https://orcid.org/0000-0002-3097-5541
Abstract
The origin of life remains one of the greatest enigmas in science. The immense leap in complexity between prebiotic soup and cellular life challenges historically “chemistry-forward” and “biology-backwards” approaches. Evolution must have bridged this gap in complexity, so understanding factors that influence evolutionary outcomes is critical for exploring life’s emergence. Here, we review insights from ecology and evolution and their application throughout abiogenesis. In particular, we discuss how ecological and evolutionary constraints shape the evolution of biological innovation. We propose an “eco-evolutionary” approach, which is agnostic towards particular chemistries or environments and instead explores the several ways that an evolvable system may emerge and gain complexity.
“But if (and oh what a big if) we could conceive in some warm little pond . . .” wrote Charles Darwin in 1871. In this pond, he imagined, life could chemically form, “ready to undergo still more complex changes [ . . . ]” (Darwin & Hooker, 1871).
Today, the origin of life remains one of the biggest open questions in biology. How did life emerge and gain complexity? For over a century, evolutionary biologists have explored the history of life on Earth, using comparative methods to disentangle the origin of novel traits, innovation, and major transitions in complexity (Szathmáry & Smith, 1995; Weiss et al., 2016). The origin of life itself, however, has eluded this comparative approach, paving the way for chemists to characterize Darwin’s “warm little pond” and build a repertoire of prebiotic processes.
As a result, most investigations have focused on life’s initial chemistry rather than its biology. These “chemistry-forward” approaches have provided many insights into the suite of organic compounds present on early Earth and the chemistries likely to facilitate life-like processes, such as energy acquisition (Sousa et al., 2013) and autocatalysis (Kauffman, 1986). However, the primary challenge of this work has been to explain the immense leap in molecular complexity between the first steps in life’s emergence and the Last Universal Common Ancestor (LUCA). How does chemistry transition to biology? To fill this gap, “biology-backward” approaches have used phylogenetic methods to explore simpler precursors to the processes and structures found in extant life (Becerra et al., 2014). However, there is little expectation that even these precursors could arise spontaneously without a prior adaptive process. Further work, therefore, has recognized the role of evolvability in passing through a series of “mesobiotic” entities toward LUCA (Baum, 2015; Shenhav et al., 2003), although it remains unlikely that “chemistry-forward” and “biology-backward” approaches will someday meet in the middle. In this paper, we consider an “eco-evolutionary” approach, where eco-evolutionary processes are considered at the outset, and insights from biological disciplines guide our exploration of life’s origins.
Insights from alternative programs
Chemistry-forward approach
While the transition from nonliving matter into life is yet unknown, progress has been made towards a solution. Seventy years ago, the Miller-Urey spark-discharge experiment synthesized amino acids under what was considered a model atmosphere for primitive Earth (Miller, 1953). This landmark experiment ushered in a new era of experimental studies on prebiotic chemistry, where researchers developed an inventory of organic compounds thought to be necessary for life (McCollom, 2013). The synthesis of life’s building blocks, however, does not elucidate a pathway toward even the simplest cellular processes. Hence, another “chemistry-forward” approach emerged through the study of macromolecular self-organization and autocatalysis, the process in which reaction products are also catalysts in the same or coupled reactions (Hordijk et al., 2012; Kauffman, 1993). In particular, autocatalytic sets demonstrate how self-sustaining, evolvable systems may emerge spontaneously in the environment—a critical step toward the evolution of greater complexity (Hordijk & Steel, 2014). Yet, a vast gulf remains between “chemistry-forward” approaches and derived biological processes.
Biology-backward approach
Modern biology can tell us a great deal about LUCA and its closest predecessors. By exploring precursors to cellular life, “biology-backward” approaches have constructed minimal cells (Hutchison et al., 2016), protocells (Adamala & Szostak, 2013), and inferred the evolution of cellular processes like protein catalysis and genetic transmission (Guseva et al., 2017). However, anticipating precise evolutionary trajectories before genetic transmission challenges the scope of phylogenetic methods (Becerra et al., 2014). To add to this challenge, our ability to predict evolution even in extant life is severely limited by the roles of chance and history (Gould, 1990; Travisano et al., 1995). While studies of parallel and convergent evolution have shown that similar phenotypes can evolve in response to similar environmental challenges (Losos et al., 1998), they may take substantially different paths toward that outcome. In the Long-Term Evolution Experiment (LTEE), for example, 12 separate populations of E. coli were founded from a single clone, evolving in controlled conditions for over 65,000 generations (Lenski et al., 1991). While the derived populations resembled one another in several ways (e.g., each population evolving greater fitness, growth rate, and cell size), the populations also diverged considerably, with each population accumulating a unique set of mutations and achieving different degrees of fitness under other conditions (Blount et al., 2018). Parallel replay experiments such as this demonstrate the sensitivity of outcomes to chance events over evolutionary history, known as “historical contingencies.” If evolutionary outcomes are contingent upon idiosyncratic events—rendering them unpredictable even in highly simplified and controlled settings where initial conditions are precisely known—how can we hope to predict evolutionary trajectories that occurred 4 billion years in the past? How can we peer behind the phylogenetic curtain and determine the path life took prior to genetic transmission? What is more, many hypotheses on the origins of life are nonfalsifiable, leading investigations down separate, inconclusive paths. Thus, we need an approach to investigating the origins of life that is not wedded to resolving each step precellular life took on primordial Earth and which incorporates insights on processes we know were present.
What’s missing?
Consistent with each proposed scenario for the origins of life is the relevance of evolutionary processes as they played out in an ecological theater. Studies of eco-evolutionary dynamics have explored the generation and maintenance of biological diversity in a range of contemporary contexts, incorporating theory on environmental heterogeneity, organism function, and niche construction (Fussmann et al., 2007). In particular, the reciprocal interaction between the environment and evolving populations—known as eco-evolutionary feedbacks (EEFs) – sits at the core of ecology and evolutionary biology today (Post & Palkovacs, 2009). At the origins of life, eco-evolutionary feedbacks were undoubtedly central in driving prebiotic complexity, yet they are rarely labeled as such in the literature. For example, vast amounts of work have explored how vesicles provide mechanisms for localization and microenvironments favorable for prebiotic reaction networks (Toparlak & Mansy, 2019) or how encapsulation may provide stability necessary for information transmission (Peng et al., 2022). Other work includes theory on environmental alteration without encapsulation (e.g., localization on a mineral surface) that drives subsequent selection (e.g., colonizing the surface) (Baum, 2018; Wächtershäuser, 1988). Despite the ubiquity of eco-evolutionary dynamics in a prebiotic world, ecologists and evolutionists have largely remained silent on the subject. Since eco-evolutionary theory may be adapted to a variety of prebiotic contexts, allowing for the exploration of multiple paths toward life, biological approaches can greatly inform our understanding of precellular evolution (Box 1). Indeed, with evolvability as a requisite to life, there is a clear role for ecological and evolutionary perspectives in exploring life’s emergence.
Here we briefly discuss definitions of life and prebiotic evolution as used in this paper. Defining life has been subject to tremendous debate. For practical purposes, we employ a working definition lately used by NASA: “a self-sustaining system capable of Darwinian evolution” (Joyce et al., 1994). Our consideration of life, in the context of life’s origin, is chemically agnostic and in general pertains to a variety of systems that could exist prior to cellular life as we know it. Our use of the term “pre-cellular” refers to systems that existed before—or are simpler than—LUCA (e.g., bacterial or archaeal life) (Becerra et al., 2014).
Our requirements for “Darwinian evolution” are as described by Lewontin (1970) (Lewontin, 1970): phenotypic variation, fitness differences, and heritability of fitness. This scheme of evolution does not assume a particular mechanism of inheritance, so we apply it generally throughout the emergence of life. “Darwinian evolution,” also known as adaptive evolution, results specifically from the process of selection, but we assume that precellular evolutionary outcomes could also be subject to processes governed by chance and history (e.g., drift and the stochastic generation of variation). “Evolvability” here is considered the ability to undergo adaptive and neutral evolution.
A few evolvable systems used in the literature:
The goal of this paper is to demonstrate how foundational insights from ecology and evolution can be applied throughout the process of abiogenesis—from prebiotic soup to cells. We will explore three insights in particular, using each to inform our understanding of the eco-evolutionary dynamics present at the origins of life, beginning with, (a) spatial and temporal dynamics of eco-evolutionary feedbacks, (b) ecophysiological constraints on evolvability, and (c) determining ancestral function from derived traits. We will survey literature that exemplify this eco-evo approach, concluding with how these perspectives can guide future work on the origins of life.
Eco-evolutionary feedbacks
I cannot consider the organism without its environment… (Mitchell, 1959)
How early cells interacted with their environment has been a central concern for several decades, leading to a search for environments that mimic life’s features (e.g., proton gradients; Sojo et al., 2016). On Earth today, the converse to the above quote is also true: it’s difficult to consider environments without organisms. Eco-evolutionary feedbacks have become critical to our understanding of biological systems, playing an important role in processes like the emergence of adaptive radiations (Habets et al., 2006), species coexistence (Loeuille, 2010), and the maintenance of community stability (Patel et al., 2018). Key parameters governing EEFs today can therefore inform the eco-evolutionary context when life began.
Eco-evolutionary feedbacks are local in time and space
There is accumulating evidence that ecological and evolutionary processes can occur at compatible timescales. Temporal constraints on EEFs require that the evolutionary response to ecological change is congruent with the subsequent ecological response, and so forth. Despite these constraints, EEFs are well-documented throughout natural and experimental systems, over long and short timescales (Post & Palkovacs, 2009). Even in highly simple systems—a single species in a constant environment—feedbacks can occur between changing population densities and corresponding strengths of selection (i.e., density-dependent selection) (Clarke, 1972; Travis et al., 2013). In more complex systems, EEFs occur across several levels of biological organization, from populations to communities or even ecosystems. For example, recent empirical work with three-spined sticklebacks observed ecosystem modifications caused by population-level changes in response to parasites, which then influenced the next generation of sticklebacks, highlighting the overlapping timescales of host–parasite and host–ecosystem interactions (Brunner et al., 2017). Thus, while EEFs may occur across several biological scales, temporal constraints remain local (Figure 1).
Eco-evolutionary feedback. EEFs are cyclical interactions between ecology and evolution. (1) Ecological interactions affect evolutionary change in a population’s traits, which then, (2) modify the nature of ecological interactions.
As EEFs occur locally, they are also spatially constrained. Even in a nonstructured single-species environment, where resources are uniformly available, EEFs can occur between population size and trait values. In a structured environment, where resources are localized or populations exist in patches, EEFs may occur among patches between regional population sizes and local trait values (Govaert et al., 2019). In doing so, EEFs can link eco-evolutionary processes at different spatial scales; for example, a selective increase in patch size can select against dispersal, which decreases the likelihood of recolonization (Poethke et al., 2011). Spatial selection can also act as an ecological filter, where faster individuals who arrive first benefit from reduced competition (Fronhofer & Altermatt, 2015). In each instance, the ecological processes at a regional scale among several patches (e.g., dispersal) can influence within-patch dynamics (e.g., competition), and likewise for evolutionary processes, where gene flow can reduce the effects of drift in a local population. In structured environments, or systems with discrete patches, dispersal is therefore a central trait, acting as both an ecological mechanism impacting densities and an evolutionary process mediating gene flow. Without dispersal, the evolutionary response to ecological change is more likely to be influenced by stochasticity (Gomulkiewicz & Holt, 1995). At the origins of life, where precellular life faced a heterogeneous abiotic environment, these mechanisms would have been crucial for system stability. As life began to construct its environment, EEFs would have further generated spatial variation in biotic interactions, making spatiotemporal constraints even more important for understanding evolutionary outcomes.
The origins of life as an eco-evolutionary feedback
Since life must have emerged from an abiotic environment, it is reasonable to consider the geochemical conditions favorable for the transition from nonliving to living matter. As soon as an evolvable system emerged, however, it would have interacted with its environment (and other individuals) in a way that impacted its likelihood of persistence. For example, the adsorption of self-replicating molecules to a mineral surface would have stabilized the local environment, making it more likely that further autocatalysis could occur there, similar to primary succession in ecology (Baum, 2018). It is therefore important to not only consider which prebiotic environments led rise to life, but which constructed environments enabled its persistence. Indeed, this idea appears in many hypotheses on the origins of life. The RNA World Theory, for instance, relies on the insight that RNA molecules can both undergo replication (without DNA) and catalyze their replication (without proteins), leading to a proposed system of self-replicating RNA molecules that preceded the flow of genetic information from DNA to RNA (Joyce, 2002). This scenario requires the eventual takeover of DNA as the primary means of heredity and control—an event that would have occurred in an RNA-constructed world, such as the microenvironment of a protocell (Szostak et al., 2001). In several proposed scenarios, encapsulation is required for this genetic takeover, demonstrating the interplay of niche construction and subsequent evolution before cellular life.
Another key consideration for EEFs at the origins of life is the role of dispersal as both an ecological and evolutionary mechanism, as discussed above. An excellent demonstration of these dynamics is a model proposed by David Baum and colleagues, explored using “chemical ecosystem selection” (Baum, 2015; Baum & Vetsigian, 2017; Vincent et al., 2019). In this system, an “ecosystem” of mutually-catalytic molecules colonize a mineral surface in an aqueous solution (e.g., the ocean). As they reproduce, they spread across the surface, competing with neighboring evolvers for space. Eventually, parts of the surface-bound networks could break away and be released into the water column. These propagules, like spores or seeds, can land on another surface and grow, enabling the colonization of more surfaces via dispersal. In this system, the ability to persist in the water column (i.e., migration) offers a competitive advantage over other surface evolvers and is selected for, offering a pathway toward precells where life may reproduce without adherence to a surface.
Another model demonstrating the eco-evolutionary dynamics of dispersal is the Hot Spring Hypothesis, which proposes a pathway for the synthesis of lipid-encapsulated polymers (Damer & Deamer, 2020). In this system, cycles of rehydration and dehydration are thought to occur in “chemically optimal” freshwater pools, where selection favors persistence throughout the cycles. Following evolution in these pools, life-like systems would disperse to other “more extreme” streams and marine systems, where they would evolve tolerance to this new suite of conditions. Similar to surface evolvers, selection in this system favors persistence through dilution and stability in a new environment. In other words, the system selects for colonization ability. As the ability to colonize increases, life facilitates its evolution. While proposed to occur in different environments, the commonalities between the Hot Spring Hypothesis and Baum’s surface evolvers demonstrate how eco-evolutionary processes like dispersal can generally be applied to research on the origins of life.
Large-scale EEFs have long been recognized for their role in the evolution of life, such as the Great Oxidation Event, where the rise of oxygenic photosynthesis transformed Earth’s eco-evolutionary landscape. At the origins of life, EEFs would have occurred locally. The local spatiotemporal constraints on feedbacks are therefore critical for understanding the evolutionary context of early life. Our understanding of abiogenesis should not be a linear progression of chemical systems in an unaltered environment, but instead the evolution of EEFs themselves (Figure 2). Future work should allow for the “deformation” of fitness landscapes, exploring how biology-induced environmental changes could have consequences for the fitness and coexistence of descendants, a phenomena observed throughout life in variable environments today (Lindsey et al., 2013; Ogbunugafor et al., 2016; Suvorov et al., 2023)
Hypothetical scenario for how EEFs can be conceptualized in a prebiotic system. As time passes, increased stability of life-like systems promotes the strength of the feedback. The origin of biological function can be understood through change in these cyclical interactions. (1) A diverse pool of self-replicating polymers exist in solution with a mineral surface. Polymers have varying abilities to bind to the surface. (2) Some polymers attach to the surface, which provides a concentration mechanism to help molecules react, grow, and propagate. (3) The localized reactants at the surface provides a selection pressure to bind to the surface. Better binders are more likely to persist and outcompete polymers in solution. (4) Bound polymers provide new structures, where other polymers can attach. More opportunities to bind to molecules results in the evolution of cooperation between different polymer “species.” (5) Networks of self-replicating polymers populate a mineral surface. Catalyzing the production of neighboring molecules provides new surfaces to bind to. (6) Some networks of polymers grow faster than others and spread across the surface. (7) Faster growers outcompete other networks and evolve competitive ability to grow and colonize surfaces. (8) Evolution of new networks provides opportunities for novel interactions with other (perhaps larger) compounds.
Our description of EEFs at the origins of life (Figure 2) emphasizes adaptation with the pervasive effects of drift during the evolution of complexity. Indeed, random and neutral processes like drift have been studied extensively (Kimura, 1983; Suvorov et al., 2023; Wright, 1931), and evolution cannot be understood without them (Lynch, 2007; Travisano et al., 1995). Natural selection alone is insufficient to explain the emergence of complexity, and in some cases, diversity and complexity increase via a series of nonadaptive steps (Muñoz-Gómez et al., 2021; Stoltzfus, 1999). Neither drift nor selection occur in isolation, complexity arising spontaneously as variation accumulates (Brandon & McShea, 2020) and as a result of adaptive benefits (Szathmáry & Smith, 1995). Some researchers have highlighted the contributions of random and neutral processes to the spread of complex traits, such as the origins of life, eukaryotes and multicellularity, that are not necessarily more adaptive than simpler alternatives (Lynch et al., 2022; Muñoz-Gómez et al., 2021). This body of work, which includes the role of bioenergetics and evolutionary cell biology (Lynch & Trickovic, 2020; Lynch et al., 2022; Yang et al., 2021) further clarifies the mechanisms by which neutral processes can generate complexity and their relevance for selection. This interplay underscores the need for an eco-evolutionary approach to the origins of life, where opportunities for stochastic effects are embraced alongside adaptive consequences.
Ecophysiological constraints on evolvability
Living cells have evolved the ability to take energy from their surroundings and transform it into forms usable inside a cell. Ecophysiological constraints on evolvability are important for understanding early forms of energy acquisition.
Modern metabolic systems follow a series of small leaps in energy
For all life on Earth, metabolism is the summation of incremental changes. Through a series of stepwise chemical reactions, electrons flow from donors to acceptors to generate ATP and establish a charge on the cell membrane. Cells then couple this flow of electrons to essential physiological activities, using enzymes to catalyze each reaction step (i.e., lower activation energies). This makes a relatively direct association between enzymatic activities and reproductive success, particularly for microorganisms. Because survival, stability, and fitness all depend on patterns of metabolism, microorganisms have evolved remarkable metabolic capacities that capture significant amounts of free energy from a substrate and use it to reproduce. This multi-step process of lowering activation energies through a series of intermediates is clearly derived—an enigma recognized by researchers of the origins of life. How did the first life-like systems acquire energy before the coordination of enzymatic machinery?
Early life-like systems lacked the metabolic capacity to make big steps small
The energy potential of a prebiotic environment must have been harnessed without incremental steps. This means that the activation energies needed to tap into the chemical energy harbored in prebiotic “food” must have been sufficiently low that they were surpassed in one or a few steps. Additionally, since homeostasis is derived, the environmental fluctuations and extremes present at life’s outset must not have been so high as to disrupt an individual’s integrity. This is why life today can exist in environments that would not have been hospitable for the earliest biological systems. For example, some microorganisms can survive the harsh and highly dynamic conditions of an active hydrothermal vent chimney, due to the versatility in their energy and carbon exploitation—but this capacity evolved over time (Le Bris et al., 2019). Thus, while it is tempting to look for environments with high-energy potentials to “jumpstart” life, early biological systems lacked the metabolic machinery to make these big leaps small. Instead, ancestral systems would only have had the ability to coopt small leaps in energy. The physiochemical gradients surrounding vents—or other more moderate environments—may have provided energy potentials that prebiotic metabolisms could overcome. From an eco-evolutionary perspective, the contexts favorable for early biological processes would also have provided enough stability for persistence yet the capacity to change and eventually disperse to other locations.
The canonical “chicken-and-egg” paradox at the origins of life has led to a variety of proposed mechanisms for self-replication without coordinated enzymes. “Metabolism-first” scenarios generally begin with life as a collective self-reproducing metabolism that emerges in a space of possible organic reactions. From this foundation, the ability to accurately transmit information via replication must emerge, a constraint that is difficult to overcome in high-energy environments (Vasas et al., 2010). This is because “metabolism-first” models often rely on information storage via composition of molecular assemblies, where compositional information is passed on following fission of vesicles (Segré et al., 2000, 2001). The evolvability of these “ensemble replicators” is limited by the inaccuracy of replicating compositional information such that fitter genomes are not necessarily maintained by selection (Vasas et al., 2010). Given this challenge—and the derived nature of modern metabolisms—an “eco-evolutionary” approach prioritizes evolvability as a requisite to harnessing large energy potentials (Figure 3).
Evolvability as a function of energy potential differs for ancestral and derived metabolic systems. Early life-like systems could harness small energy potentials in order to self-propagate and pass on fitness differences. As the environment’s energy potential increases, the ability of ancestral metabolisms to harness energy decreases. Modern life, however, uses enzymes to lower activation energies in a series of incremental steps, enabling adaptive evolution at high-energy potentials. Arrow indicates the difference in evolvability between early and modern life, at the point where energy potential is too high to be overcome without derived coordination of enzymatic machinery. Environments favorable for cellular life cannot be mapped onto ancestral systems.
Evolution of genetic inheritance
Modern genetic systems are incredibly good at expressing and transmitting biological information. How did template-based inheritance begin? Determining the origin of information transmission poses several challenges inherent in exploring ancestral states of complex traits.
Adaptation cannot be inferred from current function
Biologists have long recognized the difficulty in disentangling evolutionary trajectories, both retroactively and in predicting future trends. This is largely due to the roles of chance and history in determining evolutionary outcomes, where contingencies obscure reasons for why life evolved the way it did (Gould, 1990; Travisano et al., 1995). Beyond chance and history, there are also challenges in inferring the ancestral state of a derived trait when a trait’s current function did not serve as its causal basis (Gould & Lewontin, 1979). In these cases, a trait may have originally evolved due to selection for a different function than is observed today (Gould & Vrba, 1982). For example, feathers have typically been seen as adaptations for flight, as their flight-enhancing features appear to be the product of recent evolutionary history. The origin of feathers, however, may have occurred due to selection for a different function, such as thermal regulation, as there are early fossils of flightless—yet feathered—animals (Benton et al., 2019). At some point in its evolutionary history, selection may have switched such that feathers evolved into structures that could support flight. Hence, the function of a trait at its emergence may be different from its function today.
Evolution of pregenetic inheritance
In the case of information transmission, double-stranded DNA and single-stranded RNA have a remarkable capacity for translating information and passing that information to offspring. Since inheritance is essential for evolution, an incredible amount of work has explored how these polynucleotide chains emerged (Bhowmik & Krishnamurthy, 2019; Crick, 1968; Dounce, 1981; Robertson & Joyce, 2012). Much of this work has assumed that ancestral genes—or simpler precursors—were still comprised of linear strands of nucleotides. This assumption stems from our understanding of contemporary genetic inheritance, for which linear chains serving as templates are essential for transmitting information. In other words, researchers assume similar ancestral and derived functions for nucleic acids. These hypotheses imagine linear strands of DNA or (more commonly) RNA performing template-based replication in a prebiotic context, often as the first evolvable unit (Lincoln & Joyce, 2009). However, it is difficult to conceive how something as complex as a strand of RNA—with sufficient stability and replication fidelity—could appear without a prior adaptive process. Earlier adaptation must have occurred without digital inheritance. To fill this gap, several hypotheses have proposed a means for nongenetic inheritance, such as analog inheritance, where varying chemical compositions are passed from generation to generation (Segré et al., 2000). If this is the case—and the earliest life-forms evolved prior to genetic transmission—then how and when did nucleic acids emerge? Looking to the origin of feathers for inspiration, nucleotides may have performed a different function before they were coopted for replication. As a consequence, it is difficult to explore this prior function based on contemporary genetic structures. Just as is the case with other coopted adaptations, modern genetics should not be a guide for what heritability looked like at the origins of life.
A phylogenetic approach that extrapolates from modern biology is very limited in its ability to infer past functions. Detailing the steps in the origins of life is not a case of piecing together a string of increasingly complex structures. Instead, evolutionary processes create paths that twist, turn, and pass through a series of nonintuitive entities. Nature itself has been described as a “crafty backwoods mechanic,” reconditioning and redesigning old machines, or fashioning new ones with whatever is at hand (Wimsatt & Wimsatt, 2007). This process of reengineering has profound implications for the products of evolutionary change. Rather than examine the path toward a certain structure, exploring the many ways of enabling biological function will help us better understand how life can emerge.
An “eco-evolutionary” approach to the origins of life
Darwin’s “tangled bank” appreciates life’s remarkable complexity and codependency that is nevertheless guided by the same natural laws (Darwin, 1859). Rather than disentangling or recreating the bank, efforts to understand the origin of life should rely on understanding these natural processes. An “eco-evolutionary” approach asks: How do eco-evolutionary processes occur throughout the emergence of life? Do expectations shift for evolvable systems before cells?
Goals of an “eco-evolutionary” approach
The overarching goal of this approach is to explore the several ways that an evolvable system may emerge and gain complexity. This perspective remains agnostic towards particular chemistries or geochemical environments that may have influenced the emergence of life on early Earth and instead aims to understand eco-evolutionary contexts that facilitate multiple paths toward life. Drawing inspiration from the LTEE, where general trends were observable but pathways towards these outcomes were unpredictable (Blount et al., 2018; Travisano & Lenski, 1996), we conclude that retracing the precise trajectory life took on early Earth is unknowable, but understanding relevant processes—and their influence on the emergence of biological innovation—is possible.
This perspective is compatible with a “piecewise approximations to reality” approach to exploring complex systems (Wimsatt & Wimsatt, 2007). This view embraces the reality that even the best theories make idealizations or assumptions that fail as correct descriptions of the world. In the study of the origins of life, this is particularly poignant, as there are no “natural” systems or observational studies. Rather, every system is a model aimed at understanding historical events but ultimately failing to recapitulate the events themselves. Even the most “prebiotically plausible” or “realistic” systems will fall short at recreating primordial environments and evolutionary events, due to the limitations both inherent in the geologic record and of researchers trying to predict evolution’s capacity to tinker and reengineer. Instead, researchers could employ what is described as “local realism,” where scientists argue on certain terms that a phenomenon is real and that their approach requires them to presuppose the existence of that phenomenon. For example, studies on the origins of life argue that there must have been adaptation before cells; therefore, we can presuppose the existence of an evolvable system. In this way, we hope to focus on biological processes unfolding throughout life’s emergence, rather than connecting a stepwise series of chemical entities to cellular life as we know it (Figure 4).
Two conceptual views of the origins of life. The “traditional view” (stepwise line) depicts a series of stepwise increases in life-like properties. Each step represents a discrete transition in complexity that follows a particular hypothesis for how life emerged on early Earth, in this case the RNA World Theory. Researchers with this view of life aim to determine how each step could have occurred under “prebiotically plausible” conditions, with the overarching goal to string together each step from prebiotic soup to simple cellular life. Importantly, this perspective follows one lineage, where each step proposes particular chemical entities that fit the same narrative. Given the role of stochasticity and contingency in evolution, piecing together each step in the emergence of life is not only impossible but also an unrealistic depiction of evolution as progress. An “eco-evolutionary” view (branching lines) envisages that there are many paths towards life and explores several ways life-like systems may gradually evolve. In this view, eco-evolutionary dynamics like convergence, coexistence, and competitive exclusion play important roles throughout life’s trajectory. While “life-likeness” must have generally increased, this progression was not uniform, with some populations reverting to ancestral states or remaining stable. Tips of branches indicate extinction events, and branching nodes represent instances of divergence, where EEFs facilitate biological innovation and the evolution of greater complexity in some lineages (see Figure 2). Researchers with this view do not aim to determine each step in the origin of life but instead explore how the process of evolution can play out in a variety of systems (figure adapted from Shenhav et al., 2003).
Evolution of complexity
The origin of life requires the emergence of biological innovation and complexity. From prebiotic soup to LUCA, the production of novel organisms may not have been continuous, but complexity must have generally increased. This challenge poses several relevant questions for evolutionary biologists: How do populations avoid evolutionary plateaus or stable states? How can innovation facilitate diversification? What factors promote evolvability? Answers to these questions rely on concepts like competition, cooperation, and niche construction to help generate systems capable of open-ended evolution (Channon, 2006; Leon et al., 2018; Turner et al., 2015). For example, competition within a species can promote novel phenotypes that alter fitness landscapes to provide qualitatively different peaks (Rainey & Travisano, 1998), and cooperation within groups can harbor diversity necessary for adaptation to new environments (Preisner et al., 2016). Evolvability can even emerge as a byproduct of selection for other traits, such as modularity in the beaks of Darwin’s finches. In this case, a two-module developmental program independently regulating beak depth and width induced multi-dimensional shifts in beak morphology, enabling radiations to new niches (Mallarino et al., 2011). These and similar studies show the connections to longstanding evolutionary theory on quantitative genetic control of traits (Andersson & Shaw, 1994; Hansen & Pélabon, 2021) and the complex causal connections between plasticity and evolvability (Crozier et al., 2008; Draghi & Whitlock, 2012; Murren et al., 2015). Devising experiments that allow open-endedness has been a central concern for artificial life researchers and can be translated to research on the origins of life (Dittrich et al., 2000; Taylor et al., 2016). For instance, the digital evolution system, AVIDA, has enabled the evolution of novel, complex features (Lenski et al., 2003) and ecological interactions with greater levels of complexity (Zaman et al., 2014). An “eco-evolutionary” approach to the origins of life capitalizes on our understanding of these processes to design experiments with multiple solutions to a given selective pressure. These studies should not rely on the imagination of the researcher to predetermine the desired outcome but instead provide ecological contexts that allow several pathways for innovation to emerge (Table 1).
Summary of how three insights from ecology and evolution can guide investigations on the origins of life (OoL). Given evolvability was a requisite for transitioning from prebiotic soup to LUCA, current understanding of the ecology and evolution of living systems should guide research on the emergence of life. This “eco-evolutionary” approach explores how eco-evolutionary dynamics in extant life inform evolution before cells and is not restricted to particular chemistries, geochemical environments, or molecular entities that may have influenced the emergence of life on early Earth. Since most hypotheses on the origins of life are non-falsifiable (e.g., RNA World, Hot Spring Hypothesis, etc.), the overarching goal of this approach is to explore the several ways that an evolvable system may emerge and gain complexity.
| Eco-evolutionary insight | Significance for OoL | Approach to OoL studies with this perspective | Examples of work using this approach in non-OoL systems |
|---|---|---|---|
| Eco-evolutionary feedbacks are local in time and space | Prebiotic evolution cannot be understood without considering the spatiotemporal constraints on reciprocal interactions between organisms and their environment | Design experiments that allow evolving populations to reshape (or “deform”) fitness landscapes | |
| Ecophysiological factors constrain evolvability | Pre-cellular life had to evolve the metabolic capacity to adapt to environments with large energy potentials | Explore ecological contexts that promote evolvability in the emergence of life | |
| Adaptation cannot be inferred from current function | The structures found in cellular life (e.g., ribosomes, nucleic acids) are limited guides for determining their origin | Study the interplay of evolutionary processes (e.g., selection, drift, migration) in a variety of evolvable systems |
| Eco-evolutionary insight | Significance for OoL | Approach to OoL studies with this perspective | Examples of work using this approach in non-OoL systems |
|---|---|---|---|
| Eco-evolutionary feedbacks are local in time and space | Prebiotic evolution cannot be understood without considering the spatiotemporal constraints on reciprocal interactions between organisms and their environment | Design experiments that allow evolving populations to reshape (or “deform”) fitness landscapes | |
| Ecophysiological factors constrain evolvability | Pre-cellular life had to evolve the metabolic capacity to adapt to environments with large energy potentials | Explore ecological contexts that promote evolvability in the emergence of life | |
| Adaptation cannot be inferred from current function | The structures found in cellular life (e.g., ribosomes, nucleic acids) are limited guides for determining their origin | Study the interplay of evolutionary processes (e.g., selection, drift, migration) in a variety of evolvable systems |
Summary of how three insights from ecology and evolution can guide investigations on the origins of life (OoL). Given evolvability was a requisite for transitioning from prebiotic soup to LUCA, current understanding of the ecology and evolution of living systems should guide research on the emergence of life. This “eco-evolutionary” approach explores how eco-evolutionary dynamics in extant life inform evolution before cells and is not restricted to particular chemistries, geochemical environments, or molecular entities that may have influenced the emergence of life on early Earth. Since most hypotheses on the origins of life are non-falsifiable (e.g., RNA World, Hot Spring Hypothesis, etc.), the overarching goal of this approach is to explore the several ways that an evolvable system may emerge and gain complexity.
| Eco-evolutionary insight | Significance for OoL | Approach to OoL studies with this perspective | Examples of work using this approach in non-OoL systems |
|---|---|---|---|
| Eco-evolutionary feedbacks are local in time and space | Prebiotic evolution cannot be understood without considering the spatiotemporal constraints on reciprocal interactions between organisms and their environment | Design experiments that allow evolving populations to reshape (or “deform”) fitness landscapes | |
| Ecophysiological factors constrain evolvability | Pre-cellular life had to evolve the metabolic capacity to adapt to environments with large energy potentials | Explore ecological contexts that promote evolvability in the emergence of life | |
| Adaptation cannot be inferred from current function | The structures found in cellular life (e.g., ribosomes, nucleic acids) are limited guides for determining their origin | Study the interplay of evolutionary processes (e.g., selection, drift, migration) in a variety of evolvable systems |
| Eco-evolutionary insight | Significance for OoL | Approach to OoL studies with this perspective | Examples of work using this approach in non-OoL systems |
|---|---|---|---|
| Eco-evolutionary feedbacks are local in time and space | Prebiotic evolution cannot be understood without considering the spatiotemporal constraints on reciprocal interactions between organisms and their environment | Design experiments that allow evolving populations to reshape (or “deform”) fitness landscapes | |
| Ecophysiological factors constrain evolvability | Pre-cellular life had to evolve the metabolic capacity to adapt to environments with large energy potentials | Explore ecological contexts that promote evolvability in the emergence of life | |
| Adaptation cannot be inferred from current function | The structures found in cellular life (e.g., ribosomes, nucleic acids) are limited guides for determining their origin | Study the interplay of evolutionary processes (e.g., selection, drift, migration) in a variety of evolvable systems |
We’ve emphasized the role of selection here against a backdrop of ever-present random processes, which are essential for opening avenues for further innovation. The ability for drift and selection to act together during adaptive evolution has long been recognized in evolutionary theory (Ishida, 2017; Provine, 1986; Svensson, 2023), albeit with contention over their relative contributions (Dobzhansky & Pavlovsky, 1957; Fisher, 1930; Jensen et al., 2019; Kern & Hahn, 2018; Lynch et al., 2016). There is strong experimental evidence for how drift and selection can each act as primary drivers of evolutionary outcomes (Hochberg et al., 2020; Marques et al., 2018; Wong et al., 2012), yet entanglement of the two specifically can have profound consequences for the emergence of innovation. For instance, allowing populations to randomly explore genotype space via drift may provide opportunities to cross fitness valleys. Similarly, drift can work to maintain within-population variation upon which selection may then act. Experimental evidence for this entanglement was found during the LTEE when one population evolved the ability to use citrate, an abundant but previously unusable energy source (Blount et al., 2008). This radical innovation was made possible through long periods of drift before the population could climb a new adaptive peak (Blount et al., 2008; Burnham & Travisano, 2021). For these reasons, experiments that allow several possible outcomes are more likely to provide insights into the emergence of complexity.
Conclusion
Research on the origins of life has traditionally been conducted by chemists. We recast these questions so that biologists study them as well. The dichotomy between “chemistry-forward” and “biology-backward” approaches neglects the role of eco-evolutionary dynamics throughout the process of abiogenesis, thus our goal is to use insights from biological disciplines to guide future work on the origins of life. Hypotheses claiming particular environments or the stepwise emergence of cellular structures are nonfalsifiable and should not limit the scope of investigations. Instead, our inability to recreate primordial environments, as well as anticipate stochastic evolutionary trajectories, means that experiments should explore the processes we know were present when life began: evolution as influenced by adaptation, chance, and history. An “eco-evolutionary” approach harnesses these processes in a variety of systems to explore the many ways innovation can emerge, in a warm little pond or otherwise.
Data availability
There is no data associated with this work.
Author contributions
Conceptualization: M.K., M.T.; Visualization: M.K., M.T.; Funding acquisition: M.T.; Writing—original draft: M.K.; Writing—review & editing: M.K., M.T.
Funding
The study was supported by the National Science Foundation (Award Number: NSF-1724011) and National Aeronautics and Space Administration (Award Number: 16-IDEAS16-0002).
Conflict of interest:
The authors declare no conflict of interest.
Acknowledgment
The authors thank members of the Travisano lab for valuable comments on an early draft of the manuscript. All figures were created with BioRender.com.
