Apparent PKA activity responds to intermittent hypoxia in bone cells: a redox pathway?

Abstract

We studied hypoxia-induced dynamic changes in the balance between PKA and PKA-counteracting phosphatases in the microfluidic environment in single cells using picosecond fluorescence spectroscopy and intramolecular fluorescence resonance energy transfer (FRET)-based sensors of PKA activity. First, we found that the apparent PKA activity in bone cells (MC3T3-E1 cells) and endothelial cells (bovine aortic endothelial cells) is rapidly and sensitively modulated by the level of O(2) in the media. When the O(2) concentration in the glucose-containing media was lowered due to O(2) consumption by the cells in the microfluidic chamber, the apparent PKA activity increases; the reoxygenation of cells under hypoxia leads to a rapid ( approximately 2 min) decrease of the apparent PKA activity. Second, lack of glucose in the media led to a lower apparent PKA activity and to a reversal of the response of the apparent PKA activity to hypoxia and reoxygenation. Third, the apparent PKA activity in cells under hypoxia was predominantly regulated via a cAMP-independent pathway since 1) changes in the cAMP level in the cells were not detected using a cAMP FRET sensor, 2) the decay of cAMP levels was too slow to account for the fast decrease in PKA activity levels in response to reoxygenation, and 3) the response of the apparent PKA activity due to hypoxia/reoxygenation was not affected by an adenylate cyclase inhibitor (MDL-12,330A) at 1 mM concentration. Fourth, the immediate onset of ROS accumulation in MC3T3-E1 cells subjected to hypoxia and the sensitivity of the apparent PKA activity to redox levels suggest that the apparent PKA activity change during hypoxia and reoxygenation in this study can be linked to a redox potential change in response to intermittent hypoxia through the regulation of activities of PKA-counteracting phosphatases such as protein phosphatase 1. Finally, our results suggest that the detection of PKA activity could be used to monitor responses of cells to hypoxia in real time.

Fig. 1.

Dynamics of metabolic O₂ consumption by cells. Quenching by O₂ is reduced at lower concentrations, leading to stronger phosphorescence. O₂ exchange between the media and outside environment was inhibited at time 0. The solid line is a monoexponential fit with a time constant of 114 ± 1 min.

Fig. 2.

Dynamics of PKA activity in cells in response to depletion of O₂ and reoxygenation. The O₂ depletion was due to natural cellular metabolism. Stopping the flow through the chamber induced a rapid increase in the AKAR3 fluorescence resonance energy transfer (FRET) signal in MC3T3-E1 cells (A) and bovine aortic endothelial cells (BAECs; B), respectively. Reoxygenation by resuming the flow induced a rapid decrease in the AKAR3 FRET signal. The increased AKAR3 FRET signal indicates increased apparent PKA activity. The perfusion media was DMEM. The initial normoxic condition was maintained by keeping the flow rate at 62.5 μl/min.

Fig. 3.

Dynamics of PKA activity in cells in response to depletion of O₂ and reoxygenation. O₂ depletion was achieved by perfusion of cells with media of controlled Po₂. Perfusion with degassed media (DMEM, Po₂ < 60 mmHg) increased the AKR3 FRET signal, whereas perfusion with air-equilibrated media (DMEM, Po₂ = 160 mmHg) induced a rapid decrease in the AKAR3 FRET signal.

Fig. 4.

Dynamics of the apparent PKA activity in MC3T3-E1 cells (A) and BAECs (B) in response to 100 μM CoCl₂ treatment in DMEM. The solid lines are monoexponential fits with time constants of 0.55 h (A) and 0.97 h (B).

Fig. 5.

Dynamics of the apparent PKA response to reoxygenation in MC3T3-E1 cells in the presence of glucose. Dulbecco's PBS (DPBS) buffer with the addition of 5.5 mM glucose was used as the perfusion media.

Fig. 6.

Effect of fluid shear stress (FSS), magnitude of FSS, and flow volume on the PKA activity response to hypoxia and reoxygenation. DPBS buffer (without the addition of glucose) was used as the incubation media; cells were subjected to hypoxia by the metabolic consumption of O₂. The time evolution of the AKAR3 FRET signal was detected during reoxygenation by reperfusion with air-equilibrated DPBS buffer under different flow rates and durations (A–F).

Fig. 7.

Measurements of the decay rate of cAMP due to hydrolysis by phosphodiesterases at different concentrations of IBMX. A: no IBMX; B: 100 μM IBMX; C: 1 mM IBMX; D: 5 mM IBMX. The cAMP concentration was monitored using the ICUE3 FRET sensor. A lower FRET ratio indicates higher cAMP concentration. cAMP production was stimulated by 5 μM isoproterenol (Iso; A–D, shaded circles) and 100 pM Iso (A, open circles, normalized to the signal at the 5 μM level by 1.5 × upscaling factor).

Fig. 8.

A: dynamics of cAMP activity in cells in response to reoxygenation as detected by the ICUE3 FRET sensor. The O₂ depletion was due to natural cellular metabolism. B: the PKA response to hypoxia in the presence of the adenylate cyclase inhibitor MDL-12,330A.

Fig. 9.

A and B: the PKA response to phosphatase inhibition (A) and effect of phosphatase inhibition on the apparent PKA response to hypoxia and reoxygenation (B). When MC3T3-E1 cells were treated with 10 μM tautomycin to inhibit PKA-counteracting phosphatases, the AKAR3 FRET signal increased, indicating an increase in the apparent PKA activity due to the inhibition of phosphatases (A). The normoxic condition was maintained by keeping the flow rate at 62.5 μl/min. In the presence of 10 μM tautomycin, reoxygenation by reperfusion caused a slower decrease in the AKAR3 FRET signal (B).

Fig. 10.

A and B: dynamics of the PKA response to oxidation of the cellular environment (A) and effect of oxidation on the PKA response to hypoxia and reoxygenation (B). Treatment of MC3T3-E1 cells with 50 μM diamide selectively oxidized PKA-counteracting phosphatases, leading to an increase of the AKAR3 FRET signal (A). In the presence of 100 μM diamide, reoxygenation and hypoxia caused only small changes in the AKAR3 FRET signal (B).

Fig. 11.

Accumulation of ROS due to hypoxic conditions in response to the metabolic consumption of O₂ by cells detected using 2′,7′-dichlorofluorescein (DCF) fluorescence. O₂ exchange between the cell suspension and outside environment was inhibited at time 0. The solid lines are monoexponential fits with the time constants (τ) as indicated on the graph.

Fig. 12.

Model explaining the activity responses of PKA and phosphatases to hypoxia and reoxygenation at different redox conditions.