The ABA/LANCL1/2 Hormone/Receptor System Controls Adipocyte Browning and Energy Expenditure

Abstract

The abscisic acid (ABA)/LANC-like protein 1/2 (LANCL1/2) hormone/receptor system regulates glucose uptake and oxidation, mitochondrial respiration, and proton gradient dissipation in myocytes. Oral ABA increases glucose uptake and the transcription of adipocyte browning-related genes in rodent brown adipose tissue (BAT). The aim of this study was to investigate the role of the ABA/LANCL system in human white and brown adipocyte thermogenesis. Immortalized human white and brown preadipocytes, virally infected to overexpress or silence LANCL1/2, were differentiated in vitro with or without ABA, and transcriptional and metabolic targets critical for thermogenesis were explored. The overexpression of LANCL1/2 increases, and their combined silencing conversely reduces mitochondrial number, basal, and maximal respiration rates; proton gradient dissipation; and the transcription of uncoupling genes and of receptors for thyroid and adrenergic hormones, both in brown and in white adipocytes. The transcriptional enhancement of receptors for browning hormones also occurs in BAT from ABA-treated mice, lacking LANCL2 but overexpressing LANCL1. The signaling pathway downstream of the ABA/LANCL system includes AMPK, PGC-1α, Sirt1, and the transcription factor ERRα. The ABA/LANCL system controls human brown and "beige" adipocyte thermogenesis, acting upstream of a key signaling pathway regulating energy metabolism, mitochondrial function, and thermogenesis.

Keywords: AMPK/PGC-1α/Sirt1 signaling axis; DIO2; ERRα; OXPHOS uncoupling; UCP1/3; energy metabolism; glucose transport and oxidation; mitochondrial biogenesis and respiration; thermogenesis; thyroid receptors; β-adrenergic receptor.

Figure 1

Effect of ABA on human white and brown adipocyte differentiation. (A) representative micrographs (magnification ×20×40) of immortalized brown (TERT-hBA, upper panels) and white (TERT-hWA, lower panels) human preadipocytes at days 0, 12 and 15 of differentiation. (B) qPCR analysis; human TERT-hWA and TERT-hBA preadipocytes differentiated to adipocytes in the absence (control) or presence of 100 nM ABA and/or 1 μM rosiglitazone: at the indicated time points (day 0, 12 or 15), the mRNA levels of the indicated proteins were evaluated by qPCR. Upper panel, AMPK-1α, PGC-1α and Sirt1 mRNAs; central left panel, UCP1, PPAR-γ, CIDE-A and GLUT4 mRNAs; central right panel, mitochondrial/genomic DNA ratio (MT-DNA); lower left panel, CPT1β, MPC1 and PDHα1 mRNAs; lower right panel, LANCL1, LANCL2 and LANCL3 mRNAs. Results shown are the mean ± SD from at least 4 experiments; * p < 0.04 relative to undifferentiated untreated control, # p < 0.004 relative to differentiated control. p values are calculated by unpaired, two-tailed t-test.

Figure 2

Overexpression of LANCL1/2 proteins increases total and phosphorylated AMPK and PGC-1α protein levels in brown (TERT-hBA) and white (TERT-hWA) human preadipocytes. LANCL1 and/or LANCL2 were stably overexpressed in human TERT-hBA (A) or TERT-hWA (B) pre-adipocytes by viral infection. Upper left panels, representative Western blots of LANCL1/2 proteins in cells overexpressing LANCL1 (ovLANCL1), LANCL2 (ovLANCL2) or both LANCL1 and LANCL2 (ovLANCL1+2); upper right panels, densitometric quantitation of the LANCL proteins relative to control cells, transfected with the empty vector (PLV); total AMPK, phosphorylated (Ser473) AMPK (pAMPK), p-AMPK/AMPK ratio and PGC-1α in cells overexpressing LANCL1 (ovLANCL1), LANCL2 (ovLANCL2) or both LANCL1 and LANCL2 (ovLANCL1+2), relative to PLV. The order of the samples is the same in the quantifications and in the representative blots except for the lower right and central panels of Figure 2A which show the first 4 bands for untreated samples and the last 4 bands for ABA-treated samples. Values are normalized against vinculin, as housekeeping protein; * p < 0.03 and # p < 0.004 relative to untreated control; $ p < 0.005 relative to ABA-treated PLV-infected cells. p values are calculated by unpaired, two-tailed t-test.

Figure 3

The combined silencing of LANCL1/2 proteins reduces total and phosphorylated AMPK and PGC-1α protein levels in brown (TERT-hBA) and in white (TERT-hWA) preadipocytes. LANCL1 and LANCL2 were stably silenced in human immortalized TERT-hBA (A) or TERT-hWA (B) pre-adipocytes by viral infection. Upper left panels, representative Western blots of LANCL1/2 proteins in cells silenced for both LANCL1 and LANCL2 (shRNA-L1+L2); upper central panels, densitometric quantitation of the LANCL proteins relative to control cells, transfected with the vector containing scrambled silencing sequences (shRNA-SCR); upper right panels, LANCL1/2 mRNA levels in LANCL1/2-silenced cells relative to control; total AMPK, phosphorylated (Ser473) AMPK (pAMPK), p-AMPK/AMPK ratio and PGC-1α in double-silenced cells, relative to shRNA-SCR. Values are normalized against vinculin, as housekeeping protein; * p < 0.01 and # p < 0.002 relative to untreated control; $ p < 0.005 relative to ABA-treated shRNA-SCR-infected cells. p values are calculated by unpaired, two-tailed t-test.

Figure 4

Transcriptional effects of the overexpression of LANCL1/2 on differentiated brown and white adipocytes. Human brown (TERT-hBA) or white (TERT-hWA) preadipocytes overexpressing LANCL1 and LANCL2 (ovLANCL1+2), or control cells infected with the empty vector (PLV), were differentiated to white or brown adipocytes in the absence or in the presence of 100 nM ABA: at days 0, 12 and 15 of culture, mRNA levels of the indicated genes were evaluated by qPCR. (A) analysis of the AMPK/PGC-1α/Sirt1 signaling axis in TERT-hBA (left panel) and TERT-hWA (right panel). (B) TERT-hBA; upper left panel, GLUT4 mRNA; upper right panel, carnitine palmitoyl transferase 1β (CPT1β), mitochondrial pyruvate carrier 1 (MPC1), pyruvate dehydrogenase subunit α1 (PDHα1) and adenine nucleotide translocase 1 (ANT1) mRNAs; lower left panel, adrenergic receptor β-3 (ADRβ3), thyroid receptor α-1 (THRα1), thyroid receptor β (THRβ), insulin receptor (INSR), estrogen related receptor α (ERRα) and deiodinase 2 (DIO2) mRNAs; lower central panel, LANCL1/2 mRNAs; lower right panel, mitochondrial/genomic DNA ratio (MT-DNA and MT-ND1). * p < 0.02 relative to undifferentiated untreated control, # p < 0.006 relative to differentiated control, $ p < 0.003 relative to undifferentiated untreated overexpressing LANCL1 and LANCL2 control, ç p < 0.001 relative to differentiated overexpressing LANCL1 and LANCL2 control. (C) TERT-hWA; upper left panel, GLUT4 mRNA; upper right panel, CPT1β, MPC1, PDHα1, UCP1, UCP3 and ANT1 mRNAs; lower left panel, ADR β3, THRα1, THRβ, INSR, ERRα and DIO2 mRNAs; lower central panel, LANCL1/2 mRNAs; lower right panel, mitochondrial/genomic DNA ratio (MT-DNA and MT-ND1). Results shown are the mean ± SD from at least 5 experiments; * p < 0.02 relative to undifferentiated, untreated control cells, # p < 0.006 relative to differentiated control cells, $ p < 0.003 relative to undifferentiated, untreated cells overexpressing LANCL1/2, c, p < 0.001 relative to differentiated cells overexpressing LANCL1/2. p values are calculated by unpaired, two-tailed t-test.

Figure 5

Transcriptional effects of the combined silencing of LANCL1/2 on differentiated brown and white adipocytes. Human brown (TERT-hBA) or white (TERT-hWA) preadipocytes silenced for the expression of LANCL1 and LANCL2 (shRNA-L1+L2), or control cells infected with the vector containing scrambled sequences (shRNA-SCR), were differentiated to white or brown adipocytes in the absence or in the presence of 100 nM ABA: at days 0, 12 and 15 of culture, mRNA levels of the indicated genes were evaluated by qPCR. (A) analysis of the AMPK/PGC-1α/Sirt1 signaling axis in TERT-hBA (left panel) and TERT-hWA (right panel). (B) TERT-hBA; upper left panel, GLUT4 mRNA; upper right panel, CPT1β, MPC1, PDHα1 and ANT1 mRNAs; lower left panel, ADRβ3, THRα1, THRβ, INSR, ERRα and DIO2 mRNAs; lower central panel, LANCL1/2 mRNAs; lower right panel, mitochondrial/genomic DNA ratio (MT-DNA and MT-ND1). * p < 0.02 relative to undifferentiated untreated control, # p < 0.006 relative to differentiated control. (C) TERT-hWA; upper left panel, GLUT4 mRNA; upper right panel, CPT1β, MPC1, PDHα1, UCP1, UCP3 and ANT1 mRNAs; lower left panel, ADRβ3, THRα1, THRβ, INSR, ERRα and DIO2 mRNAs; lower central panel, LANCL1/2 mRNAs; lower right panel, mitochondrial/genomic DNA ratio (MT-DNA and MT-ND1). Results shown are the mean ± SD from at least 5 experiments; * p < 0.02 relative to undifferentiated, untreated control cells, # p < 0.006 relative to differentiated control cells. p values are calculated by unpaired, two-tailed t-test.

Figure 6

Mitochondrial respiration and uncoupling are controlled by LANCL1/2 expression levels in differentiated white and brown adipocytes. White (TERT-hWA) and brown (TERT-hBA) adipocytes, either overexpressing LANCL1/2 (ovLANCL1+2) and their controls, infected with the empty vector (PLV), or silenced for LANCL1/2 (shRNA-L1+L2) and their controls, infected with the scrambled silencing sequences (shRNA-SCR) were differentiated (up to day 12 of culture for brown adipocytes and up to day 15 of culture for white-derived “beige” adipocytes). Respiration measurements were performed using the Seahorse XFp Analyzer, with the sequential addition of oligomycin, FCCP and rotenone/antimycin A. Maximal and ATP-linked oxygen consumption rates (OCR) were measured in the absence or presence of 100 nM T3 or 100 nM ABA. (A) OCR of TERT-hBA; upper panels, LANCL1/2-overexpressing adipocytes, treated with T3 (left) or with ABA (right); lower panels, LANCL1/2-silenced adipocytes, treated with T3 (left) or with ABA (right). (B) OCR of TERT-hWA; upper panels, LANCL1/2-overexpressing adipocytes, treated with T3 (left) or with ABA (right); lower panels, LANCL1/2-silenced adipocytes, treated with T3 (left) or with ABA (right). * p < 0.05 and # p < 0.01 relative to maximal respiration without treatment (gray bar); $ p < 0.05 and § p < 0.01 relative to ATP-linked respiration without treatment (yellow bar), by unpaired t-test. Data shown are the mean ± SD of 3 experiments per group, with each value calculated in triplicate.

Figure 7

Mitochondrial number increases in LANCL1/2-overexpressing, and is conversely reduced in LANCL1/2-silenced white and brown adipocytes. The mitochondrial number was evaluated by MitoTracker analysis in TERT-hBA and TERT-hWA cells overexpressing LANCL1 and LANCL2 (ovLANC1+2) or double-silenced for the expression of both proteins (shRNA-L1+L2), and their respective controls, cells infected with the empty vector (PLV) or with the scrambled silencing sequences (shRNA-SCR), treated or not with 100 nM ABA. Representative confocal microscopy images of brown (panels A–D) and white (panels E–H)) adipocytes overexpressing (A) or silenced (B) for LANCL1 and LANCL2. The bar diagrams show the mean number of mitochondria per cell in each experiment (approx. 10 cells analyzed/cell type). The mean ± SD of the relative mitochondrial fluorescence was always calculated in at least 3 microscopic fields. * p < 0.005 relative to untreated control cells and # p < 0.01 relative to ABA-treated control cells by unpaired t-test.

Figure 8

LANCL1/2-expression levels control the mitochondrial proton gradient in human white and brown adipocytes. Fully differentiated TERT-hBA and TERT-hWA cells overexpressing LANCL1 and LANCL2 (ovLANC1+2), or double-silenced for the expression of both proteins (shRNA-L1+L2), and their respective controls, cells infected with the empty vector (PLV) or cells infected with the scrambled silencing sequences (shRNA-SCR), were loaded with the ∆Ψ-sensitive ratiometric fluorescent dye JC-1, and cultured/incubated for 4 h without or with 100 nM ABA. Representative confocal microscopy images of brown (left panels) and white (right panels) adipocytes overexpressing LANCL1/2 (A) or silenced for both proteins (B). An increase in red fluorescence indicates a higher ∆Ψ. The histograms show the mean ± SD of the red/green fluorescence ratio calculated in at least 3 microscopic fields for each experiment. * p < 0.01 relative to untreated control cells and # p < 0.05 relative to ABA-treated control cells by unpaired t-test.

Figure 9

LANCL1/2 expression levels control glucose uptake and affect triglyceride synthesis in white and brown adipocytes. Fully differentiated brown (TERT-hBA) and “beige” TERT-hWA adipocytes overexpressing LANCL1 and LANCL2 (ovLANC1+2), or double-silenced for the expression of both proteins (shRNA-L1+L2), and their respective controls, cells infected with the empty vector (PLV) or cells infected with the scrambled silencing sequences (shRNA-SCR), were analyzed for glucose transport and for triglyceride content. Cells were normalized to total cellular protein content by Bradford assay. (A) Glucose transport assays; cells were serum-starved for 12 h, then treated with 100 nM ABA for 5 min prior to incubation with the fluorescent glucose analog 2-NBDG. Results are expressed as NBDG uptake relative to the respective, ABA-untreated control (shRNA-SCR for silencing and PLV for overexpression, not shown in the figure). TERT-hBA, left panel; TERT-hWA, right panel. * p < 0.0004 relative to the respective untreated control; # p < 0.001 relative to the respective ABA-treated shRNA-SCR or PLV-infected cells. All results are the mean ± SD from at least 3 separate experiments. (B) Triglyceride content; triglycerides were stained with Oil Red O and quantitative measures were obtained by spectrophotometric absorbance of isopropanol extracts. TERT-hBA, adipocytes overexpressing LANCL1/2 (upper left panel) or silenced for both proteins (upper right panel); TERT-hWA, adipocytes overexpressing LANCL1/2 (lower left panel) or silenced for both proteins (lower right panel). * p < 0.05 relative to untreated control, # p < 0.005 relative to untreated control, $ p < 0.005 relative to ABA-treated shRNA-SCR or PLV-infected cells. All results are the mean ± SD from at least 4 separate experiments.

Figure 10

Chronic ABA treatment increases transcription of “browning” receptors, enzymes and transcription factors and mitochondrial DNA content in the BAT of LANCL2^−/− mice. Male, LANCL2^−/− mice (KO) and their wild-type siblings (WT), 5 per group, were treated without or with ABA (1 μg/kg BW/day, administered in the drinking water) for 4 weeks. At the end of treatment, mice were euthanized, and samples of brown adipose tissue were taken for qPCR analysis. Results shown are mRNA levels relative to the WT expression level and are the mean ± SD from 5 mice per group. p values are calculated by unpaired t-test. Upper left panel, LANCL2 mRNA; upper right panel, LANCL1 mRNA; central left panel, AMPK, PGC-1α and Sirt1 mRNAs; central right panel, mitochondrial/genomic DNA ratio (MT-DNA and MT-ND1); lower panel, β-adrenergic receptor 3 (ADRβ3), thyroid hormone receptor α1 (THRα1), thyroid hormone receptor β (THRβ), the enzyme deiodinase 2 (DIO2) and estrogen related receptor alpha (ERRα) mRNAs. * p < 0.008 relative to WT; # p < 0.05 relative to KO.

Figure 11

Schematic representation of the role of the ABA/LANCL1/2 hormone/receptor system in the control of energy metabolism, of adipocyte browning and of thermogenesis. Either one of the LANCL proteins can in turn activate the AMPK/PGC-1α/Sirt1/ERRα pathway, increasing glucose transport and oxidation, mitochondrial biogenesis and respiration, OXPHOS uncoupling and T3 and β-adrenergic receptors. The regulatory pathway was probed by overexpressing or silencing the LANCL proteins.