Reconciling molecular evolution and evolutionary ecology studies: a phylogenetic reconciliation method for gene-symbiont-host systems

Interactions between species are a driving force in evolution. Many organisms host symbiotic partners that live all or part of their life in or on their host. Whether they are mutualistic or parasitic, these symbiotic associations impose strong selective pressures on both partners and affect their evolutionary trajectories. In fine, they can have a significant impact on the diversification patterns of both host and symbiont lineages, with symbiotic lineages sometimes speciating simultaneously with their hosts and/or switching from one host species to another. Long-term associations between species can also result in gene transfers between the involved organisms. Those lateral gene transfers are a source of ecological innovation but can obscure the phylogenetic signals and render the process of phylogenetic reconstructions complex (Lerat et al. 2003).

Methods known as reconciliations explore similarities and differences between phylogenetic trees. They have been widely used to both compare the diversification patterns of hosts and symbionts and identify lateral gene transfers between species. Though the reconciliation approaches used in host/ symbiont and species/ gene phylogenetic studies are identical, they are always applied separately to solve either molecular evolution questions or investigate the evolution of ecological interactions. However, the two questions are often intimately linked and the current interest in multi-level systems (e.g. the holobiont concept) calls for a unique model that will take into account three-level nested organization (gene/symbiont/ host) where both symbiont and genes can transfer among hosts. 

Here Menet and collaborators (2023) provide such a model to produce three-level reconciliations. In order to do so, they extend the two-level reconciliation model implemented in “ALE” software (Szöllősi et al. 2013), one of the most used and proven reconciliation methods. Briefly, given a symbiont gene tree, a symbiont tree and a host tree, as in previous reconciliation models, the symbiont tree is mapped onto the host tree by mixing three types of events: Duplication, Transfer or Loss (DTL), with a possibility of the symbiont evolving temporarily outside the host phylogeny (in a “ghost” host lineage). The gene tree evolves similarly inside the symbiont tree, but horizontal transfers are constrained to symbionts co-occurring within the same host. Joint reconciliation scenarios are reconstructed and DTL event rates and likelihoods are estimated according to the model. As a nice addition, the authors propose a method to infer the symbiont phylogeny through amalgamation from gene trees and a host tree.

The authors then explore the diverse possibilities offered by this method by testing it on both simulated datasets and biological datasets in order to check whether considering three nested levels is worthwhile. They convincingly show that three-level reconciliation has a better capacity to retrieve the symbiont donors and receivers of horizontal gene transfers, probably because transfers are constrained by additional elements relevant to the biological systems. Using, aphids, their obligate endosymbionts, and the symbiont genes involved in their nutritional functions, they identify horizontal gene transfers between aphid symbionts that are missed by two-level reconciliations but detected by expertise (Manzano-Marín et al. 2020). The other dataset presented here is on the human pathogen Helicobacter pylori, which history is supposed to reflect human migration. They use more than 1000 H. pylori gene families, and four populations, and use likelihood computations to compare different hypotheses on the diversification of the host.

In summary, this study is a proof-of-concept of a 3-level reconciliation, where the authors manage to convey the applicability of their framework to many biological systems. Reported complexities, confirmed by reported running times, show that the method is computationally efficient. Without a doubt, the tool presented here will be very useful to evolutionary biologists who want to investigate multi-scale cophylogenies and it will move forward the study of associations between host and symbionts when symbiont genomic data are available.

REFERENCES

Lerat, E., Daubin, V., & Moran, N. A. (2003). From gene trees to organismal phylogeny in prokaryotes: the case of the γ-Proteobacteria. PLoS biology, 1(1), e19.
https://doi.org/10.1371/journal.pbio.0000019
 
Manzano-Marın, A., Coeur d'acier, A., Clamens, A. L., Orvain, C., Cruaud, C., Barbe, V., & Jousselin, E. (2020). Serial horizontal transfer of vitamin-biosynthetic genes enables the establishment of new nutritional symbionts in aphids' di-symbiotic systems. The ISME Journal, 14(1), 259-273.
https://doi.org/10.1038/s41396-019-0533-6

Menet H, Trung AN, Daubin V, Tannier E (2023) Host-symbiont-gene phylogenetic reconciliation. bioRxiv, 2022.07.01.498457, ver. 2 peer-reviewed and recommended by Peer Community in Evolutionary Biology. https://doi.org/10.1101/2022.07.01.498457

Szöllősi, G. J., Rosikiewicz, W., Boussau, B., Tannier, E., & Daubin, V. (2013). Efficient exploration of the space of reconciled gene trees. Systematic biology, 62(6), 901-912.
https://doi.org/10.1093/sysbio/syt054