Supplemental Data

Files in this Data Supplement:

• Supplemental Figure 1 - The late flowering phenotype of the gi-11 mutation can be complemented with a WT GI cDNA. Flowering time of the WT, gi-11 and two independently transformed lines of the gi-11 mutant with a 35S:GI construct, was measured under long days (16L:8D). Expression of the WT GI construct restored the WT flowering phenotype in the gi-11 background.
• Supplemental Figure 2 - Temperature compensation of leaf movement rhythms for Ler WT and Ler gi-3 Seedlings were entrained under 12L:12D cycles for 7 days after which seedlings were transferred to constant light at either 12, 17, 22 or 27°C where rhythms of leaf movement were assayed. For each temperature 25-30 WT and 25-30 gi mutant plants were assayed corresponding to 50-60 leaf movement rhythms. Graphs of period estimates for individual leaves plotted against their relative amplitude error for either 12, 17, 22 or 27°C. This experiment was performed independently two times at each of the 4 temperatures and this experiment is representative.
• Supplemental Figure 3 - Amino acid substitutions in the N-terminal half of the GI protein can lengthen or shorten circadian period but do not result in late flowering. The gi mutants gi-611 and gi-596 were isolated from an EMS-mutagenised population carrying the CAB:LUC+ transgene in the WS background, in a genetic screen for plants with altered temporal regulation of CAB:LUC+ expression. Genetic mapping (>90 chromosomes) showed that each mutation was tightly linked to GI; sequencing revealed G-A transition mutations in the GI coding region in both lines, leading to predicted amino acid changes L₂₈₁→F in gi-611 and S₁₉₁→F in gi-596. Under greenhouse conditions with extended photoperiods, both gi-611 and gi-596 flowered within the WT range of leaf numbers, in contrast to gi-11, which flowered late after producing more than three times the wild type leaf number (data not shown). ( A, B) Flowering times of gi-611, gi-11 and WT plants under long (A, 16L:8D) and short (B, 8L:16D) photoperiod conditions. gi-611 is early-flowering in short photoperiods, consistent with its shorter circadian period, but appears wild-type in long photoperiods, in contrast to gi-11. (C, D, E) Transgenic seedlings carrying the CAB:LUC reporter gene were entrained under 12L:12D cycles for 7 days at 22°C, after which the seedlings were maintained at 22°C and luminescence was monitored by low-light imaging (C, E) or in an automated counter (D). (C, D) Seedlings were transferred at the start of luminescence analysis (time 0h) to constant red + blue light. (E) seedlings were imaged during 12L:12D cycles of red light and transferred to constant red light at time 72h: open box on time axis represents red light, filled box, darkness. Period estimates for these data are (C) gi-611 22.3h, WT 23.5h; (D) gi-596 27.6h, WT 26.0h; (E) gi-611 20.1h, gi-596 23.7h, WT 20.9h. n = 12 to 50 seedlings. WT period varies due to the light fluence rates: light levels are lower in the automated counter (D) than in (C). (F) Variable weighted mean of period estimates plotted against temperature, error bars represent variance weighted standard error of the mean (SEM). (WT, filled squares; gi-611, open triangles).
• Supplemental Figure 4 - At 4°C the oscillations of key clock genes dampen rapidly. The Micro-array data was produced as part of the AtGenExpress initiative and made publicly available via TAIR (TAIR accession number ExpressionSet:1007966553, NASCArrays Experiment Reference Number: NASCARRAYS-138) . Plant material from 18 days old Arabidopsis thaliana plants of Col-0 ecotype were either transferred to constant light and 22°C or constant light and 4°C. Shoots were harvested at the following time points: 0,5h; 1h; 3h; 6h; 12h; 24h. The data was normalised using RMA and the expression of key clock genes plotted for plants at 22°C and 4°C.
• Supplemental Figure 5 - Modelling the temperature compensation mechanism at high temperatures. The interlocked feedback loop model of Locke et al (Locke et al., 2005) was used to simulate temperature effects on the TOC1 and LHY mRNA profile at 17°C and 27°C for both WT and gi backgrounds. (See Supplemental Methods) (A) Simulated TOC1 mRNA profile after 72 hours in constant light for both WT and gi backgrounds (B) A reproduction of the cognate experimental results (figure 3A) for ease of comparison. (C) Simulation of LHY mRNA levels at 17°C and at 27°C for both WT and gi backgrounds. A direct comparison to experimental results for LHY and CCA1 (Figure 3C,) is not possible, as LHY and CCA1 are grouped as a single gene, LHY, in simulations. (D) Simulation of TOC1 mRNA levels under 12;12 LD conditions at17°C and 27°C for WT background.
• Supplemental Figure 6 - Modelling the temperature compensation mechanism at low temperatures. The interlocked feedback loop model of Locke et al (Locke et al., 2005) was used to simulate temperature effects on the TOC1 and LHY mRNA profile at 17°C and 12°C for both WT and gi backgrounds. (See Supplemental Methods) (A) Simulated TOC1 mRNA profile after 72 hours in constant light at 17°C and 12°C for both WT and gi backgrounds. (B) A reproduction of the cognate experimental results (figure 3B) for ease of comparison. (C) Simulation of LHY mRNA levels at 17°C and at 12°C for both WT and gi backgrounds.
• Supplemental Methods