Bacterial Metabolism And Physiology
Ailene Goldhirsh
Bacterial obligate intracellular parasites (BOIPs) represent an exclusive group of bacterial pathogens that all depend on invasion of a eukaryotic host cell to reproduce. BOIPs are characterized by extensive adaptation to their respective replication niches, regardless of whether they replicate within the host cell cytoplasm or within specialized replication vacuoles. Genome reduction is also a hallmark of BOIPs that likely reflects streamlining of metabolic processes to reduce the need for de novo biosynthesis of energetically costly metabolic intermediates. Despite shared characteristics in lifestyle, BOIPs show considerable diversity in nutrient requirements, metabolic capabilities, and general physiology. In this review, we compare metabolic and physiological processes of prominent pathogenic BOIPs with special emphasis on carbon, energy, and amino acid metabolism. Recent advances are discussed in the context of historical views and opportunities for discovery.
Figure 4 Developmental transitions. While recent advances based on molecular genetics are starting to shed light on the mechanism(s) of developmental transitions in Coxiella and Chlamydia, the genes and regulatory networks involved remain elusive. Leading ideas for mechanisms underlying developmental transitions are highlighted. LCVs and RBs vs SCVs and EBs are indicated by larger gray cells or smaller dark calls, respectively. Infectious form relates to infectivity of cultured cells. The direction of the developmental cycle is indicated in parentheses. "PG" and "MEC" denote peptidoglycan and 2-C-methylerythritol 2,4-cyclodiphosphate, respectively.
Figure 5 Opportunities for discovery. Certain areas of research appear especially relevant given recent technical and/or scientific advances. While these areas may be scientifically distinct, regulatory mechanisms are biologically intertwined and expected to influence pathogen virulence characteristics. Indicated micronutrients include, trace metals, vitamins and co-factors, macronutrients carbon sources and amino acids, and physicochemical conditions pH, oxygen and carbon dioxide.
Copyright 2024 Mandel, Sanchez, Monahan, Phuklia and Omsland. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Bacteria are the most abundant cells on Earth. They are generally regarded as ancient, but due to striking diversity in their metabolic capacities and widespread lateral gene transfer, the physiology of the first bacteria is unknown. From 1089 reference genomes of bacterial anaerobes, we identified 146 protein families that trace to the last bacterial common ancestor, LBCA, and form the conserved predicted core of its metabolic network, which requires only nine genes to encompass all universal metabolites. Our results indicate that LBCA performed gluconeogenesis towards cell wall synthesis, and had numerous RNA modifications and multifunctional enzymes that permitted life with low gene content. In accordance with recent findings for LUCA and LACA, analyses of thousands of individual gene trees indicate that LBCA was rod-shaped and the first lineage to diverge from the ancestral bacterial stem was most similar to modern Clostridia, followed by other autotrophs that harbor the acetyl-CoA pathway.
Both from the geochemical and the biological standpoint, looking back into the earliest phases of evolution ca. 4 billion years ago is challenging. The geological challenge is that rocks of that age are generally rare, and those that bear traces of life are extremely scarce. The biological challenge is that LGT has reassorted genes across genomes for 4 billion years. As an alternative to reconstructing gene history, metabolic networks themselves harbor independent inroads to the study of early evolution25. Metabolic networks represent the set of chemical transformations that occur within a cell, leading to both energy and biomass production26. Genome-scale metabolic networks are inferred from a full genome and the corresponding full set of functional (metabolic) annotations27, allowing for predictive models of growth and insights into physiology28. Furthermore, metabolism itself is connected to the informational processing machine in the cell, because enzymes are coded in DNA, transcribed, and translated, while they also produce the building blocks of DNA and RNA and polymerize them. However, metabolism is much more versatile than information processing. Metabolic networks include multiple redundant paths, and in different species, different routes can lead to the same functional outcome. Because metabolism is far more variable across lineages than the information processing machinery, the genes coding for enzymes are not universal across genomes and are much more prone to undergo LGT than information processing genes are29. This circumstance has impaired the use of metabolic enzymes for the study of early prokaryotic evolution.
Metabolic networks and metabolic enzymes unquestionably bear witness to the evolutionary process, but methods to harness their evolutionary information are so far lacking. Here we take a simple but effective approach at inferring the metabolism of LBCA, by focusing on anaerobic genomes and genes that are widely distributed among them. We reconstruct the core metabolic network of LBCA independent of any single backbone phylogenetic tree30 for the lineages in question. In doing so, we harness the information in thousands of individual trees for gene families of anaerobic prokaryotes, analyze converging signals, and point to the modern groups most similar, in terms of metabolism, to the groups that diverged earliest from LBCA.
Metabolic interconversions encoded by 146 LBCA genes plus 9 genes present in fewer groups are shown in a bipartite graph, with 243 metabolites (circular nodes) and 130 reactions (diamond nodes). Black circles represent the 57 universal target metabolites and gray circles represent the remaining metabolites. Note, however, that some of these are also universal (e.g., NADH), but directly connected to the chosen targets (e.g., in that case NAD+). Node sizes increase according to node degree. Diamonds (reactions) are colored according to the presence of genes encoding for those reactions in different taxonomic groups: in black, reactions present in all taxa; in a gradient from purple to orange reactions added during network expansion and distributed in fewer taxa (target compounds are highlighted with the same outline color if they were introduced with network expansion). Transparent colored ellipses highlight the core of energy (red) hydride transfer (blue) and carbon (yellow) metabolism.
Analysis of 131 rooted trees of genes universally present in bacterial anaerobic taxa spanning major functional categories (sorted horizontally according to curated classifications shown on top; order as in Supplementary Data 3). Illustrative trees on the side portray the metric used in each analysis and identify the group at the root in each with yellow nodes. a Root-to-tip distance measured as node depth (normalized by the largest distance in each tree). b Root-to-tip distance measured as branch length (normalized by the largest distance in each tree).
Prokaryotic gene trees differ from the species tree due both to random phylogenetic errors and to the cumulative impact of LGT62. In the absence of LGT, gene lineages branch together (monophyletic) and the phylogenetic diversity of sister clades reflects the time since their origin, with older lineages having higher sister diversity. In the context of gene evolution with LGT, gene lineages branch into multiple clades, with the number of clades increasing with gene transfer prevalence. Because LGT is a continuous phenomenon in prokaryotic evolution, the taxonomic labels of sister lineages change dynamically, but their phylogenetic diversity gives us the means to infer the relative timing for the origin of lineages. To integrate the information of sister relation from all gene trees spanning the 25 bacterial groups, we scored the phylogenetic diversity for sister clades of each group in the individual trees permitting as many inter-group LGT as necessary in the trees (5402 trees with at least six groups, Fig. 3 and Supplementary Data 5). The analyses show Clostridia as the group with the highest sister clade diversity, measured as the maximum number of phyla in a sister clade (on average five), followed by a tie between Deltaproteobacteria, Bacilli, Actinobacteria, and Spirochaetes all with three distinct groups on average present in sister clades. The result stands when looking at the 131 universal trees only, where Clostridia has on average nine distinct sister groups, followed by Actinobacteria with seven and Deltaproteobacteria with five (Supplementary Data 6). Maximum-likelihood ancestral state reconstructions using 131 universal trees indicate that LBCA was a rod-shaped cell (Supplementary Fig. 4) and reconstructs Clostridia as the most ancestral lineage (Supplementary Fig. 5) in agreement with the previous analyses.
Sister diversity (maximum number of different groups in the sister clade) for each group (rows) for 5402 trees with at least six groups (columns). An illustrative tree portrays the question asked in the analyses, where the yellow group is the one with the highest sister diversity score and therefore inferred as most ancestral.
c80f0f1006