HSP60 associated with adiponectin receptors
Adiponectin receptors, AdipoR1 and AdipoR2, are important members in a new family of cell surface receptor, called Progestin and AdipoQ Receptor (PAQR) family . AdipoR1 is expressed ubiquitously and constitutively in most tissues and cells including adult cardiomyocytes and rat cardiac H9c2 cells [32, 33], while AdipoR2 is mainly expressed in the liver . To demonstrate the association between AdipoRs with HSP60, we firstly detected the localization of AdipoRs and HSP60 in H9c2 cells and mouse liver HepIR cells . Immunofluorescence staining revealed that endogenous HSP60 co-localized with endogenous AdipoR1 in H9c2 cells (Fig. 1a). Similar observation was also made when HSP60 was overexpressed in HepIR cells (Fig. 1b). To investigate whether HSP60 and AdipoRs associate directly, GST pull-down and co-immunoprecipitation assays were performed. As shown in Fig. 1c and 1d, endogenous AdipoR1 in H9c2 cells and AdipoR2 in HepIR cells interacted with GST-HSP60 but not with GST control proteins. Co-immunoprecipitation experiments revealed that overexpressed HSP60 interacted specifically with endogenous AdipoR1 in H9c2 cells (Fig. 1e) and endogenous AdipoR2 in HepIR cells (Fig. 1f). These findings indicate that HSP60 interacts directly with adiponectin receptors.
HSP60 mediated adiponectin action
To understand the functional role of HSP60 in regulating adiponectin action, intracellular HSP60 protein levels were increased by overexpression (OE) or decreased by siRNA knockdown (KD), respectively. The cells were then starved serum for 6 h, followed by stimulation with 1μg/ml adiponectin for 30 min. It has been reported that AdipoR1-mediated adiponectin signaling could be activated by globular adiponectin (gADPN) whereas AdipoR2 only bind with full-length adiponectin (fADPN) [31, 34]. Therefore, H9c2 cells and HepIR cells were treated with gADPN and fADPN, respectively. We found that phosphorylation of AMPK and p38 MAPK in response to adiponectin stimulation were greatly suppressed in HSP60-KD H9c2 cells (Fig. 2a and 2b) and HepIR cells (Fig. 2c and 2d) but obviously enhanced in HSP60-OE H9c2 cells (Additional file 1: Figure S1a and S1b) and HepIR cells (Additional file 1: Figure S1c and S1d), respectively. Since phosphorylation of AMPK in cardiomyocytes and p38 MAPK in hepatocytes are the markers of their activities [35, 36], our results demonstrate that HSP60 positively modulates adiponectin signaling.
In the present study, we also found that knocking down of HSP60 induced reductions of p38 MAPK and AMPK phosphorylation at basal levels (Fig. 2). Although the underlying mechanism is unclear, HSP60 has been proven to positively regulate p38 MAPK pathway in various cells [37, 38], suggesting that HSP60 plays a role in controlling p38 MAPK activity in both adiponectin-dependent and -independent mechanism. The impacts of HSP60 on AMPK activity is controversial. In cancer cells, HSP60 silencing can activate AMPK through triggering the excessive ROS production, which is beneficial for tumor progression [39, 40]. In adipose tissues, however, high-fat diet feeding induces a reduction of HSP60 protein levels and this change is not associated with any changes in AMPK activity . Our finding indicates that HSP60 deficiency reduced basal AMPK phosphorylation, suggesting that adiponectin-independent mechanism is also involved in HSP60 controlled AMPK activation. Future studies are needed to dissect the specific role of HSP60 in variety of the cells residing in fat tissues in regulating AMPK activity.
HSP60 knockdown mitigated the protective effects of adiponectin on high glucose-induced oxidative stress and cell apoptosis in H9c2 cells
Hyperglycemia is a hallmark feature of both type 1 and type 2 diabetes. Previous study has evidenced that high levels of glucose induce oxidative stress and cell apoptosis in cardiomyocytes [30, 42], which can be protected by adiponectin administration . Using this model, we wanted to further confirm the role of HSP60 in mediating adiponectin signaling.
H9c2 cells were starved serum for 6 h, and then incubated with 5.5 mM (normal glucose control) or 33 mM glucose (high glucose, HG) in the presence or absence of 1 μg/ml of gADPN for another 48 h. TUNEL and DHE staining assays were carried out to detect cell apoptosis and real-time formation of ROS, respectively. The cleaved caspase-3 was detected by western blot to confirm the progression of apoptosis.
We found that HSP60 depletion significantly increased cell apoptosis, even on normal glucose (Fig. 3a, Additional file 2: Figure S2a, and Fig. 3b). This finding is consistent with previous study showing that the deletion of HSP60 in adult cardiomyocytes results in the impairment of structure and function of cardiac muscle cells . Furthermore, adiponectin administration markedly inhibited HG-induced apoptosis in siRNA control cells (Fig. 3a, Additional file 2: Figure S2a, and Fig. 3b). However, these protective effects were almost completely diminished in HSP60-KD cells (Fig. 3a, Additional file 2: Figure S2a, and Fig. 3b). The similar effects on ROS formation were found in siRNA control or HSP60-KD cells treated with or without adiponectin (Fig. 3c and Additional file 2: Figure S2b). These findings further confirm the HSP60 regulation on adiponectin signaling.
HSP60 stabilized adiponectin receptor through a proteasome-dependent mechanism
HSP60 has been found to positively regulate insulin-like growth factor-1 (IGF-1) signaling, through maintaining the abundance of IGF-1 receptor in cardiac muscle cells . To figure out whether the similar mechanism exists in adiponectin receptors, we observed the effects of HSP60 KD on the protein levels of AdipoR1 in cardiac H9c2 cells. The cells were starved serum for 6 h, followed by stimulation with 1 μg/ml of gADPN for 18 h. We found that AdipoR1 expression was significantly reduced by HSP60 depletion but not affected by adiponectin treatment (Fig. 4a and Fig. 4b), suggesting that HSP60 depletion induced AdipoR1 degradation.
It is well-known that intracellular protein degradation is mainly induced by two cellular routes: the ubiquitin-proteasome system (UPS) and the autophagy-lysosome system . HSP60 has been reported to modulate proteasome activity and protein ubiquitination [45, 47]. We thus investigated the potential effects of HSP60 on the UPS. Indeed, HSP60 depletion markedly decreased the ubiquitination of total proteins (Fig. 4c). In addition, 20S proteasome activity was also greatly enhanced in HSP60-KD H9c2 cells (Additional file 3: Figure S3). Consistent with study performed in yeast , these findings demonstrate that HSP60 can inhibit proteasome activity in the mammalian cells.
When H9c2 cells were starved serum for 6 h, followed by incubation with 0.1 μM of MG132, a specific proteasome inhibitor for 18 h, we found that proteasome inhibition significantly restored HSP60 depletion-reduced protein levels of AdipoR1 (Fig. 4d and 4e). Proteasome inhibition also significantly increased the ubiquitination of total proteins when compared with HSP60-KD cells (Fig. 4f). These findings further suggested that HSP60 depletion-induced AdiopR1 degradation is mediated by a proteasome-dependent mechanism.