Fig. 1

FRET changes in response to AMP. a AMP-dependent variation of fluorescence emission spectra of AMPfret constructs 1.1 and 2.1 and AMPfret-negative control (CTL) (blue line: no AMP, orange line: 20 µM AMP). Spectra normalized to the isosbestic point at 512 nm. AMPfret topologies are depicted on top (bars: AMPK subunits, red box: putatively rigid helix, blue circles: cyan fluorescent protein mseCFP_Δ11, yellow circles: yellow fluorescent protein cp173-mVenus). b Normalized FRET ratio of AMPfret constructs calculated from a (same color code), and corresponding autoradiograms of in vitro kinase activity assays using acetyl-CoA carboxylase (ACC) as substrate. Data and error bars represent mean ± SEM (n ≥ 7, *p < 0.001, Student’s t-test). c Schematic drawing of AMPfret 2.1 operating mode. Within an α2β2γ1 AMPK heterotrimer (α: blue, β: orange, γ: pink), a CFP variant (blue circle) fused to a small rigid α-helix linker (red) was added at the α2-C-terminus, while a YFP variant (yellow circle) was added at the β2-C-terminus. Binding of AMP (and ADP) to the γ-subunit induces a conformational change which modifies distance and/or orientation between the fluorophore couple, thus increasing the non-radiative energy transfer (FRET; black wave) between them. Excitation/emission wavelengths are indicated. The model is based on AMPK core structures co-crystalized with either ATP (PDB ID: 4EAK) or AMP (PDB ID: 4EAI)¹⁶ and extended for missing parts by the full-length AMPK structure (PDB ID: 5ISO), thus visualizing some of the conformational changes (AMPK in schematic backbone representation, ATP and AMP in space filling representation). d Kinetics experiments of AMPfret 1.1 (left) and AMPfret control (CTL, right) responses to the addition of 10 µM AMP: the ratiometric FRET signals were measured every 10 s, corrected for cross-talk (CT = 0.20) and direct excitation (DE = 0.04) and normalized by the average signal before addition of AMP. Gray and black lines represent individual samples and their mean, respectively (n = 5)