Replaying the tape of lice

In his 1989 book, “Wonderful Life”, Stephen Jay Gould proposed a thought experiment where scientists would “replay the tape of life”¹.  After replaying the tape, scientists would find that some products of organic evolution emerge reliably, while others result from rare chance events.  Using the information, scientists could assess the relative contributions of natural selection (determinism), chance events (stochasticity), and the interaction of selection and chance events (contingency) in shaping evolutionary outcomes^1,2.  While Gould described his grand experiment in the context of restarting life, which is impossible, recent research has brought a renewed interest in the concept of replays².

In the linked article, we describe a study inspired by Gould’s concept of replays.  For this study, we examined the role of determinism, stochasticity, and contingency in shaping gene loss in bacteria.  Specifically, we focused on host-associated bacteria.  Following the transition from free-living to host-associated, bacteria lose much of their genetic material³.  While the loss of genetic material is a common evolutionary outcome, it is the result of processes that proceed over millions of years⁴.  However, by comparing the outcomes of bacteria that lost genes in parallel, we could identify both deterministic and stochastic outcomes. 

Our study focused exclusively on symbiotic bacteria that reside in parasitic lice.  These louse-associated bacteria are uniquely suited to a “replays” study, because closely related species of lice independently acquired their bacteria from a common bacterial ancestor⁵.  Furthermore, these bacteria are passed from a mother louse to her offspring via ovarian transfer (resembling patterns of mitochondrial inheritance), causing each to form a lineage isolated from lineages in other louse species⁶.  Hence, these louse-associated bacteria each originated from a common ancestor with similar genetic material and then evolved in parallel. 

By comparing bacteria isolated from different louse species, each having undergone the loss of genes in parallel, we searched for outcomes consistent with determinism, stochasticity, and contingency.  We found that most genes were either retained or lost in all bacterial lineages.  The finding highlights the relatively large contribution of deterministic processes in shaping gene loss.  However, we also discovered a number of genes that were present in some bacterial lineages but not others. The differences appear to result from stochastic losses that created contingencies constraining subsequent gene losses.  Specifically, the contingency led to the retention of the gene(s) supporting the same or similar cellular function as the gene(s) lost.  Thus, the bacteria retained corresponding functions, but have done so by using different genes.

Collectively, the results underscore the complex interaction of determinism and contingency, resulting in convergent outcomes from distinct evolutionary paths.  Furthermore, the results confirm Gould’s prediction that evolutionary processes can take different routes to reach the same destination.

 The role of fieldwork in a bioinformatics study

The project started with a discussion between two of the studies’ authors at a National Science Foundation meeting in Arlington, Virginia.  However, the project quickly grew into a highly collaborative effort involving researchers with expertise in computer science, bioinformatics, evolutionary biology, systematics, parasitology, and microbiology.  While each member of the team brought essential knowledge and a skill set to the project, it is critical to highlight the role of fieldwork and collections in such a study, which are easy to overlook when reading the research article. 

The work described in the study required access to louse species found in distant parts of the world, each parasitizing a different species of bird.  In total, we examined data from 68 individual lice collected from 66 bird species sampled in 20 different countries. The majority of these samples were collected by the experienced ornithologists and parasitologists involved in the study.  The specimens collected represent part of a growing trend to preserve tissue samples in cold storage at museums and research universities.  Archival cold storage allows for advancing technologies, common in the “-omics” fields, to quickly utilize existing research materials that would otherwise be impossible to access.  Dedicated systematists, field biologists, and curators who assemble and maintain collections, therefore, play a critical role in facilitating study of the natural world.

Future directions

Obligate parasites, through their close associations with host species, provide several advantages for studies of evolutionary and ecological processes beyond those described here⁷.  We plan to continue our research of louse-associated bacteria following the discovery of additional symbiont groups in lice parasitizing both birds and mammals.   

1. Gould, S. J. Wonderful life: The Burgess Shale and the Nature of History. (WW Norton & Company, 1990).
2. Blount, Z. D., Lenski, R. E. & Losos, J. B. Contingency and determinism in evolution: Replaying life’s tape. Science 362, eaam5979 (2018).
3. Moran, N. A. Accelerated evolution and Muller’s Rachet in endosymbiotic bacteria. Natl. Acad. Sci. U S A93, 2873–2878 (1996).
4. Wolf Y. I. et al. Genome reduction as the dominant mode of evolution. Bioessays 35, 829-837 (2013).
5. Smith, W. A. et al. Phylogenetic analysis of symbionts in feather-feeding lice of the genus Columbicola: evidence for repeated symbiont replacements. BMC Evol. Biol. 13, 109 (2013).
6. Fukatsu, T. et al. Bacterial Endosymbiont of the slender pigeon louse, Columbicola columbae, allied to endosymbionts of grain weevils and tsetse flies. Environ. Microbiol. 73, 6660–6668 (2007).
7. Clayton D. H. et al. Coevolution of Life on Hosts: Integrating Ecology and History. (Chicago University Press, 2015).