Dulbecco's modified Eagle's medium (DMEM)/F-12, DMEM, fetal bovine serum (FBS) and human insulin were all purchased from Invitrogen; Thermo Fisher Scientific, Inc. (Waltham, MA, USA). Dexamethasone (Dex), 3-isobutyl-1-methylxanthine (IBMX), rosiglitazone, isoproterenol and recombinant murine tumor necrosis factor-α (TNF-α) were purchased from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany). Mouse anti-human apoA5 antibody was obtained from Novus Biologicals, LLC (cat. no. NB400-139; Littleton, CO, USA) (6). Rabbit anti-mouse LRP1 antibody (cat. no. ab92544), mouse anti-human cidec antibody (cat. no. ab77115) and rabbit anti-human perilipin antibody (cat. no. ab50291) were all purchased from Abcam (Cambridge, UK). Goat anti-rabbit IgG antibody (bs-0295G) and rabbit anti-mouse IgG antibody (bs-0296R) were purchased from Bioss (Beijing, China). Recombinant human apoA5 was produced in Escherichia coli and isolated, as previously described (27).

Subcutaneous adipose tissue samples were obtained from 19 patients (7 male and 12 female) undergoing elective open-abdominal surgical procedures in the Second Xiangya Hospital, Central South University, China between December 2014 and November 2015. None of the patients had known diabetes or severe systemic illness, and were not taking medications known to affect adipose tissue mass or metabolism. The mean BMI was 22.9 kg/m² (range, 16.3–31.1 kg/m²), and the mean age was 40 years (range, 15–55). In particular, 3 obese patients (BMI ≥30) were recruited together with 3 matched non-obese patients (BMI <25) to obtain the subcutaneous adipose tissues for comparing the endocytosis of apoA5 by adipocytes in vivo. Tissue samples were immediately either frozen in liquid nitrogen and stored at −80°C, or digested with collagenase (Gibco; Thermo Fisher Scientific, Inc.) to isolate preadipocytes, as previously described (28). The protocol was approved by the Ethics Committee of Central South University (Hunan, China) and all patients provided written informed consent. The 3T3-L1 preadipocyte murine cell line was obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA).

Human preadipocytes were differentiated into adipocytes as previously described, with a few modifications (29). Briefly, human preadipocytes were cultured in DMEM/F-12 medium supplemented with 10% FBS, 100 U/ml penicillin and 100 U/ml streptomycin (complete medium) and incubated at 37°C in a humidified 5% CO₂/95% air atmosphere. At 100% confluence, cells were treated with differentiation medium containing DMEM/F-12, 100 U/ml penicillin, 100 U/ml streptomycin, 5% FBS, 1 µM Dex, 500 µM IBMX, 10 µg/ml insulin and 1 µM rosiglitazone for 12 days. Cells continued to be cultured in differentiation medium, but without rosiglitazone for a further 9 days. Cells were then cultured in complete medium for 2 days. These cells were used as differentiated adipocytes in all experiments. Each medium was replaced with fresh medium every 3 days.

3T3-L1 fibroblasts were cultured and differentiated into adipocytes as previously described (30,31). Cells at day 8 and day 21 post-induction of differentiation were used as non-hypertrophied and hypertrophied adipocytes, respectively (30). To render adipocytes insulin resistant, fully differentiated adipocytes were treated with 3 ng/ml recombinant murine TNF-α for 3 days, with daily replacement of medium.

Differentiated 3T3-L1 adipocytes were transfected with 5 nM small interfering RNA (siRNA) targeting LRP1 (CGC TGA CCC TAT TTG AAGA) or non-silencing control siRNA (TTC TCC GAA CGT GTC ACGT) using HiPerFect transfection reagent (Qiagen GmbH, Hilden, Germany) for up to 48 h at 37°C, according to the manufacturer's instructions. The silencing effect of LRP1 siRNA transfection was quantified by western blot analysis.

For expression studies, apoA5 was labeled with ¹²⁵I using Iodogen (Sigma-Aldrich; Merck KGaA) according to the manufacturer's instructions. Differentiated 3T3-L1 adipocytes were pulsed with ¹²⁵I-labeled apoA5 (~2×10⁶ cpm/well) in serum-free medium for 2 h at 37°C. At the end of the pulse, the medium was aspirated and cells were washed three times with ice-cold PBS, and incubated with heparin (10 mg/ml in PBS; Sigma-Aldrich; Merck KGaA) for 3 min at room temperature, and then washed three times with PBS. Fresh medium was added, and cells were chased for various time periods (0, 1, 2, 4, 6 and 24 h) at 37°C. At the end of the chase period, the medium was collected and cells were treated with heparin as described above. This heparin wash was added to the collected medium which was then precipitated with 13% trichloroacetic acid to estimate degradation of apoA5. Following heparin treatment, cells were collected in 0.1 M NaOH. Total cell protein was determined using the bicinchoninic acid assay (Novagen; Merck KGaA). Radioactivity (cpm) measured by gamma-ray counter was normalized to cell protein, and data are expressed as total cpm/mg protein at each time-point.

Differentiated human adipocytes were incubated in serum-free DMEM/F-12 in the presence of 200 ng/ml recombinant apoA5 for 48 h. The TG content of adipocytes was determined using a triglyceride quantification kit (BioVision, Inc., Milpitas, CA, USA) according to the manufacturer's instructions. For lipolysis analysis, adipocytes were washed with PBS once to remove the phenol red. Then 200 ng/ml apoA5 was added in phenol red free DMEM/F-12 for 24 h. In some experiments, after apoA5 treatment, adipocytes were subsequently incubated with 10 µM isoproterenol for 3 h. At the end of incubation, the medium was collected and glycerol content was measured using a colorimetric assay at 412 nm (BioVision, Inc.).

Mature adipocytes grown on coverslips were washed three times with cold PBS, fixed in 4% paraformaldehyde for 20 min and permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature. Oil Red O (0.5% in isopropyl alcohol; Sigma-Aldrich; Merck KGaA) was mixed with distilled water in the ratio of 60:40 and filtered through 0.45 µm filters. The freshly prepared Oil Red O solution was added to permeabilized cells for 30 min. These cells were then washed with PBS three times, incubated with 0.2 µg/ml of 4′,6-diamidino-2-phenylindole for 10 min at room temperature and washed with PBS three times. Cells were mounted on slides with ProLong Gold antifade reagent (Invitrogen; Thermo Fisher Scientific, Inc.), and visualized under a confocal microscope (model TCS SP5; Leica Microsystems, Inc., Buffalo Grove, IL, USA). The size of Oil Red O-stained lipid droplets was calculated using ImageJ software version 1.48s (National Institutes of Health, Bethesda, MD, USA).

Total protein lysate from frozen tissues or cultured cells was prepared using lysis buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.5% sodium deoxycholate and protease inhibitor (Invitrogen; Thermo Fisher Scientific, Inc.) on ice for 0.5 h. The cell lysates were centrifuged at 13,000 × g for 10 min at 4°C. The supernatants were separated by 12% SDS-PAGE and transferred onto polyvinylidene difluoride membranes. The membranes were blocked for 2 h with 5% skim milk in TBS containing 0.1% Tween-20 at room temperature and incubated overnight at 4°C with mouse anti-human apoA5 (1:1,000), mouse anti-LRP1 (1:1,000), human anti-Cidec (1:1,000) or human anti-perilipin (1:1,000). The blots were then incubated with a corresponding horseradish peroxidase-conjugated secondary antibody (goat anti-rabbit IgG, 1:5,000 or rabbit anti-mouse IgG, 1:5,000) for 1 h at room temperature. The immunoblot signals were visualized using an enhanced chemiluminescence substrate (Invitrogen; Thermo Fisher Scientific, Inc.) and quantified by ImageJ software version 1.48s (National Institutes of Health).

Total RNA from human differentiated adipocytes was isolated using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. First strand cDNA was synthesized using a cDNA synthesis kit (Fermentas; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. qPCR was performed on an ABI 7300 Real-Time PCR instrument (Applied Biosystems; Thermo Fisher Scientific, Inc.) with SYBR-Green mix (Takara Bio Inc., Otsu, Japan) and specific primers to amplify the genes. An initial denaturation step at 95°C for 30 min; 40 cycles at 95°C for 5 sec and 60°C for 31 sec. Gene expression levels of the genes encoding cidec (CIDEC), perilipin (PLIN1), diacylglycerol acyltransferase (DGAT) 1, DGAT2, stearoyl-CoA desaturase 1 (SCD1), HSL, adipose triglyceride lipase (ATGL), nuclear respiratory factor 1 (NRF1), subunit IV of cytochrome c oxidase (COXIV, officially known as COX4I1), PPARγ coactivator 1α (PGC1α), uncoupling protein 1 (UCP1) and forkhead box C2 (FOXC2), were normalized to 18S rRNA, and presented as the fold change relative to the control. The following primer sequences were used: CIDEC sense, 5′-AAGTCCCTTAGCCTTCTCTACC-3′ and antisense, 5′-CCTTCCTCACGCTTCGATCC-3′; PLIN1 sense, 5′-GACCTCCCTGAGCAGGAGAAT-3′ and antisense, 5′-GTGGGCTTCCTTAGTGCTGG-3′; DGAT1 sense, 5′-AACCTCATCAAGTATGGCATCCT-3′ and antisense, 5′-CATTGGCCGCAATAACCAGG-3′; DGAT2 sense, 5′-AAGACCCTCATAGCCGCCTA-3′ and antisense, 5′-TTGGACCTATTGAGCCAGGTG-3′; SCD1 sense, 5′-TCTAGCTCCTATACCACCACCA-3′ and antisense, 5′-TGTCGTCTTCCAAGTAGAGGG-3′; HSL sense, 5′-TCAGTGTCTAGGTCAGACTGG-3′ and antisense, 5′-AGGCTTCTGTTGGGTATTGGA-3′; ATGL sense, 5′-ATGGTGGCATTTCAGACAACC-3′ and antisense, 5′-CGGACAGATGTCACTCTCGC-3′; NRF1 sense, 5′-AACAAAATTGGGCCACGTTACA-3′ and antisense, 5′-TCTGGACCAGGCCATTAGCA-3′; COXIV sense, 5′-CAGGGTATTTAGCCTAGTTGGC-3′ and antisense, 5′-GCCGATCCATATAAGCTGGGA-3′; PGC1α sense, 5′-TCTGAGTCTGTATGGAGTGACAT-3′ and antisense, 5′-CCAAGTCGTTCACATCTAGTTCA-3′; UCP1 sense, 5′-AGGTCCAAGGTGAATGCCC-3′ and antisense, 5′-GCGGTGATTGTTCCCAGGA-3′; FOXC2 sense, 5′-CCTCCTGGTATCTCAACCACA-3′ and antisense, 5′-GAGGGTCGAGTTCTCAATCCC-3′; 18S rRNA sense, 5′-GCTTAATTTGACTCAACACGGGA-3′ and antisense, 5′-AGCTATCAATCTGTCAATCCTGTC-3′. Gene expression levels were quantified using the 2^−ΔΔCq method and normalized to the internal reference gene 18S rRNA (32).

Statistical comparisons between 2 groups were assessed by a two-tailed, unpaired Student's t-test. Comparisons between ≥3 groups were analyzed by one-way analysis of variance followed by a Newman-Keuls post hoc test. P<0.05 was considered to indicate a statistically significant difference. Data are expressed as the mean ± standard error of ≥3 independent experiments.

3T3-L1 fibroblasts were completely differentiated into adipocytes after 8 days incubation with induction media. For generating hypertrophied adipocytes, 3T3-L1 adipocytes were cultured up to 21 days after the induction of differentiation. Oil red O staining exhibited a gradual increase in lipid accumulation from day 8 to day 21 (Fig. 1). At 4 h after the addition of apoA5, internalized apoA5 was quantified by immunoblotting. During the course of adipocyte hypertrophy, apoA5 internalization was significantly lower on day 21 (15%) in hypertrophied cells than that on day 8 in non-hypertrophied control cells (Fig. 2A). Using a pulse-chase protocol in which cells were incubated with [¹²⁵I]-labeled apoA5 and chased for various time-points up to 24 h after removal of the radiolabeled sample, internalized, labeled apoA5 in hypertrophied adipocytes was significantly decreased by 66% compared with control cells (Fig. 2B). The quantity of degraded apoA5 in the medium was increased in a time-dependent manner during the chase period, reaching 63% of total apoA5 at the 24 h time-point both in hypertrophied and non-hypertrophied adipocytes (data not shown).

Similar results were observed in the case of the insulin-resistant adipocytes. An in vitro model of metabolic insulin resistance was established using TNF-α-treated 3T3-L1 adipocytes, as described previously (33). Western blot analysis revealed that TNF-α treatment decreased apoA5 internalization by 52% (P<0.05; Fig. 3A). Similar data was also confirmed by a pulse-chase experiment (Fig. 3B).

To determine the impact of obesity on apoA5 expression in human adipose tissue, we obtained subcutaneous abdominal adipose tissue biopsies from 6 patients with a wide range of BMI. The apoA5 protein was detected in these human adipose tissue samples. The amount of apoA5 was significantly reduced by 69% in the obese group as compared with the non-obese group (Fig. 4).

It was investigated whether knockdown of LRP1 expression in adipocytes prevents apoA5 internalization. After 48 h transfection of differentiated 3T3-L1 adipocytes with LRP1 siRNA, knockdown efficiency was 87% (P<0.05) as determined by western blot analysis (Fig. 5A). The reduction of LRP1 expression resulted in decreased apoA5 uptake by adipocytes (~70%), as determined by western blot analysis and the pulse-chase experiment (Fig. 5B and C).

Our previous study revealed that treatment with 200 and 600 ng/ml apoA5 for 48 h significantly decreased intracellular TG levels in human adipocytes (12). As 200 ng/ml is within the normal range of human plasma levels of apoA5 (5.4–455.6 ng/ml) (6), this concentration was used in the present study. ApoA5 treatment for 48 h led to a significant decrease in TG content in adipocytes, which was consistent with our previous findings (6) (P<0.001; Fig. 6A). Next, the effect of apoA5 on the morphological changes of lipid droplets was investigated. Oil Red O staining and confocal microscopy revealed that apoA5 treatment did not affect lipid droplet formation in adipocytes; however, the lipid droplets in apoA5-treated cells were generally smaller than those in control cells (Fig. 6B and C).

RT-qPCR analysis revealed that apoA5 decreased CIDEC mRNA expression levels in a time-dependent manner with maximal effect at 48 h (Fig. 7A). Treatment with apoA5 for 48 h decreased CIDEC mRNA expression levels by 71%. Similarly, apoA5 significantly decreased PLIN1 mRNA expression levels by 45%, with maximal effect at 24 h (Fig. 7B). Western blot analysis further confirmed that apoA5 reduced the protein expression levels of cidec and perilipin in human adipocytes (Fig. 7C).

To further address the potential regulation of TG-associated metabolic processes by apoA5, lipolysis in adipocytes was measured following apoA5 treatment. Measurement of glycerol released into the medium during culture of cells under basal conditions for 24 h revealed that lipolysis was significantly increased in apoA5-treated adipocytes. Examination of lipolysis in response to β-adrenergic stimulation revealed that treatment with apoA5 also enhanced isoproterenol-induced lipolysis in human adipocytes (Fig. 8).

Next, the effects of apoA5 on expression of genes associated with lipid synthesis and hydrolysis in adipocytes was investigated. RT-qPCR analysis was performed 48 h after intervention of apoA5 and revealed that the mRNA expression levels of DGAT1, DGAT2 and SCD1, all of which participate in lipid synthesis, did not significantly differ between apoA5-treated and control adipocytes (Fig. 9A). In addition, mRNA expression levels of HSL and ATGL, both of which contribute to TG hydrolysis in adipocytes, were not significantly different between the apoA5-treated and control cells (Fig. 9B). In addition, the expression of genes associated with mitochondrial biogenesis or oxidative phosphorylation were measured. Treatment of apoA5 for 48 h did not affect the mRNA expression levels of NRF1, COXIV and PGC1α in adipocytes (Fig. 9C). Finally, the mRNA expression levels of brown fat-specific gene UCP1 were measured, which serves a role in uncoupling oxidative phosphorylation and converts this proton gradient energy into heat to maintain normal body temperature (34). UCP1 mRNA expression levels were significantly increased (~1.87-fold; P<0.05) by apoA5 treatment for 48 h (Fig. 9D). The mRNA expression levels of FOXC2, which is an important regulator of UCP1 (35), were also significantly elevated (P<0.05; Fig. D). These results suggested that the decrease in TG accumulation in human adipocytes after apoA5 treatment may be accompanied by increased lipolysis, uncoupling or thermogenesis, but not by increased mitochondrial biogenesis or oxidative phosphorylation.