A fluorescent reporter of AMPK activity and cellular energy stress

Abstract

AMP-activated protein kinase (AMPK) is activated when the AMP/ATP ratio in cells is elevated due to energy stress. Here, we describe a biosensor, AMPKAR, that exhibits enhanced fluorescence resonance energy transfer (FRET) in response to phosphorylation by AMPK, allowing spatiotemporal monitoring of AMPK activity in single cells. We show that this reporter responds to a variety of stimuli that are known to induce energy stress and that the response is dependent on AMPK α1 and α2 and on the upstream kinase LKB1. Interestingly, we found that AMPK activation is confined to the cytosol in response to energy stress but can be observed in both the cytosol and nucleus in response to calcium elevation. Finally, using this probe with U2OS cells in a microfluidic device, we observed a very high cell-to-cell variability in the amplitude and time course of AMPK activation and recovery in response to pulses of glucose deprivation.

Figure 1. Schematic representation of AMPKAR construct

(A) Phosphorylation of the optimized AMPK substrate motif on AMPKAR induces binding to the intrinsic FHA domain, thereby prompting a conformational change that is detected by a change in fluorescence resonance energy transfer (FRET) between eCFP and an YFP variant Venus. (B) Domain structures of AMPKAR and amino acid sequence of the optimized AMPK substrate motif.

Figure 2. AMPKAR FRET response to energy stress and calcium elevation

(A) COS-7 cells expressing AMPKAR were treated with 20 mM 2-DG. (Left) Mean normalized (Y/C) emission ratio (n=32). (Middle) Representative pseudocolor images of Y/C ratio show the FRET response. (Right) Western blots showed comparable time course for AMPK activation indicated by an increase in phospho-ACC (S-79) and the corresponding phospho-AMPK (T-172) levels. (B) COS-7 cells transfected with AMPKAR were treated with 1 µM ionomycin. (Left) Mean normalized (Y/C) emission ratio (n=15). (Middle) Representative pseudocolor images of Y/C ratio show the FRET response. (Right) Western blots showed comparable time course for AMPK activation indicated by increase of phospho-ACC (S-79) and the corresponding phospho-AMPK (T-172) levels. (C) T to A mutation in the substrate peptide phosphorylation site abolishes FRET. (Left) COS-7 cells expressing either AMPKAR(T) (n=30) or mutant AMPKAR(A) (n=8) were treated with 20 mM 2-DG. (Right) Cos-7 cells expressing either AMPKAR(T) (n=14) or mutant AMPKAR(A) (n=18) were treated with 1 µM ionomycin.

Figure 3. AMPKAR FRET signal is AMPK-dependent

(A) The AMPKAR FRET response is not observed in mouse embryonic fibroblasts that lack AMPK. MEFs lacking both AMPK-α1 and α2 (DKO) or DKO cells reconstituted with AMPK α2 subunit (α2) were treated with either 20 mM 2-DG (left) (DKO, n=6; a2, n=11) or 1µM ionomycin (right) (DKO, n=4; α2, n=13). All drugs were added at time zero. (B) Western blots of DKO vs α2 MEF treated under the same conditions as those used in the imaging experiments (C) The AMPKAR FRET response to energy stress requires LKB1. AMPKAR FRET signals were compared between HeLa cells reconstituted with LKB1 (HeLa_LKB1) or vector (HeLa_vector). Cells were treated with either 20 mM 2-DG or 1 µM ionomycin.

Figure 4. Subcellular targeting of AMPKAR

(A) Schematic domain structures of nuclear and cytosol targeted AMPKAR. The asterisk represents the localization of the AMPK substrate motif. YFP fluorescence image of AMPKAR-NLS (Left) or AMPKAR-NES expressing C2C12 cells. (B) Representative emission ratio time course of cytosolic (cy, n=20) versus nuclear targeted AMPKAR in response to 1 µM ionomycin (nu, n=27). (C) Representative emission ratio time course of cytosolic (cy, n=5) versus nuclear (nu, n=16) targeted AMPKAR in response to 20 mM 2-DG treatment. (D) Representative emission ratio time course of cytosolic (cy, n=6) versus nuclear (nu, n=19) targeted AMPKAR in response to 1 mM AICAR treatment. All drugs were added at time 0 as indicated by arrows.

Figure 5. AMPKAR exhibits reversible dynamics

(A) Manipulating extracellular glucose concentrations leads to reciprocal changes of FRET signals from AMPKAR and ICGS (intracellular glucose sensor FLII12Pglu-700uDelta6). COS-7 cells were transfected with either AMPKAR (right) or ICGS (left) and seeded on the microfluidic devices. Cells were perfused with a continuous flow of regular DMEM with 25 mM glucose for 4 minutes (time points 1~5), switched to glucose free DMEM (G=0) for 20 minutes, and then switched back to regular DMEM with 25 mM glucose (G=25). Each frame shows the FRET ratio at 1 minute intervals (left to right and top to bottom). (B) Tracings of the AMPKAR FRET response (Squares; average from 30 cells) and ICGS FRET response (Circles; average from 10 cells) from cells treated as described in 5A. The AMPKAR FRET signal from the control in which cells were continuously exposed to a flow of 25 mM glucose throughout the time course is also presented (Triangles; averaged from 7 cells).

Figure 6. AMPKAR reveals cell-to-cell variability in energy stress response

(A) AMPKAR reveals cell-cell variations in energy stress response. At time 0, U2OS cells stably expressing AMPKAR were switched from media with 25 mM glucose to media with either 0 glucose (6 boxes on left) or 3 mM glucose (6 boxes on right). After 10 minutes the cells were shifted back to 25 mM glucose, at 20 minute they were shifted back to the low glucose and at 30 minutes back to high glucose. The 12 boxes are representative single cell tracings of cellular FRET response to the 2 pulses of “low glucose”. (B) Distribution of maximal AMPKAR FRET response (Amp_MAX) from individual cells exposed to zero (G0) or 3 mM glucose (G3). (C) Definition of parameters in AMPKAR activity in response to low glucose stimulation. Gw denotes glucose withdrawal and G25 denotes 25 mM glucose in media. Amp_Max denotes the maximal FRET signal (normalized y/c ratio); T_1/2up denotes the time to reach half of maximal response (minutes), T_1/2d denotes the time to return to half of maximal fluorescence during FRET decrease (minutes). Subscript 1 and 2 denotes response from 1st and 2nd pulse, respectively. (D) Cell-cell variability measured as the CV (coefficients of variance: ratio of standard deviation to mean) of different aspects of cellular response to glucose starvation. Upper panel: AMPKAR FRET response upon shift to 3 mM glucose (G3). Lower panel: AMPKAR FRET response upon shift to zero glucose (G0). (E) While the median increase in the AMPKAR FRET amplitude is greater upon complete glucose removal compared to partial glucose removal, cells exhibit a similar response in the 2^nd pulse of low glucose to the response in the 1^st pulse, and the half times for the increase and decrease of the FRET signal are relatively constant. g0 denotes zero glucose; g3 denotes 3mM glucose; p1 and p2 denotes 1^st and 2^nd pulse of stimulation, respectively. (NS: statistically non-significant)