Hypothalamic AMPK-ER Stress-JNK1 Axis Mediates the Central Actions of Thyroid Hormones on Energy Balance - PubMed (original) (raw)

. 2017 Jul 5;26(1):212-229.e12.

doi: 10.1016/j.cmet.2017.06.014.

Patricia Seoane-Collazo 1, Cristina Contreras 1, Luis Varela 1, Joan Villarroya 2, Eva Rial-Pensado 1, Xabier Buqué 3, Igor Aurrekoetxea 3, Teresa C Delgado 4, Rafael Vázquez-Martínez 5, Ismael González-García 1, Juan Roa 5, Andrew J Whittle 6, Beatriz Gomez-Santos 3, Vidya Velagapudi 7, Y C Loraine Tung 6, Donald A Morgan 8, Peter J Voshol 6, Pablo B Martínez de Morentin 1, Tania López-González 9, Laura Liñares-Pose 1, Francisco Gonzalez 10, Krishna Chatterjee 6, Tomás Sobrino 11, Gema Medina-Gómez 12, Roger J Davis 13, Núria Casals 14, Matej Orešič 15, Anthony P Coll 6, Antonio Vidal-Puig 6, Jens Mittag 16, Manuel Tena-Sempere 17, María M Malagón 5, Carlos Diéguez 1, María Luz Martínez-Chantar 4, Patricia Aspichueta 3, Kamal Rahmouni 18, Rubén Nogueiras 1, Guadalupe Sabio 19, Francesc Villarroya 20, Miguel López 21

Affiliations

Hypothalamic AMPK-ER Stress-JNK1 Axis Mediates the Central Actions of Thyroid Hormones on Energy Balance

Noelia Martínez-Sánchez et al. Cell Metab. 2017.

Abstract

Thyroid hormones (THs) act in the brain to modulate energy balance. We show that central triiodothyronine (T3) regulates de novo lipogenesis in liver and lipid oxidation in brown adipose tissue (BAT) through the parasympathetic (PSNS) and sympathetic nervous system (SNS), respectively. Central T3 promotes hepatic lipogenesis with parallel stimulation of the thermogenic program in BAT. The action of T3 depends on AMP-activated protein kinase (AMPK)-induced regulation of two signaling pathways in the ventromedial nucleus of the hypothalamus (VMH): decreased ceramide-induced endoplasmic reticulum (ER) stress, which promotes BAT thermogenesis, and increased c-Jun N-terminal kinase (JNK) activation, which controls hepatic lipid metabolism. Of note, ablation of AMPKα1 in steroidogenic factor 1 (SF1) neurons of the VMH fully recapitulated the effect of central T3, pointing to this population in mediating the effect of central THs on metabolism. Overall, these findings uncover the underlying pathways through which central T3 modulates peripheral metabolism.

Keywords: AMPK; BAT; ER stress; JNK1; SF1; VMH; autonomic nervous system; ceramides; liver; thyroid hormones.

Copyright © 2017 The Authors. Published by Elsevier Inc. All rights reserved.

PubMed Disclaimer

Figures

None

Graphical abstract

Figure 1

Figure 1

Effect of Central THs on Liver and BAT (A) Protein levels of the AMPK pathway in the liver, WAT, muscle, and BAT (n = 8–17 rats/group). (B and C) Oil Red O (20×; scale bar, 100 μm) staining analysis (B) and TG levels in the liver (n = 8–9 rats/group) (C). (D) [3H]-acetate incorporation into TGs in the liver (n = 6–7 rats/group). (E) mRNA levels of BAT genes (n = 5–7 rats/group). (F) Protein levels of UCP1 in the BAT (n = 14 rats/group). (G) 18F-FDG uptake analysis (n = 8 rats/group). (H) Lipid oxidation rate, oxygen consumption rate in the BAT, and oxygen consumption in BAT mitochondria (n = 6–7 rats/group). (I) Electron microscopy images (4,000×; scale bar, 10 μm) and quantification of lipid droplet (LD) and mitochondria number/area unit, size, and ultrastructure in the BAT (n = 4 rats/experimental group, 30 images/animal). (J) Cumulative EE, RQ, and LA (n = 5 rats/group). (K) c-FOS images (10×; scale bar, 50 μm) and staining analysis in the dorsal nucleus of the vagus (DMV) (n = 4 rats/group, 9–32 sections/animal) of rats ICV treated with vehicle or T3. (L) Protein levels of the AMPK pathway in the liver of sham or VGX rats ICV treated with vehicle or T3 (n = 11–14 rats/group). (M) Sympathetic nerve activity (SNA) in the BAT (n = 8–11 rats/group) of rats ICV treated with vehicle or T3. (N) Protein levels of the AMPK pathway in the BAT of rats ICV treated with vehicle or T3 and s.c. treated with SR59230A (n = 7 rats/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus vehicle ICV. #p < 0.05 T3 ICV vehicle s.c. versus T3 ICV SR59230A s.c. Data are expressed as mean ± SEM. The bands in gels from (A), (F), (L), and (N) have been spliced from the same original gels. CC, central canal; HN, hypoglossal nucleus; ND, non-detected; SUV, standardized uptake value. See also Figure S1.

Figure 2

Figure 2

Effect of T3 within the VMH on Liver and BAT (A and B) 14C levels in serum (A) and lipid or aqueous phases (B) of rats ICV treated with vehicle or T3 (n = 9 rats/group). (C–E) Total 14C content (C); 14C content corrected by tissue/organ weight in the liver, BAT, and WAT (D); and 14C TG content in serum, liver, and BAT (E) of rats ICV treated with vehicle or T3 (n = 9 rats/group). (F) Body weight change, food intake, and body mass change of rats treated in the VMH with vehicle or T3 (n = 16–17 rats/group; 7 rats/group for NMR analysis). (G–I) Protein levels of UPC1 in the BAT and the AMPK pathway in the BAT and liver of rats treated in the VMH with vehicle or T3 (n = 14 rats/group). (J and K) Oil Red O (20×; scale bar, 100 μm) staining analysis (J) and TG levels in the liver of rats treated in the VMH with vehicle or T3 (n = 8–9 rats/group) (K). (L–N) Protein levels of the AMPK pathway in the BAT (L) and liver (M) and UCP1 in the BAT (N) of rats treated in the VMH with adenoviruses encoding GFP or TR-DN (n = 8 rats/group). (O and P) Oil Red O (20×; scale bar, 100 μm) staining analysis (O) and TG levels in the liver of rats treated in the VMH with adenoviruses encoding GFP or TR-DN (n = 8 rats/group) (P). (Q) Protein levels of the AMPK pathway in the liver of sham or VGX rats treated in the VMH with vehicle or T3 (n = 8 rats/group). (R and S) Oil Red O (20×; scale bar, 100 μm) staining analysis (R) and TG levels in the liver of sham or VGX rats treated in the VMH with vehicle or T3 (n = 8 rats/group) (S). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus vehicle ICV, vehicle VMH, sham, or GFP. Data are expressed as mean ± SEM. The bands in gels from (H), (I), (L)–(N), and (Q) have been spliced from the same original gels. See also Figure S2.

Figure 3

Figure 3

Effect of Activation of AMPK within the VMH on the Central Actions of THs on Liver and BAT (A) Body weight change and food intake of rats treated in the VMH with adenoviruses encoding GFP or AMPKα-DN (n = 23–25 rats/group). (B–D) Protein levels of UCP1 in the BAT (B) and the AMPK pathway in the BAT (C) and liver (D) of rats treated in the VMH with adenoviruses encoding GFP or AMPKα-DN (n = 12–13 rats/group). (E and F) Oil Red O (20×; scale bar, 100 μm) staining analysis (E) and TG levels in the liver of rats treated in the VMH with adenoviruses encoding GFP or AMPKα-DN (n = 12–14 rats/group) (F). (G and H) Protein levels of the AMPK pathway in the liver (G) and BAT (H) of euthyroid and hyperthyroid rats treated in the VMH with adenoviruses encoding GFP or AMPKα-CA (n = 7 rats/group). (I and J) Protein levels of the AMPK pathway in the liver (I) and BAT (J) of rats ICV treated with vehicle or T3 and treated in the VMH with adenoviruses encoding GFP or AMPKα-CA (n = 7 rats/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus GFP, euthyroid GFP, and vehicle ICV GFP; #p < 0.05, ##p < 0.01, ###p < 0.001, hyperthyroid GFP versus hyperthyroid AMPKα-CA or T3 ICV GFP versus T3 ICV AMPKα-CA. Data are expressed as mean ± SEM. The bands in gels from (B)–(D) and (G)–(J) have been spliced from the same original gels. See also Figure S3.

Figure 4

Figure 4

Effect of Deletion of AMPKα1 in SF1 Neurons in Liver and BAT (A–C) Body weight (A), food intake (B), and fat depot mass (C) of SF1-Cre AMPKα1flox/flox mice (n = 15–25 mice/group). (D) Cumulative EE of SF1-Cre AMPKα1flox/flox mice (n = 5–6 mice/group). (E) 18F-FDG uptake analysis of SF1-Cre AMPKα1flox/flox mice (n = 5–6 mice/group). (F) Infrared thermal images and quantification of temperature of BAT skin area of SF1-Cre AMPKα1flox/flox mice (n = 9–23 mice/group). (G) BAT SNA of SF1-Cre AMPKα1flox/flox mice (n = 7–8 mice/group). (H) mRNA levels of BAT genes of SF1-Cre AMPKα1flox/flox mice (n = 8 mice/group). (I) Protein levels of thermogenic markers and proteins involved in lipolysis in the BAT of SF1-Cre AMPKα1flox/flox mice (n = 7 mice/group). (J) Protein levels of AMPK pathway in the liver of SF1-Cre AMPKα1flox/flox mice (n = 7 mice/group). (K) TG levels in the liver of SF1-Cre AMPKα1flox/flox mice (n = 9–10 mice/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, SF1-Cre AMPKα1flox/flox versus AMPKα1flox/flox. Data are expressed as mean ± SEM. The bands in gels from (I) and (J) have been spliced from the same original gels. iWAT, inguinal WAT; gWAT, gonadal WAT; vWAT, visceral WAT. See also Figure S4.

Figure 5

Figure 5

Effect of THs and AMPK on Hypothalamic Ceramides or ER Stress (A–C) Ceramide levels (A), mRNA levels of the enzymes involved in the metabolism of complex lipids (B), and protein levels of the UPR in the hypothalamus of euthyroid or hyperthyroid rats (n = 8–15 rats/group) (C). (D) Protein levels of the UPR in the MBH of rats ICV treated with vehicle or T3 (n = 7 rats/group). (E and F) Protein levels of the UPR (E) and ceramide levels in the VMH of rats treated in the VMH with vehicle or T3 (n = 7–10 rats/group) (F). (G and H) Protein levels of the UPR (G) and ceramide levels in the VMH of rats treated in the VMH with GFP-expressing adenoviruses or adenoviruses encoding a AMPKα-DN (n = 7–9 rats/group) (H). (I) Protein levels of the UPR in the VMH of hyperthyroid rats treated in the VMH with adenoviruses encoding GFP or AMPKα-CA (n = 7 rats/group). (J) Protein levels of the UPR in the VMH of SF1-Cre AMPKα1flox/flox (n = 7–14 rats/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus euthyroid, vehicle ICV, vehicle VMH, and GFP or hyperthyroid GFP versus hyperthyroid AMPKα-CA or SF1-Cre AMPKα1flox/flox versus AMPKα1flox/flox. Data are expressed as mean ± SEM. The bands in gels from (C)–(E), (G), (I), and (J) have been spliced from the same original gels. See also Figure S5.

Figure 6

Figure 6

Effect of Hypothalamic Ceramide-Induced Lipotoxicity and ER Stress on the Central Action of THs on Liver and BAT (A) Body weight change and food intake of hyperthyroid rats ICV treated with vehicle or C6 ceramide ICV (n = 9–10 rats/group). (B–D) Protein levels of UCP1 in the BAT (B) and the AMPK pathway in the BAT (C) and liver (D) of hyperthyroid rats ICV treated with vehicle or C6 ceramide ICV (n = 7 rats/group). (E) Body weight change and food intake of hyperthyroid rats treated in the VMH with adenoviruses encoding GFP or GRP78-DN (n = 22–25 rats/group). (F–H) Protein levels of UCP1 in the BAT and the AMPK pathway in the BAT and liver of hyperthyroid rats treated in the VMH with adenoviruses encoding GFP or GRP78-DN (n = 13–15 rats/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus hyperthyroid vehicle ICV or hyperthyroid GFP VMH. Data are expressed as mean ± SEM. The bands in gels from (C), (F), and (G) have been spliced from the same original gels. See also Figure S6.

Figure 7

Figure 7

Effect of JNK1 in the VMH on the Central Actions of T3 on Liver and BAT (A) Protein levels of JNK and pJNK in the hypothalamus of euthyroid and hyperthyroid rats (n = 7 rats/group). (B) Protein levels of JNK and pJNK in the MBH of rats ICV treated with vehicle or T3 (n = 7 rats/group). (C) Protein levels of JNK and pJNK in the VMH of rats treated in the VMH with vehicle or T3 (n = 7 rats/group). (D) Protein levels of JNK and pJNK in the VMH of hyperthyroid rats treated in the VMH with adenoviruses encoding GFP or AMPKα-CA (n = 7 rats/group). (E) Protein levels of JNK and pJNK in the VMH of rats ICV treated with vehicle or T3 and treated in the VMH with adenoviruses encoding GFP or AMPKα-CA (n = 11 rats/group). (F) Protein levels of JNK and pJNK in the VMH of SF1-Cre AMPKα1flox/flox mice (n = 7 mice/group). (G) Body weight change and food intake of wild-type (WT) and JNK1 KO mice ICV treated with vehicle or T3 (n = 12–16 mice/group). (H and I) Protein levels of the AMPK pathway in the liver of WT (H) or JNK1 KO mice (I) ICV treated with vehicle or T3 (n = 12–16 mice/group). (J and K) Oil Red O (40×; scale bar, 50 μm) staining analysis (J) and TG levels in the liver of WT or JNK1 KO mice ICV treated with vehicle or T3 (n = 11–15 mice/group) (K). (L) Body weight change and food intake of JNK2 KO-JNK1flox/flox mice ICV treated with vehicle or T3 and treated in the VMH with AAV expressing GFP or Cre (n = 9-12 mice/group). (M and N) Protein levels of the AMPK pathway in the liver of JNK2 KO-JNK1 floxed mice ICV treated with vehicle or T3 and treated in the VMH with AAV expressing GFP (M) or Cre (N) (n = 9–12 mice/group). (O and P) Oil Red O (40×; scale bar, 50 μm) staining analysis (O) and TG levels in the liver of JNK2 KO-JNK1flox/flox mice ICV treated with vehicle or T3 and treated in the VMH with AAV expressing GFP or Cre (n = 6–12 mice/group) (P). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus euthyroid, vehicle ICV, vehicle VMH, GFP VMH, and T3 ICV vehicle ICV. Data are expressed as mean ± SEM. The bands in gels from (A)–(F), (H), (I), (M), and (N) have been spliced from the same original gels. See also Figure S7.

References

    1. Aspichueta P., Pérez S., Ochoa B., Fresnedo O. Endotoxin promotes preferential periportal upregulation of VLDL secretion in the rat liver. J. Lipid Res. 2005;46:1017–1026. - PubMed
    1. Bartelt A., Bruns O.T., Reimer R., Hohenberg H., Ittrich H., Peldschus K., Kaul M.G., Tromsdorf U.I., Weller H., Waurisch C. Brown adipose tissue activity controls triglyceride clearance. Nat. Med. 2011;17:200–205. - PubMed
    1. Benoit S.C., Kemp C.J., Elias C.F., Abplanalp W., Herman J.P., Migrenne S., Lefevre A.L., Cruciani-Guglielmacci C., Magnan C., Yu F. Palmitic acid mediates hypothalamic insulin resistance by altering PKC-theta subcellular localization in rodents. J. Clin. Invest. 2009;119:2577–2589. - PMC - PubMed
    1. Beylot M., Martin C., Laville M., Riou J.P., Cohen R., Mornex R. Lipolytic and ketogenic fluxes in human hyperthyroidism. J. Clin. Endocrinol. Metab. 1991;73:42–49. - PubMed
    1. Breen D.M., Yue J.T., Rasmussen B.A., Kokorovic A., Cheung G.W., Lam T.K. Duodenal PKC-δ and cholecystokinin signaling axis regulates glucose production. Diabetes. 2011;60:3148–3153. - PMC - PubMed

MeSH terms

Substances

Grants and funding

LinkOut - more resources