Under homeostatic conditions, animals use well-defined hypothalamic neural circuits to help maintain stable body weight, by integrating metabolic and hormonal signals from the periphery to balance food consumption and energy expenditure1,2. In stressed or disease conditions, however, animals use alternative neuronal pathways to adapt to the metabolic challenges of altered energy demand3. Recent studies have identified brain areas outside the hypothalamus that are activated under these ‘non-homeostatic’ conditions4,5,6, but the molecular nature of the peripheral signals and brain-localized receptors that activate these circuits remains elusive. Here we identify glial cell-derived neurotrophic factor (GDNF) receptor alpha-like (GFRAL) as a brainstem-restricted receptor for growth and differentiation factor 15 (GDF15). GDF15 regulates food intake, energy expenditure and body weight in response to metabolic and toxin-induced stresses; we show that Gfral knockout mice are hyperphagic under stressed conditions and are resistant to chemotherapy-induced anorexia and body weight loss. GDF15 activates GFRAL-expressing neurons localized exclusively in the area postrema and nucleus tractus solitarius of the mouse brainstem. It then triggers the activation of neurons localized within the parabrachial nucleus and central amygdala, which constitute part of the ‘emergency circuit’ that shapes feeding responses to stressful conditions7. GDF15 levels increase in response to tissue stress and injury, and elevated levels are associated with body weight loss in numerous chronic human diseases8,9. By isolating GFRAL as the receptor for GDF15-induced anorexia and weight loss, we identify a mechanistic basis for the non-homeostatic regulation of neural circuitry by a peripheral signal associated with tissue damage and stress. These findings provide opportunities to develop therapeutic agents for the treatment of disorders with altered energy demand.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Clemmensen, C. et al. Gut-brain cross-talk in metabolic control. Cell 168, 758–774 (2017)
Myers, M. G., Jr & Olson, D. P. Central nervous system control of metabolism. Nature 491, 357–363 (2012)
Wang, A. et al. Opposing effects of fasting metabolism on tissue tolerance in bacterial and viral inflammation. Cell 166, 1512–1525.e1512 (2016)
Wu, Q., Boyle, M. P. & Palmiter, R. D. Loss of GABAergic signaling by AgRP neurons to the parabrachial nucleus leads to starvation. Cell 137, 1225–1234 (2009)
Carter, M. E., Soden, M. E., Zweifel, L. S. & Palmiter, R. D. Genetic identification of a neural circuit that suppresses appetite. Nature 503, 111–114 (2013)
Wu, Q., Clark, M. S. & Palmiter, R. D. Deciphering a neuronal circuit that mediates appetite. Nature 483, 594–597 (2012)
Morton, G. J., Meek, T. H. & Schwartz, M. W. Neurobiology of food intake in health and disease. Nat. Rev. Neurosci. 15, 367–378 (2014)
Tsai, V. W., Lin, S., Brown, D. A., Salis, A. & Breit, S. N. Anorexia-cachexia and obesity treatment may be two sides of the same coin: role of the TGF-b superfamily cytokine MIC-1/GDF15. Int. J. Obes. 40, 193–197 (2016)
Hsiao, E. C. et al. Characterization of growth-differentiation factor 15, a transforming growth factor beta superfamily member induced following liver injury. Mol. Cell. Biol. 20, 3742–3751 (2000)
Breit, S. N. et al. The TGF-β superfamily cytokine, MIC-1/GDF15: a pleotrophic cytokine with roles in inflammation, cancer and metabolism. Growth Factors 29, 187–195 (2011)
Johnen, H. et al. Tumor-induced anorexia and weight loss are mediated by the TGF-beta superfamily cytokine MIC-1. Nat. Med. 13, 1333–1340 (2007)
Li, Z. et al. Identification, expression and functional characterization of the GRAL gene. J. Neurochem. 95, 361–376 (2005)
Airaksinen, M. S. & Saarma, M. The GDNF family: signalling, biological functions and therapeutic value. Nat. Rev. Neurosci. 3, 383–394 (2002)
Treanor, J. J. et al. Characterization of a multicomponent receptor for GDNF. Nature 382, 80–83 (1996)
Jing, S. et al. GDNF-induced activation of the ret protein tyrosine kinase is mediated by GDNFR-alpha, a novel receptor for GDNF. Cell 85, 1113–1124 (1996)
Asai, N., Murakami, H., Iwashita, T. & Takahashi, M. A mutation at tyrosine 1062 in MEN2A-Ret and MEN2B-Ret impairs their transforming activity and association with shc adaptor proteins. J. Biol. Chem. 271, 17644–17649 (1996)
Parkash, V. et al. The structure of the glial cell line-derived neurotrophic factor-coreceptor complex: insights into RET signaling and heparin binding. J. Biol. Chem. 283, 35164–35172 (2008)
Wang, X., Baloh, R. H., Milbrandt, J. & Garcia, K. C. Structure of artemin complexed with its receptor GFRalpha3: convergent recognition of glial cell line-derived neurotrophic factors. Structure 14, 1083–1092 (2006)
Leppänen, V. M. et al. The structure of GFRalpha1 domain 3 reveals new insights into GDNF binding and RET activation. EMBO J. 23, 1452–1462 (2004)
Tsai, V. W. et al. The anorectic actions of the TGFβ cytokine MIC-1/GDF15 require an intact brainstem area postrema and nucleus of the solitary tract. PLoS One 9, e100370 (2014)
Contreras, R. J., Fox, E. & Drugovich, M. L. Area postrema lesions produce feeding deficits in the rat: effects of preoperative dieting and 2-deoxy-D-glucose. Physiol. Behav. 29, 875–884 (1982)
Kott, J. N., Ganfield, C. L. & Kenney, N. J. Area postrema/nucleus of the solitary tract ablations: analysis of the effects of hypophagia. Physiol. Behav. 32, 429–435 (1984)
Tsuzuki, T. et al. Spatial and temporal expression of the ret proto-oncogene product in embryonic, infant and adult rat tissues. Oncogene 10, 191–198 (1995)
Quartu, M. et al. Tissue distribution of Ret, GFRalpha-1, GFRalpha-2 and GFRalpha-3 receptors in the human brainstem at fetal, neonatal and adult age. Brain Res. 1173, 36–52 (2007)
Campos, C. A., Bowen, A. J., Schwartz, M. W. & Palmiter, R. D. Parabrachial CGRP neurons control meal termination. Cell Metab. 23, 811–820 (2016)
Campos, C. A. et al. Cancer-induced anorexia and malaise are mediated by CGRP neurons in the parabrachial nucleus. Nat. Neurosci. 20, 934–942 (2017)
Richard, J. E. et al. GLP-1 receptor stimulation of the lateral parabrachial nucleus reduces food intake: neuroanatomical, electrophysiological, and behavioral evidence. Endocrinology 155, 4356–4367 (2014)
Holland, R. A., Leonard, J. J., Kensey, N. A., Hannikainen, P. A. & De Jonghe, B. C. Cisplatin induces neuronal activation and increases central AMPA and NMDA receptor subunit gene expression in mice. Physiol. Behav. 136, 79–85 (2014)
Alhadeff, A. L. et al. Excitatory hindbrain-forebrain communication is required for cisplatin-induced anorexia and weight loss. J. Neurosci. 37, 362–370 (2017)
Anders, J., Kjar, S. & Ibáñez, C. F. Molecular modeling of the extracellular domain of the RET receptor tyrosine kinase reveals multiple cadherin-like domains and a calcium-binding site. J. Biol. Chem. 276, 35808–35817 (2001)
Ferrannini, E. The theoretical bases of indirect calorimetry: a review. Metabolism 37, 287–301 (1988)
Yamazaki, T., Okawa, S. & Takahashi, M. The effects on weight loss and gene expression in adipose and hepatic tissues of very-low carbohydrate and low-fat isoenergetic diets in diet-induced obese mice. Nutr. Metab. (Lond.) 13, 78 (2016)
Agarwal, R., Bonanno, J. B., Burley, S. K. & Swaminathan, S. Structure determination of an FMN reductase from Pseudomonas aeruginosa PA01 using sulfur anomalous signal. Acta Crystallogr. D 62, 383–391 (2006)
Dauter, Z., Dauter, M., de La Fortelle, E., Bricogne, G. & Sheldrick, G. M. Can anomalous signal of sulfur become a tool for solving protein crystal structures? J. Mol. Biol. 289, 83–92 (1999)
Otwinowski, Z. & Minor, W.  Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326 (1997)
Sheldrick, G. M. Experimental phasing with SHELXC/D/E: combining chain tracing with density modification. Acta Crystallogr. D 66, 479–485 (2010)
McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007)
Cowtan, K. The Buccaneer software for automated model building. 1. Tracing protein chains. Acta Crystallogr. D 62, 1002–1011 (2006)
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D 66, 486–501 (2010)
Vagin, A. A. et al. REFMAC5 dictionary: organization of prior chemical knowledge and guidelines for its use. Acta Crystallogr. D 60, 2184–2195 (2004)
We thank N. Maddox, H. Tran, J. Oeffinger and R. Suriben for cloning, sequencing and genotyping, M. Bailey for purification of recombinant protein, R. Suto for help with solving the crystal structures and S. Talukdar, D. Kaplan, R. Suriben, S. Katewa and J.-L. Chen for critical reading of the manuscript.
D.L. and A.W. declare no direct competing financial interests. All other authors are or were employees of NGM Biopharmaceuticals or Merck Research Labs and may hold stock or stock options in these companies. NGM Biopharmaceuticals, Inc. has filed a non-provisional patent application entitled “Compositions and Methods for Modulating Body Weight” (PCT/US2017/020753), which discloses the GFRAL–GDF15 protein complex, and methods of identifying agents that may modulate the protein complex interaction(s) and potential uses of those agents to control body weight. J.-Y.H., S.C., J.H., J.T., W.D.S., Y.A.C. and H.T. are listed as inventors.
Reviewer Information Nature thanks R. Palmiter, M. Saarma and the other anonymous reviewer(s) for their contribution to the peer review of this work.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
a, Phylogenetic tree of GDF15, TGFβ and GDNF superfamily members. b, Sequence conservation (%) of GDF15 versus individual superfamily members. c, Binding of 125I-labelled TGFβ or 125I-labelled GDF15 (30 pM) to COS7 cells transfected with combinations of type I and type II TGFβ family receptors. Data are mean ± s.d. Source data
a, Crystal structure of GDF15 dimer at 1.97 Å resolution. b, Intramolecular disulfides (pink and yellow) within the cysteine knot core shown on a backbone trace of the GDF15 monomer subunit. c, GDF15 sequence alignment and disulfide arrangement versus solved structures of 7-cysteine and 9-cysteine BMP, TGFβ, GDF and GDNF superfamily members (PDB accession codes indicated). GDF15 contains a novel (1→2, 3→7) disulfide arrangement (blue) not previously identified for other 9-cysteine family members (1→3, 2→7) (red).
a, Mature GFRAL extracellular domain sequence alignment and disulfide arrangement against mature GFRα(1–4) family members. GFRAL (blue) maintains the conserved disulfide bonding network observed in all other family members (red) in the ligand recognition domain 2. GFRAL contains a novel arrangement of disulfide bonds in the crystal structure when compared with solved structures of GFRα1 and GFRα3 in domain 3, which may impact binding or activity of co-receptor RET. b, Phylogenetic tree of GFRAL versus GFRα family members. c, Sequence conservation (%) of GFRAL versus individual GFRα family members.
a, GDF15 physically interacts with GFRAL and GFRAL–RET on the cell surface. HEK293T cells (transfected with Flag–GFRAL or Flag–GFRAL + RET) bound with125I-labelled GDF15 were crosslinked by BS3 and then immunoprecipitated using anti-Flag. Unlabelled GDF15 added at 100× molar excess of 125I-labelled GDF15 blocked radio-ligand-receptor complex formation. Asterisks mark unique bands present only in the 125I-labelled GDF15 crosslinked complex in Flag–GFRAL + RET cells. b, RET is required for GDF15-induced signalling. HEK293T cells transfected with indicated plasmids were treated with 10 nM GDF15 for 30 min at 37 °C. pERK or RAB11 (loading control) in whole-cell lysates were detected by western blotting. A signalling-defective RET mutant (Y1062F) failed to induce pERK activation by GDF15 (top). Co-expression of GFRAL and RET is required for GDF15-induced pERK activation (bottom). c, Specific binding of 125I-labelled GDF15 to HEK293T cells co-transfected with GFRAL and RET. d, GDF15 does not activate RET through other GFRα family receptors. HEK293T cells co-transfected with RET and GFRα1 (i), GFRα2 (ii), GFRα3 (iii) or GFRα4 (iv) were treated with GDF15, GDNF, neurturin, artemin or persephin, as indicated. e, GDF15 binds GFRAL, but not GFRα family receptors. HEK293T cells transfected with GFRAL, GFRα1, GFRα2, GFRα3 or GFRα4 were bound by Fc–GDF15 and detected by a fluorescent anti-Fc secondary antibody. f, Non-iodinated GDF15 (but not GDNF, neurturin, artemin or persephin) competes with the binding of 125I-labelled GDF15 (150 pM) to GFRAL-transfected HEK293T cells. g, RET-dependent reporter gene is activated by GDF15 but not by GDNF, neurturin, artemin or persephin in HEK293T cells co-transfected with GFRAL and RET. Data are mean ± s.d. Data are representative of two (b, c) or three (e) experiments. Source data
a, Ribbon comparison of co-crystal of GDF15–GFRAL (light blue and yellow) with solved family members GDNF–GFRα1 (green and magenta; PDB code 2E5V) and artemin–GFRα3 (blue and red; PDB code 2GH0) through superimposition of the ligands. In all cases, each receptor specifically contacts its cognate ligand along the finger domains of the ligand. b, Diagram showing residue-specific intermolecular contacts (≤ 4.5 Å) between ligand and receptor for solved structures of GDF15, GDNF, and artemin with cognate receptors. The conserved RRR motif common to all GFRα1–4 receptors corresponds to SKE in GFRAL (yellow). c, Close-up view of the hydrogen bonding network comprising the RRR motif common to all GFR receptors and ligands, demonstrating the divergent nature of GDF15–GFRAL interactions. d, Electrostatic surface rendering (superimposition of domains 2 and 3) of GFRAL, GFRα1 and GFRα3 with monomer subunit of GDF15, GDNF and artemin, respectively (green backbone), demonstrating the unique hydrophobic and electrostatic recognition of each ligand’s finger domain for each receptor.
a, Schematic representation of genetic targeting used to create Gfral knockout mice (TF3754), and representative genotyping data identifying wild-type (+/+), heterozygous (+/−) and knockout (−/−) mice. b, Indirect calorimetry measurement of RER in 16-week-old Gfral+/+ and Gfral−/− mice on chow diet. n = 4 animals per group. c, Body weight of Gfral+/+ and Gfral−/− mice on chow diet at three months of age (n = 4 wild-type and n = 6 knockout animals) and six months of age (n = 5 animals per group). All data are ± s.e.m. Source data
a, Gene expression profile of Gfral mRNA in mouse tissues measured by quantitative PCR with reverse transcription. Data are mean ± s.d. n = 6 individual dorsal medulla tissue samples. b, GFRAL immunofluorescence in area postrema and NTS in a Gfral wild-type mouse (left) but not in a Gfral knockout mouse (right). c, FOS in area postrema and NTS neurons 1 hour after administration of GDF15 to a Gfral wild-type mouse (left) but not in a Gfral knockout mouse (right). d, FOS in central amygdala neurons 1 hour after administration of GDF15 to wild-type mice. e, No increase in FOS in hypothalamic neurons 1 hour after administration of GDF15 to wild-type mice. Data are representative of two independent experiments with n = 3 mice per experiment (b, c), and four independent experiments with n = 3 (PBS, d) or n = 5 (GDF15, e) mice per experiment. Source data
a, Body weight reduction after an injection of recombinant Fc–GDF15 or vehicle in wild-type (Glp1r+/+) or knockout (Glp1r−/−) mice. n = 6 animals. b, Body weight reduction after daily injections of GLP1 (Victoza) in wild-type (Gfral+/+) or knockout (Gfral−/−) mice. n = 9 animals per group. All data are mean ± s.e.m. Source data
a, Changes in whole body lean mass in wild-type and Gfral knockout mice treated with cisplatin (4 mg kg−1) once per week for six weeks. n = 8 (wild-type), n = 6 (knockout + vehicle) n = 7 (knockout + cisplatin) animals. P values obtained by two-tailed Holm–Sidak t-test (α = 0.05). b, Changes in wet weight of tissues isolated from wild-type and Gfral knockout mice treated with cisplatin (4 mg kg−1) once per week for six weeks. n = 6 animals. P values were calculated using two-tailed Mann–Witney test (95% confidence interval). All data are mean ± s.e.m. Source data
About this article
Cite this article
Hsu, JY., Crawley, S., Chen, M. et al. Non-homeostatic body weight regulation through a brainstem-restricted receptor for GDF15. Nature 550, 255–259 (2017). https://doi.org/10.1038/nature24042
The FASEB Journal (2021)
Molecular Metabolism (2021)
“Treasure Your Exceptions”—Studying Human Extreme Phenotypes to Illuminate Metabolic Health and Disease: The 2019 Banting Medal for Scientific Achievement Lecture
Serum growth differentiation factor 15 is closely associated with metabolic abnormalities in Chinese pregnant women
Journal of Diabetes Investigation (2021)
Hippocampal and Prefrontal Cortical Brain Tissue Levels of Irisin and GDF15 Receptor Subunits in Children
Molecular Neurobiology (2021)