Multiscale analysis of autotroph-heterotroph interactions in a high-temperature microbial community

Abstract

Interactions among microbial community members can lead to emergent properties, such as enhanced productivity, stability, and robustness. Iron-oxide mats in acidic (pH 2-4), high-temperature (> 65 °C) springs of Yellowstone National Park contain relatively simple microbial communities and are well-characterized geochemically. Consequently, these communities are excellent model systems for studying the metabolic activity of individual populations and key microbial interactions. The primary goals of the current study were to integrate data collected in situ with in silico calculations across process-scales encompassing enzymatic activity, cellular metabolism, community interactions, and ecosystem biogeochemistry, as well as to predict and quantify the functional limits of autotroph-heterotroph interactions. Metagenomic and transcriptomic data were used to reconstruct carbon and energy metabolisms of an important autotroph (Metallosphaera yellowstonensis) and heterotroph (Geoarchaeum sp. OSPB) from the studied Fe(III)-oxide mat communities. Standard and hybrid elementary flux mode and flux balance analyses of metabolic models predicted cellular- and community-level metabolic acclimations to simulated environmental stresses, respectively. In situ geochemical analyses, including oxygen depth-profiles, Fe(III)-oxide deposition rates, stable carbon isotopes and mat biomass concentrations, were combined with cellular models to explore autotroph-heterotroph interactions important to community structure-function. Integration of metabolic modeling with in situ measurements, including the relative population abundance of autotrophs to heterotrophs, demonstrated that Fe(III)-oxide mat communities operate at their maximum total community growth rate (i.e. sum of autotroph and heterotroph growth rates), as opposed to net community growth rate (i.e. total community growth rate subtracting autotroph consumed by heterotroph), as predicted from the maximum power principle. Integration of multiscale data with ecological theory provides a basis for predicting autotroph-heterotroph interactions and community-level cellular organization.

Fig 1. Conceptual representation of the multiscale metabolic interactions between a primary autotroph (M. yellowstonensis) and heterotroph (Geoarchaeum str. OSPB) present in high-temperature (70–80°C) acidic Fe-oxide mats.

A) Modeling of these interactions used genome annotations and physiological studies to identify the enzymes for electron donor and acceptor utilization, as depicted here for iron and sulfur oxidation by the primary autotroph. Electron flow is indicated by white dashed lines. B) These enzymes provided the basis for cellular-level models to quantify resource requirements for the primary autotroph and heterotroph to produce biomass and cellular energy from reduced inorganic electron donors and organic carbon, respectively. C) The total resource requirements of each population and in silico approaches provided bases to test interaction hypotheses (community-level modeling) using in situ measurements, including relative population abundances, carbon isotope fractionation, Fe(III) deposition rates, and oxygen flux into the mat. The system volume considered in these analyses is defined as the aerobic region of the mat (0.07 cm by 1 cm by 1 cm) moving vertically up as the mat grows. The effect of high and low oxygen associated with the top and bottom of the relatively homogeneous aerobic region provided bounds for electron acceptor stresses. Abbreviations: I—energy conserving NADH:ubiquinone oxidoreductase; Cyt ba–cytochrome/quinol oxidase; Fox AB and FoxCDG–Fe(II) oxidation complexes; MCO–Multicopper protein; Q–quinone; QH₂ –quinol; IV–ATP synthase; Sqr–Sulfur-quinone reductase; Hdr–Heterodisulfide reductase; Rdh–Rhodanese-related sulfur transferase; Tqo–Thiosulfate-quinone oxidoreductase; and Dox–Sulfur associated terminal oxidase.

Fig 2. Analysis of electron donor and acceptor resource requirements to produce biomass indicate the autotroph requires more oxygen, and oxygen availability affects carbon requirements of the heterotroph.

A) Moles of oxygen and electron donor required to produce one Cmole of M. yellowstonensis biomass for all biomass producing elementary flux modes, note the log scales. Clusters of electron donors, which include Fe(II), sulfide, sulfur, sulfite, and thiosulfate, are circled by dashed lines. B) Moles of oxygen and autotroph biomass required to produce one Cmole of Geoarchaeum str. OSPB biomass. Optimizations for carbon- and oxygen-limited scenarios are marked by points A and B, respectively. C) Relative oxygen consumption plotted as a function of biomass production for autotroph, M. yellowstonensis, utilizing Fe(II) or elemental sulfur, and for heterotroph, Geoarchaeum str. OSPB, utilizing autotroph biomass generated from Fe(II) oxidation or exogenous sources of reduced organic carbon (i.e. landscape carbon).

Fig 3. Overlaying simulation of various nutrient limitation, relative population abundance, and in situ parameters established a lower limit for the total concentration of biomass in a mat community.

Three oxygen fluxes, representing 50, 100, and 200%, of the average observed oxygen flux (420 nmol O₂ cm^-2 h^-1) provided upper and lower bounds of resource availability [12,14]. Each modeled oxygen flux was analyzed using the predicted metabolic strategy for most efficient utilization of oxygen by the autotroph (dotted line) as well as oxygen (solid line) and carbon (dashed line) for the heterotroph. Minimum and maximum active biomass concentrations (top and bottom panels, respectively, note the log scales) and a constant aerobic volume allowed for the prediction of specific growth rates. Consistent with in situ sequence data, the predicted relative population abundance was 0.3 to 0.5 autotroph to heterotroph (vertical boxed area) based on maximum in vitro specific growth rate (horizontal boxed area) and community stability.

Fig 4. Overlaying simulation of various nutrient limitation, relative population abundance, and in situ parameters A) indicates a maximization of total community growth rates and B) established an upper limit for the total concentration of biomass in the studied mat community.

Three oxygen fluxes, representing 50, 100, and 200%, of the average observed oxygen flux (420 nmol O₂ cm^-2 h^-1) provided upper and lower bounds of resource availability [12,14]. Each modeled oxygen flux was analyzed using the predicted most efficient metabolic strategy for utilization of oxygen (solid line) and carbon (dashed line). Consistent with in situ sequence data, the predicted relative population abundance was 0.3 to 0.5 autotroph to heterotroph (vertical boxed area) based on maximum in vitro specific growth rate and community stability.

Fig 5. Sensitivity analysis of system behavior as a function of geochemical variation provides physiological context for in situ measurements.

Three scenarios (middle panel) predicted the metabolic space feasible for autotroph-heterotroph interactions of the entire community (top panel). Scenario A analyzes a community composed solely of an Fe(II) oxidizing autotroph (represented by R1 or Fe) and autotroph-consuming heterotroph (R4 or AC), which are present at relative population abundances of 0.3 and 0.5 autotroph to heterotroph. Scenario B analyzes the impact of additional carbon sources for the heterotroph (R5), such as the consumption of landscape carbon (LC), while enforcing 42 to 99% of total biomass carbon was autotroph in origin based on the in situ relative system carbon isotope signature. The decrease in slope toward the bottom of scenario B (lower panel) is due to mathematical resolution and the required carbon fractionation in biomass. Scenario C determines the impact of an additional electron donor (sulfide) for the autotroph (R2 or S) in addition to a landscape carbon source for the heterotroph. The net production of autotroph is denoted by R3. Maximum specific growth rate and system volume were set to 0.1 h^-1 and 0.07 cm^-3, respectively. Biomass and cellular energy yields were determined by elementary flux mode analysis, which were then used as inputs to flux balance analysis to predict rates. Averages of observed Fe(III)-oxide deposition (0.054 and 0.36 μmol cm^-3 h^-1 for Beowulf and OSP, respectively) and oxygen uptake rates (0.38 and 0.46 μmol cm^-3 h^-1 for Beowulf and OSP, respectively) [12,14] (dotted lines) are shown for comparison.