Circadian clock genes, ovarian development and diapause

Ikeno et al. suppressed the expression of the clock genes cycle and period by injecting bean bugs with double-stranded RNA (RNAi), showing by northerns that this achieved at least partial downregulation of the targeted gene, and also a reduction in per mRNA after cyc RNAi treatment, which would be expected as a secondary consequence of reduced levels of CYC (Figure 1 in [6]). As a read-out for the function of the circadian clock they looked at the layering of the cuticle, which appears as alternating bright and dark layers under polarized light, and is laid down in a rhythm they established as showing the classic features of being regulated by a circadian clock (Figure 1 in [6]). Injection of cyc RNAi resulted in the loss of alternate layering and instead the deposition of a single bright layer. Injection of per RNAi also resulted in the loss of alternate layering, but the outcome was a single dark layer (Figure 3 in [6]). Thus, in both cases there was a loss of circadian rhythm, but with a different phenotypic consequence, and the authors interpret this as an arrest of the clock in two different phases, leading to the activation of distinct downstream cascades.

The remarkable result was that per and cyc RNAi also affected ovarian diapause. Normally R. pedestris undergoes an ovarian diapause that is controlled by the length of day (Figure 2). Long days promote ovarian maturation by inhibiting neurohormones from the brain that prevent synthesis or secretion of juvenile hormone from the corpora allata [8]. Short days permit expression of the inhibitory neurohormones and, ultimately, result in non-developing ovaries (diapause). When there were significant differences from controls, per RNAi increased the incidence of ovarian development and cyc RNAi decreased the incidence of ovarian development (Figure 4 in [6]). Consistent with these results, and indicating an action on the regulatory cascade leading to diapause at the level of juvenile hormone or upstream, per and cyc RNAi had opposite effects on expression of genes known to be up- or down-regulated by juvenile hormone (Figure 5 in [6]). Furthermore, the application of a juvenile hormone analog (methoprene) restored ovarian development in cyc RNAi bugs (Figure 6 in [6]). Thus, cyc is involved in regulating diapause somewhere between the input of light and the secretion of juvenile hormone in the corpora allata (Figure 2).

Ikeno et al. see parallels in their results that lead them to claim support for Bünning's hypothesis. They suggest that having targeted the principal positive regulator of the bean bug's clock (cyc) and a negative regulator (per), they have stopped the circadian clock at different phases of its oscillating cycle. It is unable to oscillate in response to the night-day cycle, but having become stuck in opposing phases, output signals are still delivered to give phenotypes that correspond, in the case of the cuticle, to those normally associated with opposite swings of the night-day pendulum. In the case of diapause, the differences in sensitivity to day length are interpreted as disruption of the photoperiodic timer, with the switch for diapause stuck in either one of two opposing positions. Thus, they see support for Bünning's proposition that the circadian clock mechanism lies at the heart of photoperiodicity.

These are, however, speculative suggestions only, venturing beyond the data, and in our view biased by the intrinsic appeal of connecting the circadian clock and photoperiodic time measurement. Based on the data, we would make the following points. First, both RNAi treatments lead to a reduction of per expression, so that any phenotypic differences observed must be attributed largely to the difference in cyc expression. Second, both treatments rendered the circadian clock dysfunctional. Third, all of the phenotypes that varied between the two RNAi treatments did so in the consistent absence of a functional circadian clock. If there is consistently no functional circadian clock in both RNAi treatments, then phenotypic differences between those treatments cannot be ascribed to the circadian clock. We would therefore conclude that the circadian clock as a functional unit (module) does not provide the essential clockworks for photoperiodic time measurement in R. pedestris.

Just because the circadian clock is dysfunctional does not mean that individual clock genes have no other pleiotropic effects [9]. The diapause response of bugs exposed to long and short days (Figure 4 in [6]) can be considered in terms of the presence of cyc expression (per RNAi treatment) or absence of cyc expression (cyc RNAi treatment). When cyc is expressed, there is an increase in ovarian development (non-diapause) compared to when cyc is not expressed, regardless of day length. The application of a juvenile hormone analog tells us that cyc is not acting at the level of the ovaries themselves. However, these results do not tell us where cyc is exerting its effect. This could be anywhere in the cascade of events (Figure 2), from the input of light [10] to the secretion of juvenile hormone in the corpora allata. Given that cyc encodes a positive transcriptional regulator, multiple pleiotropic effects are possible, and even likely.

In sum, Bünning's hypothesis remains in contention but unproven in insects. Nonetheless, Ikeno et al. [6] have made significant advances in the understanding of the expression of diapause. They have shown in R. pedestris that the circadian clock gene cyc plays an important role in ovarian development or diapause, and that it is involved somewhere in the upstream part of the pathway of juvenile hormone and not at the level of the ovaries themselves (Figure 2). Finally, they have added cycle in Heteroptera to timeless in Diptera [5] as core circadian clock genes that also have independent effects - on insect diapause. Their research establishes cyc as an important and interesting focus for future research.