Skip to main content
Figure 2 | BMC Biology

Figure 2

From: Real-time determination of intracellular oxygen in bacteria using a genetically encoded FRET-based biosensor

Figure 2

In vivo analysis of FluBO fluorescence lifetime. (A) FLIM pictures of E. coli cells expressing either FbFP or FluBO. Cells were cultivated until they reached the stationary growth phase. Fluorescence was observed at the maximum of FbFP (λobs < 500 nm; λexc = 760 nm). Two-photon excitation of FluBO fluorescence was tested at different wavelengths and was found best at 760 nm, corresponding the doubled wavelength of the normal one-photon excitation optimum (380 nm, see Figure 1D). The figure shows the fluorescence lifetimes (τave) of FbFP (left panel) and FluBO (right panel) in living E. coli cells. (B, C) Analysis of the fluorescence decays of FluBO and FbFP expressed in E. coli under aerobic conditions (data are derived from images shown in Figure 2A). The fluorescence decay of FbFP was satisfactorily analyzed with a monoexponential decay function (F(t) = a·exp(-τ1/t); τave = τ = 2.73 ns) while the fluorescence decay of FluBO needed a biexponential decay for correct description (F(t) = a1·exp(-τ1/t)+ a2·exp(-τ2/t); τave = (a1·τ1+a2·τ2)/(a1+a2) = 1.74 ns); τ1 = 1.03 ns, τ2 = 2.72 ns. Due to FRET, the average fluorescence lifetime τave of FbFP in FluBO is reduced by 0.99 ns compared to FbFP alone. Using the average lifetimes of FbFP and FluBO and the equation E = 1-τave, FluBOave, FbFP an apparent FRET efficiency of 37% is calculated (B). The shorter lifetime (mean τ1 = 1.03 ns) determined form detailed analysis of the FluBO-FLIM experiments exhibited a major amplitude of 67%. Hence, a FRET efficiency of approximately 62% within the FluBO proteins that undergo FRET can be deduced from τ1 (E = 1-τ1ave, FbFP). The remaining 33% of FluBO molecules behave like single isolated FbFP molecules with a long lifetime of 2.72 ns (C).

Back to article page