprintDB. High-throughput experimental assays suggest that the statistical energy can model the TFBSs of a significant number of TFRs reliably. Thus the energy function may be applied to explore for new TFBSs in respective genomes. It is possible to extend our approach to other transcriptional factor families with sufficient structural information.

The unphosphorylated form of the PilR two-component system regulates pilA gene expression in Geobacter sulfurreducens

In Geobacter sulfurreducens, metal reduction and generation of bioelectricity require the participation of several elements, and among them, the type IV pili has an essential role. The pilus is composed of multiple PilA monomers. Expression of pilA gene depends mainly on the σ54 factor and the response regulator protein PilR. In this work, we characterized the role of the PilS-PilR two-component system in the regulation of the pilA gene expression. Experimental evidence indicates that PilS is autophosphorylated at the His-334 residue, which in turn is transferred to the conserved Asp-53 in PilR. Contrary to other PilS-PilR systems, substitution D53N in PilR resulted in higher activation of the pilA gene. By using a pilA::luxCDABE fusion with different promoter fragments and in vitro DNA-binding assays, we demonstrated the existence of multiple functional PilR binding sites. A regulatory model in which the non-phosphorylated PilR protein directs activation of pilA expression by binding to two sites in the promoter region of this gene is presented.

Phylogenetic footprinting: a boost for microbial regulatory genomics

Phylogenetic footprinting is a method for the discovery of regulatory elements in a set of homologous regulatory regions, usually collected from multiple species. It does so by identifying the best conserved motifs in those homologous regions. There are two popular sets of methods-alignment-based and motif-based, which are generally employed for phylogenetic methods. However, serious efforts have lacked to develop a tool exclusively for phylogenetic footprinting, based on either of these methods. Nevertheless, a number of software and tools exist that can be applied for prediction of phylogenetic footprinting with variable degree of success. The output from these tools may get affected by a number of factors associated with current state of knowledge, techniques and other resources available. We here present a critical apprehension of various phylogenetic approaches with reference to prokaryotes outlining the available resources and also discussing various factors affecting footprinting in order to make a clear idea about the proper use of this approach on prokaryotes.

Molecular analysis of phosphate limitation in Geobacteraceae during the bioremediation of a uranium-contaminated aquifer

Nutrient limitation is an environmental stress that may reduce the effectiveness of bioremediation strategies, especially when the contaminants are organic compounds or when organic compounds are added to promote microbial activities such as metal reduction. Genes indicative of phosphate-limitation were identified by microarray analysis of chemostat cultures of Geobacter sulfureducens. This analysis revealed that genes in the pst-pho operon, which is associated with a high-affinity phosphate uptake system in other microorganisms, had significantly higher transcript abundance under phosphate-limiting conditions, with the genes pstB and phoU upregulated the most. Quantitative PCR analysis of pstB and phoU transcript levels in G. sulfurreducens grown in chemostats demonstrated that the expression of these genes increased when phosphate was removed from the culture medium. Transcripts of pstB and phoU within the subsurface Geobacter species predominating during an in situ uranium-bioremediation field experiment were more abundant than in chemostat cultures of G. sulfurreducens that were not limited for phosphate. Addition of phosphate to incubations of subsurface sediments did not stimulate dissimilatory metal reduction. The added phosphate was rapidly adsorbed onto the sediments. The results demonstrate that Geobacter species can effectively reduce U(VI) even when experiencing suboptimal phosphate concentrations and that increasing phosphate availability with phosphate additions is difficult to achieve because of the high reactivity of this compound. This transcript-based approach developed for diagnosing phosphate limitation should be applicable to assessing the potential need for additional phosphate in other bioremediation processes.

Genome-wide analysis of the RpoN regulon in Geobacter sulfurreducens

BACKGROUND

The role of the RNA polymerase sigma factor RpoN in regulation of gene expression in Geobacter sulfurreducens was investigated to better understand transcriptional regulatory networks as part of an effort to develop regulatory modules for genome-scale in silico models, which can predict the physiological responses of Geobacter species during groundwater bioremediation or electricity production.

RESULTS

An rpoN deletion mutant could not be obtained under all conditions tested. In order to investigate the regulon of the G. sulfurreducens RpoN, an RpoN over-expression strain was made in which an extra copy of the rpoN gene was under the control of a taclac promoter. Combining both the microarray transcriptome analysis and the computational prediction revealed that the G. sulfurreducens RpoN controls genes involved in a wide range of cellular functions. Most importantly, RpoN controls the expression of the dcuB gene encoding the fumarate/succinate exchanger, which is essential for cell growth with fumarate as the terminal electron acceptor in G. sulfurreducens. RpoN also controls genes, which encode enzymes for both pathways of ammonia assimilation that is predicted to be essential under all growth conditions in G. sulfurreducens. Other genes that were identified as part of the RpoN regulon using either the computational prediction or the microarray transcriptome analysis included genes involved in flagella biosynthesis, pili biosynthesis and genes involved in central metabolism enzymes and cytochromes involved in extracellular electron transfer to Fe(III), which are known to be important for growth in subsurface environment or electricity production in microbial fuel cells. The consensus sequence for the predicted RpoN-regulated promoter elements is TTGGCACGGTTTTTGCT.

CONCLUSION

The G. sulfurreducens RpoN is an essential sigma factor and a global regulator involved in a complex transcriptional network controlling a variety of cellular processes.

The relative value of operon predictions

For most organisms, computational operon predictions are the only source of genome-wide operon information. Operon prediction methods described in literature are based on (a combination of) the following five criteria: (i) intergenic distance, (ii) conserved gene clusters, (iii) functional relation, (iv) sequence elements and (v) experimental evidence. The performance estimates of operon predictions reported in literature cannot directly be compared due to differences in methods and data used in these studies. Here, we survey the current status of operon prediction methods. Based on a comparison of the performance of operon predictions on Escherichia coli and Bacillus subtilis we conclude that there is still room for improvement. We expect that existing and newly generated genomics and transcriptomics data will further improve accuracy of operon prediction methods.

Using the Gibbs Motif Sampler for phylogenetic footprinting

The Gibbs Motif Sampler (Gibbs) is a software package used to predict conserved elements in biopolymer sequences. Although the software can be used to locate conserved motifs in protein sequences, its most common use is the prediction of transcription factor binding sites (TFBSs) in promoters upstream of gene sequences. We will describe approaches that use Gibbs to locate TFBSs in a collection of orthologous nucleotide sequences, i.e., phylogenetic footprinting. To illustrate this technique, we present examples that use Gibbs to detect binding sites for the transcription factor LexA in orthologous sequence data from representative species belonging to two different proteobacterial divisions.

Genome-wide expression profiling in Geobacter sulfurreducens: identification of Fur and RpoS transcription regulatory sites in a relGsu mutant

Rel(Gsu) is the single Geobacter sulfurreducens homolog of RelA and SpoT proteins found in many organisms. These proteins are involved in the regulation of levels of guanosine 3', 5' bispyrophosphate, ppGpp, a molecule that signals slow growth and stress response under nutrient limitation in bacteria. We used information obtained from genome-wide expression profiling of the rel(Gsu) deletion mutant to identify putative regulatory sites involved in transcription networks modulated by Rel(Gsu) or ppGpp. Differential gene expression in the rel(Gsu) deletion mutant, as compared to the wild type, was available from two growth conditions, steady state chemostat cultures and stationary phase batch cultures. Hierarchical clustering analysis of these two datasets identified several groups of operons that are likely co-regulated. Using a search for conserved motifs in the upstream regions of these co-regulated operons, we identified sequences similar to Fur- and RpoS-regulated sites. These findings suggest that Fur- and RpoS-dependent gene expression in G. sulfurreducens is affected by Rel(Gsu)-mediated signaling.

Importance of c-Type cytochromes for U(VI) reduction by Geobacter sulfurreducens

Background

In order to study the mechanism of U(VI) reduction, the effect of deleting c-type cytochrome genes on the capacity of Geobacter sulfurreducens to reduce U(VI) with acetate serving as the electron donor was investigated.

Results

The ability of several c-type cytochrome deficient mutants to reduce U(VI) was lower than that of the wild type strain. Elimination of two confirmed outer membrane cytochromes and two putative outer membrane cytochromes significantly decreased (ca. 50–60%) the ability of G. sulfurreducens to reduce U(VI). Involvement in U(VI) reduction did not appear to be a general property of outer membrane cytochromes, as elimination of two other confirmed outer membrane cytochromes, OmcB and OmcC, had very little impact on U(VI) reduction. Among the periplasmic cytochromes, only MacA, proposed to transfer electrons from the inner membrane to the periplasm, appeared to play a significant role in U(VI) reduction. A subpopulation of both wild type and U(VI) reduction-impaired cells, 24–30%, accumulated amorphous uranium in the periplasm. Comparison of uranium-accumulating cells demonstrated a similar amount of periplasmic uranium accumulation in U(VI) reduction-impaired and wild type G. sulfurreducens. Assessment of the ability of the various suspensions to reduce Fe(III) revealed no correlation between the impact of cytochrome deletion on U(VI) reduction and reduction of Fe(III) hydroxide and chelated Fe(III).

Conclusion

This study indicates that c-type cytochromes are involved in U(VI) reduction by Geobacter sulfurreducens. The data provide new evidence for extracellular uranium reduction by G. sulfurreducens but do not rule out the possibility of periplasmic uranium reduction. Occurrence of U(VI) reduction at the cell surface is supported by the significant impact of elimination of outer membrane cytochromes on U(VI) reduction and the lack of correlation between periplasmic uranium accumulation and the capacity for uranium reduction. Periplasmic uranium accumulation may reflect the ability of uranium to penetrate the outer membrane rather than the occurrence of enzymatic U(VI) reduction. Elimination of cytochromes rarely had a similar impact on both Fe(III) and U(VI) reduction, suggesting that there are differences in the routes of electron transfer to U(VI) and Fe(III). Further studies are required to clarify the pathways leading to U(VI) reduction in G. sulfurreducens.

Operon prediction in Pyrococcus furiosus

Identification of operons in the hyperthermophilic archaeon Pyrococcus furiosus represents an important step to understanding the regulatory mechanisms that enable the organism to adapt and thrive in extreme environments. We have predicted operons in P.furiosus by combining the results from three existing algorithms using a neural network (NN). These algorithms use intergenic distances, phylogenetic profiles, functional categories and gene-order conservation in their operon prediction. Our method takes as inputs the confidence scores of the three programs, and outputs a prediction of whether adjacent genes on the same strand belong to the same operon. In addition, we have applied Gene Ontology (GO) and KEGG pathway information to improve the accuracy of our algorithm. The parameters of this NN predictor are trained on a subset of all experimentally verified operon gene pairs of Bacillus subtilis. It subsequently achieved 86.5% prediction accuracy when applied to a subset of gene pairs for Escherichia coli, which is substantially better than any of the three prediction programs. Using this new algorithm, we predicted 470 operons in the P.furiosus genome. Of these, 349 were validated using DNA microarray data.

Genome-wide similarity search for transcription factors and their binding sites in a metal-reducing prokaryote Geobacter sulfurreducens

The knowledge obtained from understanding individual elements involved in gene regulation is important for reconstructing gene regulatory networks, a key for understanding cellular behavior. To study gene regulatory interactions in a model microorganism, Geobacter sulfurreducens, which participates in metal reduction and energy harvesting, we investigated the presence of 59 known Escherichia coli transcription factors and predicted transcription regulatory sites in its genome. The supplementary material, available at http://www.geobacter.org/research/genomescan/, provides the results of similarity comparisons that identified regulatory proteins of G. sulfurreducens and the genome locations of the predicted regulatory sites, including the list of putative regulatory elements in the upstream regions of every predicted operon and singleton open reading frame. Regulatory sequence elements, predicted using genome similarity searches to matrices of established transcription regulatory elements from E. coli, provide an initial insight into regulation of genes and operons in G. sulfurreducens. The predicted regulatory elements were predominantly located in the upstream regions of operons and singleton open reading frames. The validity of the predictions was examined using a permutation approach. Sequence similarity searches indicate that E. coli transcription factors ArgR, CytR, DeoR, FlhCD (both FlhC and FlhD subunits), FruR, GalR, GlpR, H-NS, LacI, MetJ, PurR, TrpR, and Tus are likely missing from G. sulfurreducens. Phylogenetic analysis suggests that one HU subunit is present in G. sulfurreducens as compared to two subunits in E. coli, while each of the two E. coli IHF subunits, HimA and HimD, have two homologs in G. sulfurreducens. The closest homolog of E. coli RpoE in G. sulfurreducens may be more similar to FecI than to RpoE. These findings represent the first step in the understanding of the regulatory relationships in G. sulfurreducens on the genome scale.

Role of RelGsu in stress response and Fe(III) reduction in Geobacter sulfurreducens

Geobacter species are key members of the microbial community in many subsurface environments in which dissimilatory metal reduction is an important process. The genome of Geobacter sulfurreducens contains a gene designated rel(Gsu), which encodes a RelA homolog predicted to catalyze both the synthesis and the degradation of guanosine 3',5'-bispyrophosphate (ppGpp), a regulatory molecule that signals slow growth in response to nutrient limitation in bacteria. To evaluate the physiological role of Rel(Gsu) in G. sulfurreducens, a rel(Gsu) mutant was constructed and characterized, and ppGpp levels were monitored under various conditions in both the wild-type and rel(Gsu) mutant strains. In the wild-type strain, ppGpp and ppGp were produced in response to acetate and nitrogen deprivation, whereas exposure to oxygen resulted in an accumulation of ppGpp alone. Neither ppGpp nor ppGp could be detected in the rel(Gsu) mutant. The rel(Gsu) mutant consistently grew to a higher cell density than the wild type in acetate-fumarate medium and was less tolerant of oxidative stress than the wild type. The capacity for Fe(III) reduction was substantially diminished in the mutant. Microarray and quantitative reverse transcription-PCR analyses indicated that during stationary-phase growth, protein synthesis genes were up-regulated in the rel(Gsu) mutant and genes involved in stress responses and electron transport, including several implicated in Fe(III) reduction, were down-regulated in the mutant. The results are consistent with a role for Rel(Gsu) in regulating growth, stress responses, and Fe(III) reduction in G. sulfurreducens under conditions likely to be prevalent in subsurface environments.

Computational prediction of RpoS and RpoD regulatory sites in Geobacter sulfurreducens using sequence and gene expression information

RpoS, the sigma S subunit of RNA polymerase, is vital during the growth and survival of Geobacter sulfurreducens under conditions typically encountered in its native subsurface environments. We investigated the conservation of sites that may be important for RpoS function in G. sulfurreducens. We also employed sequence information and expression microarray data to predict G. sulfurreducens genome sites that may be related to RpoS regulation. Hierarchical clustering identified three clusters of significantly downregulated genes in the rpoS deletion mutant. The search for conserved overrepresented motifs in co-regulated operons identified likely -35 and -10 promoter elements upstream of a number of functionally important G. sulfurreducens operons that were downregulated in the rpoS deletion mutant. Putative -35/-10 promoter elements were also identified in the G. sulfurreducens genome using sequence similarity searches to matrices of -35/-10 promoter elements found in G. sulfurreducens and in Escherichia coli. Due to a sufficient degree of sequence similarity between -35/-10 promoter elements for RpoS, RpoD, and other sigma factors, both the sequence similarity searches and the search for conserved overrepresented motifs using microarray data may identify promoter elements for both RpoS and other sigma factors.

Predicting cis-acting elements of Lactobacillus plantarum by comparative genomics with different taxonomic subgroups

Cis-acting elements in Lactobacillus plantarum were predicted by comparative analysis of the upstream regions of conserved genes and predicted transcriptional units (TUs) in different bacterial genomes. TUs were predicted for two species sets, with different evolutionary distances to L.plantarum. TUs were designated ‘cluster of orthologous transcriptional units’ (COT) when >50% of the genes were orthologous in different species. Conserved DNA sequences were detected in the upstream regions of different COTs. Subsequently, conserved motifs were used to scan upstream regions of all TUs. This method revealed 18 regulatory motifs only present in lactic acid bacteria (LAB). The 18 LAB-specific candidate regulatory motifs included 13 that were not described previously. These LAB-specific different motifs were found in front of genes encoding functions varying from cold shock proteins to RNA and DNA polymerases, and many unknown functions. The best-described LAB-specific motif found was the CopR-binding site, regulating expression of copper transport ATPases. Finally, all detected motifs were used to predict co-regulated TUs (regulons) for L.plantarum, and transcriptome profiling data were analyzed to provide regulon prediction validation. It is demonstrated that phylogenetic footprinting using different species sets can identify and distinguish between general regulatory motifs and LAB-specific regulatory motifs.

Reconstruction of regulatory and metabolic pathways in metal-reducing delta-proteobacteria

A study of the genetic and regulatory factors in several biosynthesis, metal ion homeostasis, stress response, and energy metabolism pathways suggests that phylogenetically diverse δ-proteobacteria have homologous regulatory components.

Background

Relatively little is known about the genetic basis for the unique physiology of metal-reducing genera in the delta subgroup of the proteobacteria. The recent availability of complete finished or draft-quality genome sequences for seven representatives allowed us to investigate the genetic and regulatory factors in a number of key pathways involved in the biosynthesis of building blocks and cofactors, metal-ion homeostasis, stress response, and energy metabolism using a combination of regulatory sequence detection and analysis of genomic context.

Results

In the genomes of δ-proteobacteria, we identified candidate binding sites for four regulators of known specificity (BirA, CooA, HrcA, sigma-32), four types of metabolite-binding riboswitches (RFN-, THI-, B12-elements and S-box), and new binding sites for the FUR, ModE, NikR, PerR, and ZUR transcription factors, as well as for the previously uncharacterized factors HcpR and LysX. After reconstruction of the corresponding metabolic pathways and regulatory interactions, we identified possible functions for a large number of previously uncharacterized genes covering a wide range of cellular functions.

Conclusions

Phylogenetically diverse δ-proteobacteria appear to have homologous regulatory components. This study for the first time demonstrates the adaptability of the comparative genomic approach to de novo reconstruction of a regulatory network in a poorly studied taxonomic group of bacteria. Recent efforts in large-scale functional genomic characterization of Desulfovibrio species will provide a unique opportunity to test and expand our predictions.