Sex as an algal antiviral strategy

Emiliana huxleyi is a tiny eukaryotic alga that plays a huge role in the ecology and biogeochemistry of the world's oceans. Under favorable conditions it forms extensive blooms that are visible to mariners as milky-colored ocean waters, and that appear from space in LANDSAT images as chalky swirls that can cover thousands of square kilometers. The white color is caused by light scattered by numerous calcareous plates, called coccoliths, that armor the cell surface of E. huxleyi and give members of this group of algae their common name, the coccolithophores. The enormous amount of calcium carbonate sequestered in coccoliths during extensive algal blooms makes these algae important carbon pumps in the global carbon cycle. In this issue of PNAS, Frada et al. (1) describe a novel strategy used by E. huxleyi to evade viruses that contribute to the alga's boom-and-bust population cycles.

Algal-Viral Dynamics

Viruses play a major role in the dynamics of marine algal blooms, including the prominent ones formed by E. huxleyi. Recent estimates suggest that a large fraction of the primary productivity attributed to oceanic algae is lost to viruses that infect and kill algal cells (2). Other evidence suggests that viruses can cause much of the mortality of algal cells that ultimately causes the collapse of marine algal blooms (2). E. huxleyi is no exception to this, because it is known to be infected by a group of giant phycodnaviruses called EhVs (3). Given the intense selective pressure that EhVs should impose during the collapse of algal blooms, there should be strong selection for resistance to viral attack by the most abundant algae, and equally strong selection on viruses to circumvent algal resistance. Comparable rapid evolution of resistance to viruses is well known in laboratory studies of interacting populations of bacteria and phage (4). Such coevolutionary arms races, fancifully termed “Red Queen” dynamics by Lee Van Valen (5), form the core of traditional thinking about the evolutionary and population dynamics of predators and prey, diseases and hosts, and similar sets of species that act as exploiters and victims. The net result of Red Queen dynamics should be rapid evolution that simply maintains the ecological status quo. However, Frada et al. (1) show in their recent paper that there is another fascinating way for E. huxleyi to evade their viral enemies. This mechanism involves an induced transition from diploid bloom-forming cells to a morphologically distinct haploid life history stage that is apparently immune to some of the viruses that infect diploid cells.

E. huxleyi has a complex life cycle that alternates between nonmotile diploid coccolith-bearing cells that form extensive blooms, and motile flagellated haploid cells that are completely covered by organic scales but that lack coccoliths. Frada et al. (1) show that the haploid phase of the complex life cycle provides a refuge from viral attack, creating a resistant reservoir of haploid cells that can mate, repopulate the diploid phase of the life cycle, and presumably go on to create new blooms. Instead of rapidly evolving immunity to viral attack, as would be predicted by Red Queen dynamics, the production of haploid cells effectively makes E. huxleyi invisible to viral attack, leading Frada et al. to coin the term “Cheshire Cat dynamics” for the resulting decoupled interaction between haploid algae and viruses. Cheshire Cat dynamics have interesting implications for understanding the dynamics of boom-and-bust cycles in dominant oceanic algae, and also illustrate a previously unappreciated set of conditions that can select for the maintenance of sexual reproduction in organisms that remain capable of very rapid clonal population growth.

The production of haploid cells effectively makes E. huxleyi invisible to viral attack.

Red Queen Dynamics

Red Queen dynamics offer one of the more compelling explanations for the evolution and maintenance of sexual reproduction. Although many ideas have been proposed to explain the selective advantage conferred by sexual reproduction (6–8), much of the empirical evidence seems to support ideas that invoke the Red Queen. That evidence takes the form of correlations between the frequency of parasitism and sexual reproduction in populations that contain a mixture of sexually and asexually reproducing individuals (9), correlated spatial and temporal shifts in the genetic composition of parasites and their hosts (10, 11), and the potential for more rapid genetic change in sexually reproducing organisms (12, 13). Cheshire Cat dynamics provide a different and previously unrecognized advantage to sexual reproduction, as the induction of the haploid phase of the life cycle confers an invulnerable refuge against viruses that are known to attack the diploid phase of the life cycle. This advantage would also tend to oppose the disadvantage imposed by deleterious recessive alleles exposed to selection in the haploid phase of the life cycle. This purging of deleterious mutations has been proposed as another important consequence of sexual reproduction (14).

The results of Frada et al. (1) raise interesting questions about the dynamics of natural E. huxleyi blooms. One curious feature of E. huxleyi dynamics is that the same genetically distinct strains of diploid cells tend to recur in different blooms that form over time. This pattern is inconsistent with Red Queen dynamics, which would produce genetically different resistant strains over time, with each strain eventually being decimated by viruses that evolve ways to circumvent the resistance. However, if bouts of sexual reproduction reshuffle the combinations of genes in the E. huxleyi genome, how do the same recurrent dominant diploid strains reassemble from a pool of haploid gametes? One must imagine that an extraordinary number of diploid progeny are produced to reassemble those strains by chance, certainly possible given the enormous population sizes of microbes. Or, perhaps the recurrent genetic strains that consistently form blooms are cryptic species with rather restricted ranges of genetic variation and adaptation, despite the presence of sexual reproduction. Obviously, more study is needed to resolve the apparent contradiction presented by the genetic diversity that should be produced by sexual recombination and the reduced genetic diversity in bloom-forming strains.

Another interesting problem concerns the factors that prevent bloom formation by the haploid phase of the life cycle. Frada et al. (1) challenged haploid and diploid algae only with viruses that have been isolated from diploid algae. This raises the question of whether haploid E. huxleyi may have their own, as yet unrecognized, set of viral pathogens that limit their ability to form blooms. Of course, it is also possible that blooms of haploid E. huxleyi occur but remain unrecognized because of the current limits of oceanographic sampling methods.

The mechanisms that confer viral resistance on the haploid phase of the E. huxleyi life cycle also remain uncertain. Frada et al. (1) suggest that the organic scales that closely cover the surface of haploid cells provide a possible mechanical defense against attachment by viral particles. As the authors point out, other examples of evolved resistance to viral attack usually involve changes in the composition of cell surface sites that promote attachment by viral particles. It is unclear how the simple meiotic transition from diploid to haploid would accomplish such a genetic change.

E. huxleyi is only one of a host of organisms that possess a complex life cycle that alternates between haploid and diploid phases. It will be fascinating to learn whether the advantages of Cheshire Cat dynamics are a general feature of most organisms with similar life cycles, or whether they are peculiar to E. huxleyi.