Acute stimulation of white adipocyte respiration by PKA-induced lipolysis

Abstract

Objective: We examined the effect of β-adrenergic receptor (βAR) activation and cAMP-elevating agents on respiration and mitochondrial uncoupling in human adipocytes and probed the underlying molecular mechanisms.

Research design and methods: Oxygen consumption rate (OCR, aerobic respiration) and extracellular acidification rate (ECAR, anaerobic respiration) were examined in response to isoproterenol (ISO), forskolin (FSK), and dibutyryl-cAMP (DB), coupled with measurements of mitochondrial depolarization, lipolysis, kinase activities, and gene targeting or knock-down approaches.

Results: ISO, FSK, or DB rapidly increased oxidative and glycolytic respiration together with mitochondrial depolarization in human and mouse white adipocytes. The increase in OCR was oligomycin-insensitive and contingent on cAMP-dependent protein kinase A (PKA)-induced lipolysis. This increased respiration and the uncoupling were blocked by inhibiting the mitochondrial permeability transition pore (PTP) and its regulator, BAX. Interestingly, compared with lean individuals, adipocytes from obese subjects exhibited reduced OCR and uncoupling capacity in response to ISO.

Conclusions: Lipolysis stimulated by βAR activation or other maneuvers that increase cAMP levels in white adipocytes acutely induces mitochondrial uncoupling and cellular energetics, which are amplified in the absence of scavenging BSA. The increase in OCR is dependent on PKA-induced lipolysis and is mediated by the PTP and BAX. Because this effect is reduced with obesity, further exploration of this uncoupling mechanism will be needed to determine its cause and consequences.

FIG. 1.

Elevated cAMP levels acutely induce respiration in white adipocytes. Representative measurements of the percent increase in OCR (A) or ECAR (C), respectively, relative to baseline rates in response to ISO, FSK, or DB in human adipocytes. As indicated, at 0 min ISO (1 μmol/l), FSK (10 μmol/l), DB (1 mmol/l), or DMEM (CONT) were injected. OCR (A) or ECAR (C) measurements before drug injection were set as 100%. Adipocytes were pooled from five different human subjects, and each data point is a mean of 4–6 wells. Histograms summarizing the average maximal percent increase of OCR (B) or ECAR (D), respectively, over their baseline rates in response to ISO, FSK, DB, and FCCP (0.6 μmol/l). Data are collected from 8 to 10 experiments, using a total of 20 subjects (BMI 21.7–35.5 kg/m²). All experiments included wells that received ISO, FSK, or DB and wells that received FCCP. *P < 0.001; #P < 0.01 compared with CONT.

FIG. 2.

Elevated cAMP levels induce Oligo-insensitive OCR. Representative measurements of the percent change in OCR in human adipocytes after injections of DMEM (open symbols) or the inhibitors (closed symbols) ATP synthase inhibitor oligomycin (Oligo (1 μg/ml) or the complex 1 inhibitor rotenone (3 μmol/l), as indicated by the dashed vertical lines. Oligo was injected 30–40 min after ISO (1 μmol/l) (A), FSK (10 μmol/l) (B), DB (1 mmol/l) (C), or DMEM (CONT). OCR before Oligo injection was set as 100%. The results are from adipocytes pooled from five subjects, and each data point is a mean of 4–5 wells. D: Histogram summarizing the percent decrease in OCR in response to Oligo or rotenone injections. Data are relative to OCR levels before Oligo injection (and after ISO, FSK, or DB injections). The data are an average of 3–5 experiments using a total of 15 subjects (BMI 21.7–35.5 kg/m²). *P < 0.001, compared with untreated CONT; ∧P < 0.01, compared with untreated ISO; #P < 0.001, compared with CONT + Oligo. E: The relative changes in OCR between Oligo and rotenone are expressed as % uncoupling (calculated from the results presented in D, dotted arrows). Rot, rotenone.

FIG. 3.

Elevated cAMP levels induce mitochondrial depolarization. A: Microscopy images of human adipocytes stained with TMRM and visualized under confocal microscopy or their phase-contrast (Ph-Cont) mode. Cells were treated with ISO (1 μmol/l, 30 min) or FSK (10 μmol/l, 60 min) in DMEM + 10% FBS, stained with TMRM and immediately analyzed, as described in research design and methods. The scale bar in the images represents 20 microns. B: Histogram summarizing TMRM fluorescence intensity of untreated (CONT) and ISO-, FSK-, or FCCP- (40 μmol/l, 60 min) treated cells of 50 images collected from three independent experiments. TMRM intensity in the untreated cells (CONT) was set as 1. *P < 0.001, compared with CONT. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 4.

cAMP-induced respiration is PKA-dependent. Representative measurements of the percent increase in OCR (A) and ECAR (C), relative to baseline rates, in response to ISO or FSK in human adipocytes pretreated or not with H89. At 0 min, ISO (1 μmol/l), FSK (10 μmol/l), or DMEM (CONT) was injected to cells that were pretreated (closed symbols) or not (open symbols) with H89 (10 μmol/l, 1 h). Each data point is the mean of four wells, and error bars not visible are contained within the symbols. Histograms summarizing the maximum percent increase of OCR (B) or ECAR (D), over their baseline rates in response to DMEM, ISO, or FSK. Cells were pretreated with or without H89 or Rp-cAMPS (0.5 mmol/l, 1 h). Results are the average of a total of 11 subjects (BMI 21.7–35.5 kg/m²), measured in three independent experiments. *P < 0.05, compared with untreated respective control; ∧P < 0.05, compared with respective ISO or FSK without H89; #P < 0.05, compared with respective ISO or FSK without Rp-cAMPS.

FIG. 5.

cAMP-induced OCR is FA-dependent. Representative measurements of the percent increase in OCR (A) and ECAR (C), relative to baseline rates, in response to ISO or FSK in adipocytes in the presence or absence of FA-free BSA. At 0 min, ISO (1 μmol/l), FSK (10 μmol/l), or DMEM (CONT) was injected to cells that were pretreated (closed symbols) or not (open symbols) with BSA (5%, 1 h). Each data point is the mean of four wells, and error bars not visible are contained within the symbols. Histograms summarizing the maximum percent increase of OCR (B) or ECAR (D) in response to DMEM (CONT), ISO, FSK or DB (1 mmol/l) over their baseline rates. Data are an average of 11 subjects (BMI 21.7–35.5 kg/m²), measured in three independent experiments. *P < 0.05, compared with CONT; #P < 0.05, compared with respective samples without BSA.

FIG. 6.

cAMP-induced OCR is dependent upon ATGL. Human adipocytes were transfected with siRNA targeted to ATGL or scrambled sequence (SCR), as described in research design [scap]and [scap]m[scap]ethods, which resulted in 50% suppression (supplementary Fig. 5B). Representative measurements of the percent increase in OCR (A) and ECAR (C) relative to baseline rates in response to FSK (10 μmol/l) in human adipocytes treated with SCR (open symbols) or ATGL siRNA (closed symbols). As indicated, at 0 min FSK or DMEM (CONT) was injected. OCR measurements before FSK injection were set as 100%. Adipocytes were pooled from five different human subjects, and each data point is a mean of six wells. Histograms summarizing the maximum percent increase of OCR (B) or ECAR (D), respectively, relative to respective baseline rates, collected from three experiments using a total of 20 subjects (BMI 21.7–35.5 kg/m²). *P < 0.01, compared with respective control samples; #P < 0.05, compared with FSK + SCR.

FIG. 7.

ISO-induced OCR is reduced with obesity. Representative measurements of the percent increase in OCR (A) and ECAR (C), in response to ISO of adipocytes taken from lean or obese humans. Cells were pooled from five lean subjects (BMI 21.7–24.6 kg/m²) or five obese subjects (BMI 30–35.5 kg/m²). As indicated, at 0 min ISO (1 μmol/l, open symbols) or DMEM (closed symbols) was injected. Each data point is a mean of 6–8 wells, and error bars not visible are contained within the symbols. Histograms summarizing the maximum percent increase of OCR (B) or ECAR (D) in response to ISO and relative to baseline rates. Results are presented as fold vs. lean. Data are the mean of 12 lean and 9 obese samples; each was measured in three different experiments, containing six to eight replicates. *P < 0.001, compared with lean.

FIG. 8.

cAMP-induced mitochondrial uncoupling is mediated by BAX and the PTP. A and B: Human adipocytes were transfected with scrambled (SCR) or BAX siRNAs (30 nmol/l, 48 h). A: BAX gene expression was measured by RT-PCR and normalized to GAPDH. B: Representative Western blots (one of four replicates) of BAX protein level compared with GAPDH. C and E: Representative measurements of the percent increase in OCR relative to baseline rates, in response to ISO or FSK. Human adipocytes were transfected with SCR or BAX siRNAs (C) or pretreated or not with CSA (5 μg/ml, 72 h) (E). At 0 min, ISO (1 μmol/l), FSK (10 μmol/l), or DMEM (CONT) was injected into cells. Each data point is the mean of four wells, and error bars not visible are contained within symbols. D and F: Histograms summarizing the maximum percent increase of OCR over the baseline rates in response to DMEM (CONT), ISO, or FSK. In D, results are the average of six subjects measured in three independent experiments. *P < 0.001, compared with respective control; #P < 0.001, compared with ISO + SCR. In F, results are the average of a total of 12 subjects measured in five independent experiments. *P < 0.001, compared with CONT; ∧P < 0.05, compared with respective ISO without CSA; #P < 0.001, compared with respective FSK without CSA.