Extension of a genetic network model by iterative experimentation and mathematical analysis

Abstract

Circadian clocks involve feedback loops that generate rhythmic expression of key genes. Molecular genetic studies in the higher plant Arabidopsis thaliana have revealed a complex clock network. The first part of the network to be identified, a transcriptional feedback loop comprising TIMING OF CAB EXPRESSION 1 (TOC1), LATE ELONGATED HYPOCOTYL (LHY) and CIRCADIAN CLOCK ASSOCIATED 1 (CCA1), fails to account for significant experimental data. We develop an extended model that is based upon a wider range of data and accurately predicts additional experimental results. The model comprises interlocking feedback loops comparable to those identified experimentally in other circadian systems. We propose that each loop receives input signals from light, and that each loop includes a hypothetical component that had not been explicitly identified. Analysis of the model predicted the properties of these components, including an acute light induction at dawn that is rapidly repressed by LHY and CCA1. We found this unexpected regulation in RNA levels of the evening-expressed gene GIGANTEA (GI), supporting our proposed network and making GI a strong candidate for this component.

Figure 1

The single-loop LHY/CCA1–TOC1-X network. Left panel: Network diagram. LHY and CCA1 are modelled as a single gene, LHY (genes are boxed). Nuclear and cytoplasmic protein levels are grouped for clarity (shown encircled) and degradation is not shown. Light acutely activates LHY transcription at dawn and activates TOC1 transcription throughout the day. TOC1 activates a putative gene X, which in turn activates LHY. Nuclear LHY protein represses TOC1 transcription. Right panel: Simulation of mRNA levels for the optimal parameter set. In all figures, filled box above the panel represent dark interval and open or no box represent light interval. LHY mRNA (dotted line) peaks at dawn in LD12:12 and TOC1 (solid line) falls after dusk, due to the loss of light activation.

Figure 2

Expression of CCR2:LUC+ in WT (filled diamonds) and cca1-11;lhy-21 double mutant (open diamonds) plants in LL (top) and DD (bottom). Luminescence of each seedling was normalised to its mean value over the entire time course. Data are averages of normalised luminescence from WT seedlings in LL n=16, in DD n=18, cca1;lhy seedlings in LL n=13, in DD n=15. Error bars represent one s.e.m., often within symbols.

Figure 3

PTC for WT (left panel) and cca1;lhy double mutant (right panel). Red light pulses (15 μmol m⁻² s⁻¹ for 1 h) were administered at 3 h intervals to CCR2:LUC+ plants in DD. The new phase of the rhythm induced by the light pulses was converted to circadian time (CT, 24ths of the free-running period) and plotted against the circadian time of light treatment (solid lines). Simulated phase responses are represented by dashed lines, and show simulated response of the interlocked feedback loop model to a 1 h light pulse. Phase marker for simulation was TOC1 mRNA peak, compared to CCR2:LUC+ peak in data.

Figure 4

The interlocked feedback loop network. Left panel: Network diagram. Compared to Figure 1, TOC1 is activated by light indirectly via hypothetical gene Y. Y activates TOC1 transcription and both LHY and TOC1 repress Y transcription, forming a second feedback loop. Right panel: Simulation of LHY (dashed line) and TOC1 (solid line) mRNA levels for the optimal parameter set, representing WT (top) and cca1;lhy double mutant (bottom) in DD. Translation rate of LHY mRNA in simulated mutant is 1/1000 WT value. Period of WT in DD is 26 h and period of mutant is 17 h.

Figure 5

Comparison of interlocked feedback loop simulations (dashed line) under LD to data (solid line). (A) TOC1 mRNA levels in WT plants entrained to LD12:12, left axis; TOC1 mRNA levels relative to UBIQUITIN (UBQ) (Makino et al, 2000), right axis. (B) LHY mRNA levels in WT plants entrained to LD12:12, left axis; data from Kim et al (2003), right axis. (C–E) TOC1 mRNA levels in WT (C), cca1;lhy mutant (D) and cca1 mutant (E) entrained to LD16:8; data from Mizoguchi et al (2002). Translation rate of LHY in simulated cca1 is set to 1/2 WT value. Highest value of data and simulation is set to 1, for each panel. (F) TOC1 protein levels for WT simulation entrained to LD12:12; TOC1 fusion protein data from Mas et al (2003b).

Figure 6

GI is a candidate gene for Y. Simulated Y mRNA levels under LD12:12 and LL (dashed line). Data for GI mRNA levels (crosses), assayed by quantitative RT–PCR relative to the ACT2 control, from samples harvested at the times indicated. Left panel, WT; right panel, cca1;lhy. Highest value of data and simulation is set to 1, for each panel.