Other related taxa include uncultured clones from a variety of engineered and natural environments, including: wastewater digesters in France and the United States, estuarine sediments in Taiwan, peat wetlands in Japan, sinkholes in Mexico, and hydrothermal vents in the Pacific Ocean . Whether or not the capacity for DPO is a common feature of this clade remains to be seen, as there is evidence of lateral acquisition of phosphite oxidation genes in FiPS-3 as well as in several APO-capable bacteria . Indeed, whether Phox-21 itself is capable of growth by DPO coupled to CO2 reduction in pure culture has yet to be confirmed, since efforts to isolate this organism have so far proven unsuccessful.The work presented in Chapter 3 revealed the presence in my wastewater enrichments of a novel bacterium from an uncultured clade within the Deltaproteobacteria whose abundance was strongly correlated with DPO activity. Based on this evidence, I hypothesized that this bacterium, designated as strain Phox-21, was the organism responsible for phosphite oxidation in my cultures. However, as my attempts to isolate Phox-21 were unsuccessful, I was unable to confirm that it was capable of carrying out this metabolism in pure culture. Therefore, I decided to perform a metagenomic analysis of some of the DNA samples I had previously collected for 16S rRNA gene analysis in order to look for possible functional markers of DPO, such as the ptx-ptd genes,grow hydroponic in the genome of Phox-21. In addition to establishing a more conclusive link between Phox-21 and DPO, I also hoped that genomic profiling of this organism would reveal its broader metabolic characteristics and provide insights that could aid in its isolation and culturing.
Furthermore, I expected the metagenomic dataset to shed light on the metabolic capabilities of other enrichment community members, thus providing a wider ecological context for the role of DPO in this system.Coates Genomics Sequencing Laboratory at UC Berkeley for sequencing on an Ilumina HiSeq 2000 . Ilumina sequencing reads were trimmed for quality and filtered using Sickle v1.33 with a quality threshold value of 28 and good-quality paired-end reads were then merged using IDBA-UD v1.0 . Merged reads from all samples were combined and assembled using MEGAHIT v1.0.2 with default parameters . MEGAHIT is an assembler developed specifically for metagenomic reads that uses succinct de Bruijn graphs and an iterative multiple k-mer size strategy. Merged reads from each sample where then mapped backed to the combined assembly using BWA-MEM v0.7.10 with default parameters in order to assess sequencing coverage . Contigs from the combined assembly were binned into individual genomes using the Anvi’o v1.1.0 platform . Anvi’o generates hierarchical clusters of related contigs using both tetranucleotide frequency and coverage across samples as the clustering parameters. The platform also provides a visualization interface that allows the user to further refine the contig clusters into genome bins based on coverage, GC content, and phylogenetic marker genes. Furthermore, Anvi’o assigns taxonomic lineages to the genome bins based on the presence of phylogenetic marker genes. Genome bins generated with Anvi’o were subsequently assessed for completeness and contamination based on the presence of lineagespecific, conserved, single-copy marker genes using the automated bin evaluation tool CheckM v1.0.1 . CheckM calculates ‘completeness’ based on the number of expected marker genes that are present in a given bin and ‘contamination’ based on the number of marker genes that are present in multiple copies and have less than 90% amino acid identity to each other.
High-quality genomes were submitted to the Integrated Microbial Genomes database for gene calling and annotation . IMG utilizes Prodigal v2.50 for identification of protein-coding genes, which are then functionally annotated using a custom, manually-curated pipeline based on BLAST and HMMER searches against multiple protein databases .The presence of a ptx-ptd gene cluster in the genome of Phox-21, as well as its higher abundance during phosphite-oxidizing conditions, clearly indicates that this is the organism responsible for DPO in our enrichments. In addition, the observed CO2 dependence of DPO in our enrichments coupled with the fact that no other terminal electron acceptors were added to our media implies that Phox-21 is capable of growing by coupling phosphite oxidation to CO2 reduction. The presence of a formate dehydrogenase complex FdhAB similar to that of M. thermoacetica in Phox-21 provides a putative means by which CO2 reduction could occur. However, the absence of key WLP genes suggests that this organism is unable to generate acetyl-CoA from CO2 alone and therefore is not a true autotroph. Furthermore, it lacks an electron transport chain and thus appears to be incapable of energy conservation through oxidative phosphorylation. Instead, we propose that Phox-21 couples phosphite oxidation to CO2 reduction to formate by means of FdhAB and uses the energy generated by this reaction to assimilate organic carbon sources such as acetate . Based on thermodynamic calculations and physiological evidence, Schink et al. have previously proposed that FiPS-3 is able to conserve energy during DPO by directly generating ATP as well as NADH from the oxidation of phosphite. This putative substrate-level phosphorylation step during DPO is likely mediated by PtdFHI and would allow for energy conservation in the absence of membrane-associated electron transport .
ATP produced in this manner could be used by Phox-21 to incorporate acetate into biomass via AcsM and the partial TCA cycle as well as to run the proton and sodium translocating ATP synthases in reverse in order to establish an ion motive force across the cell membrane . The resulting sodium ion gradient could drive the RNF complex to reduce ferredoxin, which could then serve as an electron donor for pyruvate synthesis by the PFOR enzyme as well as for NADPH production by the NfnAB complex . The FocA transporter could serve both to import acetate for assimilation and to export formate from the cell as a metabolic waste product . However,mobile grow system in the presence of nitrite, formate could be re-oxidized to CO2 by FdoGHI coupled to the reduction of nitrite by NrfAH . This reaction would contribute to the maintenance of a proton motive force and would also yield ammonia, which could be imported into the cell via the AmtB transporter to serve as a nitrogen source. Our metabolic model predicts that Phox-21 should require an organic carbon substrate such as acetate for growth and should also excrete formate into the medium as a product of CO2 reduction. A requirement for organic carbon may at least partly explain the stimulatory effect of rumen fluid on DPO, since rumen fluid has been shown to contain as much as 60 mM acetate in addition to various carbohydrates, organic acids, amino acids, and fatty acids . Additionally, Phox-21 is predicted to be incapable of synthesizing alanine, histidine, threonine, or THF, all of which were absent from our original growth media but may be present in rumen fluid. However, attempts to grow our enrichments in media supplemented with acetate, THF, and amino acids but lacking rumen fluid have so far resulted in substantially lower phosphite oxidation rates, indicating that there may be other components in the rumen fluid that promote the growth of Phox-21. Acetate present in rumen fluid amended cultures could also have served as a growth substrate for the Tepidanaerobacter strains and methanogens present in the communities.
Both Tepidanaerobacter genomes have all the genes for the carbonyl branch of the WLP but lack a formate dehydrogenase and an uptake hydrogenase, which is consistent with previous genomic analysis of Tepidanaerobacter acetatoxydans and indicates that these organisms are capable of syntrophic acetate oxidation but not of autotrophic growth . Likewise, both Methanoculleus sp. EBM-46 and Methanococcoides sp. EBM-47 appear to be capable of acetoclastic methanogenesis, although only EBM-46 has the genes necessary for growth on H2/CO2 and formate as well. The remaining enrichment community members had no WLP genes and mostly lacked genes involved in respiratory processes and were therefore likely involved in the fermentation of organic acids, amino acids, and carbohydrates present in the rumen fluid. DPO was discovered over a decade ago in the marine sediment isolate D. phosphitoxidans FiPS-3, but there had so far been no additional reports of this metabolism in other environments. This study, therefore, represents only the second ever observation of DPO and the first ever description of a community structure and metagenome related to this metabolism. Furthermore, the organism responsible for phosphite oxidation in our system, Phox-21, is a novel bacterium belonging to a candidate order within the Deltaproteobacteria that currently has no cultured representatives. Although my attempts to isolate Phox-21 were unsuccessful, metagenomic analysis revealed the presence of a ptx-ptd cluster in its genome, similar to the one found in FiPS-3. Previous work has shown that ptxD and ptdC are necessary for phosphite oxidation in FiPS-3 . Additionally, as part of this study, I found that ptdFGHI are significantly upregulated in FiPS-3 in the presence of phosphite. That ptdCFHI are also found in Phox-21, but not in any other genome currently available in the IMG database, is further evidence that these genes play an important role in DPO. Interestingly, ptdG is present in FiPS-3 but not in Phox-21, which suggests that this gene may not be essential for DPO in every organism, although it may still be required by FiPS-3. However, much work is still needed in order to elucidate how energy for growth is conserved during DPO. The involvement of PtdFHI in this process has yet to be experimentally confirmed and the mechanism of action of these enzymes has not been determined. Whether ATP is indeed produced from substrate-level phosphorylation during DPO is another outstanding question that requires additional experiments, both in vivo and in vitro, in order to conclusively address. It is assumed to function as a phosphite/phosphate antiporter based on its homology to known antiporters, but this assertion still needs to be tested. The presence of an incomplete Wood-Ljungdahl Pathway in Phox-21 was unexpected, since we had previously assumed that it was capable of growing autotrophically. Nonetheless, this predicted requirement for organic carbon could at least partially explain why the growth of Phox-21 improved so markedly when rumen fluid was added to the enrichments. Although there have been previous reports of organisms with incomplete WLPs , to our knowledge, this study provides the first genomic evidence of an organism capable of using CO2 as a terminal electron acceptor but not as a carbon source. Due to its extremely low redox potential, phosphite is the only known biological electron donor that could drive the reduction of CO2 to formate while generating enough energy to produce ATP for the incorporation of acetate into biomass. As such, this unprecedented metabolism may be unique to DPO-capable organisms. However, as discussed above, the feasibility of this proposed metabolism hinges on whether or not DPO can generate ATP through substrate-level phosphorylation, which is still an open question. Still, the apparent lack of any electron transport chain components in Phox-21 suggests that under phosphite-oxidizing, CO2-reducing conditions, energy conservation in this organism would have to proceed exclusively by means of substrate-level phosphorylation. Ultimately, though, this metabolic model still needs to be experimentally confirmed. I have so far been unable to detect significant formate production or acetate consumption under DPO conditions in my enrichments. However, it may not be possible to detect these processes in enrichment cultures due to the presence of other community members capable of using formate and producing acetate. Obtaining a pure culture of Phox-21 would therefore greatly facilitate any future investigations of its physiology. The environmental prevalence and phylogenetic diversity of DPO-capable organisms remains unclear. Like FiPS-3, Phox-21 belongs to the Deltaproteobacteria, but the sample size of known DPO-capable organisms is still far too small to allow us to determine whether this metabolism is restricted to a specific phylogenetic group. There is evidence that the ptx-ptd genes in FiPS-3 were acquired by lateral gene transfer, but we do not know how common it is for these genes to be horizontally propagated in the environment or what the phylogenetic range of these events might be.