Nature should be the model for microbial sciences

ABSTRACT Until recently, microbiologists have relied on cultures to understand the microbial world. As a result, model organisms have been the focus of research into understanding Bacteria and Archaea at a molecular level. Diversity surveys and metagenomic sequencing have revealed that these model species are often present in low abundance in the environment; instead, there are microbial taxa that are cosmopolitan in nature. Due to the numerical dominance of these microorganisms and the size of their habitats, these lineages comprise mind-boggling population sizes upward of 1028 cells on the planet. Many of these dominant groups have cultured representatives and have been shown to be involved in mediating key processes in nature. Given their importance and the increasing need to understand changes due to climate change, we propose that members of Nitrosophaerota (Nitrosopumilus maritimus), SAR11 (Pelagibacter ubique), Hadesarchaeia, Bathyarchaeia, and others become models in the future. Abundance should not be the only measure of a good model system; there are other organisms that are well suited to advance our understanding of ecology and evolution. For example, the most well-studied symbiotic bacteria, like Buchnera, Aliivibrio, and Rhizobium, should be models for understanding host-associations. Also, there are organisms that hold new insights into major transitions in the evolution of life on the planet like the Asgard Archaea (Heimdallarchaeia). Innovations in a variety of in situ techniques have enabled us to circumvent culturing when studying everything from genetics to physiology. Our deepest understanding of microbiology and its impact on the planet will come from studying these microbes in nature. Laboratory-based studies must be grounded in nature, not the other way around.

H istorically, our understanding of microbes has been based on laboratory cultures.
Much of what we know at a mechanistic level is based on "model organisms" which are species that readily grow in laboratory conditions.Model organisms are defined as nonhuman species that serve as subjects in studies aimed at generating knowledge applicable to other species (1).In the realm of microbiology, the species Escherichia coli (2) has been the go-to model organism since its discovery in 1884.E. coli quickly became the bacterium of choice for laboratory training and experiments.With its rapid growth, adaptability to varying growth conditions, low pathogenicity, and ubiquity in the human microbiome, it was the obvious choice when the molecular biology revolution began in the 1950s.Studies involving E. coli played pivotal roles in describing the DNA replication machinery (3), demonstrating the stochastic nature of mutations (4), and serving as the foundation for various genetic engineering technologies, such as molecular cloning (5).It should be noted that we now have thousands of genomes belonging to this species; however, there is considerable genomic variation within these organisms (6).Thus, the concept of E. coli as a model is not well defined.
Another bacterium widely employed as a model organism is Bacillus subtilis.Like E. coli, B. subtilis has proven to be amenable to experimental manipulation and boasts a low pathogenicity profile.The prominence of B. subtilis as a model stemmed from the need to study Gram-positive bacteria and their distinctive cell wall properties, especially in a medical context.B. subtilis also offers unique advantages for the study of cell differentia tion, owing to its sporulation capability and ease of biofilm formation (7).In fact, some traits that make them the focus of study are the result of laboratory domestication (8).
In the late 1970s, the emerging technology of nucleic acid sequencing led to the discovery of a new group of methane-producing microorganisms that defied organism classification as eukaryotes or Bacteria (9).This groundbreaking revelation eventually led to the identification of these unique microorganisms now known as Archaea (10,11).It is worth noting that, even before this taxonomic reclassification, researchers had already collected archaeal cells from acidic ponds (12).Archaea had played pivotal roles in key breakthroughs in cellular biology, such as the discovery of bacteriorhodopsin (13).However, during that era, they were mistakenly grouped together with Bacteria.
Within the domain of Archaea, species from the genus Sulfolobus within the TACK superphylum have emerged as the model organisms of choice (14).Moreover, genetic analyses often involve alternatives from the Euryarchaeota phylum, including Haloferax volcanii, Methanosarcina acetivorans, Methanococcus maripaludis, and Thermococcus gammatolerans, each with its own unique strengths and limitations (15).Unlike bacterial model organisms, the success of Sulfolobus as a model organism can be attributed to its rich array of mobile genetic elements, such as insertion sequences, viruses, and plasmids (16).Understanding and characterizing these genetic elements are pivotal for the development of genetic tools, which are key for exploring the intricacies of archaeal cell biology.
Advances in sequencing technologies and computational approaches in the last 20 years have enabled microbiologists to obtain genomes directly from nature.This has revealed a vast diversity and community structure of dominant lineages previously overlooked and uncultured.Simultaneously, it revealed that the model organisms traditionally employed do not hold a central role in natural ecosystems.Take, for instance, E. coli, which is far from being a dominant species in water or soil samples and is often used as a marker for anthropogenic contamination in such environments.
This finding has underscored the pressing need for the development of new model organisms that better represent the key players in our ecosystems.
Here, we argue that it is important to focus on the dominant microbes on our planet and how they persist in their natural habitats.Classically, the concept of a model organism relied on an individual species.The concept of a species in microbiology is controversial, and often in nature, populations of species or broad groups (e.g., a phylum) have strong environmental importance.In some cases, the model taxa we describe here have no culture representatives, or there are stable laboratory cultures that rely on symbiotic partners.Thus, it is important to remember that a model for understanding the biology of life on the planet can consist of an individual cultured isolate, mixtures of distinct lineages, or a broad taxonomic group.There is no need to limit studies of models to one of these; ideally, one can study cultures and link those techniques to examine natural populations.

KEY MICROBES IN EARTH ECOSYSTEMS
From the subsurface to the atmosphere, our understanding of the composition of microbial communities has advanced considerably with the advent of diversity surveys using both DNA sequencing and microscopy.When considering ideal targets for model organisms on our planet, we think it is important to choose Bacteria and Archaea that are ubiquitous, dominant members of the community, and are involved in key biogeochem ical processes of that habitat.For example, when we consider the surface oceans, we should focus on groups capable of photosynthesis and carbon cycling.
Despite localized variations, it has become clear that the surface oceans are dominated by Bacteria, and Archaea dominate the deep oceans (Fig. 1).These organisms have streamlined cell volumes and genomes that are adapted to survive in low-nutrient open oceans.The upper waters consist primarily of a group referred to as "SAR11" (17,18), cyanobacteria belonging to genera Prochlorococcus (19) and Synechococcus (20) (Fig. 1).Considering the volume of the oceans, it is easy to see these lineages are among the most dominant on the planet.SAR11 has a global estimated population size of 2.4 × 10 28 cells and comprises ~25% of the cells in the ocean (21) Table 1.Cultured representatives of all these bacteria exist; however, SAR11, named Pelagibacter spp., is not easily cultivated due to their slow growth and other factors including the fact that commonly used plastics for laboratory culturing are toxic to them (22).
Strikingly, archaea constitute the majority (up to 69%) of the cells in the deep sea, with members of the Thaumarchaeota (now named the Nitrosophaerota phylum) dominating with up to 39% of the cells in the deep sea (32,33).The first member of Thaumarchaeota cultured, N. maritimus, confirmed that they are capable of ammonia oxidation (23).Ammonia oxidation is coupled to nitrite oxidation to produce nitrate, a key nutrient for primary productivity.Isotopic analyses have demonstrated that these archaea may be responsible for 30% of the nitrous oxide in the world's atmosphere (34).In freshwater lakes, there is also a clear dominant group of Bacteria within the Actinobacteria, group acI (35).Like other bacterioplankton (bacteria that inhabit the water column), the acI are relatively challenging to culture and require high throughput approaches (26,27).Remarkably, these bacteria are dominant in lakes around the world, and their ability to harness light via actinorhodopsins and organic carbon utilization makes them important players in terms of climate change (36).
Soils and marine sediments are where things become less clear in terms of a single dominant group.These environments tend to be more heterogeneous and certainly considerably more diverse (37).However, there are certain phyla that are ubiquitous such as Bathyarchaeia on the ocean floor and Proteobacteria in soils (38).However, the Proteobacteria is one of the most diverse groups in the Bacteria, and it is not clear that there is one specific lineage that stands out in distribution and abundance.Due to the diversity of soils, more work needs to be done to determine which taxa fit this billing.One could agree this is true in the subsurface as well; however, there are certain taxa that are ubiquitous and often dominate key niches.For example, Bathyarchaeia could be considered an ideal model group for sediments as it is known to be a key player in the recycling of detrital organic matter from the overlying water column (39).Recently, the first pure culture of Bathyarchaeia was obtained from anaerobic lake sediment, Bathyarchaeum tardum, which supported dependency on detrital organic matter, specifically oligomers (24).
The deep subsurface is a dynamic and heterogeneous environment that starts a few meters below the surface.One ubiquitous archaeal group that can thrive in a broad range of anoxic subsurface environments (4°C-80°C) is the phylum Hadesarchaeota (25).These archaea are uniquely adapted to utilize substrates generated via subsur face geological processes, carbon monoxide, and hydrocarbons (40).In contrast to the deep subsurface, the atmosphere also contains microbial communities, though these communities are likely far less metabolically active (41).Bacteria found in the atmos phere are likely involved in hydrogen, carbon (specifically methane, carbon dioxide, and carbon monoxide), and nitrogen cycling (42,43).Recent studies have shown that the dominant microbes present in the atmospheric boundary layer are Alphaproteobacteria and Gammaproteobacteria (42,43).

MODEL MICROBES IN HOST-ASSOCIATION INTERACTIONS
Interactions between microbes and eukaryotes, as well as other prokaryotes, have shaped the ecology and evolution of the planet (44).The diversity of host associations in nature makes it challenging to define good model systems to advance our understand ing of the symbiotic processes.Furthermore, complicating this multiple microbial species can inhabit hosts, like Wolbachia and insects (45).We now know that huge swaths of microbial diversity (CPR [candidate phyla radiation] and DPANN archaea) consist of small cells that likely require another species to survive (46,47).However, the host range of most of these lineages has not been determined.There is one member of the Saccharibacteria (formerly TM7) that has been shown to be a ubiquitous human oral episymbiont.A strain belonging to this group has been cultured Nanosynbacter lyticus, making interaction experiments more feasible (48,49).As more of these enigmatic small organisms are studied, our understanding of this vast diversity will reveal new mechanisms of symbiotic interactions.Some symbiotic bacteria have been extensively studied, and we suggest would be models for different types of interactions.Notably, Aliivibrio fischeri, which interacts with squid and fish (28), and Buchnera, the obligate endosymbionts of aphids (29,30).These interactions are fundamentally different, and Buchnera is vertically inherited and is mostly beneficial to the host, while A. fischeri colonizes the host at each generation.That makes the mechanisms distinct, and both should be models.Perhaps the most well studied of all interactions is between Rhizobium and plant roots, which makes this another potential system on which decades of studies can be founded (31).

UNDERSTANDING KEY TRANSITION IN THE ORIGINS OF COMPLEX LIFE
Not only should prominence in nature be a metric for a model organism but also significance in the evolution of life on the planet.For example, the origins of organelles like mitochondria from Alphaproteobacteria (50) and chloroplasts from cyanobacteria via endosymbiosis.Certainly, one of the major transitions of life was the origin of complex cellular organization in eukaryotes.The past decade has revealed that a group of Archaea named Asgards (now phylum Asgardarchaeota) were pivotal in this event (51).The first to be described was Lokiarchaeia, then soon after several others were identified, including Odinarchaeia, Thorarchaeia, and Heimdallarchaeia (52).Two cultures of Lokiarchaeia representatives from this clade have been obtained (53,54); however, these are difficult to maintain as well.Not only are these Archaea related to eukaryotes on the tree of life, but they also have a wide variety of proteins that had previously been thought to be exclusive to eukaryotes.These building blocks of complex life from uncultured Asgards have been heterologically expressed, and their functions have been determined (55)(56)(57).These novel proteins from these uncultured lineages have been functionally characterized at the structural level using crystallization (58).

CULTURE-INDEPENDENT TECHNIQUES TO UNDERSTAND MODEL ENVIRON MENTAL ORGANISMS
Classically, model organisms are easy to maintain in cultures, readily manipulated (via genetics), cryopreserved, and broadly available to everyone.This is not the case for many of the organisms discussed above and other dominant lineages.While we now have cultured representatives of them, they are difficult to grow in the lab, and they all lack established genetic systems.Regardless, having cultures of these globally impor tant organisms is extremely valuable.Much has been learned from cultures of Prome theoarchaeum syntrophicum (Lokiarchaeia), Nitrosopulmilus spp.(Thaumarchaeota), and Pelagibacter spp.(SAR11-Proteobacteria).For example, high-resolution images have been generated of the N. maritimus surface layer (S-layer) using electron cryotomog raphy and single-particle cryomicroscopy revealing that they can sieve/concentrate ammonium in oligotrophic waters (59).This combined with these archaea's remarkable versatility to utilize a variety of nitrogen substrates is what makes them predominant players in ocean nutrient cycling (60,61).This reminds us that we must always have an environmental perspective of what we discover.Studies on physiology, biochemistry, regulation, signaling, interactions, pathogenesis, evolution, and other aspects of an organism in isolation in a laboratory are going to differ from nature.
The distinction between culture-based and culture-independent is not black and white.It is common to run experiments using inoculum from the environment to study mixed communities, e.g., mesocosms.All the approaches discussed in this section used to study microbes in situ can be used to examine mesocosms.One common technique is to add isotopically labeled substrates to track the uptake and flow of carbon and nutrients by extracting the heavy DNA, referred to as DNA-SIP (stable isotope probing).For example, this has been used to resolve the community structure and metabolic pathways involved in organic carbon cycling in soils (62) and oil degradation in the Gulf of Mexico (37).
The most common approach to studying uncultivated microorganisms is metage nomics.Despite the large number of discoveries that have resulted from the reconstruc tion of metagenomic-assembled genomes (MAGs), there is often skepticism regarding the accuracy of these as true representations of a genome from a cell in nature.This issue is largely due to their reconstruction from mixed communities.They can contain assembly errors, and reads from distinct strains can be mixed into an individual MAG assembly, unless they are carefully curated (63,64).Also, metagenomic binning can lead to contigs from different species being incorporated into MAGs, usually shorter fragments <20 kb in length.This is not a new concern, and similar discussions occurred with the sequencing of individual 16S rRNA genes decades ago (65).The analyses of MAGs in environmental microbiology are now the norm.Despite this, there has not been a major misrepresentation of communities using this approach.One reason for this is that we rarely obtain 1-2 closely related MAGs; instead, we are analyzing dozens that have a similar genetic composition.This results in a statistically reliable set of genes within a clade.
It is much more powerful to have data from the community rather than a single species that is often not a dominant member of that community.For example, we do not see the presence of finescale genetic heterogeneity in MAGs as being an issue.If handled correctly, this can make it possible to track strain-level shifts in populations over time (64).Similar to what was done in the long-term evolution experiment, but in nature (66).Moreover, metagenomics circumvents the reliance on using cultivation to infer the biology of microbes in nature.The microbiology community acknowledges metagenom ics as a powerful resource to study uncultivated organisms.This is particularly true given the large amount of data generated, decreasing costs, wide accessibility, and numer ous experimental validations of predictions generated through metagenomic analyses (67)(68)(69)(70).Understanding the existing limitations of culture-independent approaches are important and should not drive us away from them but allow us to utilize them more effectively and responsibly.
In the past, we have lacked techniques to understand cell biology and activity in situ or even know what organisms are present in nature.The advances in metagenomics, transcriptomics, proteomics, microscopy, and geochemistry techniques have enabled us to examine individual cells to entire communities independent of culturing.We can now obtain complete genomes (71), track metabolic activity (72,73), image biomolecules (46,74), express genes, confirm function (57), and examine physiologies of uncultured cells (37).
There is an ever-growing toolbox for studying microbes in nature.One powerful example is the use of Raman spectroscopy which measures the inelastic scatter ing of light to provide detailed information about the molecular composition and metabolic activities of individual cells.This technique enables researchers to analyze microbial cells directly in their natural environments, identifying specific biomole cules, tracking metabolic changes, and monitoring interactions within communities (75).This technique can be combined with other culture-independent approaches to enhance our understanding of the microbial diversity, ecological roles, and func tional dynamics of uncultivated microorganisms across various environments.These complementary technologies include nano-scale secondary ion mass spectrometry (nanoSIMS), fluorescence in situ hybridization (FISH), cell sorting, and genomic sequencing.
The combination of Raman spectroscopy and nanoSIMS allows us to map the distribution of specific elements within individual cells and semi-quantify microbial metabolic activities (76).When combined with FISH, Raman can identify target microbes by linking spectroscopy with taxonomic information (77).Raman-activated cell sorting enables the collection of target microbes from environmental samples based on specific biological activity of interest (78).

SUMMARY
Advancements in technology now allow us to study microbes beyond pure cultures.Culture-independent techniques play a crucial role in achieving this goal.When combined, these techniques provide valuable information about the metabolism and physicochemical conditions necessary for the survival of uncultivated organisms.Thus, we hope that there can be a shift from studying classic organisms in culture to more studies looking at the ecologically dominant microbes in the natural habitats to understand them at a molecular level and their impact on the ecosystems.

TABLE 1
Dominant microbes across key ecosystems and cultured representatives proposed here as potential model organisms in the future, given their significant roles in their given ecosystem