Steatohepatitis/Metabolic Liver Disease
Resistin reduces mitochondria and induces hepatic steatosis in mice by the protein kinase C/protein kinase G/p65/PPAR gamma coactivator 1 alpha pathway†
Article first published online: 25 JAN 2013
Copyright © 2012 American Association for the Study of Liver Diseases
Volume 57, Issue 4, pages 1384–1393, April 2013
How to Cite
Zhou, L., Yu, X., Meng, Q., Li, H., Niu, C., Jiang, Y., Cai, Y., Li, M., Li, Q., An, C., Shu, L., Chen, A., Su, H., Tang, Y., Yin, S., Raschke, S., Eckardt, K., Eckel, J. and Yang, Z. (2013), Resistin reduces mitochondria and induces hepatic steatosis in mice by the protein kinase C/protein kinase G/p65/PPAR gamma coactivator 1 alpha pathway. Hepatology, 57: 1384–1393. doi: 10.1002/hep.26167
Potential conflict of interest: Nothing to report.
- Issue published online: 8 APR 2013
- Article first published online: 25 JAN 2013
- Accepted manuscript online: 23 NOV 2012 04:33AM EST
- Manuscript Accepted: 31 OCT 2012
- Manuscript Revised: 17 OCT 2012
- Manuscript Received: 10 JUL 2012
- National Key Basic Research Program of China. Grant Number: 2012CB124702
- Major Projects of National Science and Technology. Grant Number: 2009ZX08009-151B
- European Foundation for the Study of Diabetes/Chinese Diabetes Society/Lilly research grant. Grant Number: 948 Program (2012-S13)
- Specialized Research Fund for the Doctoral Program of Higher Education. Grant Number: 20110146130002
- National Natural Science Foundation of China. Grant Numbers: 31172093, 30970356
- National Innovation Experiment Program for University Students. Grant Number: 101050410
- Scientific Research Foundation for Returned Scholars of Ministry of Education of China
Obesity is associated with many severe chronic diseases and deciphering its development and molecular mechanisms is necessary for promoting treatment. Previous studies have revealed that mitochondrial content is down-regulated in obesity, diabetes, and nonalcoholic fatty liver disease (NAFLD) and proposed that NAFLD and diabetes are mitochondrial diseases. However, the exact mechanisms underlying these processes remain unclear. In this study, we discovered that resistin down-regulated the content and activities of mitochondria, enhanced hepatic steatosis, and induced insulin resistance (IR) in mice. The time course indicated that the change in mitochondrial content was before the change in fat accumulation and development of insulin resistance. When the mitochondrial content was maintained, resistin did not stimulate hepatic fat accumulation. The present mutation study found that the residue Thr464 of the p65 subunit of nuclear factor kappa B was essential for regulating mitochondria. A proximity ligation assay revealed that resistin inactivated peroxisome proliferator activated receptor gamma coactivator 1 alpha (PGC-1α) and diminished the mitochondrial content by promoting the interaction of p65 and PGC-1α. Signaling-transduction analysis demonstrated that resistin down-regulated mitochondria by a novel protein kinase C/protein kinase G/p65/PGC-1α-signaling pathway. Conclusion: Resistin induces hepatic steatosis through diminishing mitochondrial content. This reveals a novel pathway for mitochondrial regulation, and suggests that the maintenance of normal mitochondrial content could be a new strategy for treatment of obesity-associated diseases. (HEPATOLOGY 2013)
Obesity is a global health problem and is associated with many chronic diseases, such as nonalcoholic fatty liver disease (NAFLD),1 cardiovascular diseases (CVD),2 type 2 diabetes,3 hypertension,4 and certain cancers.5 Although many related genes and signaling pathways have been revealed, an understanding of the development of obesity remains limited. As a cellular energy source, mitochondria are involved in the regulation of fatty acid (FA) oxidation (FAO) and apoptosis. Accumulated evidence indicates that mitochondrial content is down-regulated in obesity,6 diabetes,7 and NAFLD.8, 9 Many antidiabetic therapies have been shown to enhance mitochondrial biogenesis.10, 11 Although how and why these mechanisms are regulated remains unclear, they give support to the idea that obesity-associated diseases are significantly inversely related to mitochondrial status.
Resistin was first described in 2001 and is secreted from adipose tissue12 and was first thought to be related to the development of insulin resistance (IR) and therefore named resistin. Resistin is up-regulated during the course of adipocyte differentiation and fat deposition13 and is expressed at higher levels in obese humans and mice, compared to lean controls.14 Its levels are positively correlated with body mass index15 and NAFLD16 and are significantly decreased upon weight loss.17 These results suggest that a higher resistin level is associated with higher levels of glucose and fat, supporting a positive correlation between resistin and obesity.
Previous studies have shown that mitochondrial content was down-regulated in obesity, diabetes, and NAFLD. However, the exact mechanisms underlying these processes remain unclear. Existing evidence has demonstrated that during the development of obesity-associated diseases, changes in mitochondrial content and resistin levels show opposite trends. To clarify whether increased resistin signals are associated with decreased mitochondrial content, we analyzed the regulatory effect of resistin on mitochondria in vivo and in vitro and investigated the molecular mechanisms by which resistin exerts its biological effect.
Materials and Methods
TRIzol reagent was purchased from TaKaRa (Dalian, China), Lipofectamine 2000 was purchased from Invitrogen (Carlsbad, CA), and a mammalian cell protein extraction kit was purchased from Beyotime (Jiangsu, China). Antibodies (Abs) against peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC-1α) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). p65, phosphorylated protein kinase B (Akt), and tubulin Abs were supplied by Cell Signaling Technology (Danvers, MA, USA). Recombinant human and mouse resistin were purchased from PeproTech, Inc. (Rocky Hill, NJ). Reactive oxygen species (ROS) fluorescent probe dihydroethidine (DHE) and the ATP-Lite Assay Kit were purchased from Vigorous Biotechnology (Beijing, China). Bovine serum albumin (BSA; fraction V, FA free) was obtained from Roche (Basel, Switzerland). A mouse insulin enzyme-linked immunosorbent assay (ELISA) kit was purchased from Wuhan Xinqidi Biological Technology Co., Ltd. (Wuhan, China). A glucose determination kit and triacylglycerol (TAG) assay kit were purchased from Applygen Technologies Co., Ltd. (Beijing, China). Pyrrolidine dithiocarbamate (PDTC), PD98059, aminoimidazole carboxamide ribonucleotide (AICAR), and rosiglitazone were purchased from Sigma-Aldrich (St. Louis, MO). BPIPP, NS2028, H89, phloretin, KT5823, phorbol 12-myristate 13-acetate (PMA), and 8-bromo-cGMP (cyclic guanosine monophosphate) were purchased from Santa Cruz Biotechnology. Palmitic acid (PA) and oleic acid (OA) were provided by Sigma-Aldrich. Small interfering RNA was synthesized by IBS Bio (Shanghai, China). Pierce bicinchoninic acid (BCA) protein quantitative assay kits were purchased from Thermo-Fisher Scientific (Waltham, MA), and a plasmid extraction kit was purchased from Tiangen Biotech Co. Ltd. (Beijing, China).
Male C57BL/6J mice (8 weeks old) were purchased from Huafukang Biotech (Beijing, China) and housed in individual plastic cages on a 12-hour light/dark cycle with free access to water and food at room temperature. Mice were given standard chow and water and given a daily vena caudalis injection for 6 days with or without resistin (400 ng/day). Mice were sacrificed on day 7. All procedures were approved by the Hubei Province Committee on Laboratory Animal Care.
DNA Isolation and Quantitative Real-Time Polymerase Chain Reaction Analysis of Mitochondria.
Genomic DNA (gDNA) was isolated from cultured cells or mouse tissues using the Qiagen DNA extraction kit (Qiagen, Hilden, Germany). Relative content of mitochondrial DNA (mtDNA) was determined by quantitative real-time polymerase chain reaction (qPCR). The ratio of mtDNA to nuclear DNA (nDNA) reflects the content of the mitochondria. Primers for mtDNA and nDNA qPCR are shown in Supporting Table 1.
Real-Time Reverse-Transcription PCR.
Real-time reverse-transcription PCR was used to determine messenger RNA levels of genes with a SYBR Green PCR Kit (TaKaRa) using β-actin as an internal control. Sequences of primers and accession numbers for each gene are shown in Supporting Table 2.
Cell Culture and Treatment.
HepG2 cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum under a 5% CO2 atmosphere at 37°C. The control group was cultured without recombinant resistin, whereas the treatment group was cultured with recombinant resistin (25 ng/mL). Cells were collected 24 hours after treatment to isolate their proteins, RNA, or gDNA.
Flow Cytometry Analyses.
Intracellular ROS level was determined using DHE, as previously described.18 Briefly, cultured cells were incubated with DHE (10 μM) for 30 minutes at 37°C under dark conditions and then washed with phosphate-buffered saline (PBS). Cells were then harvested with 0.025% trypsin/ethylenediaminetetraacetic acid, washed with PBS, and finally resuspended in PBS. Samples were analyzed using the FACSCalibur flow cytometer with CellQuest software (BD Biosciences, Franklin Lakes, NJ). Mitochondrial membrane potential (MMP; Δψm) was determined using an MMP assay kit (Beyotime). Briefly, cultured cells were incubated with a buffer containing 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolcarbocyanine iodide (JC-1; 1:200) for 20 minutes at 37°C. Cells were then washed twice with staining buffer on ice to remove excess probe. The ΔΨm was assessed using the FACSCalibur flow cytometer (BD Biosciences).
Measurement of TAG and Glycerol.
HepG2 cells were cultured in serum-free DMEM supplemented with free FAs (FFAs; 0.2 mM of OA, 0.1 mM of PA, and 0.075 mM of BSA)19 or FFA plus resistin (0.2 mM of OA, 0.1 mM of PA, 0.075 mM of BSA, and 25 ng/mL of resistin). TAG and glycerol were measured using a TAG assay kit and a glycerol assay kit (Applygen Technologies Co. Ltd., Beijing, China), respectively. Values were normalized to protein concentrations using the Pierce BCA protein quantitative assay kit (Thermo-Fisher Scientific).
Data are presented as means ± standard deviation (SD). Statistical analysis was performed using the unpaired two-tailed t test (for two groups) and analysis of variance (for multiple groups). P values <0.05 were considered statistically significant.
Resistin Down-regulated Mitochondrial Content.
Analysis of the ratio of mtDNA to nDNA in HepG2 cells showed that the direct addition of resistin markedly diminished mitochondrial content in a dose-dependent manner (Fig. 1A). The time course studied indicated that the effect of resistin reached significance after incubation for 4 hours (Fig. 1B). Subsequently, the in vivo study also confirmed this finding. C57BL/6J mice were treated with or without resistin for 6 days. qPCR showed that mitochondrial content in livers of resistin-treated animals was significantly lower (Fig. 1C). Moreover, flow cytometry data verified the change of mitochondrial content (Fig. 1D). These data proved our hypothesis and confirmed that increased resistin signaling down-regulated mitochondrial content. The in vivo study also indicated that resistin significantly stimulated levels of blood glucose, insulin, and TAG. Data of the homeostasis model assessment of IR (HOMA-IR) revealed resistin-induced IR (Table 1).
|Control Group11||Resistin-Administered Group10|
|Glucose, mM||4.59 ± 0.52||5.87 ± 0.55**|
|Insulin, mU/L||35.22 ± 3.25||48.12 ± 3.75**|
|TAG, mM||0.58 ± 0.06||0.83 ± 0.05**|
|HOMA-IR||7.18 ± 0.94||12.56 ± 1.61**|
Resistin Impaired Mitochondrial Function.
To investigate the effect of resistin on mitochondrial function, HepG2 cells were cultured with or without 25 ng/mL of resistin for 24 hours, followed by measurement of Δψm and intracellular ROS and adenosine triphosphate (ATP) content. Resistin diminished Δψm and ATP levels substantially, but had little effect on ROS levels (Figs. 2A-C). The study of transcription levels indicated that resistin stimulated ucp2 expression, but did not influence sod2 RNA levels (Fig. 2D). Moreover, genes in the tricarboxylic acid (TCA) cycle and electron transport chain (ETC) were also assayed. Resistin inhibited the expressions of citrate synthase (belonging to TCA) and the cytochrome c oxidase subunit Vα (belonging to ETC complex IV) and had no effect on the expression of others (Figs. 2E,F). These results implied that resistin diminished ATP levels through increasing the uncoupling effect and impairing the functions of TCA and ETC.
Resistin Enhanced Hepatic Steatosis.
Decrease in mitochondria content was correlated to changes in fat metabolism. Subsequently, mouse epididymal fat and liver were collected and analyzed. Histomorphological results indicated that there was no difference in weight and cell size of epididymal fat between the control and the resistin-administered groups (Fig. 3A); however, there were more, and larger, vacuoles in the hepatic cytoplasm of the resistin-administered group (Fig. 3B). Furthermore, TAG levels were significantly higher in the resistin-administered group, compared to the control group (Fig. 3B). To understand the role of resistin in hepatic fat accumulation, HepG2 cells were cultured with FAs (as described above) and with or without 25 ng/mL of resistin for 24 or 48 hours. Cells were then harvested to measure TAG and glycerol contents. Results demonstrated that resistin increased TAG levels and decreased glycerol levels (Fig. 3C,D). The result also showed that resistin inhibited the activity of acyl-CoA (coenzyme A) dehydrogenase (CAD), which catalyzes the first reaction of FA β-oxidation (Fig. 3E). However, after 24 hours of treatment, resistin did not change the phosphorylation level of Akt (Ser473) (Fig. 3F).
Resistin Activated PKG by PKC.
To clarify the signal transduction of resistin, the second messengers, cyclic adenosine monophosphate (cAMP) and cGMP, were measured. Resistin stimulated intracellular cAMP, but had no effect on cGMP (Fig. 4A). The cAMP-dependent protein kinase (PKA) inhibitor (H89) (50 nM) was added and was expected to block the effect of resistin, but the results indicated that inhibition did not occur (Supporting Fig. 1A). A higher concentration of H89 (5 μM) blocked the effect of resistin (Fig. 4B); however, at this concentration, it also inhibited protein kinase C (PKC) and cGMP-dependent protein kinase (PKG).20, 21 To distinguish the protein kinases involved in resistin action, the inhibitors, phloretin (a PKC inhibitor) and KT5823 (a PKG inhibitor), were both found to inhibit decreases observed in the mitochondria (Fig. 4C). Subsequently, to explore the upstream signal transduction of PKC, U73122 (a PLC inhibitor) was used, but could not block the effect of resistin (Supporting Fig. 1B). cGMP is a classic agonist for PKG, and cellular cGMP production is dependent on two kinds of guanylyl cyclases (GCs). The first is located on the plasma membrane and termed particulate guanylyl cyclase (pGC), whereas the second is located in the cytoplasm and termed soluble guanylyl cyclase (sGC).22 Neither BPIPP (a pGC inhibitor) nor NS2028 (an sGC inhibitor) could maintain mitochondrial content (Supporting Fig. 1C,D). The direct addition of 8-bromo-cGMP (a cGMP analog) did not change mitochondrial content (Fig. 4D). However, PMA (a PKC agonist) mimicked the effect of resistin and diminished mitochondrial content. Its effect was blocked by KT5823 (Fig. 4D). These data indicated that activation of PKG by resistin is independent of cGMP and that resistin activates PKG by PKC. Furthermore, inhibition of PKG blocked the action of resistin (Fig. 4E), indicating that resistin functions through PKG. Because the 48-hour treatment of resistin increased fat accumulation (Fig. 3C), we cultured cells with resistin and KT5823 and detected TAG content after incubation for 48 hours. The data showed that when mitochondrial content was maintained by KT5823 (Fig. 4C), cellular TAG was restored to normal levels (Fig. 4F).
Resistin-Regulated Mitochondria by p65.
Bioinformatic analysis predicted that resistin functions through the nuclear factor kappa B (NF-κB)-, insulin-, adenosine-monophosphate–activated protein kinase (AMPK)-, and extracellular signal-related kinase 1/2 (Erk1/2)-signaling pathways (described in Supporting Tables 3 and 4). To confirm this prediction, AICAR (an AMPK agonist), PDTC (an NF-κB antagonist), PD98059 (an Erk1/2 antagonist), and rosiglitazone (an insulin sensitizer) were used to test which one blocked the effect of resistin. The data showed that PDTC reversed the effect of resistin (Fig. 5A), but the other molecules had no effects (Supporting Fig. 1E-G), indicating that resistin functions by the NF-κB-signaling pathway. Assay of expression level showed that resistin enhanced p65 expression (Fig. 5B). RNA interference (RNAi) of p65 destroyed the effect of resistin and restored mitochondrial content (Fig. 5C). On the contrary, overexpression of p65 diminished mitochondrial content (Fig. 5D). Further investigations indicated that KT5823 inhibited the regulatory effect of p65 on mitochondria (Fig. 5D), revealing that the role of p65 in mitochondrial biogenesis is dependent on PKG activation. Previous studies have reported that p65 was phosphorylated by PMA in the region between amino acids 442 and 47023 and that PKG activated NF-κB by phosphorylating p65.24 Based on our data, we presumed that PMA phosphorylates p65 by activating PKG and discovered that there are four potential phosphorylation sites in p65 (Fig. 5E). To clarify whether p65 regulates mitochondria through these sites, we first constructed two mutants: M1 (S457A and T458A) and M2 (T464A and S468A). Results showed that mutations of Thr464 and Ser468 abolished the effect of p65 (Fig. 5F). A further mutation study discovered that M3 (T464A) did not decrease mitochondrial content, implying that Thr464 residue of p65 was essential for regulating mitochondria and a potential phosphorylation site for PKG (Fig. 5F). Based on these data, we concluded that the signal-transduction pathway is resistinPKCPKGp65.
Resistin Functioned Through PGC-1α.
PGC-1α plays a crucial role in mitochondrial biogenesis.25 Our data showed that resistin inhibited PGC-1α expression; however, KT5823 blocked the role of resistin and restored its expression (Fig. 6A). Because KT5823 maintained mitochondrial content (Fig. 4C) and PGC-1α level, we presumed that resistin regulated mitochondria through inhibiting PGC-1α expression. The result of PGC-1α overexpression verified our hypothesis, because it blocked resistin action and maintained normal mitochondrial content (Fig. 6B). Moreover, RNAi of p65 was found to stimulate PGC-1α expression, whereas p65 overexpression impaired PGC-1α expression (Fig. 6C). Through PLA, it was demonstrated that p65 interacted with PGC-1α. Resistin promoted the interaction of these proteins, but KT5823 inhibited their interaction (Fig. 6D). The interaction of p65 and PGC-1α is inversely related to mitochondrial content. Because PGC-1α is able to activate its own transcription,26, 27 interacting with p65 may impair self-activation and expression of PGC-1α. To prove this point, promoter activity of PGC-1α was measured. Both resistin and p65 suppressed the transcriptional activity of PGC-1α; however, cotransfection of p65 and PGC-1α restored the transcriptional activity of PGC-1α (Fig. 6E,F). Therefore, we concluded that resistin inactivated PGC-1α and inhibited mitochondrial biogenesis by promoting the interaction of p65 and PGC-1α. Our data suggest the following signaling scenario: First, resistin activates PKG by PKC; second, PKG activates p65 by phosphorylating its Thr464 and promotes the interaction of p65 and PGC-1α; and, finally, this interaction inactivates PGC-1α, diminishes mitochondrial content, and induces hepatic fat accumulation. Taken together, resistin diminishes mitochondria and induces hepatic steatosis through the PKC/PKG/p65/PGC-1α pathway.
Both in animal models and humans, accumulated evidence supports the notion that mitochondrial content is down-regulated in obesity-associated diseases.6, 7, 11, 28, 29 However, it remains unclear whether the change in mitochondria content or obesity is the initial event in these processes. In this study, we found that resistin down-regulated mitochondrial content and impaired mitochondrial function. After 24 hours of treatment, resistin slightly increased fat accumulation (Fig. 3C) and did not affect phosphorylation of Akt (Fig. 3F), but diminished mitochondrial content markedly (Fig. 1A). Hence, mitochondria diminution occurs before the change in fat accumulation and development of IR, implying that the change in the mitochondria occurs before NAFLD and diabetes. Moreover, when mitochondrial content was maintained by KT5823 (Fig. 4C), resistin did not stimulate hepatic TAG accumulation (Fig. 4F). Therefore, mitochondrial diminution may be an inducing factor for NAFLD. Some other groups also discovered that mitochondrial abnormalities are closely related to the pathogenesis of NAFLD and diabetes and proposed that NAFLD and diabetes are mitochondrial diseases8, 30, 31; however, the exact mechanisms underlying these processes remain unclear. Here, we discovered that resistin induces hepatic steatosis through diminishing mitochondrial content and propose that maintenance of normal mitochondrial content may present a new strategy for the therapy of obesity-associated diseases.
Mitochondria are sites of FAO. A decrease in mitochondria content and activities will inhibit lipolysis and promote fat deposition. Our data showed that hepatic TAG levels were significantly higher in the resistin-treated group (Fig. 3B). Compared with subcutaneous fat, excessive visceral fat is more detrimental to health. A further study showed that resistin decreased intracellular glycerol levels and impaired CAD activity (Fig. 3D,E). Based on these data, we presumed that resistin promoted hepatic fat deposition through suppression of lipolysis.
In conclusion, we report that resistin down-regulated mitochondria by a novel PKC/PKG/p65/PGC-1α-signaling pathway and aggravated hepatic steatosis by diminishing mitochondrial content. Our data link mitochondria to NAFLD by resistin and provide some novel targets (e.g., PKC, PKG, and p65) to regulate mitochondria and hepatic fat accumulation.
- 20Inhibition of forskolin-induced neurite outgrowth and protein phosphorylation by a newly synthesized selective inhibitor of cyclic AMP-dependent protein kinase, N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide (H-89), of PC12D pheochromocytoma cells. J Biol Chem 1990; 265: 5267-5272., , , , , , et al.
Additional Supporting Information may be found in the online version of this article.
|HEP_26167_sm_SuppFig1.tif||376K||Supporting Information Figure 1.|
|HEP_26167_sm_SuppTables.doc||78K||Supporting Information Tables.|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.