Evolutionary mitochondrial biology in titisee

In the spring of 2017, we had the privilege of organizing and chairing the 115th International Titisee Conference (ITC). Since 1962, these invitation-only conferences, sponsored by the Boehringer Ingelheim Fonds, have been held at Lake Titisee, in the Black Forest of southern Germany (https://www.bifonds.de/titisee-conferences/about-itcs.html). The ITCs have explored a wide variety of themes, typically with the goal of bringing together researchers from different disciplines who might not normally have an opportunity to meet together. The presentation, in an intimate setting, of cutting-edge research by both senior and more junior scientists is aimed at stimulating discussion that will lead to new concepts and new approaches, as well as forging new collaborations. The theme of the 115th ITC was Evolutionary mitochondrial biology: molecular, biochemical, and metabolic diversity (https://www.bifonds.de/titisee-conferences/past-conferences/past-conference/items/id-115th-evolutionary-mitochondrial-biology-molecular-biochemical-and-metabolic-diversity.html). The conference featured investigators studying mitochondria from a phylogenetically broad range of eukaryotes, with the goal of exploring the mechanistic basis and physiological consequences of their diversity. The evident enthusiasm of participants and success of the conference encouraged us to suggest a special issue of IUBMB Life that would focus on the theme of the 115th ITC and capture some of the work and ideas discussed. Senior investigators presenting at the conference were invited to submit Critical Reviews or original Research Communications, all peer-reviewed, based on their ITC presentation. We are gratified by the response, and present herein a sampling of the work discussed at or emanating from the 115th ITC. As the cover illustration emphasizes, mitochondria come in a wide variety of “flavors,” from conventional aerobic organelles, through anaerobic mitochondria using terminal electron acceptors other than oxygen, to mitochondrion-related organelles (MROs) that dispense with part or all of the mitochondrial genome, the “missing” portions including genes that encode key parts of the system that generates ATP through coupled electron transport/oxidative phosphorylation (OXPHOS). Depending on the degree of reduction in mitochondrial function, particular MROs may generate ATP via substrate-level phosphorylation (as in the hydrogenosome, the first such MRO to be discovered), or they may have completely lost the ability to form ATP (e.g., mitosome). A recent review by Roger et al. 1 provides a comprehensive treatment of the mitochondrion in its various functional and evolutionary forms. Mitochondria were already present in the last eukaryotic common ancestor 2 and, over 1–2 billion years, have undergone tremendous lineage-specific evolution, today exhibiting remarkable molecular, biochemical, and metabolic diversity. While animal mitochondria contain a high-copy circular genome, the mitochondrial genomes of other eukaryotes display a wide array of physical forms, gene arrangements and modes of expression 3. Mitochondria can exhibit variant configurations of the respiratory chain, branching with alternative electron donors or acceptors, all designed to preserve bioenergetics, redox balance, and promote survival under diverse environmental conditions. Examples of this diversity in its various forms appear throughout this special issue. The origin of mitochondria has been the subject of intense interest and spirited debate ever since Lynn Margulis revived the moribund endosymbiont hypothesis in the late 1960s 4. The discovery of DNA in mitochondria and the characterization and comparison of the genes encoded in that genome not only solidified the view that the mitochondrial genome is of bacterial origin, but also pointed clearly to a specific lineage, α-Proteobacteria, as the closest extant relatives of mitochondria 5. Although an α-proteobacterial origin of mitochondria is widely accepted, a complicating factor is that only a small percentage of the mitochondrial proteome (the set of proteins making up the functional mitochondrion) clearly traces to α-Proteobacteria. In fact, the mitochondrial proteome is a mosaic of proteins having differing phylogenetic affinities and seemingly of differing evolutionary origins 6. How to reconcile the various conflicting features of the accumulated data is a challenge for proponents of the widely different models that aim to describe the biological process(es) underpinning the initial acquisition of an α-proteobacterial endosymbiont by a host cell, and its subsequent conversion to a key eukaryotic organelle 7, 8. In the first article in this issue, Gabaldón 9 addresses two specific issues relating to the acquisition of mitochondria within the overall evolution of the eukaryotic cell (eukaryogenesis). The first issue relates to the timing of the acquisition: whether mitochondria appeared early in eukaryogenesis, their acquisition in fact playing a pivotal role in eukaryotic cell evolution, as proposed in mitochondria-early models; or, whether mitochondria were a later addition to a host cell that had already acquired many of the hallmarks of a eukaryotic cell, in particular endocytosis, as mitochondria-late models argue. The second issue has to do with the nature of the host cell: whether a simple prokaryote, as envisioned in mitochondria-early models, or an already complex protoeukaryote, as in mitochondria-late models. Gabaldón reviews the evidence favoring a late acquisition of mitochondria, and proposes that a series of cryptic endosymbioses, predating the mitochondrial endosymbiosis, may have seeded the host with proteins from a variety of phylogenetically disparate sources, with some of these proteins subsequently used to populate the mitochondrial proteome. Two articles in this issue graphically illustrate the exceptional variability in structure and expression that characterizes mitochondrial DNA (mtDNA) across the diversity of eukaryotes. Burger and Valach 10 review the extraordinary organization of the mitochondrial genome in diplonemids, which together with euglenids and kinetoplastids comprise Euglenozoa. This clade, which exhibits some of the most unusual patterns of mitochondrial genome organization and expression uncovered to date 11, is a member of a eukaryotic supergroup that also contains the most gene-rich, bacteria-like and least derived mitochondrial genomes yet described 12. In diplonemid mitochondria, protein-coding genes are fragmented into a number of subgenic modules, encoded by separate DNA molecules. Module transcripts are trans-spliced to form translatable mRNAs, a process that also involves a limited amount of 3′-U addition and, for some transcripts, substantial C-to-U and A-to-I editing. The mechanisms by which this complicated mitochondrial mRNA processing occurs, and the enzymatic machinery mediating it, remain to be elucidated 10. In kinetoplastids, mitochondrial genome organization is quite different than in diplonemids, with a low-copy maxicircle DNA encoding genes whose transcripts typically require internal U addition/deletion editing. Editing sites are defined by complementary base pairing with small guide RNAs, encoded by high-copy minicircle DNAs, concatenated with one another and with maxicircle DNA into a massive three-dimensional complex of kinetoplast DNA 13. However, as Lukeš et al. 14 demonstrate here, despite important differences in the structure and expression of their mitochondrial genomes, diplonemids, and kinetoplastids are alike in that mtDNA content is highly inflated relative to the respective nuclear genomes. The authors discuss the cause(s) and consequences of this bloating, suggesting that a ratchet-like neutral evolutionary model leads to mtDNA hyperinflation. The articles by Burger and Valach 10 and Lukeš et al. 14, among numerous other reports in the literature, serve to emphasize the incredible variety of ways, some quite bizarre, in which mitochondrial genes can be encoded and expressed. These observations give the impression that the mitochondrion is a molecular biological playground, able to find very different solutions to the problem of organellar genetic information flow in different eukaryotes. Why this should be so is unclear, but perhaps relaxed evolutionary constraints, related to the relatively small number of genes encoded in the mitochondrial genome, underlies this versatility. It is the case, generally, that the smaller the number of genes encoded by a particular mitochondrial system, the more likely that system is to depart from the ancestral, bacteria-like pattern of gene organization and expression. To support mitochondrial translation, a full complement of mitochondrial tRNAs is required. Some mitochondrial genomes (e.g., those of most mammals) encode a set of tRNAs just sufficient to support mitochondrial protein synthesis. In other cases, fewer tRNAs than necessary are encoded by mtDNA, the remainder being nucleus-encoded tRNAs that are imported into the organelle. In still other cases (e.g., kinetoplastids), the organelle genome encodes no tRNAs whatsoever: all are imported from the cytosol to function in mitochondrial protein synthesis. Imported tRNAs invariably contain the post-transcriptional modifications characteristic of cytosolic tRNAs. Here, Paris and Alfonzo 15 review how such modifications may affect mitochondrial function, arguing that mitochondrial evolution must have been influenced by the combination of tRNA import and compartmentalization of tRNA modification enzymes. In 1975, John and Whatley 16 demonstrated that Paracoccus denitrificans (an α-proteobacterium) “resembles a mitochondrion more closely than do other bacteria, in that it effectively assembles in a single organism those features of the mitochondrial respiratory chain and OXPHOS which are otherwise distributed at random among most other aerobic bacteria.” Ten years later, Yang et al. 17 provided the first molecular evidence specifically linking a mtDNA-encoded gene with its α-proteobacterial counterpart. In this issue, Ferguson 18 reviews similarities and differences between bacterial and mitochondrial respiratory systems (with an emphasis on P. denitrificans), noting the loss of particular bacterial respiratory enzymes in the course of mitochondrial evolution. Assembly of c-type cytochromes is discussed as an example of an α-proteobacterial system that has been retained in the mitochondria of some eukaryotes but lost and replaced by a different system in others. Mitochondrial carrier (MC) proteins mediate the transport and exchange of a variety of ions and small metabolites across the inner mitochondrial membrane. MC proteins, which comprise a large family of structurally and evolutionarily related members, appear to have arisen specifically within eukaryotes via repeated duplication of a single ancestral sequence 19. In this issue, Harborne and Kunji 20 discuss an unusual member of the MC family, the ATP-Mg/Pi carrier (APC), responsible for the calcium-dependent regulation of net adenosine nucleotide concentrations in the mitochondrial matrix. This structure of this carrier is unusual in that it consists of three separate, unrelated domains, one of which is a calmodulin-like regulatory domain containing EF-hand calcium-binding motifs. The authors review the identification, structure and physiological role of the APC, commenting on the evolution of this interesting protein. The advent of the CRISPR-Cas9 DNA editing system has generated much excitement and stimulated much discussion and experimentation, to date focused almost exclusively on the eukaryotic nuclear genome as target. Here, Loutre et al. 21 discuss the prospects of adapting this system for the modification of mtDNA. Their mitochondrial CRISPRization system makes use of mitochondrion-targeted versions of the Cas9 protein plus a set of small guide RNAs, targeted to mitochondria through a pathway that has been shown previously to import small RNAs into the organelle. The authors present results that provide encouragement that this approach might be possible, while at the same time cautioning that the application of the system to mtDNA is evidently more complicated than in the case of nuclear DNA. In all animals, mitochondrial OXPHOS consists of dozens of proteins that are encoded by both mtDNA and the nuclear genome. The expression and assembly of proteins originating from these two genomes must be coordinated in space and time to ensure robust cellular bioenergetics and whole organism homeostasis. Rand et al. 22 explore such “mitonuclear” genetic interactions by employing Drosophila models in which it is possible to maintain a fixed nuclear genetic background while systematically varying the mtDNA genotype. By using development time and global gene expression as sensitive phenotypes, they are able to demonstrate just how pervasive mitonuclear genetic interactions can be, and how these genetic interactions then interact in turn with environmental variables, notably oxygen. An important result from their reanalysis of a previously reported dataset is that there tends to be greater than expected overlap in the transcripts whose abundance is influenced by mitonuclear genetic interactions and gene–environment interactions. Continuing on this broad theme of mitonuclear interactions, van Esveld and Huynen 23 explore the question of whether evolution of mtDNA gene expression drives the evolution of the nucleus-encoded mitochondrial proteome. They focus on the organization of the mtDNA in animals, which can show variation in gene order, organization of RNA genes, or GC content. They identify nuclear proteins that appear to have arisen in evolution coincident with variation in these mtDNA properties, raising new mechanistic hypotheses about the function of these proteins. Muthye and Lavrov 24 use computational analysis to explore the mitochondrial proteome of nonbilaterian animals. To date, most proteomic investigations of animal mitochondria have been focused on mammals 25, and it is generally assumed that mitochondrial proteomes across animals are uniform. However, as the authors point out, mammals represent only one phylum of five major lineages within animals. The authors use computational analysis (sequence similarity, mitochondrial targeting sequence prediction) of genomes from 23 animal species representing all four nonbilaterian phyla to predict the mitochondrial proteome from these organisms. They report 513 orthologous groups of proteins common across all animal lineages, and discuss the categories of proteins common to all lineages, and those that are lineage-specific, of course emphasizing that these remain computational predictions. The Titissee conference also explored how variation in mitochondria can underlie physiological adaptations that enable animals to cope with extreme environments. Hand et al. 26 discuss mechanisms employed by Artemia franciscana (brine shrimp) as they enter diapause—a programmed form of “metabolic arrest” that can protect against anhydrobiosis or extreme cold, permitting survival for years. The authors review mitochondrial bioenergetics during diapause, noting that oxidative metabolism drops by 99%, and also review in depth the special role of late embryogenesis abundant (LEA) proteins—which are disordered in the presence of water, but then become ordered during dessication—to protect lipid membranes during anhydrobiosis. Finally Moussa-Hadj et al. 27 explore the transcriptional, biochemical, and signaling basis of metabolic rate depression in mammalian hibernation, highlighting important gaps in our knowledge at this time. Throughout the 115th ITC meeting, we were continually awed by the seemingly endless variety in mitochondrial molecular biology and biochemistry across the tree of life. We are grateful to the Boehringer Ingelheim Fonds, which provided us with the opportunity to convene this most stimulating conference, and hope that that this issue of IUBMB Life will spark continued inquiry in this field.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.