Genomes to Life Contractor-Grantee Workshop II
February 29-March 2, 2004, Washington, D.C.
Genomics:GTL Program Projects
University of Massachusetts, Amherst
Analysis of the Genetic Potential and Gene Expression of Microbial Communities Involved in the in situ Bioremediation of Uranium and Harvesting Electrical Energy from Organic Matter
17
Analysis of Predominant Genome Sequences and Gene Expression During In Situ Uranium Bioremediation and Harvesting Electricity from Waste Organic Matter
Stacy Ciufo1*, Dawn Holmes1, Zhenya Shelbolina1, Barbara Methé2, Kelly Nevin1, and Derek Lovley1 (dlovley@microbio.umass.edu)
*Presenting author
1Department of Microbiology, University of Massachusetts, Amherst, MA and 2The Institute for Genomic Research, Rockville, MD
Field studies have demonstrated that stimulating dissimilatory metal reduction in uranium-contaminated subsurface environments is an effective, simple, and inexpensive method for removing uranium from contaminated groundwater. Molecular analyses, which avoid any culture bias, have demonstrated that Geobacter species are the predominant microorganisms in a variety of subsurface environments in which dissimilatory metal reduction is an important process. For example, during in situ bioremediation of a uranium-contaminated site in Rifle, Colorado Geobacter species accounted for as much as 80% of the microbial community in the groundwater during uranium bioremediation. In a similar manner, it has been demonstrated that Geobacter species are the predominant microorganisms on electrodes used to harvest electricity from waste organic matter. In order to determine whether models of Geobacter physiology derived from pure culture studies are applicable to as-yet-uncultured Geobacters living in uranium-contaminated subsurface environments or on the surface of electrodes, it is necessary to determine the relative similarities of the genome sequence and gene expression patterns of as-yet-uncultured Geobacters and the pure cultures.
One strategy to evaluate the genetic potential of the Geobacters that predominate in the environments of interest is to sequence genomic DNA directly extracted from the environment. Genomic DNA was extracted from the sediments of a uranium-contaminated aquifer, located in Rifle, Colorado, in which the activity of Geobacters had been stimulated with the addition of acetate to promote precipitation of uranium. The DNA was cloned in bacterial artificial chromosomes (BACs) with an average insert size of ca. 40 kbp. Large scale sequencing of the BAC inserts resulted in the recovery of 4.2 mbp of environmental genomic DNA sequence. The sequence data was assembled with BACPACK, an algorithm we specifically developed for this purpose. A contig of over 580 kbp of sequence was assembled as were several other contigs of 25-475 kbp. Analysis for highly conserved Geobacter genes indicated that these contigs were from a Geobacter species. The uncultured Geobacter had a 16S rRNA gene sequence identical to a sequence that predominated in the groundwater during uranium bioremediation. The environmental genome sequence was similar to that of pure Geobacter species in that it had a lower percentage of putative proteins predicted to be localized in the cytoplasm and more proteins targeted to the inner membrane, periplasm, and outer membrane than has been found in non-Geobacteraceae. The uncultured Geobacter species had a high percentage of Geobacter signature genes and gene arrangements similar to those found in cultured species. However, there were also genes of unknown function in the uncultured Geobacter that have not been identified in the genomes of any pure cultures. These results suggest that although the uncultured Geobacter species involved in uranium bioremediation are clearly not identical to the pure cultures that are being intensively studied, there are many genomic similarities and thus models of pure cultures may have applicability to Geobacter-dominated subsurface environments.
Another strategy to determine the genome sequences of the Geobacters predominating in environments of interest is to adapt culture conditions to permit culturing of these organisms. A medium in which the clay-fraction of subsurface sediments was the source of Fe(III) oxide was developed. With this medium a Geobacter species with a 16S rRNA gene sequence identical to one of the sequences that predominated during uranium bioremediation was isolated. Cultivation of this organism required the addition of groundwater from the site to the medium. Sufficient quantities of this organism have now been cultured so that its genome can be sequenced. Of further interest is the finding that one of the large BAC contigs from the uranium bioremediation site has the same 16S rRNA gene sequence. Thus, it will be possible to compare results from direct sequencing of environmental genomic DNA with the strategy of isolation in culture followed by genome sequencing.
One test of the environmental applicability of the current physiological models of Geobacter metabolism is to determine whether Geobacters in environments of interest have patterns of gene expression that are similar to those in pure cultures. Therefore, methods for effectively extracting mRNA from aquifer sediments and the surface of energy-harvesting electrodes were developed. Initial studies on the metabolic state of Geobacter species in aquifer sediments demonstrated that the natural populations of Geobacters were highly expressing genes for nitrogen fixation, suggesting that they were limited for fixed nitrogen. Addition of 100 µM ammonium to the sediment repressed expression of the nitrogen fixation genes. These results demonstrated that it is possible to evaluate the in situ metabolic state of Geobacters in subsurface environments. The next step will be to evaluate the expression of a larger suite of genes involved in nutrient uptake and stress response with microarrays.
Evaluation of gene expression in Geobacter sulfurreducens growing on the surface of energy-harvesting electrodes suggested that environmental analysis with whole-genome DNA microarrays is feasible. A microarray was used to compare mRNA levels of G. sulfurreducens growing on electrodes with mRNA levels of planktonic cells. Up-regulation of several genes was significant on the electrode. For example, mRNA levels for several outer-membrane cytochromes were 40-80 fold higher in cells growing on the electrodes. This suggests that these cytochromes play an important role in electron transfer to the electrode surfaces. There was also an upregulation of genes annotated as encoding for heavy-metal efflux proteins. This may reflect the presence of heavy metal contaminants in the electrode material. Many genes of unknown function were also down-regulated. The most prominently down-regulated genes were related to oxygen respiration and/or oxygen toxicity. These included a cytochrome oxidase as well as thioredoxin peroxidase and superoxide dismutase. This is indicative of an important change in the respiratory pathway. These results demonstrate that electron transfer to electrodes is associated with significant shifts in gene expression and provide the first insights into the mechanisms for this novel form of respiration.
In summary, these initial results demonstrate that it will be possible not only to determine the genetic potential of the Geobacter species actually involved in subsurface bioremediation or in harvesting electricity from waste organic matter, but also to broadly assess their metabolic state. This will significantly improve the development of in silico models for predicting the metabolic responses of Geobacter species under different environmental conditions and provide information on how to most effectively optimize these applications of Geobacter.
18
Functional Analysis of Genes Involved in Electron Transport to Metals in Geobacter sulfurreducens
Maddalena Coppi1*, Eman Afkar1, Tunde Mester1, Daniel Bond1, Laurie DiDonato1, Byoung-Chan Kim1, Richard Glaven1, Ching Leang1, Winston Lin1, Jessica Butler1, Teena Mehta1, Susan Childers1, Barbara Methé2, Kelly Nevin1, and Derek Lovley1 (dlovley@microbio.umass.edu)
*Presenting author
1Department of Microbiology, University of Massachusetts, Amherst, MA and 2The Institute for Genomic Research, Rockville, MD
As noted in a companion abstract, ecological studies have demonstrated that Geobacter species are the predominant microorganisms in a variety of subsurface environments in which dissimilatory metal reduction is an important process, including during in situ uranium bioremediation. Therefore, in order to effectively model in situ bioremediation of uranium and develop strategies for improving this process it is necessary to understand the factors controlling the growth and activity of Geobacter species. Most important in this regard is information on electron transfer not only to U(VI), but also to Fe(III), because most of the energy supporting the growth of Geobacter species in uranium-contaminated subsurface environments is derived from electron transfer to Fe(III).
Functional analysis of electron transfer to metals in Geobacter sulfurreducens has initially focused on the c-type cytochromes which are abundant in the genome, as well as outer-membrane proteins of previously unknown function. For example, analysis of the outer-membrane proteins of G. sulfurreducens with MALDI-TOF mass spectrometry revealed that the most abundant protein, designated OmpA, had a predicted amino acid sequence without any significant homology with previously described genes. OmpA is predicted to have a hydrophobic leader sequence, consistent with export to the outer membrane, and a b-barrel structure. No heme c or metal binding motifs were detected. When the gene was deleted with the single gene replacement method, the ompA-deficient mutant grew the same as wild type with fumarate as the electron acceptor, but it could not grow with Fe(III) or Mn(IV) oxides as the electron acceptor. Although the total heme c content in the mutant and the wild type were comparable, the mutant had only ca. 50% of the heme c content in the outer membrane as the wild type. There was a corresponding substantial increase in the total heme c content in the cytoplasmic membrane and soluble fraction of the ompA mutant. These results suggest that OmpA plays an important role in localizing c-type cytochromes in the outer membrane of G. sulfurreducens via a novel mechanism not previously described in any microorganism.
A mutation in a novel secretory system in G. sulfurreducens specifically eliminated its ability to reduce Fe(III) oxides, but not soluble electron acceptors, including chelated Fe(III). Comparison of the proteins in the periplasm of this mutant with wild type cells indicated that several proteins were accumulating in the periplasm of the mutant. Analysis of peptide fragments of one of these proteins revealed a gene, designated ompB, which encodes for a 1303 amino acid protein, with 23 transmembrane amino acids and 1275 amino acids predicted to be exposed outside the cell. There are four putative metal-binding sites. This gene is found in the four Geobacteraceae genome sequences that are available, but not in any other organisms. The ompB mutant did not grow on Fe(III) oxide, but grew on soluble electron acceptors. These results suggest that OmpB plays an important role in cell-Fe(III) oxide contact or in sequestering Fe(III) from Fe(III) oxides, prior to Fe(III) reduction. This is a novel concept for dissimilatory Fe(III) oxide reduction.
Our previous studies have suggested that c-type cytochromes are important in electron transfer to Fe(III) in Geobacter sulfurreducens. However, elucidating which cytochromes are involved in Fe(III) reduction is not trivial because the genome of G. sulfurreducens contains genes for over 100 c-type cytochromes, at least 25 of which are predicted to be localized in the outer membrane where Fe(III) reduction is likely to take place. Therefore, our initial strategy has been to focus on cytochromes predicted to be localized in the outer membrane, as well as cytochromes that are specifically expressed during growth on Fe(III). Functional analysis of nearly all of the outer-membrane c-type cytochromes has led to the surprising result that, in many instances, the deletion of just one of the cytochrome genes severely inhibits Fe(III) reduction. This suggests that many of the multiple outer-membrane cytochromes do not serve duplicative functions, but act in concert to bring about electron transfer to Fe(III).
There are also many periplasmic cytochrome genes that are highly similar. Mutants were generated in order to evaluate their role in electron transfer to metals. It was found that PpcA, PpcB, and PpcC are required for soluble Fe(III) reduction but that mutants that could no longer produce PpcD or PpcE grew better on Fe(III) than the wild type, as did a double mutant lacking PpcB and PpcC. These results demonstrate that despite their apparent similarities in size and heme content, these periplasmic cytochromes have some different functions in electron transfer in G. sulfurreducens and that it is possible to make mutations that will enhance electron transfer to metals.
It has been proposed that, based on analogy to our previous findings in Desulfovibrio vulgaris, c-type cytochromes are also important electron carriers for U(VI) reduction. Analysis of over 15 c-type cytochrome mutants suggested that the small periplasmic c-type cytochromes in G. sulfurreducens, which are most closely related to the c3 cytochrome responsible for U(VI) reduction in D. vulgaris, were not responsible for U(VI) reduction. However, knockout mutations in several outer-membrane cytochromes inhibited U(VI) reduction. These results suggest that the mechanisms for U(VI) reduction in G. sulfurreducens are significantly different than for D. vulgaris and indicate that even though U(VI) is soluble, and could potentially be reduced in the periplasm, reduction by G. sulfurreducens is more likely to take place primarily at the outer membrane surface.
Sequencing of the G. sulfurreducens genome revealed the presence of genes predicted to be involved in oxygen respiration, which was surprising because no Geobacter species had ever been found to grow on oxygen. However, growth conditions under which G. sulfurreducens can grow at oxygen concentrations that are 50% or less of atmospheric levels have now been identified. Knockout mutation studies demonstrated that growth on oxygen is dependent upon a cytochrome oxidase. The ability of Geobacter species to grow at low oxygen levels helps explain how they survive in aerobic subsurface environments and then rapidly respond to the development of anaerobic conditions during metals bioremediation.
Functional analysis of proteins important in central metabolism, such as a novel eukaryotic-like citrate synthase and a bifunctional succinate dehydrogenase/fumarate reductase, has also been completed. These studies are rapidly improving the understanding the physiology of G. sulfurreducens. This information will permit more informed decisions on strategies to optimize bioremediation and energy harvesting applications of Geobacter species.
19
Adapting Regulatory Strategies for Life in the Subsurface: Regulatory Systems in Geobacter sulfurreducens
Gemma Reguera1*, Cinthia Nunez1, Richard Glaven1, Regina O’Neil1, Maddalena Coppi1, Laurie DiDonato1, Abraham Esteve-Nunez1, Barbara Methé2, Kelly Nevin1, and Derek Lovley1 (dlovley@microbio.umass.edu)
*Presenting author
1Department of Microbiology, University of Massachusetts, Amherst, MA and 2The Institute for Genomic Research, Rockville, MD
As outlined in accompanying abstracts, Geobacter sulfurreducens serves as a pure culture model for the Geobacter species that are responsible for in situ uranium bioremediation in contaminated subsurface environments and that harvest electricity from waste organic matter. In order to predictively model the activity of Geobacters involved in bioremediation and energy harvesting it is necessary to understand how electron transport to metals as well as central metabolism are regulated under different environmental conditions.
Of particular relevance for bioremediation and energy harvesting applications of Geobacter species is understanding regulation of gene expression under the sub-optimal growth conditions typically encountered in subsurface environments. For example, the genome of G. sulfurreducens contains a homolog of the E. coli stationary-phase sigma factor, RpoS, which is of interest because growth in the subsurface is likely to be analogous to the stationary phase of cultures. Survival in stationary phase, aerotolerance, growth on oxygen, and reduction of insoluble Fe(III) were diminished in an rpoS mutant, but there was no apparent impact on response to high temperature or alkaline pH stress, as seen in E. coli. In order to further elucidate the rpoS regulon, gene expression in the rpoS mutant and the wild type were compared with whole genome DNA microarray and proteomics approaches. These studies demonstrated that RpoS controls genes involved in Fe(III) reduction, oxygen tolerance, and oxygen respiration. This study represents the first characterization of RpoS in a member of the d subclass of the Proteobacteria and suggests that RpoS plays an important role in regulating metabolism of Geobacter species under the stressful conditions found in subsurface environments.
RpoS negatively regulates another sigma factor, RpoE, which modulates a distinct regulon also involved in oxygen tolerance and repair of oxidative stress damage. RpoE also was found to have an important role in controlling attachment to Fe(III) oxide and electrode surfaces, two key processes for the environmental success of Geobacter species. Genome-wide transcriptional profiles of G. sulfurreducens biofilms grown on Fe(III) oxide surfaces versus their planktonic counterparts, as well as transcriptional profiling of an rpoE mutant, suggested that RpoE regulates the transition from planktonic to biofilm conditions as well as maintenance of the biofilm mode of growth and electron transfer to Fe(III). These results demonstrate that RpoE and RpoS act coordinately to finely tune the adaptive responses that enable Geobacters to survive and outcompete many other organisms in subsurface environments.
RelA is another regulatory protein that could be important in influencing growth in the subsurface as a mutant in the putative relA gene in G. sulfurreducens grew faster than wild type under nutrient limitation. Microarray analyses of the relA mutant demonstrated that, as in E. coli, ribosomal proteins and chaperones are negatively regulated by RelA, while stress response genes are positively controlled, further suggesting that RelA may play a critical role in slow growth and stress response. In addition, RelA also appeared to positively regulate proteins required for the reduction of insoluble Fe(III) reduction, thus illustrating that in G. sulfurreducens RelA has unique targets that link the regulation of growth rate to metal reduction.
Analysis of the G. sulfurreducens genome revealed that this organism is highly attuned to its environment with 5.2% of the open reading frames in the genome dedicated to two-component proteins and 1.9% dedicated to chemotaxis. A combination of genomics and proteomics approaches identified putative histidine kinases and response regulators and results of microarray analyses of wild type and selected mutants enabled preliminary characterization and pairing of 16 two-component signal transduction proteins that previously were of unknown function and classified as “orphans”. Histidine kinase knockouts in G. sulfurreducens over-expressing the cognate response regulators produced information on the environmental signals triggering regulatory cascades and provided further support for the role of two component systems in integrating responses to environmental stimuli with electron transfer.
Geobacter species generally live in environments high in dissolved Fe(II) and have unusually high requirements for iron due to their high cytochrome content. In G. sulfurreducens, concentrations of dissolved Fe(II) as high as 100 µM were found to be required for optimal growth and acetate uptake and the cellular iron content greatly exceeded that of E. coli, suggesting that mechanisms to regulate iron uptake and iron overload in Geobacters may be different than in other, previously studied organisms. A homolog of the E. coli Fe(II)-dependent ferric uptake regulator, Fur, was identified in the G. sulfurreducens genome. As in E. coli, expression of fur in G. sulfurreducens was repressed in the presence of Fe(II) and the phenotype of a fur-knockout mutant suggested that Fur has a key role in responding to changes in Fe(II) concentration in the environment. Only a small fraction of Fur-regulated genes identified by microarray analysis were preceded by a recognizable Fur box. Surprises in the genes under Fur control included proteins required for Fe(III) oxide reduction.
These studies, as well as other ongoing studies on novel regulatory strategies in G. sulfurreducens, suggest that models of regulation that have been developed in previously studied microorganisms can help in identifying some of the regulatory components in G. sulfurreducens but, in many instances, regulation patterns and mechanisms in G. sulfurreducens have been modified in order to adapt to life in the subsurface. Results from these regulation studies will be incorporated into the expanding in silico model of G. sulfurreducens in order to better predict the likely response of Geobacter species during attempts to optimize bioremediation and energy harvesting strategies.
See Also
