A fast, robust and tunable synthetic gene oscillator - The BioCircuits ...

doi: 10.1038/nature07389 SUPPLEMENTARY INFORMATION
A
B
1200
Fluorescence
(AU)
Fluorescence
(AU)
200
0 60 120
Time (min)
180
15
6
0 60 120
Time (min)
180
Supplementary Figure 5: Single-cell fluorescence trajectories for A, MG1655Z1/pZE12-yemGFP-ssrA
cells expressing LacI constitutively and containing neither positive or negative feedback loops (induced with
2 mM IPTG), or B, JS013 cells containing the negative feedback oscillator (induced with 0.6 mM IPTG).
Fluorescence measurements are given in arbitrary units that are consistent between the two experiments
shown. Trajectories are smoothed with a Savitsky-Golay filter. Note that the cells without feedback are
much brighter and do not show periodic fluorescence dynamics.
3.5 h at 37 ◦ C. Samples were washed with PBS by centrifugation, and flow cytometry analysis
was performed. Our expectation was that, given a circuit that produced oscillating cellular
fluorescence levels, some of the cells in an unsynchronized population would have fluorescence
levels intermediate to bright and dim levels. In contrast, a monostable circuit would result in a
well-defined unimodal distribution and a bistable circuit would result in a bimodal distribution
with few intermediate cells. A survey of inducer space using this method revealed potentially
oscillatory regions that we subsequently investigated in greater detail using flow cytometry and
microscopy (Supplementary Figs. 6–7).
Additional flow cytometry was performed by one of two similar protocols. The temperaturedependence
experiments followed a continuous timecourse flow cytometry protocol, in which a
single culture was induced at the initial timepoint and samples were removed for flow cytometry
analysis over the course of the experiment. An overnight culture of JS011 was diluted in growth
www.nature.com/nature
9
240