Trade-offs predicted by metabolic network structure give rise to evolutionary specialization and phenotypic diversification

David M Ekkers; Sergio Tusso; Stefany Moreno-Gamez; Marina C Rillo; Oscar P Kuipers; G Sander van Doorn

doi:10.1093/molbev/msac124

Trade-offs predicted by metabolic network structure give rise to evolutionary specialization and phenotypic diversification

Mol Biol Evol. 2022 Jun 9;39(6):msac124. doi: 10.1093/molbev/msac124. Online ahead of print.

Authors

David M Ekkers^{1

2}, Sergio Tusso^{3

4}, Stefany Moreno-Gamez¹, Marina C Rillo⁵, Oscar P Kuipers², G Sander van Doorn¹

Affiliations

¹ Groningen Institute for Evolutionary Life Sciences (GELIFES), University of Groningen, Nijenborgh 7, 9747 AG, Groningen, The Netherlands.
² Molecular Genetics Group, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen, Nijenborgh 7, 9747 AG, Groningen, The Netherlands.
³ Division of Evolutionary Biology, Faculty of Biology, LMU Munich, Grosshaderner Str. 2, 82152 Planegg-Martinsried, Germany.
⁴ Science for Life Laboratories and Department of Evolutionary Biology, Norbyvägen 18D, 75236 Uppsala University, 75236 Uppsala, Sweden.
⁵ Institute for Chemistry and Biology of the Marine Environment, Carl von Ossietzky University Oldenburg, Schleusenstr. 1, 26382, Wilhelmshaven, Germany.

Abstract

Mitigating trade-offs between different resource-utilization functions is key to an organism's ecological and evolutionary success. These trade-offs often reflect metabolic constraints with a complex molecular underpinning; therefore, their consequences for evolutionary processes have remained elusive. Here, we investigate how metabolic architecture induces resource utilization constraints and how these constraints, in turn, elicit evolutionary specialization and diversification. Guided by the metabolic network structure of the bacterium Lactococcus cremoris, we selected two carbon sources (fructose and galactose) with predicted co-utilization constraints. By evolving L. cremoris on either fructose, galactose or a mix of both sugars, we imposed selection favoring divergent metabolic specializations or co-utilization of both resources, respectively. Phenotypic characterization revealed the evolution of either fructose or galactose specialists in the single-sugar treatments. In the mixed sugar regime, we observed adaptive diversification: both specialists coexisted, and no generalist evolved. Divergence from the ancestral phenotype occurred at key pathway junctions in the central carbon metabolism. Fructose specialists evolved mutations in the fbp and pfk genes that appear to balance anabolic and catabolic carbon fluxes. Galactose specialists evolved increased expression of pgmA (the primary metabolic bottleneck of galactose metabolism) and silencing of ptnABCD (the main glucose transporter) and ldh (regulator/enzyme of downstream carbon metabolism). Overall, our study shows how metabolic network architecture and historical contingency serve to predict targets of selection and inform the functional interpretation of evolved mutations. The elucidation of the relationship between molecular constraints and phenotypic trade-offs contributes to an integrative understanding of evolutionary specialization and diversification.

Keywords: Lactococcus cremoris; adaptive diversification; central carbon metabolism; experimental evolution; metabolic network analysis; metabolic trade-offs.