Metabolic compatibility and the rarity of prokaryote endosymbioses

Abstract

The evolution of the mitochondria was a significant event that gave rise to the eukaryotic lineage and most large complex life. Central to the origins of the mitochondria was an endosymbiosis between prokaryotes. Yet, despite the potential benefits that can stem from a prokaryotic endosymbiosis, their modern occurrence is exceptionally rare. While many factors may contribute to their rarity, we lack methods for estimating the extent to which they constrain the appearance of a prokaryotic endosymbiosis. Here, we address this knowledge gap by examining the role of metabolic compatibility between a prokaryotic host and endosymbiont. We use genome-scale metabolic flux models from three different collections (AGORA, KBase, and CarveMe) to assess the viability, fitness, and evolvability of potential prokaryotic endosymbioses. We find that while more than half of host-endosymbiont pairings are metabolically viable, the resulting endosymbioses have reduced growth rates compared to their ancestral metabolisms and are unlikely to gain mutations to overcome these fitness differences. In spite of these challenges, we do find that they may be more robust in the face of environmental perturbations at least in comparison with the ancestral host metabolism lineages. Our results provide a critical set of null models and expectations for understanding the forces that shape the structure of prokaryotic life.

Keywords: endosymbiosis; eukaryogenesis; evolution; metabolic model; prokaryote.

Fig. 1.

Stages in the evolutionary rise of endosymbioses. From its inception, a nascent endosymbiosis faces different barriers that challenge the success of its lineage. This schematic organizes these barriers into three broad stages corresponding to initial viability, persistence, and evolvability. Metabolic compatibility influences barriers in each of these stages. In the viability stage, both host and endosymbiont must be able to grow and reproduce such that the pair can produce offspring host–endosymbiont pairs. In the persistence stage, the population of endosymbioses must persist (avoid extinction) by surviving environmental perturbations and competing successfully with other species, including their ancestors. In the evolvability stage, the endosymbiosis must be able to access a sufficient number and caliber of beneficial mutations to foster adaptation to various environments. The successful navigation of evolutionary trajectories through these stages determines the abundance and diversity of endosymbioses in the biosphere.

Fig. 2.

Viability of prokaryotic endosymbioses. (A) A schematic shows the nested compartment structure of the two possible host–endosymbiont pairs considered in our analyses. Each cell has a cytoplasm compartment [c] and can exchange compounds with its external environment (indicated by arrows). In endosymbioses, the extracellular compartment of the endosymbiont is the cytoplasm compartment of its host. (B) A bar graph with an inset of a violin plot shows the percentages of paired metabolisms that form viable endosymbioses for each of the three metabolic model collections. The percentages are the means of 100 samples of 1,000 pairs and are of similar magnitude, 56.2 to 66.2%; however, an ANOVA confirms the means do differ across collections (P < .001). (C) A bar graph shows the percentage of pairings in which neither, only one, or both configurations of endosymbioses are viable. There is at least one configuration viable for 84 to 92% of pairs in each collection and the most common scenario is where only one configuration is viable.

Fig. 3.

Paths to fixing nonviable endosymbioses. (A) A schematic shows the two causes of nonviability in endosymbioses using metabolic models: 1. (Top) missing transport of an extracellular compound into the host’s cytoplasm and 2. (Bottom) no access to compounds that remain in the extracellular compartment. For each, we show a typical form of the missing essential reaction and the percentage of 15,000 nonviable endosymbioses that can be made viable by providing the necessary type of reaction (colors indicate model collections, following the legend in B). Between the two causes, the first is a more frequent source of nonviability. Note that within a collection the percentages for the two causes do not sum to 100% because they are not mutually exclusive. (B) The graph shows the minimum number of compounds whose transport needs to be provided in order to fix nonviable endosymbioses. Each point is the mean of 100 samples of 100 fixable, nonviable endosymbioses, and the error bars are the standard deviations of those samples. For each collection, the most common case is that the endosymbiosis can be fixed by transporting a single compound. The pie chart inset shows how often that single compound is a specific compound, i.e., there is only one such compound whose transport makes the endosymbiosis viable.

Fig. 4.

Survival competition between endosymbiosis and ancestral metabolisms in response to environmental degradation. (A) The distributions show the robustness of 10,000 endosymbiosis metabolisms (labeled “pair”) and their ancestral metabolisms (labeled “host” or “endo”) sampled from AGORA. Robustness is quantified as the proportion of environmental perturbations survived by a metabolism. The distributions have similar single peaked shapes with > 80% of the population lying between the range of 85 to 95%. SI Appendix, Fig. S3 shows a similar figure for the other collections. (B) The bar graph shows the percentage of environmental perturbations for which all three metabolisms are viable, nonviable, or mixed across the three model collections. In only 3 to 7% of perturbations is there a difference in the viability of the endosymbiosis metabolism compared to at least one of its ancestral metabolisms. (C) The star plot displays the relative frequency of the possible outcomes for the mixed cases from B. Of the 6 possible scenarios for mixed viability, the two most frequent feature different survival between the ancestral endosymbiont metabolism compared to the other two, though which is more robust depends on the model collection (SI Appendix, Fig. S4). The star plot also shows that endosymbioses survive more environmental perturbations than their ancestral host metabolisms in all collections.

Fig. 5.

Growth-rate competition between endosymbioses and ancestral metabolisms. (A) The bar graph shows the result of a growth-rate competition between endosymbioses and their ancestral host metabolisms. The host grows faster for a majority of comparisons (85 to 94%). (B) The histogram shows the relative growth rates of endosymbioses versus their ancestral host metabolisms, sampled from AGORA (SI Appendix, Fig. S6 for KBase and CarveMe). When the endosymbiosis grows faster than its ancestral host, the fitness advantage is often smaller in magnitude compared to its fitness disadvantage when it grows more slowly. (C) The bar graph is similar to A but the comparison is between endosymbioses and their ancestral endosymbiont metabolisms. Again endosymbioses grow more slowly in the majority of comparisons (88 to 92%). (D) The histogram shows the relative growth rates of endosymbioses versus their ancestral endosymbiont metabolisms, sampled from KBase (SI Appendix, Fig. S7 for AGORA and CarveMe). As in B, the typical growth advantage is smaller in magnitude than the growth disadvantage for endosymbioses. (E) The bar graph shows the result of a growth-rate competition between the endosymbiosis and its ancestral metabolisms when all share the same environment. In both collections, the most likely scenario is that the endosymbiosis grows more slowly than both ancestral metabolisms (88%). (F) The relative fitness of the endosymbiosis versus its ancestral metabolisms from E is plotted using CarveMe models (SI Appendix, Fig. S8 for AGORA).

Fig. 6.

Evolvability of endosymbioses versus ancestral metabolisms. (A) The histogram shows the difference in the number of beneficial mutations between the endosymbiosis and its ancestral host metabolism (data for A & B from AGORA, SI Appendix, Figs. S9 and S10 for CarveMe). All mutations occur in host metabolism reactions and are deemed beneficial if they increase the growth rate. A sign-rank test supports the null hypothesis that the endosymbiosis and its ancestral host metabolism do not differ in the number of beneficial mutations. (B) The histogram is similar to A but compares the endosymbiosis to its ancestral endosymbiont metabolism for mutations in their shared reactions. Here, a sign-rank test rejects the null hypothesis, such that the ancestral endosymbiont metabolism has more beneficial mutations (P < .01). (C) A bubble chart shows the frequency of mutations, in endosymbioses vs. their ancestral host metabolisms, that increase the growth rate above that of the ancestral host metabolism (data for C & D from CarveMe, SI Appendix, Figs. S11 and S12 for AGORA). There are more such mutations in ancestral host metabolisms than in endosymbioses. (D) The plot is similar to C except it is in relation to the ancestral endosymbiont metabolism. Again, there are more such mutations in ancestral endosymbiont metabolisms than in endosymbioses.