Conceived and designed the experiments: MH PH. Performed the experiments: MH EVD. Analyzed the data: MH EVD PH. Contributed reagents/materials/analysis tools: CM. Wrote the paper: MH PH.
The authors have declared that no competing interests exist.
Low adiponectin, a well-recognized antidiabetic adipokine, has been associated with obesity-related inflammation, oxidative stress and insulin resistance. Globular adiponectin is an important regulator of the interleukin-1 receptor-associated kinase (IRAK)/NFκB pathway in monocytes of obese subjects. It protects against inflammation and oxidative stress by inducing IRAK3. microRNA (miR)-146b-5p inhibits NFκB-mediated inflammation by targeted repression of IRAK1 and TNF receptor-associated factor-6 (TRAF6). Therefore, we measured the expression of miR-146b-5p in monocytes of obese subjects. Because it was low we determined the involvement of this miR in the anti-inflammatory, antioxidative and insulin signaling action of globular adiponectin.
miR-146b-5p expression in monocytes of obese subjects was determined by qRT-PCR. The effect of miR-146b-5p silencing on molecular markers of inflammation, oxidative stress and insulin signaling and the association with globular adiponectin was assessed in human THP-1 monocytes.
miR-146b-5p was downregulated in monocytes of obese persons. Low globular adiponectin decreased miR-146b-5p and IRAK3 in THP-1 monocytes, associated with increased mitochondrial reactive oxygen species (ROS). Intracellular ROS and insulin receptor substrate-1 (IRS1) protein were unchanged. Silencing of miR-146b-5p with an antisense inhibitor resulted in increased expression of IRAK1 and TRAF6 leading to more NFκB p65 DNA binding activity and
miR-146b-5p, decreased in monocytes during obesity, is a major mediator of the anti-inflammatory action of globular adiponectin. It appears not to be involved in insulin signaling possibly by protective response of IRAK3 and lack of mitochondrial ROS production.
Over the past decade, it has become quite clear that chronic, low-grade inflammation and oxidative stress play a key role in the initiation, propagation and development of obesity and associated metabolic disorders like insulin resistance, type 2 diabetes, metabolic syndrome and cardiovascular disease
The increased adipose tissue macrophage populations during obesity require an influx of circulating monocytes. Indeed, obesity is associated with increased levels of blood monocytes
microRNAs (miRs) are a class of small endogenous non-coding RNAs (∼22 nt), which function as important regulators of a wide range of cellular processes by modulating gene expression
Here, we show that miR-146b-5p is decreased in circulating monocytes of obese subjects. Functionally, miR-146b-5p is regulated by variable globular adiponectin concentrations and acts as an inhibitor of NFκB-mediated inflammation. Furthermore, miR-146b-5p is necessary for the anti-inflammatory action of high levels of globular adiponectin. In contrast, miR-146b-5p appears not to be involved in the insulin signaling in monocytes possibly due to the protective response of IRAK3 and lack of mROS production.
The study cohort comprised 14 lean controls (29% male) and 21 morbidly obese individuals (33% male), without clinical symptoms of cardiovascular disease. Obese patients were more frequently diabetic and treated with a statin. They had higher IL-6, high sensitivity C-reactive protein (hs-CRP), leptin and glucose levels, and lower adiponectin levels, indicating chronic systemic inflammation. The higher levels of circulating oxidized LDL indicated systemic oxidative stress. Furthermore, insulin and triglyceride concentrations were higher; HDL-cholesterol was lower. Obese individuals had higher systolic and diastolic blood pressure. Insulin resistance, calculated by a homeostasis model assessment (HOMA-IR), was 86% higher in obese subjects (
Lean controls (n = 14) | Obese patients (n = 21) | |
Gender (% male) | 29 | 33 |
Smoking (%) | 7 | 19 |
T2DM (%) | 0 | 38 |
Statin use (%) | 0 | 33 |
Age (years) | 33±3 | 39±3 |
BMI (kg/m2) | 21±1 | 44±1 |
Leptin (ng/ml) | 8.7±1.4 | 65.6±8.0 |
Adiponectin (µg/ml) | 10.9±1.8 | 3.9±0.6 |
Glucose (mg/dl) | 83±2 | 111±7 |
Insulin (mU/l) | 10.3±1.8 | 16.5±2.1 |
HOMA-IR | 2.1±0.4 | 3.9±0.5 |
Triglycerides (mg/dl) | 80±7 | 132±11 |
LDL-C (mg/dl) | 110±9 | 85±6 |
HDL-C (mg/dl) | 64±4 | 49±3 |
SBP (mmHg) | 120±3 | 137±3 |
DBP (mmHg) | 75±3 | 86±2 |
IL-6 (pg/ml) | 1.8±0.2 | 4.8±0.4 |
Hs-CRP (mg/l) | 0.49±0.10 | 5.65±1.13 |
Ox-LDL (IU/l) | 50±5 | 71±4 |
Data shown are means ± SEM.
We determined the expression level of miR-146b-5p in isolated monocytes of 14 lean controls and 21 obese subjects.
Activation of monocytes in the circulation is characterized by increased activation of the inflammatory toll-like receptor-2 (TLR2)/NFκB signaling pathway. Activation of this pathway will elicit pro-inflammatory cytokine release (e.g. tumor necrosis factor α (TNFα)), mitochondrial and intracellular ROS production and impairment of the insulin sensitivity at the molecular level. More mitochondrial ROS production is associated with an increased expression of the mitochondrial antioxidant
(
Lean controls (n = 14) | Obese patients (n = 21) | |
|
0.99±0.04 | 0.65±0.02 |
|
0.99±0.04 | 1.21±0.04 |
|
0.98±0.04 | 0.49±0.03 |
|
1.00±0.18 | 0.93±0.15 |
|
1.05±0.16 | 1.01±0.06 |
|
1.00±0.09 | 1.36±0.04 |
|
1.01±0.05 | 0.92±0.03 |
|
1.00±0.05 | 2.65±0.28 |
|
0.99±0.08 | 1.54±0.06 |
|
1.05±0.09 | 2.18±0.31 |
|
0.98±0.32 | 2.26±0.25 |
Data shown are means ± SEM.
Previously, we showed that differential expressions of targets within the gene cluster depended on adiponectin and glucose levels
Short-term exposure of THP-1 monocytes to high glucose, comparable to those in obese patients, did not affect miR-146b-5p expression (
We transfected THP-1 monocytes with a LNA-modified miR inhibitor that targets miR-146b-5p. Depletion of miR-146b-5p in THP-1 monocytes increased IRAK1 and TRAF6 protein (
(
The association between miR-146b-5p and globular adiponectin (
(
Although recent evidence shows that adiponectin enhances insulin sensitivity and protects against chronic inflammation and oxidative stress in monocytes, the involvement of miRs in these protective effects of adiponectin has yet to be determined
Mechanistically, miRs have been implicated as negative regulators of inflammatory processes at the post-transcriptional level
The role of miR-146b-5p in protection against inflammation was further supported by the finding that after sequestration of miR-146b-5p the cells lost their potency to raise their anti-inflammatory action in response to high levels of adiponectin. The insulin sensitizing action of adiponectin was improved after knockdown of miR-146b-5p, possibly due to the increase in IRAK3. The loss of anti-inflammatory properties after miR-146b-5p depletion is also important in regard of adversative findings about the protective effects of adiponectin. Indeed, previous reports indicate that adiponectin suppresses inflammatory responses induced by hyperglycemia or TNFα
In summary, we have demonstrated a bidirectional relation between globular adiponectin and miR-146b-5p in monocytes: the expression level of miR-146b-5p is regulated by globular adiponectin and miR-146b-5p, at his turn, is necessary for the anti-inflammatory action of high levels of globular adiponectin. If deletion of miR-146b-5p leads to an increase of IRAK3, mROS production and insulin signaling appear to be controlled even in the presence of TNFα-mediated inflammation.
All chemicals were obtained from Sigma-Aldrich (Bornem, Belgium), unless stated otherwise. Monoclonal antibodies against β-ACT (13E5), IRAK1 (D51G7), IRS1 (D23G12) and TRAF6 (D21G3) were purchased from Cell Signaling Technology (Bioké, Leiden, Netherlands), and polyclonal anti-IRAK3 antibody from Rockland (Tebu-bio, Boechout, Belgium). Human THP-1 monocytic cells (TIB-202) were obtained from ATCC (LGC Standards, Molsheim, France).
This study complies with the Declaration of Helsinki and the Medical Ethics Committee of the KU Leuven approved the study protocol. All human participants gave written informed consent. The patient cohort comprised 14 lean control (29% male; WCF<80 cm) and 21 obese individuals (33% male; WCF: 128±11 cm, mean ± SEM). The samples were collected at the Division of Endocrinology between March 29th, 2005 and May 30th, 2006. All participants were without symptoms of clinical atherosclerotic cardiovascular disease.
Blood samples were collected and after removal of the plasma fraction, peripheral blood mononuclear cells (PBMCs) were isolated using gradient separation on Histopaque-1077. Cells were washed three times in Ca2+- and Mg2+-free Dulbecco's (D)-PBS. PBMCs were incubated with CD14 microbeads (20 µl/1×107 cells) for 15 min at 4°C. Cells were washed once and re-suspended in 500 µl Ca2+- and Mg2+-free DPBS containing 0.5% BSA/1×108 cells. The suspension was then applied to an LS column in a MidiMACS Separator (Miltenyi, Leiden, Netherlands)
Human blood samples were centrifuged to prepare plasma samples for analysis. Total and HDL-cholesterol and triglyceride levels were determined with enzymatic methods (Boehringer Ingelheim, Mannheim, Germany). LDL-cholesterol levels were calculated with the Friedewald formula. Insulin resistance was calculated by a homeostasis model assessment (HOMA-IR) = fasting plasma insulin (mU/l)×fasting blood glucose (mM)/22.5. Plasma glucose was measured with the glucose oxidase method (on Vitros 750XRC, Johnson & Johnson, New Brunswick, NJ, USA), and insulin with an immunoassay (Biosource Technologies, Invitrogen, Gent, Belgium). Ox-LDL
Total RNA was extracted with TRIzol reagent (Invitrogen) and purified on miRNeasy Mini Kit columns (Qiagen, Venlo, Netherlands). The RNA quality was assessed with the RNA 6000 Nano assay kit using the Agilent 2100 Bioanalyzer; all samples achieved an RNA Integrity Number (RIN) score more than 8.0.
Total RNA (25 ng) was reverse transcribed in 20 µl reactions using the miRCURY LNA Universal RT miR cDNA synthesis kit (Exiqon, Vedbæk, Denmark), and the cDNA was diluted 80-fold. Each PCR was carried out in duplicate on a 7500 Fast Real-Time PCR system (Applied Biosystems, Gent, Belgium) in a total volume of 20 µl by using 8 µl of the diluted cDNA, 2 µl of LNA PCR primer set (hsa-miR-146b-5p: Exiqon, catalog no. 204553) and 10 µl of miRCURY LNA SYBRGreen master mix (Exiqon) according to the manufacturer's instructions.
For investigation of mRNA expressions, total RNA (0.5 µg) was reverse transcribed using SuperScript VILO cDNA synthesis kit (Invitrogen) as recommended by the manufacturer. The cDNA was diluted 50-fold. Primers used for qRT-PCR analysis are shown in Supplementary
Human THP-1 monocytic cells were subcultured in RPMI 1640 (Gibco, Invitrogen) as described previously in detail
THP-1 cells were transiently transfected with a chemical synthesized mIRCURY LNA miR-146b-5p Power Inhibitor (
To detect mitochondrial and intracellular ROS (mROS and iROS respectively) formation in treated THP-1 cells, measurements of MitoSOX Red and CellROX Deep Red (Invitrogen) fluorescence were performed by flow cytometry (Becton, Dickinson and Company, Aalst, Belgium). Cells were incubated with PBS containing 5 µM MitoSOX for 10 min or 2.5 µM CellROX for 30 min at 37°C and 5% CO2. The labeled cells were washed twice with PBS and then suspended in warm PBS for analysis by flow cytometry.
NFκB p65 DNA binding activity was assessed on isolated nuclear extracts of transfected THP-1 monocytes by ELISA using the TransAM NFκB p65 transcription factor assay kit according to the manufacturer's protocol (Active Motif, La Hulpe, Belgium). Briefly, 2 µg of nuclear extract diluted in complete lysis buffer was used in the p65 binding assay. The samples (in duplicate) were shaken for 1 h at room temperature in 30 µl binding buffer. After washing, anti-p65 antibody diluted 1∶1000 was applied to the wells for 1 h at room temperature. Specific binding was estimated by spectrophotometry after incubation with a horseradish peroxidase-conjugated antibody (1 h at room temperature, 1∶1000 diluted) at 450 nm wave length.
Conditioned medium was harvested after centrifugation of treated THP-1 monocytes. Quantification of levels of soluble TNFα protein in the conditioned medium was performed using CBA Human TNFα Flex Set kit according to the manufacturer's instructions (Becton, Dickinson and Company). Briefly, undiluted medium was incubated with TNFα capture beads for 1 h at room temperature. Next, PE-conjugated detection antibody was added to the medium and the mixture was incubated for 1 h at room temperature. Finally, cytokine-bound beads were washed twice, and analyzed by flow cytometry.
Western blot analysis was performed with 20 µg of total protein. Protein was electrophoresed through a 10–20% SDS-polyacrylamide gel (Bio-Rad, Nazareth Eke, Belgium) and transferred to a polyvinylidene difluoride membrane (Millipore, Brussel, Belgium). Membranes were processed according to standard Western blotting procedures. To detect protein levels, membranes were incubated with primary antibodies against β-ACT, IRAK1, IRAK3, IRS1 and TRAF6. The membranes were then incubated with horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology, Tebu-bio, Boechout, Belgium) and developed with SuperSignal chemiluminescent substrate (Pierce, Thermo Fisher Scientific, Aalst, Belgium). A PC-based image analysis program was used to quantify the intensity of each band (Bio-1D, Vilber Lourmat, Marne-la-Vallée, France). Data were normalized to the housekeeping protein β-ACT.
Lean and obese subjects were compared with an unpaired t-test (two-tailed);
Primers used in qRT-PCR.
(DOC)
The authors would like to thank Roxane Menten for her excellent technical assistance.