The Centrality of a Single-Cell Stage

The life cycle of all organisms sooner or later passes through at least once single-celled stage. Surely this is more than just a coincidence. There are at least three biological hypotheses as to why a single-cell bottleneck would be advantageous. First, Crow (1988), among others (e.g., Bell and Koufopanou 1991), has argued that the single-cell sets essentially a tabula rasa for a new generation. The alternative of reproducing by vegetative (multicellular) propagules or organs leads to lineages where somatic mutations accumulate irreversibly. Selection to eliminate loaded individuals will occur but would not be very effective because it is acting on some mean of mutations per cell of the lineage overall.

In contrast, single cells (nucleus or spore) will be exposed to selection independently; the variance in fitness among individual cells exceeds that of the means of a collective of cells, so selection acts more efficiently in this case (Bell and Koufopanou 1991). Impaired cells will be quickly eliminated from the population. Moreover, in many lineages the single cell is haploid, thus deleterious alleles are not masked. For the above reasons, when reproduction occurs—whether by sexual (zygote) or asexual (spore) means—it is advantageous in evolutionary terms to package the products as single cells.

The second explanation relates broadly to competition among cell lineages in an individual and the additional source (beyond mutation) of biological heterogeneity introduced by fusion of cell lines (chimera) and presence of intracellular or extracellular symbionts, including pathogens (Grosberg and Strathmann 1998, 2007). The occurrence of multiple sub- organismal levels of selection has been a recurrent evolutionary theme (e.g., Buss 1982, 1985, 1987; Maynard Smith 1988). The replicating units subject to selection range from genetic element (transposons, etc.; Chap. 2) through gene, organelle (plastids, mitochondria), to cell lineage.

Passage through a unicellular bottleneck can reduce conflicts of interest and enhance compatibility among replicators by eliminating some levels (i.e., those operating extracellularly) of variation and aligning the interests of those units that remain. They have a commonality of descent. As Dawkins (1982, p. 264) has worded it ... “[But] a relationship of specially intimate mutual compatibility has grown up between subsets of replicators that share cell nuclei and, where the existence of sexual reproduction makes expression meaningful, share gene- pools’’

The third reason for a single-cell stage is a developmental one and has been articulated most imaginatively again by Dawkins (1982, pp. 253-262). He asks whether it is better for a growing organism to expand indefinitely as a single entity or as multiple small entities. The first strategy represents growth (G) as in the case of a hypothetical lily pad-like organism several miles wide on a sea of nutrients. The alternative is a plant that stops growing at a diameter of one foot and reproduces by single-celled propagules (sexual or asexual) cast to the wind. Each initiates a new thallus and repeats the process indefinitely. This strategy represents reproduction (R). So G only grows, while R alternately grows and reproduces. Which is better developmentally?

Dawkins argues that descendants of R can evolve complex features in a manner that those of G cannot. Each young, expanding R unit can differentiate and potentially outperform its parental colony, and at each turn of the cycle new variations can be introduced. This is a specific illustration of the principle that a complex state must evolve from a less complex condition. In contrast, each cell of G is destined to assume only a small role subordinated amidst the massive, expanding lily pad. Novelties can be made at the cellular but not the multicellular level because there is no passage through a structurally simple stage that would enable more complicated, organ-level innovation to occur.

Dawkins’ metaphor is that whereas it takes only a little tinkering to transform a Bentley into a Rolls Royce late on the assembly line, to go from a Ford to a Rolls Royce you have to start with new blueprints. Indeed Dawkins argues that the large, complex multicellular organism is possible because of recurrent life cycles that include a single cell where mutations can effect wholesale change, a process that works only by "repeated returns to the drawing board during evolutionary time" (p. 262).