Genetic details, optimization and phage life histories

https://doi.org/10.1016/j.tree.2003.10.008Get rights and content

Abstract

Optimality models assume that phenotypes evolve by natural selection largely independently of underlying genetic mechanisms. This neglect of genetic mechanisms is considered an advantage by some evolutionary biologists but a fatal flaw by others. The controversy has gone unresolved, in part, from a lack of complex phenotypes that meet optimality criteria and for which the underlying genetic mechanisms are known. Here, we look at both perspectives for lysis time in bacteriophages. We find that the basic assumptions of the optimality model are compatible with the genetic details, but the optimality model is limited in its ability to accommodate lysis time plasticity because the mechanistic underpinnings of plasticity are poorly known.

Section snippets

A model of optimal lysis timing

Optimal lysis time was studied using a variant of the marginal value theorem [14] that derives the optimal period of time a phage should wait to lyse its host by comparing the marginal fecundity gained if the phage remains in the host with the fecundity gained by releasing its progeny in search of other hosts. Wang et al. [15] used this model to determine how various life-history parameters influenced the optimal lysis time (Box 3). The phage life history is partitioned into three phases in

Key assumptions and the relevance of genetic details

Although the Wang et al. [15] model lays the foundation for a theory of optimal lysis times, it rests on three crucial assumptions. We must be able to understand the genetics of lysis to assess the plausibility of these assumptions.

The genetics of lysis

Given that several key assumptions of the optimality model depend on the genetic details underlying lysis time, we now consider what has been discovered about the genetics of lysis. All bacteria have at least an inner membrane and an outer cell wall; the molecular skeleton for the cell wall is referred to as the peptidoglycan. (Gram-negative bacteria such as E. coli also have a second membrane, outside of the cell wall, and the region between the two membranes is called the periplasmic space.)

The impact of genetic details on optimality models

To illustrate how genetic details might augment optimality models, we return to the three assumptions that underlie the optimality model. In doing so, we ask how knowledge of the genetics of lysis informs the model of optimal lysis time in Box 3.

Conclusions

The major assumptions of the optimal lysis time model are largely compatible with genetic details of lysis. Thus, intimate knowledge of genetic details might not always be a prerequisite for informative optimality models. However, we have shown that genetic details can substantially refine optimality models by revealing which phenotypes are feasible and by uncovering constraints on evolution. Thus, phages illustrate the benefits of combining genetic mechanisms with an optimality approach.

Our

Acknowledgements

We thank S. Frank, E. Charnov, B. Kerr, H. Wichman, K. Pfennig, S. Abedon, J. Kingsolver, T. Juenger, B. Podolsky, K. Reeve, R. Heineman, D. Hall, J. Sachs and H. Tsitrone for comments on this paper. This work was supported by NIH GM57756 to J.J.B., who is also supported as the J. Friedrich Miescher Professor at the University of Texas.

Glossary

Glossary

Bacteriophage (phage):
a bacterial virus.
Capsid:
a shell of protein enclosing the genome of a virus particle.
Genetic details:
aspects of genetic systems (e.g. genes, epistasis, pleiotropy, genetic correlations, mutation rates) that may influence expression of phenotypes.
Holin:
a membrane protein encoded by the phage genome that regulates the timing of lysis.
Lysis:
a type of viral reproduction in which new viral particles are made inside the host cell and eventually burst out of the cell, killing the

References (41)

  • J. Maynard Smith

    Evolution and the Theory of Games

    (1982)
  • D.W. Stephens et al.

    Foraging Theory

    (1986)
  • A. Grafen

    Modelling in behavioral ecology

  • S.H. Orzack et al.

    Optimization models and the test of adaptationism

    Am. Nat.

    (1994)
  • J. Seger et al.

    Optimization and adaptation

  • P. Kitcher

    Vaulting Ambition

    (1985)
  • P. Hammerstein

    What is evolutionary game theory?

  • H.K. Reeve et al.

    Why we need evolutionary game theory

  • G.A. Parker et al.

    Animal behavior as a strategy optimizer: evolution of resource assessment strategies and optimal emigration thresholds

    Am. Nat.

    (1976)
  • I.N. Wang

    The evolution of phage lysis timing

    Evol. Ecol.

    (1996)
  • Cited by (61)

    • Coevolution, Bacterial-Phage

      2016, Encyclopedia of Evolutionary Biology
    • Evolutionary comparison between viral lysis rate and latent period

      2014, Journal of Theoretical Biology
      Citation Excerpt :

      According to classic competition theory (Tilman, 1982), this would indicate that viral populations releasing continuously their offspring (e.g. shedding viruses) are better competitors for any feasible combination of b and L. However, the fact that this strategy performs better in isolation does not necessarily mean that it can out-compete the delayed strategy when present in the same environment (Abedon et al., 2003). Moreover, the fact that delayed lysis and not shedding dominates in, e.g. marine environments highlights the limitation of these simplified models to produce reliable ecological and evolutionary predictions without the proper modifications (Bull et al., 2004; Komarova, 2007). On the other hand, the DM results in a smaller fitness than the rate-based release for any value of the latent period, including the ESS (Fig. 6, right).

    • Trade-offs in resource allocation in the intracellular life-cycle of hepatitis C virus

      2010, Journal of Theoretical Biology
      Citation Excerpt :

      Life-history theory aims to understand the trade-offs in the allocation of resources by an organism, including how materials and energy are acquired, processed and utilised (Stearns, 1992). Although viruses are often not considered living organisms (Campbell et al., 2006), they have complex replication machinery, which are subject to constraints (Bull et al., 2004; Krakauer and Komarova, 2003). Viruses must allocate resources to different stages of its intracellular life-cycle to maintain a productive infection.

    View all citing articles on Scopus
    View full text