Conjugated linoleic acid induces human adipocyte delipidation: autocrine/paracrine regulation of MEK/ERK signaling by adipocytokines

Abstract

Dietary conjugated linoleic acid (CLA) reduces body fat in animals and some humans. Here we show that trans-10, cis-12 CLA, but not cis-9, trans-11 CLA, when added to cultures of stromal vascular cells containing newly differentiated human adipocytes, caused a time-dependent decrease in triglyceride content, insulin-stimulated glucose and fatty acid uptake, incorporation into lipid, and oxidation compared with controls. In parallel, gene expression of peroxisome proliferator-activated receptor-gamma and many of its downstream targets were diminished by trans-10, cis-12 CLA, whereas leptin gene expression was increased. Prior to changes in gene expression and metabolism, trans-10, cis-12 CLA caused a robust and sustained activation of mitogen-activated protein kinase kinase/extracellular signal-related kinase (MEK/ERK) signaling. Furthermore, the trans-10, cis-12 CLA-mediated activation of MEK/ERK could be attenuated by pretreatment with U0126 and pertussis toxin. In parallel, pretreatment with U0126 blocked the ability of trans-10, cis-12 CLA to alter gene expression and attenuate glucose and fatty acid uptake of the cultures. Intriguingly, the induction by CLA of MEK/ERK signaling was linked to hypersecretion of adipocytokines interleukin-6 and interleukin-8. Collectively, these data demonstrate for the first time that trans-10, cis-12 CLA decreases the triglyceride content of newly differentiated human adipocytes by inducing MEK/ERK signaling through the autocrine/paracrine actions of interleukins-6 and 8.

Fig. 1. Trans-10, cis-12 CLA decreases TG content

Cultures of SV cells containing newly differentiated adipocytes were treated (TRT) continuously for 1, 2, or 3 weeks with either a BSA vehicle or 30 μm linoleic acid (LA), cis-9, trans-11 CLA (,), or trans-10, cis-12 CLA (,). A, after 1, 2, or 3 weeks of treatment, cultures were harvested, and TG content was determined using a colorimetric assay. Data are expressed as μg of TG/mg of total cellular protein. Means (± S.E.; n = 8) with asterisks differ significantly (p < 0.05) from the 1 week controls. B, after 3 weeks of treatment, cultures were stained with oil red O, and phase-contrast photomicrographs were taken using an Olympus inverted microscope with a 20× objective.

Fig. 2. Isomer-specific regulation by CLA of insulin-stimulated glucose uptake and metabolism

Cultures of SV cells containing newly differentiated human adipocytes were treated (TRT) continuously for 72 h with either a BSA vehicle or 30 μm cis-9, trans-11 CLA (,) or trans-10, cis-12 CLA (,). A, basal and insulin-stimulated uptakes of 4 nmol of 2-[³H]deoxyglucose were measured after a 90-min incubation in the absence (−) or presence (+) of 100 nm insulin; the basal control (BSA, –insulin) rate of uptake was ~100 pmol/(h·mg of protein). B, basal and insulin-stimulated de novo conversions of 2.2 nmol of [¹⁴C]glucose into [¹⁴C]lipid were measured after a 3-h incubation in the absence or presence of 100 nm insulin; the basal control rate was ~10 pmol/(h·mg of protein). C, basal and insulin-stimulated [¹⁴C]CO₂ production from 2.2 nmol of [¹⁴C]glucose were measured after a 3-h incubation in the absence or presence of 100 nm insulin; the basal control rate was ~11 pmol/(h·mg of protein). All data are expressed as a percentage of basal vehicle control (BSA, – insulin) rate. Means (± S.E.; n = 6) not sharing common superscript differ, p < 0.05.

Fig. 3. Isomer-specific regulation of FA uptake and metabolism by CLA

Cultures of SV cells containing newly differentiated human adipocytes were treated (TRT) continuously for 72 h with either a BSA vehicle, 30 μm cis-9, trans-11 CLA (,) or trans-10, cis-12 CLA (,). A, [¹⁴C]oleic acid (12.5 nmol) uptake was measured after a 2-h incubation; the control rate of uptake was ~15 nmol/(h·mg of protein). B, [¹⁴C]oleic acid (12.5 nmol) incorporation into [¹⁴C]lipid was measured after a 2-h incubation; the control rate was ~ 12 nmol/(h·mg of protein). C, [¹⁴C]CO₂ production from [¹⁴C]oleic acid was measured after a 2-h incubation; the control rate was ~100 pmol/(h·mg of protein). Data are expressed as a percentage of vehicle control (BSA) rate. Means (± S.E.; n = 6) not sharing a common superscript differ, p < 0.05.

Fig. 4. Chronic trans-10, cis-12 CLA treatment alters gene expression

Cultures of SV cells containing newly differentiated human adipocytes were treated continuously for 8, 24, 72, or 216 h with either a BSA (♦) vehicle, or 30 μm cis-9, trans-11 CLA (•), or trans-10, cis-12 CLA (▴). After treatment, total RNA was harvested and used for first strand cDNA synthesis. Real time quantitative PCR analyses were performed to examine the expression of acyl-CoA-binding protein (ACBP), adiponectin, ap2, CAP, C/EBP-α, GLUT4, GPDH, leptin, lipoprotein lipase (LPL), perilipin, and PPAR-γ1 and PPAR-γ2. Results shown are representative of three separate experiments from independent human subjects.

Fig. 5. Isomer-specific regulation of MEK/ERK signaling by CLA

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and then treated (TRT) for the indicated time points with the indicated treatments. A, cultures were treated with 30 μm trans-10, cis-12 CLA for 30 min, 4 h, or 24 h. A 30-min treatment with TNF-α was used as a positive control for MAPK activation. Cell extracts were immunoblotted for the phosphorylated forms of p38 MAPK, JNK-MAPK, or p44/p42 (ERK1/2) MAPK. B, cultures were treated acutely (≤24 h) or chronically (12–72 h) with either a BSA vehicle or 30 μm trans-10, cis-12 CLA complexed to BSA. Cell extracts were immunoblotted for the phosphorylated forms of the MAPK kinase (P-MEK) and p44/p42 extracellular signal-related MAPK (P-ERK1/2) and subsequently were stripped and reprobed with antibodies recognizing total MEK (MEK), and total ERK (ERK1/2). C, cultures were treated with either a BSA vehicle (B), 30 μm cis-9, trans-11 CLA (), or 30 μm trans-10, cis-12 CLA () complexed to BSA for 24–72 h. Cell extracts were immunoblotted for P-MEK, P-ERK1/2, total MEK, and total ERK1/2. Data shown are representative of three to seven independent experiments for each panel.

Fig. 6. Effect of pharmacological inhibitors on trans-10, cis-12 CLA-mediated activation and nuclear accumulation of P-ERK1/2

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and then treated as follows. A, cultures were pretreated with the MEK inhibitor U0126 (10 μm), the GPCR-G_i/o coupling inhibitor PTX (100 ng/ml), the c-SRC kinase inhibitor PP2 (1 μm), or the protein kinase C inhibitor calphostin C (Cal. C, 200 nm) for 1 h and subsequently treated with either a BSA vehicle (B) or 30 μm trans-10, cis-12 CLA complexed to BSA (C) for an additional 24 h. Cell extracts were immunoblotted for the active phosphorylated forms of MEK (P-MEK) and ERK (P-ERK1/2) and subsequently were stripped and reprobed with antibodies recognizing total MEK (MEK), and total ERK (ERK1/2). Data shown are representative of two to three independent experiments for each panel. B, cultures were pretreated with a vehicle (dimethyl sulfoxide) for 1 h, 10 μm U0126 for 1 h, 100 ng/ml PTX for 1 h, or 1 μm BRL49653 for 24 h, and subsequently treated with a BSA vehicle (B), 30 μm cis-9, trans-11 CLA complexed to BSA (), or 30 μm trans-10, cis-12 CLA complexed to BSA () for an additional 24 h. Active ERK was then immunoprecipitated from total cell extracts and used in an in vitro kinase reaction with its known substrate, recombinant ELK-1. The resulting kinase reaction was subjected to SDS-PAGE and probed for phosphorylated ELK-1 using a phospho-specific antibody. A 30-min TNF-α treatment (100 ng/ml) of human adipocytes and active ERK-2 (ERK) was used as positive control for enzyme activation. C, cultures were pretreated with either a BSA vehicle control, 30 μm cis-9, trans-11 CLA (,), or 30 μm trans-10, cis-12 CLA (,) for 24 h. A 30-min treatment with TNF-α was used as a positive control for MAPK activation. Active ERK1/2 was then detected using immunofluorescence microscropy. Data shown are representative of two to three independent experiments for each panel.

Fig. 7. Trans-10, cis-12 CLA-induced alterations in adipocyte gene expression are blocked by pretreatment with the MEK inhibitor U0126

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and pretreated with or without 10 μm U0126 for 1 h. Subsequently, cultures were treated with either a BSA vehicle or 30 μm trans-10, cis-12 CLA for an additional 24 h. After 24 h of treatment, total RNA was harvested and used for first strand cDNA synthesis. Real time quantitative PCR analyses were performed to examine the expression of adiponectin, aP2, C/EBP-α, GLUT4, GPDH, leptin, lipoprotein lipase (LPL), perilipin, and PPAR-γ1 and PPAR-γ2. Results shown are representative of four separate experiments from independent human subjects.

Fig. 8. Trans-10, cis-12 CLA-induced alterations in glucose and FA uptake are blocked by pretreatment with U0126 and PTX

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and pretreated with or without 10 μm U0126 or 100 ng/ml PTX for 1 h. Subsequently, cultures were treated with either a BSA vehicle or 30 μm trans-10, cis-12 CLA complexed to BSA for an additional 24 h. A, insulin-stimulated uptake of 4 nmol of 2-[³H]deoxyglucose was measured after a 90-min incubation in the absence or presence of 100 nm insulin; the control rate of uptake was ~130 pmol/(h·mg of protein). B, [¹⁴C]oleic acid (12.5 nmol) uptake was measured after a 2-h incubation; the control rate of uptake was ~11 nmol/(h·mg of protein). Data are expressed as a percentage of vehicle control (BSA) rate. Means (± S.E.; n = 6) not sharing common superscripts differ, p < 0.05.

Fig. 9. Trans-10, cis-12 CLA induces MEK/ERK-dependent cytokine secretion and mRNA expression

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and pretreated with or without 10 μm U0126 for 1 h. Subsequently, cultures were treated with either a BSA vehicle or 30 μm trans-10, cis-12 CLA for an additional 24 h. A, conditioned medium was collected and utilized to detect the secretion of multiple cytokines using protein array technology. Positive control spots (used to normalize between membranes) are located in the upper left corner (n = 4) and the lower right corner (n = 2), and CLA-induced cytokines spotted in duplicate are indicated as IL-6 and IL-8. B, total RNA was harvested and used for first strand cDNA synthesis. Real time quantitative RT-PCR analyses were performed to examine the expression of IL-6 and IL-8. Results shown are representative of two (A) or four (B) separate experiments from independent human subjects.

Fig. 10. CLA-induced IL-6 and IL-8 production predominates from nonadipocyte SV cells

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and pretreated with 1 μg/ml brefeldin A for 1 h to prevent cytokine or chemokine secretion. Subsequently, cultures were treated with either a BSA vehicle or 30 μm trans-10, cis-12 CLA for an additional 12 h. Cultures were double stained for aP2 (rhodamine-conjugated anti-rabbit IgG = red) and subsequently for individual cytokines IL-6 or IL-8 (fluorescein isothiocyanate-conjugated anti-mouse IgG = green). Hoechst staining was conducted to identify nuclei positively. Fluorescent images (magnification, ×20) were captured as described under “Experimental Procedures.” Results shown are representative of two separate experiments from independent human subjects.

Fig. 11. The autocrine/paracrine actions of IL-6 are necessary for trans-10, cis-12 CLA-mediated induction of MEK/ERK signaling

Cultures of SV cells containing newly differentiated human adipocytes were serum starved for 24 h and A, treated with human recombinant IL-6, IL-8, or TNF-α (all 100 ng/ml) for 0–30 min; or B, pretreated with or without neutralizing antibodies (Ab) for IL-6 (50 μg/ml), IL-8 (50 μg/ml), or TNF-γ (20 μg/ml) and subsequently treated with either a BSA vehicle (B) or 30 μm trans-10, cis-12 CLA for an additional 24 h or recombinant IL-6, IL-8, or TNF-α for an additional 30 min. Total cell extracts were immunoblotted for P-ERK1/2. Data shown are representative of two to three independent experiments for each panel.

Fig. 12. Model of trans-10, cis-12 CLA-mediated adipocyte delipidation: metabolic control through adipocytokine-initiated, MEK/ERK-dependent repression of PPAR-γ

CLA can either enter into the adipocyte, where it may have direct effects or enter into the supporting SV cells to initiate an autocrine/paracrine signaling network. In the SV cell, CLA through a currently unidentified mechanism, increases the mRNA expression and secretion of IL-6 and IL-8. The resulting nascent IL-6 binds to its obligate transmembrane receptor (IL-6R), both on the surface of the adipocyte and the SV cell, where it activates MEK/ERK signaling in both cell populations. In addition, nascent IL-8 binds to its obligate heptahelical receptor (CXCR1), a PTX-sensitive GPCR only expressed in adipocytes, to amplify further MEK/ERK signaling in the adipocyte. The autocrine actions of nacent IL-6 in SV cells result in sustained activation of MEK/ERK signaling, which augments IL-6 and IL-8 mRNA expression, thereby feed-forwarding their synthesis and secretion. The collective paracrine actions of both IL-6 and IL-8 in the adipocyte results in sustained MEK and ERK hyperphosphorylation. Hyperphosphorylated ERK is then shuttled into the nucleus where it can phosphorylate multiple transcription factors (TFs) including ELK-1 and potentially other unidentified transcription factors. The ERK-dependent phosphorylation of other transcription factors may repress the expression of PPAR-γ itself. Collectively, ERK-dependent repression of PPAR-γ gene expression blocks the ability of PPAR-γ to modulate its traditional downstream target genes. The final result is decreased expression of genes involved in FA uptake and metabolism such as perilipin, acyl-CoA binding protein (ACBP), aP2, GPDH, and lipoprotein lipase (LPL) and genes involved in glucose uptake and metabolism such as GLUT4, CAP, and adiponectin. In addition, the ability of ligand-bound PPAR-γ to repress leptin expression is alleviated by CLA, thereby augmenting leptin, a critical regulator of lipid metabolism.