Evaluating the Adaptive Fitness of Circadian Clocks and their Evolution

Abstract

Surely most chronobiologists believe circadian clocks are an adaptation of organisms that enhances fitness, but are we certain that this focus of our research effort really confers a fitness advantage? What is the evidence, and how do we evaluate it? What are the best criteria? These questions are the topic of this review. In addition, we will discuss selective pressures that might have led to the historical evolution of circadian systems while considering the intriguing question of whether the ongoing climate change is modulating these selective pressures so that the clock is still evolving.

Keywords: chronobiology; circadian; competition assay; cyanobacteria; evolution; fitness; phase angle.

Figure 1.

The key significance of phase angle and examples of competition assays. (a) For circadian clocks, phase angle (the phase relationship of the biological clock to the environmental cycle) is the key to fitness. (b) Phase angle is a function of the FRP of a circadian rhythm, here shown as the phase relationship of a rhythm to a consistent 12:12 light/dark cycle (LD12:12) as FRP is varied. (c) Phase angle of a circadian rhythm is a function of the period (T) of an environmental cycle, here shown as the phase relationship of a circadian rhythm with an intrinsic FRP = 24.2 h as it entrains to environmental cycles of different T (T-20 = LD10:10, T-24 = LD12:12, T-30 = LD15:15). (d) A classic test of the effect of T-cycles (T-12, T-24, and T-48) on the growth of tomato plants (Highkin and Hanson, 1954), reprinted by permission of Plant Physiology. (e) Competing cyanobacteria: when strains of cyanobacteria with different FRPs (FRPs of 23, 25, and 30 h) are competed under different T-cycles (T-22 and T-30), the strains whose FRPs “resonate” with the environmental cycles (adopt optimal phase angles) outcompete other strains. Mutant strains are able to outcompete wild-type strains if their FRPs harmonize more appropriately with the T-cycle than the FRP of the wild-type strain (Ouyang et al., 1998; Woelfle et al., 2004). See the “The Competition Assay applied to Cyanobacteria” section. (f) Competing plants: similar to the case of cyanobacteria, the growth and mortality of Arabidopsis plants are promoted under aligned combinations of FRP and T-cycles (Dodd et al., 2005). See “The Competition Assay Applied to Arabidopsis” section. (g) Competing mice: In a competition experiment, the Csnk1e^tau allele frequency—causing FRP to shorten by 2 h in heterozygote mice and by 4 h in homozygote mice—gradually decreases in mouse populations over multiple generations (Spoelstra et al., 2016). Figures depict proportions of the genotype in replicated populations in outdoor enclosures. See the “The competition assay applied to mice” section. Abbreviation: FRP = free-running period.

Figure 2.

Supporting evidence for the adaptiveness of clocks. A schematic depicting various assays used for inferring the evolutionary fitness of clocks ordered in a hierarchy based on the information they provide on reproductive success (vertical) and variations in these assays based on experimental rigor involved (horizontal). More information on advantages and drawbacks is included in Table 1.

Figure 3.

Competition assay applied to cyanobacteria. (a) Competing cyanobacteria: when strains of cyanobacteria with different FRPs are competed under different T-cycles, the strains whose FRPs “resonate” with the environmental cycles are predicted to outcompete the strains whose FRPs do not resonate (Ouyang et al., 1998; Woelfle et al., 2004; Ma et al., 2013). (b) Competing cyanobacteria: When a wild-type strain of cyanobacteria is competed against a clock-less mutant, the wild-type strain is predicted to win in T-24 LD conditions. In LL, wild type should win if maintenance of an ITO is important (intrinsic), or has no selective impact in LL if ITO is less important (extrinsic). (c) The winner of a competition experiment can be determined through hard selection (top panel) and/or soft selection (bottom panel) on the basis of comparing single-strain populations versus mixed-strain populations. In the case of hard selection, the environment (LD12:12) negatively impacts the fitness of a strain (clock-less), but not that of wild type, and this effect can be observed in single-strain, non-competing population. This results in lower growth of the clock-less strain under the selective LD12:12 condition but not in the non-selective LL and can eventually lead to wild type outcompeting the clock-less strain in a mixed culture. On the other hand, no difference in growth is observed between LD12:12 and LL in single-strain populations under soft selection. Only when the two strains are competed against each other in mixed-strain populations is the growth and fitness of the clock-less strain reduced. (d) Competing cyanobacteria: when strains of cyanobacteria with different FRPs (FRPs of 23, 25, and 30 h) are competed under different T-cycles (T-22 and T-30), whichever strain has the FRP that “resonates” the most with the environmental cycle (and likely adopts the optimal phase angle of entrainment) is the winner of the competition (Ouyang et al., 1998; Woelfle et al., 2004). Abbreviations: FRP = free-running period; ITO = internal temporal order.

Figure 4.

Circadian clocks contribute to the fitness of Arabidopsis plants. (a) Decreased growth and biomass accumulation of arhythmic (CCA1-ox) plants compared to the wild type, grown under T-24 (LD12:12). (b) For Arabidopsis plants grown in single-strain monocultures, there was decreased growth and biomass accumulation when the endogenous circadian period length differed from the environmental T-cycle length. (c) For Arabidopsis period-length mutants grown in competition, there was also decreased growth and biomass accumulation, combined with greater mortality, when the endogenous circadian period differed from the T-cycle length. Diagram shows representative plants. Abbreviation: FRP = free-running period.

Figure 5.

Circadian clocks contribute to the fitness of mice. (a) Survival curve of mice with mutant “tau” (Csnk1e^tau) alleles in a competition experiment in outdoor enclosures. (a) The 3 lines show the surviving proportion of the 3 genotypes: wild-type (+/+) mice with a ~24-h FRP, heterozygote (tau/+) mice with a ~22-h FRP, and homozygote (tau/tau) mice with a ~20-h FRP. The latter genotype lives significantly shorter. (b-d) Laboratory (running wheel) activity patterns of the 3 mouse genotypes (which are highly similar to those of tau-mutant hamsters; Loudon et al., 2007) in LD12:12 conditions during day 1-15 and free-running in DD between day 15 and 25, clearly showing the effect of the mutation on the FRP. The misalignment of heterozygous mice to dissonant LD conditions strongly compromises longevity. (e-g) Outdoor enclosure activity patterns (transponder recording at feeding stations) of the three genotypes (tested over 424 days). Here, rhythms are less pronounced and in contrast to heterozygous mice in the laboratory, and longevity is compromised in homozygous mutant mice. This may be caused by other selective pressures, for example, by aerial predation. Abbreviation: FRP = free-running period.