Skip to main content


Non-steroidal anti-inflammatory drugs activate NADPH oxidase in adipocytes and raise the H2O2 pool to prevent cAMP-stimulated protein kinase a activation and inhibit lipolysis

Article metrics

  • 6872 Accesses

  • 9 Citations



Non-steroidal anti-inflammatory drugs (NSAIDs) —aspirin, naproxen, nimesulide, and piroxicam— lowered activation of type II cAMP-dependent protein kinase A (PKA-II) in isolated rat adipocytes, decreasing adrenaline- and dibutyryl cAMP (Bt2cAMP)-stimulated lipolysis. The molecular bases of insulin-like actions of NSAID were studied.


Based on the reported inhibition of lipolysis by H2O2, catalase was successfully used to block NSAID inhibitory action on Bt2cAMP-stimulated lipolysis. NSAID, at (sub)micromolar range, induced an H2O2 burst in rat adipocyte plasma membranes and in whole adipocytes. NSAID-mediated rise of H2O2 was abrogated in adipocyte plasma membranes by: diphenyleneiodonium, an inhibitor of NADPH oxidase (NOX); the NOX4 antibody; and cytochrome c, trapping the NOX-formed superoxide. These three compounds prevented the inhibition of Bt2cAMP-stimulated lipolysis by NSAIDs. Inhibition of aquaporin-mediated H2O2 transport with AgNO3 in adipocytes allowed NOX activation but prevented the lipolysis inhibition promoted by NSAID: i.e., once synthesized, H2O2 must reach the lipolytic machinery. Since insulin inhibits adrenaline-stimulated lipolysis, the effect of aspirin on isoproterenol-stimulated lipolysis in rat adipocytes was studied. As expected, isoproterenol-mediated lipolysis was blunted by both insulin and aspirin.


NSAIDs activate NOX4 in adipocytes to produce H2O2, which impairs cAMP-dependent PKA-II activation, thus preventing isoproterenol-activated lipolysis. H2O2 signaling in adipocytes is a novel and important cyclooxygenase-independent effect of NSAID.


Interest in salicylates has prompted their use for lowering blood glucose in patients with diabetes since 1876 [1]. Although salicylate treatment of diabetes never gained wide application, the molecular mechanism of the hypoglycemic activity of aspirin has acquired renewed interest because it inhibits IκB kinase-β (IKK-β) [2]. From these results, Schulman hypothesized that salicylates might prevent lipid-induced activation of the serine kinase cascade involving IKK-β [3]: serine phosphorylation of insulin receptor substrate (IRS)-1 by activated IKK-β will decrease the ability of IRS-1 to activate phosphatidylinositol 3-kinase (PI3K), an important mediator of insulin signaling and action, thus leading to insulin resistance. By reversing IKK-β kinase activation, salicylates might enhance insulin sensitivity. Results supporting this proposal include the prevention of lipid-induced insulin resistance by salicylates in IKK-β heterozygous mice and in IKK-β knockout mice without salicylate treatment [3]. Furthermore, aspirin-treated mice bearing an heterozygous deletion in the gene for the IKK-β exhibited improved insulin sensitivity and reduced plasma glucose levels [4]. Activation of additional serine kinases promotes the development of insulin resistance by a similar mechanism [e.g., [5, 6]] and, for some of these kinases, salicylates inhibited their activation and improved the effects of insulin [79]. This paper presents evidence of an alternative pathway employed by aspirin and other NSAID to enhance insulin action, by impairing the physiological activation of a specific protein kinase. In cell-free extracts of isolated adipocytes, we have shown that aspirin, naproxen, nimesulide, and piroxicam inhibited cAMP-mediated PKA activation, decreasing PKA activity and reducing translocation of hormone-sensitive lipase from cytosol to fat droplets [10, 11].

A number of insulin effects on adipocytes are mimicked by H2O2[1218], including inhibition of stimulated lipolysis [1921]. Furthermore, it has been shown that insulin activates NADPH oxidase, which produces superoxide that spontaneously dismutates to H2O2[14, 21], transiently increasing the concentration of cellular H2O2[17, 20], and a role of H2O2 as a second messenger has been hypothesized since 1977 – 1980 [14, 16, 1922]. A new wave of data to enlarge the same topic appeared years later, i.e., H2O2 is produced by an NADPH oxidase (NOX) isoenzyme during physiological insulin transduction in adipose cells [23]. A substantial advance was made by Goldstein’s group, who showed that insulin causes rapid formation of H2O2 in 3T3-L1 adipocytes, a redox signal that enhances the early insulin-stimulated cascade of tyrosine phosphorylation by reversible oxidative inactivation of thiol-dependent protein-tyrosine phosphatase (PTPase) 1B [24] and other enzymes [25, 26], which pointed to a novel regulatory mechanism complementing the early steps in insulin amplification signaling. A more recent report on insulin signaling via H2O2 during lipolysis showed that H2O2—either generated by insulin or added—reversibly inhibited the lipolysis rates activated by epinephrine or Bt2cAMP [27]. This effect took place by means of the H2O2 mediated oxidation of two sulfhydryl groups from the PKA holoenzyme: Cys 97 from regulatory IIα or IIβ subunits, and Cys 199 from the catalytic α subunit, which formed a disulfide bond that impaired cAMP activation of the holoenzyme, thus preventing PKA-stimulated lipolysis [27]. This information together with the inhibition of stimulated lipolysis by NSAID [10, 11] led us to propose H2O2 as the missing molecule generated by NSAID in adipocyte plasma membranes. Thus, the aim of this paper was to get insight on the molecular bases of insulin-like actions of NSAIDs.



Acetylsalicylic acid (aspirin), naproxen, nimesulide, piroxicam, Bt2cAMP, guanosine 5′-3-O-(thio)triphosphate (GTPγS), HEPES, MES, MOPS, NADPH, cAMP, insulin, collagenase type II, Bovine serum albumin fraction V (BSA), catalase, Diphenyleneiodonium chloride (DPI), Cytochrome c (Cyt c), adenosine, and trichloroacetic acid (TCA) were obtained from Sigma-Aldrich (St. Louis, MO, USA, The protease inhibitor cocktail was obtained from MP Biomedicals (Solon, OH, USA, The Amplex Red kit was purchased from Molecular Probes, Inc. (Eugene, OR, USA, H2O2 was obtained from Merck (Darmstadt, Germany, AgNO3 was purchased from Baker (México,, polyclonal antibodies against the PKA catalytic α subunit (sc-903) and NOX4 (sc-21860) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA,, and secondary antibodies were purchased from Pierce (Rockford, IL, USA, All other reagents were of the highest purity available commercially.


Male Wistar rats weighing 200–240 g fed ad libitum with a commercial diet (Purina, México) and with free access to water were used. All experiments were conducted in accordance with the Federal Regulations for Animal Care and Use (NOM-062-ZOO-1999, Ministry of Agriculture, México) and were approved by the Ethics Committee of the Facultad de Medicina, Universidad Nacional Autónoma de México (UNAM).

Adipocyte isolation and measurement of lipolysis

To isolate adipocytes with low cAMP endogenous levels, animals were fasted for 16 h as recommended by Londos [28]. Animals were sacrificed by decapitation and the epididymal fat pads were immediately removed. Fat pads from two rats were used in each experiment. In brief, Krebs-Ringer buffer was enriched with 25 mM HEPES, 2.5 mM CaCl2, 2 mM glucose, 200 nM adenosine, and fatty acid-free BSA either at 1 or 4%, as detailed later; pH was adjusted to 7.4. One gram of minced fat pads was digested in 10 ml of collagenase (1 mg/ml) for 30 min at 37°C, with shaking at 160 cycles/min in the Krebs-Ringer-enriched buffer supplemented with 1% BSA. Cells were filtered through nylon cloth and washed three times by centrifugation (1 min each) at 220 × g. Wet-packed adipocytes were weighed to report glycerol release by wet weight as an index of lipolysis, which was assayed using 100 μl of packed adipocytes incubated for 30 min at 37°C in a total volume of 1 ml of Krebs-Ringer-enriched buffer supplemented with 4% BSA, in which Bt2cAMP, isoproterenol, catalase, insulin, NSAID, DPI, anti-NOX4 antibody, Cyt c, and AgNO3, were dissolved to reach the final concentrations indicated in the figures. Adipocytes were maintained dispersed during incubation by shaking at 160 cycles/min. Lipolysis was stopped by transferring tubes from 37°C to an ice bath for 5 min. Tubes were immediately centrifuged at 10,000 × g at 4°C for 10 min. A 300-μl aliquot from the solution lying below the fat cake was utilized to measure released glycerol [29].

Measurement of H2O2 generation in isolated adipocytes

One hundred μl of packed rat adipocytes were incubated for 10 min (unless another time is indicated) at 37°C, with shaking at 160 cycles/min in a total 1-ml volume of Krebs-Ringer-enriched buffer supplemented with 4% BSA in which insulin, NSAID, DPI, Cyt c, anti-NOX4 antibody, and AgNO3 were dissolved to reach the final concentrations indicated in the figures. H2O2 generation was stopped by the addition of 100 μl of TCA 6 M, and the tubes were immediately centrifuged at 10,000 × g at 4°C for 10 min to measure H2O2 with the method of Zhou et al. [30], utilizing the Amplex Red hydrogen peroxide assay kit (Molecular Probes; A22188) according to the manufacturer’s instructions.

NADPH-dependent H2O2 generation system activity

The procedure described to measure NADPH oxidase system activity in adipocytes was followed [23, 27]. In brief, 100 μl of packed rat adipocytes were suspended in 900 μl of ice-cold lysis medium containing 20 mM MES pH 5.8, 2 mM MgCl2, 1 mM CaCl2, 5 mM KCl, and 100 μl of protease inhibitor cocktail. Cells were lysed after vigorous mixing for 5 min in a vortex. Lysed cells were spun at 1,000 × g for 20 min at 4°C, the supernatant was discarded, and the precipitate with plasma membrane was suspended in the activation buffer containing 30 mM MOPS, pH 7.5, 120 mM NaCl, 1.4 mM CaCl2, 5 mM MgCl2, and 10 mM NaHCO3. Centrifugation was repeated, the supernatant was discarded, and the precipitate was suspended in the activation buffer supplemented or not with MnCl2, guanosine 5′-3-O-(thio)triphosphate (GTPγS), NSAID, or insulin, as detailed in the figure legends. Adipocyte plasma membranes containing the NADPH oxidase system were incubated in activation buffer at 25°C for 25 min. Then, the samples were centrifuged under the same conditions, the supernatant was discarded, and the precipitate was suspended and washed twice in catalysis buffer containing 30 mM MES, pH 5.8, 120 mM NaCl, 4 mM MgCl2, 1.2 mM KH2PO4, 1 mM NaN3, 10 mM FAD, and supplemented when indicated with DPI, Cyt c, anti-NOX4 antibody, and AgNO3. Samples were spun again, the supernatant was discarded, and these were suspended in the same buffer without supplements; the catalytic reaction was started with 250 μM NADPH and incubated for 30 min at 37°C. The reaction was stopped by placing tubes in an ice bath for 5 min, and a 5-μl aliquot from the mix reaction was employed to measure H2O2 using the Amplex Red hydrogen peroxide assay kit.


Data points shown are means ± Standard error of the mean (SEM). All statistical analyses were performed using SigmaPlot ver. 11 software (Systat Software, Inc., San Jose, CA, USA, Statistical differences were determined employing Student’s t tests or one-way Analysis of variance (ANOVA) followed by the Dunnett or Kruskal-Wallis test. Minimum level of significance was set at p <0.05.


Role of H2O2 on the inhibitory action of NSAID

On the basis of the data available, we propose that the H2O2 generated by NSAID is the intermediary that prevents PKA-stimulated lipolysis. This putative role of H2O2 was explored by adding exogenous catalase to intact isolated adipocytes challenged with Bt2cAMP to activate lipolysis (i.e., glycerol release). As expected, the results showed that aspirin, naproxen, nimesulide, and piroxicam at 10–6 M inhibited Bt2cAMP-activated lipolysis (p <0.05) (Figure 1a). In contrast, catalase significantly enhanced Bt2cAMP-activated lipolysis, either in the absence of the cyclic nucleotide or in its presence, at all concentrations tested (Figure 1b). Because lipolysis inhibition elicited by the four selected NSAID at 10–6 M was observed when glycerol release was activated by 10–5 to 10–2 M Bt2cAMP, i.e., at concentrations 10 – 10,000-fold higher than the concentration of the aspirin-like drugs (p <0.05) (Figure 1a), direct interaction between NSAID and Bt2cAMP can be discarded. Furthermore, in all cases, the addition of exogenous catalase impaired NSAID-mediated inhibition of lipolysis (Figure 1c).

Figure 1

Effect of catalase and selected NSAID on Bt 2 cAMP-stimulated glycerol release in isolated rat adipocytes. Panel a, concentration-response curve for Bt2cAMP without NSAID, or plus 10–6 M of each of the NSAID indicated in the figure. Panel b, concentration-response curve for Bt2cAMP alone or in the presence of 1,000 units of catalase. Panel c, concentration-response curve for Bt2cAMP in the presence of 1,000 units of catalase without NSAID or plus 10–6 M of each of the NSAID indicated in the figure. Basal glycerol release in the absence of Bt2cAMP is indicated by an open circle. Each point represents the average ± Standard error (SE) of three to five independent experiments. *,p < 0.05 vs Bt2cAMP alone.

NSAID increased H2O2 generation through a NOX system

The next experiment was to test the ability of NSAID to generate sufficient H2O2 in isolated adipocytes, in order to amplify and substantiate the inhibitory action of aspirin-like drugs on stimulated lipolysis [11]. The selected NSAID employed at 10–6 M produced a linear but transient rise in the content of H2O2, reaching a maximum concentration at 10 min of incubation followed by its rapid disappearance (not shown), indicative of a rapid turnover in the H2O2 pool, as expected for a regulatory signal. Based on these data, the 10-min incubation period was chosen to conduct further experiments. Isolated adipocytes generated H2O2 with a similar concentration-response pattern and with a peak at 10–6 M for each NSAID (Figure 2a). The transient rise in H2O2 induced by NSAID is quantitatively similar to that observed with 10–8 M insulin (Figure 2a), a hormone that follows a redox signal transduction pathway, which reversibly inhibited lipolysis [27]. Cell membranes prepared from adipocytes were incubated in an enriched medium with NADPH to generate H2O2 by the NOX; under these experimental conditions, NSAID increased the production of H2O2 (Figure 2b). A concentration-response curve of these compounds in the presence of Mn2+ showed an increase in the endogenous synthesis of H2O2, with a peak at 10−6 M for NSAID, except for aspirin, for which a value of 10−5 M was observed; higher concentrations of NSAID failed to increase H2O2 generation further. We have no explanation for this last observation; however, bell-shaped dose response relationships have been previously reported for other NSAID effects (e.g., [3133]), pointing out the diverse and complex action mechanisms of NSAIDs. On the other hand, the decrease in H2O2 production at higher concentrations of NSAIDs cannot be explained by a toxic effect of NSAIDs on the cells, since the same type of response is obtained in both, whole cells (Figure 2a) and isolated plasma membranes (Figure 2b). Thus, the data suggest that NSAIDs effect is on NADPH oxidase system. An estimated IC50 near 10−7 M was obtained for these aspirin-like drugs [11] (Figure 2). The enzymatic system responsible for H2O2 generation in adipocytes has been identified previously as a NOX4 isoform [34], which can be activated by Mn2+ or GTP prior to interaction with hormones [23]. Besides NOX4, no other isoforms have been detected in adipocytes [34]. Results in isolated membranes of rat adipocytes showed that NOX activity was low in the absence of Mn2+, but that it was stimulated by all four NSAID (Figure 3a). After NOX activation by Mn2+ or GTPγS (a GTP analogue), NSAID produced greater stimulation (Figure 3b and 3c). The response observed with NSAID is similar to the response pattern obtained with insulin-challenged adipocyte plasma membranes (Figure 2b), which utilizes H2O2 as a second messenger [2327].

Figure 2

Concentration-response curve for NSAID on H 2 O 2 generation in rat adipocytes, and in plasma membranes from adipocytes. Panel a, cells were incubated for 10 min at 37°C in Krebs-Ringer-enriched buffer supplemented with 4% BSA plus 10−8 M insulin or NSAID at the concentrations indicated in the figure. Panel b, freshly prepared cell membranes were incubated for 10 min at 37°C in the activation buffer enriched with 3 mM MgCl2 and 4% BSA plus 10−8 M insulin or NSAID at the concentrations indicated in the figure. Insulin-induced H2O2 generation in the absence of NSAID is indicated by an open circle. Assay of H2O2 was conducted as described in experimental procedures and the catalytic reaction was started with 250 μM NADPH. Each point represents the average ± Standard error (SE) of three to five independent experiments. *, p < 0.05 none vs NSAID or insulin.

Figure 3

Effect of Mn2+and GTPγS on insulin- and NSAID-mediated H 2 O 2 generation through a NADPH oxidase (NOX) system in isolated plasma membranes from rat adipocytes. Freshly prepared cell membranes were incubated in activation buffer supplemented with the following: none (panel a); 3 mM MnCl2 (panel b), or 10−5 M GTPγS (panel c). In each instance, 10−8 M insulin or 10−6 M of each of the selected NSAID were included in the incubation mixture. Assay of H2O2 was conducted as described in Experimental Procedures. Each bar represents the average ± Standard error (SE) of three independent experiments. *, p <0.05 no addition vs insulin or NSAID.

NSAID-activated NOX4 impairs Bt2cAMP-stimulated lipolysis

Experiments were designed to identify the source of the pool of H2O2 impairing Bt2cAMP-activated lipolysis in adipocytes. Figure 4 shows that the stimulatory action of insulin and NSAID on NOX to raise H2O2 in isolated plasma membranes was prevented by DPI, a non-specific NOX inhibitor [35], by the anti-NOX4 antibody, and by oxidized Cyt c, which traps the electron from the superoxide ion [36] produced by NOX, which in turn might dismutate spontaneously to form H2O2 in a non-enzymatic reaction. Based on the fact that specific aquaporins facilitate H2O2 diffusion across membranes [37] and that Ag+ ions are potent inhibitors of these transporters [38], AgNO3 was tested to prevent H2O2 transport across the plasma cell membrane. Indeed, as can be observed in Figure 4, AgNO3 did not modify H2O2 synthesis by NOX. Figure 5 shows that inhibition of glycerol release by aspirin-like drugs disappeared with the three compounds, impairing H2O2 synthesis, as well as with AgNO3 (Figure 5), which allows H2O2 generation but interferes with its uptake by aquaporins [38]. In all of these experiments, Bt2cAMP-activating glycerol release prevailed over the antilipolytic action of NSAID (Figure 5).

Figure 4

Effect of DPI, Cyt c, anti-NOX4 antibody, and AgNO 3 on insulin- and NSAID-mediated H 2 O 2 generation through a NOX4 system in isolated plasma membranes from rat adipocytes. Freshly prepared membranes were incubated for 10 min at 37°C in the activation buffer enriched with 4% BSA and 10−5 M GTPγS. Activation buffer was additionally supplemented with the following: none (control), or 5 ×10−5 M DPI, or 8 × 10−5 M Cyt c, or 1:6,400 dilution anti-NOX4 antibody, or 3 × 10−5 M AgNO3. In each set of experiments, 10−8 M insulin or 10−6 M of each NSAID was added to the incubation mixture. The H2O2 assay was conducted as described in Experimental Procedures. Each bar represents the average ± Standard error (SE) of three to six independent experiments. The letters (a, b, c, d, e, f) represent a significant difference among groups with different treatments by employing one-way Analysis of variance (ANOVA) at p <0.01. Groups with the same letter did not show a significant difference at p <0.05.

Figure 5

Effect of DPI, Cyt c, anti-NOX4 antibody, and AgNO 3 on Bt 2 cAMP-stimulated glycerol release in rat adipocytes incubated in the presence of each of the selected NSAID. Cells were incubated for 30 min at 37°C in Krebs-Ringer supplemented with 4% BSA, 10−3 M Bt2cAMP, and 10−6 M of each NSAID used, as indicated in the Figure. Krebs-Ringer was additionally supplemented with the following: none (control), or 5 × 10−5 M DPI, or 8 × 10−5 M Cyt c, or a 1:6,400 dilution of anti-NOX4 antibody, or 3 × 10−5 M AgNO3. Each bar represents the average ± Standard error (SE) of three to six independent experiments. The letters (a, b, c, d) represent a significant difference among groups with different treatments by employing one-way Analysis of variance (ANOVA) at p <0.01. Groups with the same letter did not show a significant difference at p <0.05.

Aspirin inhibition of isoproterenol-activated lipolysis

Since insulin inhibits adrenaline-stimulated lipolysis [27], the effect of aspirin (used as an NSAID prototype) on isoproterenol-stimulated lipolysis in rat adipocytes was studied. As expected, isoproterenol-mediated lipolysis was blunted by both insulin and aspirin (Figure 6). This agrees with previously published results showing that NSAIDs inhibit adrenaline-stimulated lipolysis in isolated adipocytes [10]. Because NSAIDs did not modify the binding of adrenergic agonist to their receptor [10], and inhibited Bt2cAMP-activated lipolysis (Figure 1a), it is clear that the antagonistic effect of NSAIDs on isoproterenol-stimulated lipolysis is located downstream the cAMP production.

Figure 6

Effect of insulin and aspirin on isoproterenol-stimulated glycerol release in rat adipocytes. Cells were incubated for 30 min at 37°C in Krebs-Ringer supplemented with 4% BSA and with the reagents indicated in the Figure. Each bar represents the average ± Standard error (SE) of three to six independent experiments. The letters (a, b, c) represent a significant difference among groups with different treatments by employing one-way Analysis of variance (ANOVA) at p <0.01. Groups with the same letter did not show a significant difference at p <0.05.


NSAID are the most widely used drugs [3941]. Their canonical molecular action inhibiting cyclooxygenases (COX) has been enlarged by numerous COX-independent actions; among these, we reported an inhibition of cAMP-mediated PKA activation in adipocytes [10, 11]. Results in this paper supply details on the molecular mechanism of this inhibition, which was obtained with NSAID concentrations within the micromolar range, near or even below the reported levels found in human blood after administration of these compounds for therapeutic purposes [4245]. However, the goal of this paper was not to study NSAID antidiabetic actions [13, 46], but to gain insights into the molecular bases of insulin-like actions of NSAIDs on the metabolic regulation in adipose cells. Sufficient information hinted at H2O2 as the intermediate molecule between aspirin and the inhibition of stimulated lipolysis [10, 11, 27]. Results in Figure 1 not only show that Bt2cAMP-stimulated lipolysis was decreased with aspirin, but that this inhibitory action was shared by naproxen, nimesulide, and piroxicam, and, therefore, this action might be considered as a common property of NSAIDs. Results also suggest a physiological role of H2O2 in the regulation of stimulated lipolysis, because H2O2 disappearance by supplementation with catalase permitted extra synthesis of glycerol at all doses of Bt2cAMP (Figure 1b). The proposal that H2O2 is produced by NOX after its activation with NSAID was inspired by the reported action of insulin on adipocytes [2327]. Indeed, submicromolar concentrations of four selected NSAID raised the H2O2 pool, either in isolated adipocytes (Figure 2a) or in plasma membranes from adipocytes (Figure 2b). Products generated by NOX activation—O2 and H2O2—have multiple actions in signaling processes [for a review, see Stone and Yang [47]]. Currently, specific NOX inhibitors are not available [48]. However, our experiments strongly support that H2O2 was generated by the NSAID-activated NOX4 isoform based on the following pieces of independent direct or indirect evidence: i) NOX4 is the only NOX isoform expressed in adipocytes [34], ii) the enzymatic system responsible for H2O2 generation was inhibited with DPI (Figure 4), the classical and most frequently used NOX inhibitor [35]; iii) H2O2 synthesis blockade and subsequent inhibition of the antilipolytic action of NSAIDs was observed after the addition of either exogenous catalase or exogenous Cyt c (Figure 4), agents that decrease the H2O2 concentration resulting from NOX catalytic activity [36]; iv) Mn2+ and GTPγS-activated H2O2 synthesis in the membranes of rat adipocytes (Figure 3), as shown previously for activation of NOX in human adipocytes by Mn2+ and GTPγS [23]; v) AgNO3 which allows H2O2 generation (Figure 4), interferes with its antilipolytic action in whole adipocytes by inhibiting aquaporins (Figure 5), showing that the enzymatic system responsible for H2O2 generation (which is stimulated by NSAIDs) is located in the plasma membrane and releases H2O2 outside the cell, and vi) a very diluted solution of NOX4 antibody impaired H2O2 synthesis (Figure 4). This last inhibitory action of NOX4 antibodies over NADPH oxidase activity has been previously reported in both cell-free [49, 50] and intact cells assays [51, 52]. Thus, although none of the experiments described above by itself provides conclusive evidence of NOX4 activation by NSAIDs, to our knowledge there is no enzymatic system, besides NOX4, responsible for H2O2 generation at the plasma membranes of isolated adipocytes that could explain simultaneously all the results described above.

The association of H2O2 with the lipolysis in adipocytes can be supported by abundant experimental evidence. An elevated pool of H2O2 in adipocytes—as observed after incubation with insulin [2325, 34], added H2O2[27], monoamine oxidase substrates [53], and NSAID (Figures 1 and 5)—resulted in inhibition of stimulated lipolysis. This inhibition of stimulated lipolysis disappeared when the pool of H2O2 was lowered with catalase [27] (Figure 1), DPI, anti-NOX4 antibody, or Cyt c (Figure 5). One exception merits special mention. It was shown that elevated production of H2O2 in AgNO3-treated rat adipocytes (Figure 4) was not followed by inhibition of the stimulated lipolysis (Figure 5). These results suggest that the production of H2O2 by NOX occurs outside the cell and that its subsequent uptake into the cell requires the participation of AQP3 [54]. These facts are in complete agreement with previous findings by Miller et al., who showed that the downstream intracellular effects of H2O2 can be regulated across cell membranes [54]. Our results with catalase (Figure 1) and Cyt c (Figure 4) in preventing NSAID-mediated inhibition of lipolysis (Figure 5) support this proposal. It is noteworthy within this context that three different aquaporins, AQP3, AQP7, and AQP9, are expressed in adipose tissue and that all of these are upregulated by insulin [55]. Interestingly, one of these aquaporins (AQP3) is capable of mediating H2O2 uptake [54].

We reported previously that H2O2 generated by insulin in adipose cells oxidizes two Cys residues in the type II PKA holoenzyme [27]. In fact, formation of a disulfide bond between Cys-199 in the catalytic α subunit and Cys-97 in the regulatory β subunit produces an inactive holoenzyme resistant to activation by cAMP, and the thioredoxin/thioredoxin reductase system is responsible for the disulfide bond reduction [27]. therefore, with the results obtained in this work it is possible to propose as hypothesis that H2O2 generated by NSAIDs impairs PKA catalytic function in the same way as occurs in insulin-treated adipocytes [27].

A recognized action of NSAID on phagocytic cells is the antagonizing effect on the production of reactive oxygen species (ROS) during the inflammatory process [5658]. The effect described here for NSAID, i.e., NOX4 activation and higher production of H2O2, was observed in a non-phagocytic cell in which H2O2 mediates the physiological response to insulin [34]; the significance of this action might be enhanced in such cells because, as shown in this paper, PKA is an additional target molecule for H2O2. Opposite results have been described for the H2O2-mediated oxidation of other PKA types, i.e., whereas oxidation of type I PKA in skeletal muscle resulted in its activation [59] and type II PKA oxidation of rat adipocyte and bovine heart holoenzyme resulted in a lack of activation, even in the presence of activators [27]. Of great significance is the fact described in this paper that NSAID actions include the physiological amplification cascades utilized by hormones. Here we described two hormonal second messengers—H2O2 and cAMP—that are associated with NSAID effects.

Within a broad context, a synergistic role can be hypothesized for H2O2 by the convergence of two sets of facts: on the one hand, the H2O2 inhibitory effect on PTPase and other phosphatases as documented by the Goldstein group [2426], and on the other hand, H2O2-mediated prevention of kinase activation, as shown for PKA in this paper and for kinases that might be inactivated by salicylates [29]; when taken together, all of these explain the NSAID effect that enhances insulin action in adipose tissue and the hypoglycemic effect of high doses of salicylates in the treatment of diabetes [3, 4, 46]. Also, this allows a reassessment of previously described antagonism between epinephrine and NSAID actions in rat hepatocytes [60, 61]. Furthermore, NOX4, AQP3, and type II PKA (PRKAR2A) possess wide tissue distribution according to microarray expression data found in the Gene Atlas project [62] (data not shown).


NSAIDs activate NOX4 in adipocytes to produce H2O2, which impairs cAMP-dependent PKA-II activation, preventing isoproterenol-activated lipolysis. H2O2 production for signaling in adipocytes is a novel COX-independent effect of NSAID, which opens a wide horizon to decipher some of their multiple molecular actions.



Adipocyte plasma membranes


DiButyryl cAMP


Diphenyleneiodonium chloride


Guanosine 5′-3-O-(thio)triphosphate


IκB kinase-β


Insulin receptor substrate


NADPH oxidase


Non-steroidal anti-inflammatory drugs


cAMP-dependent Protein kinase A


Protein-tyrosine phosphatase.


  1. 1.

    Ebstein W: Zur therapie des diabetes mellitus, insbesondere über die Answendung des salicylsauren natron bei demselben. Berlin Klin Wochenschrift. 1876, 13: 337-340.

  2. 2.

    Yin MJ, Yamamoto Y, Gaynor RB: The anti-inflammatory agents aspirin and salicylate inhibit the activity of I(kappa)B kinase-beta. Nature. 1998, 396: 77-80. 10.1038/23948.

  3. 3.

    Kim JK, Kim YJ, Fillmore JJ, Chen Y, Moore I, Lee J, Yuan M, Li ZW, Karin M, Perret P, Shoelson SE, Shulman GI: Prevention of fat-induced insulin resistance by salicylate. J Clin Invest. 2001, 108: 437-446.

  4. 4.

    Yuan M, Konstantopoulos N, Lee J, Hansen L, Li ZW, Karin M, Shoelson SE: Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkbeta. Science. 2001, 293: 1673-1677. 10.1126/science.1061620.

  5. 5.

    Tanti JF, Jager J: Cellular mechanisms of insulin resistance: role of stress-regulated serine kinases and insulin receptor substrates (IRS) serine phosphorylation. Curr Opin Pharmacol. 2009, 9: 753-762. 10.1016/j.coph.2009.07.004.

  6. 6.

    Zheng Y, Zhang W, Pendleton E, Leng S, Wu J, Chen R, Sun XJ: Improved insulin sensitivity by calorie restriction is associated with reduction of ERK and p70S6K activities in the liver of obese Zucker rats. J Endocrinol. 2009, 203: 337-347. 10.1677/JOE-09-0181.

  7. 7.

    Gao Z, Zuberi A, Quon MJ, Dong Z, Ye J: Aspirin inhibits serine phosphorylation of insulin receptor substrate 1 in tumor necrosis factor-treated cells through targeting multiple serine kinases. J Biol Chem. 2003, 278: 24944-24950. 10.1074/jbc.M300423200.

  8. 8.

    Jiang G, Dallas-Yang Q, Liu F, Moller DE, Zhang BB: Salicylic acid reverses phorbol 12-myristate-13-acetate (PMA)- and tumor necrosis factor alpha (TNFalpha)-induced insulin receptor substrate 1 (IRS1) serine 307 phosphorylation and insulin resistance in human embryonic kidney 293 (HEK293) cells. J Biol Chem. 2003, 278: 180-186.

  9. 9.

    Park E, Wong V, Guan X, Oprescu AI, Giacca A: Salicylate prevents hepatic insulin resistance caused by short-term elevation of free fatty acids in vivo. J Endocrinol. 2007, 195: 323-331. 10.1677/JOE-07-0005.

  10. 10.

    Zentella de Piña M, Vázquez-Meza H, Piña-Zentella G, Pimentel L, Piña E: Non-steroidal anti-inflammatory drugs inhibit epinephrine- and cAMP-mediated lipolysis in isolated rat adipocytes. J Pharm Pharmacol. 2002, 54: 577-582. 10.1211/0022357021778709.

  11. 11.

    Zentella de Piña M, Vázquez-Meza H, Agundis C, Pereyra MA, Pardo JP, Villalobos-Molina R, Piña E: Inhibition of cAMP-dependent protein kinase A: a novel cyclo-oxygenase-independent effect of non-steroidal anti-inflammatory drugs in adipocytes. Auton Autacoid Pharmacol. 2007, 27: 85-92. 10.1111/j.1474-8673.2007.00392.x.

  12. 12.

    Czech MP, Lawrence JC, Lynn WS: Evidence for the involvement of sulfhydryl oxidation in the regulation of fat cell hexose transport by insulin. Proc Natl Acad Sci U S A. 1974, 71: 4173-4177. 10.1073/pnas.71.10.4173.

  13. 13.

    Czech MP, Lawrence JC, Lynn WS: Evidence for electron transfer reactions involved in the Cu2+ −dependent thiol activation of fat cell glucose utilization. J Biol Chem. 1974, 249: 1001-1006.

  14. 14.

    Lawrence JC, Larner J: Activation of glycogen synthase in rat adipocytes by insulin and glucose involves increased glucose transport and phosphorylation. J Biol Chem. 1978, 253: 2104-2113.

  15. 15.

    Livingston JN, Gurny PA, Lockwood DH: Insulin-like effects of polyamines in fat cells. Mediation by H2O2 formation. J Biol Chem. 1977, 252: 560-562.

  16. 16.

    May JM, de Haën C: The insulin-like effect of hydrogen peroxide on pathways of lipid synthesis in rat adipocytes. J Biol Chem. 1979, 254: 9017-9021.

  17. 17.

    Mukherjee SP: Mediation of the antilipolytic and lipogenic effects of insulin in adipocytes by intracellular accumulation of hydrogen peroxide. Biochem Pharmacol. 1980, 29: 1239-1246. 10.1016/0006-2952(80)90280-4.

  18. 18.

    Taylor WM, Halperin ML: Stimulation of glucose transport in rat adipocytes by insulin, adenosine, nicotinic acid and hydrogen peroxide. Role of adenosine 3':5'-cyclic monophosphate. Biochem J. 1979, 178: 381-389.

  19. 19.

    Little SA, de Haën C: Effects of hydrogen peroxide on basal and hormone-stimulated lipolysis in perifused rat fat cells in relation to the mechanism of action of insulin. J Biol Chem. 1980, 255: 10888-10895.

  20. 20.

    Mukherjee SP, Lynn WS: Reduced nicotinamide adenine dinucleotide phosphate oxidase in adipocyte plasma membrane and its activation by insulin. Possible role in the hormone’s effects on adenylate cyclase and the hexose monophosphate shunt. Arch Biochem Biophys. 1977, 184: 69-76. 10.1016/0003-9861(77)90327-7.

  21. 21.

    Mukherjee SP, Lane RH, Lynn WS: Endogenous hydrogen peroxide and peroxidative metabolism in adipocytes in response to insulin and sulfhydryl reagents. Biochem Pharmacol. 1978, 27: 2589-2594. 10.1016/0006-2952(78)90332-5.

  22. 22.

    May JM, de Haën C: Insulin-stimulated intracellular hydrogen peroxide production in rat epididymal fat cells. J Biol Chem. 1979, 254: 2214-2220.

  23. 23.

    Krieger-Brauer HI, Medda PK, Kather H: Insulin-induced activation of NADPH-dependent H2O2 generation in human adipocyte plasma membranes is mediated by Galphai2. J Biol Chem. 1997, 272: 10135-10143. 10.1074/jbc.272.15.10135.

  24. 24.

    Mahadev K, Zilbering A, Zhu L, Goldstein BJ: Insulin-stimulated hydrogen peroxide reversibly inhibits protein-tyrosine phosphatase 1b in vivo and enhances the early insulin action cascade. J Biol Chem. 2001, 276: 21938-21942. 10.1074/jbc.C100109200.

  25. 25.

    Mahadev K, Wu X, Zilbering A, Zhu L, Lawrence JT, Goldstein BJ: Hydrogen peroxide generated during cellular insulin stimulation is integral to activation of the distal insulin signaling cascade in 3T3-L1 adipocytes. J Biol Chem. 2001, 276: 48662-48669. 10.1074/jbc.M105061200.

  26. 26.

    Mahadev K, Wu X, Motoshima H, Goldstein BJ: Integration of multiple downstream signals determines the net effect of insulin on MAP kinase vs. PI 3'-kinase activation: potential role of insulin-stimulated H(2)O(2). Cell Signal. 2004, 16: 323-331. 10.1016/j.cellsig.2003.08.002.

  27. 27.

    Zentella de Piña M, Vázquez-Meza H, Pardo JP, Rendon JL, Villalobos-Molina R, Riveros-Rosas H, Piña E: Signaling the signal, cyclic AMP-dependent protein kinase inhibition by insulin-formed H2O2 and reactivation by thioredoxin. J Biol Chem. 2008, 283: 12373-12386. 10.1074/jbc.M706832200.

  28. 28.

    Honnor RC, Dhillon GS, Londos C: cAMP-dependent protein kinase and lipolysis in rat adipocytes. I. Cell preparation, manipulation, and predictability in behavior. J Biol Chem. 1985, 260: 15122-15129.

  29. 29.

    Warnick GR: Enzymatic methods for quantification of lipoprotein lipids. Methods Enzymol. 1986, 129: 101-123.

  30. 30.

    Zhou M, Diwu Z, Panchuk-Voloshina N, Haugland RP: A stable nonfluorescent derivative of resorufin for the fluorometric determination of trace hydrogen peroxide: applications in detecting the activity of phagocyte NADPH oxidase and other oxidases. Anal Biochem. 1997, 253: 162-168. 10.1006/abio.1997.2391.

  31. 31.

    Niederberger E, Tegeder I, Vetter G, Schmidtko A, Schmidt H, Euchenhofer C, Brautigam L, Grosch S, Geisslinger G: Celecoxib loses its anti-inflammatory efficacy at high doses through activation of NF-kappaB. FASEB J. 2001, 15: 1622-1624.

  32. 32.

    Frean SP, Cambridge H, Lees P: Effects of anti-arthritic drugs on proteoglycan synthesis by equine cartilage. J Vet Pharmacol Ther. 2002, 25: 289-298. 10.1046/j.1365-2885.2002.00404.x.

  33. 33.

    Francischi JN, Chaves CT, Moura AC, Lima AS, Rocha OA, Ferreira-Alves DL, Bakhle YS: Selective inhibitors of cyclo-oxygenase-2 (COX-2) induce hypoalgesia in a rat paw model of inflammation. Br J Pharmacol. 2002, 137: 837-844. 10.1038/sj.bjp.0704937.

  34. 34.

    Mahadev K, Motoshima H, Wu X, Ruddy JM, Arnold RS, Cheng G, Lambeth JD, Goldstein BJ: The NAD(P)H oxidase homolog Nox4 modulates insulin-stimulated generation of H2O2 and plays an integral role in insulin signal transduction. Mol Cell Biol. 2004, 24: 1844-1854. 10.1128/MCB.24.5.1844-1854.2004.

  35. 35.

    Cross AR, Jones OT: The effect of the inhibitor diphenylene iodonium on the superoxide-generating system of neutrophils. Specific labelling of a component polypeptide of the oxidase. Biochem J. 1986, 237: 111-116.

  36. 36.

    Pick E: Microassays for superoxide and hydrogen peroxide production and nitroblue tetrazolium reduction using an enzyme immunoassay microplate reader. Methods Enzymol. 1986, 132: 407-421.

  37. 37.

    Bienert GP, Moller AL, Kristiansen KA, Schulz A, Moller IM, Schjoerring JK, Jahn TP: Specific aquaporins facilitate the diffusion of hydrogen peroxide across membranes. J Biol Chem. 2007, 282: 1183-1192.

  38. 38.

    Niemietz CM, Tyerman SD: New potent inhibitors of aquaporins: silver and gold compounds inhibit aquaporins of plant and human origin. FEBS Lett. 2002, 531: 443-447. 10.1016/S0014-5793(02)03581-0.

  39. 39.

    Fosbol EL, Gislason GH, Jacobsen S, Abildstrom SZ, Hansen ML, Schramm TK, Folke F, Sorensen R, Rasmussen JN, Kober L, Madsen M, Torp-Pedersen C: The pattern of use of non-steroidal anti-inflammatory drugs (NSAIDs) from 1997 to 2005: a nationwide study on 4.6 million people. Pharmacoepidemiol Drug Saf. 2008, 17: 822-833. 10.1002/pds.1592.

  40. 40.

    Tenenbaum J: The epidemiology of nonsteroidal anti-inflammatory drugs. Can J Gastroenterol. 1999, 13: 119-122.

  41. 41.

    Inotai A, Hanko B, Meszaros A: Trends in the non-steroidal anti-inflammatory drug market in six Central-Eastern European countries based on retail information. Pharmacoepidemiol Drug Saf. 2010, 19: 183-190. 10.1002/pds.1893.

  42. 42.

    Brogden RN, Heel RC, Speight TM, Avery GS: Piroxicam. A reappraisal of its pharmacology and therapeutic efficacy. Drugs. 1984, 28: 292-323. 10.2165/00003495-198428040-00002.

  43. 43.

    Levy G: Clinical pharmacokinetics of salicylates: a re-assessment. Br J Clin Pharmacol. 1980, 10 (Suppl 2): 285S-290S.

  44. 44.

    Singla AK, Chawla M, Singh A: Nimesulide: some pharmaceutical and pharmacological aspects–an update. J Pharm Pharmacol. 2000, 52: 467-486. 10.1211/0022357001774255.

  45. 45.

    Vree TB, Van DB-M, Verwey-Van Wissen CP, Vree ML, Guelen PJ: The pharmacokinetics of naproxen, its metabolite O-desmethylnaproxen, and their acyl glucuronides in humans. Effect of cimetidine. Br J Clin Pharmacol. 1993, 35: 467-472. 10.1111/j.1365-2125.1993.tb04171.x.

  46. 46.

    Reid J, MacDougall AI, Andrews MM: Aspirin and diabetes mellitus. Br Med J. 1957, 2: 1071-1074. 10.1136/bmj.2.5053.1071.

  47. 47.

    Stone JR, Yang S: Hydrogen peroxide: a signaling messenger. Antioxid Redox Signal. 2006, 8: 243-270. 10.1089/ars.2006.8.243.

  48. 48.

    Aldieri E, Riganti C, Polimeni M, Gazzano E, Lussiana C, Campia I, Ghigo D: Classical inhibitors of NOX NAD(P)H oxidases are not specific. Curr Drug Metab. 2008, 9: 686-696. 10.2174/138920008786049285.

  49. 49.

    Altenhofer S, Kleikers PW, Radermacher KA, Scheurer P, Rob Hermans JJ, Schiffers P, Ho H, Wingler K, Schmidt HH: The NOX toolbox: validating the role of NADPH oxidases in physiology and disease. Cell Mol Life Sci. 2012, 69: 2327-2343. 10.1007/s00018-012-1010-9.

  50. 50.

    Zhang L, Nguyen MV, Lardy B, Jesaitis AJ, Grichine A, Rousset F, Talbot M, Paclet MH, Qian G, Morel F: New insight into the Nox4 subcellular localization in HEK293 cells: first monoclonal antibodies against Nox4. Biochimie. 2011, 93: 457-468. 10.1016/j.biochi.2010.11.001.

  51. 51.

    Campion Y, Paclet MH, Jesaitis AJ, Marques B, Grichine A, Berthier S, Lenormand JL, Lardy B, Stasia MJ, Morel F: New insights into the membrane topology of the phagocyte NADPH oxidase: characterization of an anti-gp91-phox conformational monoclonal antibody. Biochimie. 2007, 89: 1145-1158. 10.1016/j.biochi.2007.01.010.

  52. 52.

    Im YB, Jee MK, Choi JI, Cho HT, Kwon OH, Kang SK: Molecular targeting of NOX4 for neuropathic pain after traumatic injury of the spinal cord. Cell Death Dis. 2012, 3: e426-10.1038/cddis.2012.168.

  53. 53.

    Marti L, Morin N, Enrique-Tarancon G, Prevot D, Lafontan M, Testar X, Zorzano A, Carpéné C: Tyramine and vanadate synergistically stimulate glucose transport in rat adipocytes by amine oxidase-dependent generation of hydrogen peroxide. J Pharmacol Exp Ther. 1998, 285: 342-349.

  54. 54.

    Miller EW, Dickinson BC, Chang CJ: Aquaporin-3 mediates hydrogen peroxide uptake to regulate downstream intracellular signaling. Proc Natl Acad Sci U S A. 2010, 107: 15681-15686. 10.1073/pnas.1005776107.

  55. 55.

    Rodríguez A, Catalán V, Gómez-Ambrosi J, García-Navarro S, Rotellar F, Valentí V, Silva C, Gil MJ, Salvador J, Burrell MA, Calamita G, Malagón MM, Frühbeck G: Insulin- and leptin-mediated control of aquaglyceroporins in human adipocytes and hepatocytes is mediated via the PI3K/Akt/mTOR signaling cascade. J Clin Endocrinol Metab. 2011, 96: E586-E597. 10.1210/jc.2010-1408.

  56. 56.

    Costa D, Gomes A, Reis S, Lima JL, Fernandes E: Hydrogen peroxide scavenging activity by non-steroidal anti-inflammatory drugs. Life Sci. 2005, 76: 2841-2848. 10.1016/j.lfs.2004.10.052.

  57. 57.

    Costa D, Gomes A, Lima JL, Fernandes E: Singlet oxygen scavenging activity of non-steroidal anti-inflammatory drugs. Redox Rep. 2008, 13: 153-160. 10.1179/135100008X308876.

  58. 58.

    Fernandes E, Costa D, Toste SA, Lima JL, Reis S: In vitro scavenging activity for reactive oxygen and nitrogen species by nonsteroidal anti-inflammatory indole, pyrrole, and oxazole derivative drugs. Free Radic Biol Med. 2004, 37: 1895-1905. 10.1016/j.freeradbiomed.2004.09.001.

  59. 59.

    Brennan JP, Bardswell SC, Burgoyne JR, Fuller W, Schroder E, Wait R, Begum S, Kentish JC, Eaton P: Oxidant-induced activation of type I protein kinase A is mediated by RI subunit interprotein disulfide bond formation. J Biol Chem. 2006, 281: 21827-21836. 10.1074/jbc.M603952200.

  60. 60.

    Nascimento EA, Yamamoto NS, Bracht A, Ishii-Iwamoto EL: Naproxen inhibits hepatic glycogenolysis induced by Ca(2+)-dependent agents. Gen Pharmacol. 1995, 26: 211-218. 10.1016/0306-3623(94)00161-F.

  61. 61.

    Riveros-Rosas H, Zentella de Piña M, Guinzberg R, Saldaña-Balmori Y, Julián-Sánchez A, Saavedra-Molina A, Piña E: Antagonism between the metabolic responses induced by epinephrine and piroxicam on isolated rat hepatocytes. Eur J Pharmacol. 1999, 380: 49-59. 10.1016/S0014-2999(99)00521-X.

  62. 62.

    Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, Cooke MP, Walker JR, Hogenesch JB: A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci U S A. 2004, 101: 6062-6067. 10.1073/pnas.0400782101.

Download references


This work was partially supported by CONACyT-México grant 166733 to EP, UNAM-DGPA-PAPIIT grants IN216008 to MZ-P, IN214812 to RV-M, and IN216513 to HR-R. This study was performed in partial fulfillment of the requirements for the PhD degree in Biological Sciences of HV-M. We thank Enrique Moreno Hernández, Veterinary and Animal Husbandry Physician, for technical support, Alejandra Palomares for her secretarial contribution, and Maggie Brunner, M. A., and José Luis Pérez García, M.D., (Fac. Medicina, División de Investigación, UNAM) for helping in the correct usage of English in this manuscript.

Author information

Correspondence to Héctor Vázquez-Meza.

Additional information

Competing interest

The authors declare that there are no conflicts of interest.

Authors’ contributions

HV-M, EP and MZ-P designed the experimental strategy for this study and HV-M performed the experiments. HR-R and HV-M performed the statistical analysis. EP, JPP, HR-R, RV-M, HV-M and MZ-P analyzed and interpreted the data; EP, HR-R, HV-M, JPP, and RV-M wrote the manuscript. All the authors read and approved the final manuscript.

Authors’ original submitted files for images

Rights and permissions

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article


  • Non-steroidal anti-inflammatory drug(s) (NSAID)
  • Protein kinase A (PKA)
  • H2O2
  • Lipolysis
  • Acetylsalicylic acid
  • Aspirin