Direct evidence for protein phosphatase-catalyzed dephosphorylation/deactivation of hormone-sensitive lipase from adipose tissue

Adipokinetic hormone-induced mobilization of fat body triglyceride stores in Manduca sexta: role of TG-lipase and lipid droplets

Triglycerides (TG) stores build up in the insect fat body as lipid droplets at times of excess of food. The mobilization of fat body triglyceride (TG) is stimulated by adipokinetic hormones (AKH). The action of AKH involves a rapid activation of cAMP-dependent protein kinase (PKA). Recent in vitro studies have shown that PKA phosphorylates and activates the TG-lipase substrate, the lipid droplets. Conversely, purified TG-lipase from Manduca sexta fat body is phosphorylated by PKA in vitro but is not activated. This study was directed to learn whether or not AKH promotes a change in the state of phosphorylation of the lipase in vivo, and what are the relative contributions of cytosol and lipid droplets to the overall increase of lipolysis triggered by AKH. TG-lipase activity of fat body cytosols isolated from control and AKH-treated insects was determined against the native substrate, in vivo [3H]-TG radiolabeled lipid droplets, obtained from control and AKH-treated insects. The lipase activity of the system composed of AKH-cytosol and AKH-lipid droplets (11.1 +/- 2.1 nmol TG/min-mg) was 3.1-fold higher than that determined with control cytosol and lipid droplets (3.6 +/- 0.5 nmol TG/min-mg). Evaluation of the role of AKH-induced changes in the lipid droplets on lipolysis showed that changes in the lipid droplets are responsible for 70% of the lipolytic response to AKH. The remaining 30% appears to be due to AKH-dependent changes in the cytosol. However, the phosphorylation level of the TG-lipase was unchanged by AKH, indicating that phosphorylation of the TG-lipase plays no role in the activation of lipolysis induced by AKH.

Activation of the lipid droplet controls the rate of lipolysis of triglycerides in the insect fat body

The hydrolysis of triglyceride (TG) stored in the lipid droplets of the insect fat body is under hormonal regulation by the adipokinetic hormone (AKH), which triggers a rapid activation cAMP-dependent kinase cascade (protein kinase A (PKA)). The role of phosphorylation on two components of the lipolytic process, the TG-lipase and the lipid droplet, was investigated in fat body adipocytes. The activity of purified TG-lipase determined using in vivo TG-radiolabeled lipid droplets was unaffected by the phosphorylation of the lipase. However, the activity of purified lipase was 2.4-fold higher against lipid droplets isolated from hormone-stimulated fat bodies than against lipid droplets isolated from unstimulated tissue. In vivo stimulation of lipolysis promotes a rapid phosphorylation of a lipid droplet protein with an apparent mass of 42-44 kDa. This protein was identified as "Lipid Storage Droplet Protein 1" (Lsdp1). In vivo phosphorylation of this protein reached a peak approximately 10 min after the injection of AKH. Supporting a role of Lsdp1 in lipolysis, maximum TG-lipase activity was also observed with lipid droplets isolated 10 min after hormonal stimulation. The activation of lipolysis was reconstituted in vitro using purified insect PKA and TG-lipase and lipid droplets. In vitro phosphorylation of lipid droplets catalyzed by PKA enhanced the phosphorylation of Lsdp1 and the lipolytic rate of the lipase, demonstrating a prominent role PKA and protein phosphorylation on the activation of the lipid droplets. AKH-induced changes in the properties of the substrate do not promote a tight association of the lipase with the lipid droplets. It is concluded that the lipolysis in fat body adipocytes is controlled by the activation of the lipid droplet. This activation is achieved by PKA-mediated phosphorylation of the lipid droplet. Lsdp1 is the main target of PKA, suggesting that this protein is a major player in the activation of lipolysis in insects.

cAMP-dependent protein kinase of Manduca sexta phosphorylates but does not activate the fat body triglyceride lipase

cAMP-dependent-protein kinase (PKA) is a central player of the adipokinetic signal that controls the mobilization of stored lipids in the fat body. Previous studies showed that adipokinetic hormone (AKH) rapidly activates PKA from the fat body of Manduca sexta (Arrese et al. (J. Lipid. Res. 40(3): 556)). As a part of our investigation on lipolysis in insects, here we report the purification and characterization of the catalytic subunit of PKA from the fat body of M. sexta and its role in the direct activation of the TG lipase in vitro. PKA was purified to apparent homogeneity and the identity of the protein was confirmed by MALDI-TOF and Western blot analysis. The enzyme showed a high affinity for Mg-ATP (Km = 39 microM) and Kemptide (Km = 31 microM) and was strongly inhibited by the PKA specific inhibitors PKI 5-24 and H89. Manduca sexta PKA only recognized serine residues as phosphate acceptor; theronine or tyrosine containing peptides were not phosphorylated. Purified fat body TG-lipase proved to be a good substrate of the purified kinase. However, phosphorylation of the lipase did not enhance the lipolytic activity of the enzyme in vitro. These results suggest that, besides lipase phosphorylation, the mechanism of AKH-induced activation of the lipolysis requires the involvement of other proteins and/or signals.

Desensitization of beta-adrenergic responses in adipocytes involves receptor subtypes and cAMP phosphodiesterase

Acute exposure of isolated adipocytes to isoproterenol induces the desensitization of lipolytic responses to norepinephrine and selective beta 1-, beta 2- and beta 3-adrenoceptor agonists, as well as the adrenocorticotropic hormone 1-24 fragment (ACTH). Forskolin and 8-bromo-cAMP responses are also desensitized. When lipolysis was measured in the presence of OPC 3911 [N-cyclohexyl-N-2-hydroxyethyl-4(6-(1,2-dihydro-2- oxoquinolyloxy))butyramide], a specific inhibitor of the cAMP phosphodiesterase of adipocytes, the desensitization of all lipolytic agents--except the beta 2-adrenoceptor agonist--was abolished. Isoproterenol induced a similar loss (35%) of both membrane beta 1- and beta 2-adrenoceptors and an uncoupling of beta 1-adrenoceptors, but did not modify the weak coupling of control beta 2-adrenoceptors. These data suggest that isoproterenol induced (i) an activation of the cAMP phosphodiesterase, which is solely responsible for the desensitization of norepinephrine response as well as beta 1- and beta 3-adrenoceptor mediated responses and (ii) an additional desensitization of the sole beta 2-adrenergic signaling system which suggests a subtype-selective pattern of regulating processes.

Pyrazinoylguanidine downregulates the glucose fatty-acid cycle in hypertensive, hyperinsulinemic diabetic patients

Eight hypertensive patients with noninsulin dependent diabetes mellitus (NIDDM) were administered the experimental drug pyrazinoylguanidine (PZG) either alone or in combination with calcium-channel or beta-blockers. This treatment appeared to "downregulate" the glucose fatty acid cycle and reduced both systolic and diastolic blood pressures and mean body weight. Patients served as their own controls in this dose-escalation study, which included placebo treatment (baseline) 3 weeks, 300 mg PZG for 3 weeks and 600 mg for 3 weeks. PZG reduced increased serum concentrations of free fatty acids (FFA), glucose, and triglycerides (TG). TG concentrations correlated inversely with serum HDL-cholesterol concentrations. The beta-blockers used by several patients increased their FFA, glucose, insulin and TG concentrations, as well as blunting their response to PZG. The calcium-channel blockers exerted these effects to a much lesser extent. PZG reduced or abolished glycosuria, related to PZG's capacity to decrease hyperglycemia. Withdrawal of PZG restored glycosuria, as blood sugar increased. PZG was well tolerated. No patient reported any adverse effect or missed a weekly clinic visit (12 weeks). PZG deserves further study as supplementary and/or replacement therapy in NIDDM patients who are hypertensive and hyperlipidemic.

Insulin-induced dephosphorylation of hormone-sensitive lipase. Correlation with lipolysis and cAMP-dependent protein kinase activity

The effect of insulin on the state of phosphorylation of hormone-sensitive lipase, cellular cAMP-dependent protein kinase activity and lipolysis was investigated in isolated adipocytes. Increased phosphorylation of hormone-sensitive lipase in response to isoproterenol stimulation was closely paralleled by increased lipolysis. Maximal phosphorylation and lipolysis was obtained when the cAMP-dependent protein kinase activity ratio was greater than or equal to 0.1, and this corresponded to a 50% increase in the state of phosphorylation of hormone-sensitive lipase. Insulin (1 nM) reduced cAMP-dependent protein kinase activity and also reduced lipolysis with both cAMP-dependent and cAMP-independent antilipolytic effects up to an activity ratio of approximately 0.4, above which the antilipolytic effect was lost. Insulin caused a decrease in the state of phosphorylation of hormone-sensitive lipase at all levels of cAMP-dependent protein kinase activity. Under basal conditions, with cAMP-dependent protein kinase activity at a minimum, this reflected a dephosphorylation of the basal phosphorylation site of hormone-sensitive lipase in a manner not mediated by cAMP. When the cAMP-dependent protein kinase was stimulated to phosphorylate the regulatory phosphorylation site of hormone-sensitive lipase, the insulin-induced dephosphorylation occurred both at the basal and regulatory sites. At low levels of cAMP-dependent protein kinase activity ratios (0.05-0.1), dephosphorylation of the regulatory site correlated with reduced cAMP-dependent protein kinase activity, but not at higher activity ratios (greater than 0.1). Stimulation of cells with isoproterenol produced a transient (1-5 min) peak of cAMP-dependent protein kinase activity and of phosphorylation of hormone-sensitive lipase. The state of phosphorylation also showed a transient peak when the protein kinase was maximally and constantly activated. In the presence of raised levels of cellular cAMP, insulin (1 nM) caused a rapid (t1/2 approximately 1 min) dephosphorylation of hormone-sensitive lipase. In unstimulated cells the reduction in phosphorylation caused by insulin was distinctly slower (t1/2 approximately 5 min). These findings are interpreted to suggest that insulin affects the state of phosphorylation of hormone-sensitive lipase and lipolysis through a cAMP-dependent pathway, involving reduction of cAMP, and through a cAMP-independent pathway, involving activation of a protein phosphatase activity that dephosphorylates both the regulatory and basal phosphorylation sites of hormone-sensitive lipase.

The regulatory and basal phosphorylation sites of hormone-sensitive lipase are dephosphorylated by protein phosphatase-1, 2A and 2C but not by protein phosphatase-2B

The activity of hormone-sensitive lipase, the rate-limiting enzyme in adipose tissue lipolysis, is controlled by cAMP-mediated phosphorylation at a specific regulatory phosphorylation site. The lipase is also phosphorylated at a site, termed basal, without any effects on its activity [Strålfors et al. (1984) Proc. Natl Acad. Sci. USA 81, 3317-3321]. The capacity of protein phosphatase-1, 2A, 2B and 2C to dephosphorylate the lipase, selectively phosphorylated by glycogen synthase kinase-4 and cAMP-dependent protein kinase at the basal and regulatory phosphorylation sites, was compared with that towards glycogen phosphorylase and phosphorylase kinase (alpha subunit). Protein phosphatase-1, 2A and 2C were found to dephosphorylate both phosphorylation sites of hormone-sensitive lipase, while protein phosphatase-2B had no measureable activity towards any of the sites. When the activities of protein phosphatase-1, 2A and 2C were normalized with respect to the reference substrates, they were found to dephosphorylate the lipase regulatory site in the approximate relations of 1:4:3 and the basal site in the approximate relations of 1:6:4. Protein phosphatase-1 showed 20% higher and protein phosphatase-2A and 2C 80% higher activity towards the basal site compared to the regulatory site. The two phosphorylation sites of the lipase were comparable to good substrates for protein phosphatase-2A and 2C, but relatively poor substrates for protein phosphatase-1. Protein phosphatase-2C activity towards the lipase was completely dependent on Mg2+ with a half-maximal effect at 3 mM. Protamine increased the lipase dephosphorylation by protein phosphatase-1 3-5-fold with half-maximal effect at 0.6 microgram/ml, and by protein phosphatase-2A about 2-fold with half-maximal effect at 3-5 micrograms/ml, thus illustrating the potential for control of these lipase phosphatase activities.

Hormone-sensitive lipase from bovine adipose tissue

Hormone-sensitive lipase has been purified to near homogeneity from bovine perirenal adipose tissue. The purification method involves isoelectric precipitation at pH 5.0, followed by partial solubilisation in Triton N-101 and ion-exchange chromatography on DE-52. After additional solubilisation, the enzyme is further purified by chromatography on phenyl-Sepharose and heparin-Sepharose. This procedure can be completed within three working days and yields approx. 30 units of enzyme with a specific activity of 30 U/mg. The enzyme has been identified as a polypeptide of Mr 84 000 by affinity labelling with [3H]diisopropyl fluorophosphate. This polypeptide comprises approx. 60-80% of the protein in the final preparation, as judged by scanning densitometry of SDS-polyacrylamide gels stained with silver or with Coomassie blue R. The polypeptide of Mr 84 000 serves as a substrate for cyclic AMP-dependent protein kinase, phosphorylation correlating with activation of the lipase. Polyclonal antibody to the lipase has been raised in a rabbit and shown to specifically cross-react with the Mr 84 000 subunit.

Stimulation of the hormone-sensitive triacylglycerol lipase from adipose tissue by phosphatidylethanolamine

The activity of a pigeon adipose tissue hormone-sensitive triacylglycerol lipase preparation was increased from 2- to 5-fold by the presence of phosphatidylethanolamine in assays with three different methods of preparing triolein substrates. Phosphatidylethanolamine from egg yolk produced the greatest stimulation of lipase activity; the stimulation was concentration-dependent but was not time-dependent. A comparable increase in triacylglycerol lipase activity due to phosphatidylethanolamine was also observed with enzyme preparations from chicken and rat adipose tissue. Phosphatidylcholine, phosphatidylserine, phosphatidylinositol, phosphatidic acid, cardiolipin, sphingomyelin, Triton X-100 and sodium dodecyl sulfate all inhibited enzyme activity. Phosphatidylethanolamine had no effect on acid lipase activity in the pigeon adipose tissue preparation. Preincubation of the pigeon adipose tissue lipase with ATP, cyclic AMP and protein kinase resulted in a 2.15-fold activation of hydrolase activity determined in the absence of phosphatidylethanolamine. In contrast, non-activated and protein kinase-activated forms of the lipase were characterized as having very nearly the same activity in assays with substrate preparations containing phosphatidylethanolamine. The phosphatidylethanolamine-dependent stimulation of lipase activity was characterized kinetically as being due to an increase in maximal velocity. The modulation of the adipose tissue hormone-sensitive lipase activity by phospholipids could be involved in the hormonal regulation of lipolysis.