Naomi Marcil Levine

As the world’s oceans change in response to climate change, phytoplankton communities will adapt to warmer, more stratified surface waters via plasticity, evolution, and range shifts. Current global ocean models assume that size structured phytoplankton communities have fixed trait relationships, and as a result generally predict that smaller size classes will become more dominant globally. However, this general expectation fails to consider how intra-species trait tradeoffs may operate orthogonally from large-scale inter-species tradeoffs—allowing for alternative evolutionary pathways given the limits and/or possibilities available to ancestral populations. To identify evolutionary pathways phytoplankton populations might take, we develop a novel modeling framework that combines a trait-based phytoplankton quota model with stochastic evolution (ecoTRACE). EcoTRACE explicitly decouples key phytoplankton traits from interspecific allometric relationships, allowing for novel phenotypes to emerge. We validated ecoTRACE against a long-term artificial size selection experiment on Dunaliella tertiolecta. We show that ecoTRACE captures multi-dimensional evolved phenotypes that quota models based on interspecific relationships fail to reproduce. Under fluctuating multi-stressor growth, model populations evolve phenotypic plasticity that deviates from predicted interspecific allometric relationships. EcoTRACE provides a framework for generating hypotheses as to the evolutionary trajectories that phytoplankton will experience in a warmer, more variable ocean.

In the last decade, numerous laboratory experiments have demonstrated that when marine phytoplankton are exposed to thermal stress, they can evolve high temperature tolerance in a short time (months). This evolutionary potential may ensure the persistence of marine phytoplankton species under current and future global warming. However, the effect of such adaptation on the phytoplankton's interaction with the environment and other organisms depends on how cellular metabolism shifts during the evolutionary process. In order to elucidate which cellular strategies allow the emergence of thermo‐tolerant lines, we analysed the proteomic changes in a marine diatom ( Chaetoceros simplex ) following both thermal acclimation and evolutionary adaptation. We found that the high temperature‐tolerant lines exhibit a conservative cellular strategy when acclimated to high, above‐optimal temperature, where nitrogen recycling, reallocation of carbon and nitrogen, and protein repair or protection are favoured at the expense of new proteins and photosynthetic structures biosynthesis and rapid growth. While this strategy gives the lines that evolved high temperature tolerance an advantage under thermal stress, the shift to resource reallocation may explain the absence of high‐temperature adaptation when cells are exposed to low nitrate availability.

Ocean microbial communities are made up of thousands of diverse taxa whose metabolic demands set the rates of both biomass production and degradation. Thus, these microscopic organisms play a critical role in ecosystem dynamics, global carbon cycling, and climate. While we have frameworks for relating phytoplankton diversity to rates of carbon fixation, our knowledge of how variations in heterotrophic microbial populations drive changes in carbon cycling is in its infancy. Here, we leverage global metagenomic datasets and metabolic models to identify a set of metabolic niches with distinct growth strategies. These groupings provide a simplifying framework for describing microbial communities in different oceanographic regions and for understanding how heterotrophic microbial populations function. This framework, predicated directly on metabolic capability rather than taxonomy, will enable us to tractably link heterotrophic diversity directly to biogeochemical rates in large scale ecosystem models.

Warming and nutrient limitation are major stressors that affect primary production in the ocean, with cascading impacts on the food web. Yet, we lack a mechanistic understanding of how phytoplankton manage the combined stress of heat-damage and nutrient-limitation and the implications of these responses for phytoplankton biogeography. By combining theory, proteome allocation modeling, and climate projections, we identified two potential competing strategies for multi-stressor growth: (1) increase growth temperature optima through higher nutrient uptake efficiency and smaller cells, or (2) invest in heat-mitigation mechanisms achieving higher thermal tolerance at the cost of growth and larger cells. By simulating the optimal metabolic strategies of different phytoplankton functional types across a latitudinal gradient, we found that Prochlorococcus are more vulnerable in warmer, ‘heat-stressed’, tropical regions due to greater heat sensitivity and lower storage capacity, indicating a potential ecological niche for larger phytoplankton with lower sensitivity to oxidative stress, such as Synechococcus and picoeukaryotes. Our findings advocate for the inclusion of phytoplankton heat-stress responses in global models to more accurately predict their ecological niches as the climate warms.

Heterotrophic bacteria and archaea (“heteroprokaryotes”) drive global carbon cycling, but how to quantitatively organize their functional complexity remains unclear. We generated a global-scale understanding of marine heteroprokaryotic functional biogeography by synthesizing genetic sequencing data with a mechanistic marine ecosystem model. We incorporated heteroprokaryotic diversity into the trait-based model along two axes: substrate lability and growth strategy. Using genetic sequences along three ocean transects, we compiled 21 heteroprokaryotic guilds and estimated their degree of optimization for rapid growth (copiotrophy). Data and model consistency indicated that gradients in grazing and substrate lability predominantly set biogeographical patterns, and we identified deep-ocean “slow copiotrophs” whose ecological interactions control the surface accumulation of dissolved organic carbon.

Microbial community assembly is governed by the flow of carbon sources and other primary metabolites between species. However, central metabolism represents only a small fraction of the biosynthetic repertoire of microbes: metabolites such as antimicrobial compounds, signaling molecules, and co-factors are underexplored in their potential to shape microbial communities. Here, we focus on B vitamin exchange in marine bacterial communities that degrade polysaccharides, a key component of particulate organic matter. We found that in a screen of 150 natural isolates, almost a third were auxotrophs for one or more B vitamins. By measuring physiological parameters such as uptake affinities and comparing those to ambient seawater concentrations, we showed that marine bacteria live at the edge of vitamin limitation in the environment. To understand how auxotrophs survive in the open oceans, we used our experimental data to model vitamin cross-feeding on particles through both secretion and lysis. Our results highlight the importance of vitamin auxotrophies in shaping microbial community assembly and succession, adding another layer of complexity to the trophic structure of particle-associated communities.

Ocean microbial communities are made up of thousands of diverse taxa whose metabolic demands set the rates of both biomass production and degradation. Thus, these microscopic organisms play a critical role in ecosystem dynamics, global carbon cycling, and climate. While we have frameworks for relating phytoplankton diversity to rates of carbon fixation, our knowledge of how variations in heterotrophic microbial populations drive changes in carbon cycling is in its infancy. Here, we leverage global metagenomic datasets and metabolic models to identify a set of metabolic niches with distinct growth strategies. These groupings provide a simplifying framework for describing microbial communities in different oceanographic regions and for understanding how heterotrophic microbial populations function. This framework, predicated directly on metabolic capability rather than taxonomy, enables us to tractably link heterotrophic diversity directly to biogeochemical rates in large scale ecosystem models.

The oceans sequester a vast amount of carbon thus playing a central role in the global carbon cycle. Assessing how carbon cycling will be impacted by climate change requires an improved understanding of microbial dynamics, which are responsible for most carbon transformations in the oceans. Current numerical models used for predicting future states represent simplified microbial phenotypes and thus may not produce robust predictions of microbial communities. We propose reframing approaches for studying microbial trait change to allow for selection on multi-trait phenotypes. Integrating statistical approaches and trait-based models will allow for the incorporation of evolution into carbon cycle predictions.