Our study demonstrated that the sequential events of PKA and
Our study demonstrated that the sequential events of PKA and AMPK activation were involved in kinsenoside-mediated lipolysis. Within 1 h, PKA transiently inhibited AMPK activation by mitigating LKB1-mediated AMPK phosphorylation at Thr172, and PKA reduced AMPK-mediated phosphorylation at HSL-Ser565 (an inactive form). In both of these events, AMPK was subsequently activated by kinsenoside-mediated phosphorylation at Thr172 for up to 3 h of treatment (Fig. 3B); this finding is consistent with a previous finding that lipolysis is enhanced by long-term treatment with AICAR, an AMPK activator (Gaidhu et al., 2009). Thus, the PKA–HSL–perilipin and AMPK–ATGL–CPT1 axes were successively activated by kinsenoside, and were responsible for the lipolytic effect of kinsenoside in adipocytes (depicted in Fig. 6). ATGL and HSL are responsible for more than 90% of TG hydrolysis in adipose tissue (Schweiger et al., 2006). Although some related findings have been somewhat controversial (Ryden et al., 2007), ATGL is considered responsible for the initial step of lipolysis in human adipocytes, and HSL is the rate-limiting enzyme for the catabolism of DG (Bezaire et al., 2009). We demonstrated that HSL was activated by PKA before ATGL was activated by AMPK in kinsenoside-mediated lipolysis. This may occur because in terms of specificity, HSL is a considerably less discriminate lipase than ATGL and can catalyze multiple substrates (e.g., TGs, DGs, MGs, cholesteryl esters, retinyl esters, and short-chain carbonic (R)-PFI 2 hydrochloride esters) (Fredrikson et al., 1986, Wei et al., 1997). Thus, HSL may play a crucial role in lipolysis. However, the general conception of fatty-acid hydrolysis is that ATGL is first activated to hydrolyze TG to DG, and subsequently, HSL hydrolyzes DG to MG. In addition, potential modulations in cAMP level, AMP-to-ATP ratio, and other unidentified factors during the switch between PKA-induced and AMPK-induced activation in lipolysis should be further examined. Evidence has shown off-target effects of H89 on other kinases (Murray, 2008). To validate that HSL and perilipin are activated in a PKA-dependent manner, a more specific PKA agonist, 6-Bnz-cAMPs (100 µM), and an antagonist, Rp-8-Br-cAMPs (100 µM), were used in the present study. The effects of the agonist and antagonist were compared with the lipolytic effect of kinsenoside through Western blot analysis and oil-red O staining. Similar to the lipolytic effect of kinsenoside, 6-Bnz-cAMPs increased lipolysis through the phosphorylation of HSL-Ser660/563 and perilipin; 6-Bnz-cAMPs also reduced lipid accumulation. Conversely, similar to H89, Rp-8-Br-cAMPs reversed the effects of increased lipolysis and reduced lipid accumulation. In further similar with H89, Rp-8-Br-cAMPs inhibited PKA activity, reactivated AMPK through phosphorylation at Thr172, and then increased the induction of ATGL and CPT1. These effects were mitigated by 6-Bnz-cAMPs. Although PKA was found to transiently suppress AMPK activation (reflected in the increased levels of the inactivated form of pAMPK-Ser173 and the reduced levels of pHSL-Ser565 [Fig. 3]), kinsenoside-mediated lipolysis occurred on the PKA–HSL–perilipin and AMPK–pPPARα–PGC1α–ATGL–CPT1 axes. Thus, activation of the PKA and AMPK pathways occurred successively, and PKA negatively modulated AMPK activity. Ultimately, PKA and the subsequently activated AMPK synergistically increased lipolysis in kinsenoside-treated adipocytes. The results of our study demonstrated that HSL activation by kinsenoside-mediated PKA activation was coordinated with perilipin phosphorylation and facilitated efficient lipolysis, resulting in increased glycerol release and decreased lipid accumulation. The localization and interplay of perilipin and HSL in lipolysis remain unclear; the literature presents contradictory results on this topic. Studies have reported that phosphorylated perilipin dissociates from LDs before HSL catalyzes hydrolysis. However, other research groups comparing these perilipin and HSL interactions to those in isoproterenol stimulation in young and old mice have stated that perilipin's dissociation from and HSL's binding to the surface of LDs are not necessarily sequential events. Such groups have observed that lipid hydrolysis requires only HSL to dock on the surface of LDs, independent of perilipin subcellular localization. Additionally, studies have proposed that perilipin may serve as a docking protein for the function of HSL (Egan et al., 1992), although other studies have shown that no such protein–protein interaction occurs in a Co-IP and yeast two-hybrid system (Clifford et al., 2000, Shen et al., 1999). Furthermore, one study showed that norepinephrine induced lipolysis in rat fat cells and that this induction of lipolysis was dependent on the localization of pHSL to the LD surface and independent of the level and activity of pHSL-Ser660/563 (Morimoto et al., 2001). Hence, the mechanism underlying the interplay of HSL and perilipin in TG hydrolysis requires further verification.