Bone-derived hormone suppresses appetite

The glycoprotein lipocalin 2 is released from the bones of mice in a nutrient-dependent manner and binds to receptors in the brain to suppress appetite. This is the first example of bone-derived signals mediating hunger. See Article p.385

On pape 385, Mosialou et al.1 reveal a completely unexpected role for a secreted glycoprotein called lipocalin 2 (LCN2), which is known for its role in immune responses2,3. The authors show that LCN2 also acts as a hormone that mediates hunger. The glycoprotein is produced by bone-forming cells called osteoblasts in a nutrient-sensitive manner, and stimulates hormone-receptor proteins on neurons in the brain to suppress appetite.

Mosialou et al. first examined mice that could not produce LCN2 in osteoblasts. These animals ate more food and gained more weight than wild-type mice, and their ability to metabolize glucose was impaired. Mice lacking LCN2 expression in all their cells showed the same traits, substantiating the importance of LCN secretion from bone. On the basis of these observations, the researchers hypothesized that LCN2 acts as a hormone to reduce appetite. LCN2 was first identified more than 20 years ago4, and mice that lack it were generated 13 years ago to study its role in immunity3, but the mild obesity in these mutant mice apparently went unnoticed.

The authors found that secretion of LCN2 from bone increased several-fold following a meal, indicating that its release into the blood is sensitive to nutritional signals. To demonstrate that LCN2 is really a hormone (a signalling molecule that is transported between organs by the circulatory system), Mosialou and colleagues gave the animals daily injections of purified LCN2 to achieve concentrations in the blood similar to those reached after a meal. This treatment suppressed the appetites of wild-type mice, which lost weight compared with untreated control animals. Daily LCN2 injections in obese mice greatly reduced their weight gain and improved the ability of insulin to stimulate glucose uptake into cells.

Because feeding behaviour is controlled by the brain, the authors measured the amount of LCN2 in various brain regions, and found that it was most abundant in the brainstem and hypothalamus — regions that control feeding behaviour. When LCN2 was injected directly into the brain, it suppressed feeding as effectively as it did in the blood. The researchers concluded that LCN2 produced in bones circulates in the blood, crosses the blood–brain barrier and becomes selectively enriched in regions of the brain associated with appetite suppression (anorexia).

The discovery of a new hormone, especially one derived from bone, is itself intriguing, but Mosialou et al. set out to complete the story by identifying the receptor protein responsible for LCN2-induced anorexia. Clues to where this receptor might be found came from the authors' observation of LCN2 enrichment in the hypothalamus, combined with the fact that one of the best-established anorexia-promoting pathways is the signalling pathway involving α-melanin-stimulating hormone (α-MSH)5. In this pathway, α-MSH is produced by neurons in the hypothalamus and suppresses appetite by interacting with the melanocortin 4 receptor (MC4R), a member of a protein family called G-protein-coupled receptors.

Thus, the researchers explored the possibility that LCN2 somehow mimics α-MSH signalling. Indeed, they found that, in vitro, LCN2 stimulated production of an intracellular signalling molecule, cyclic AMP, by cells that expressed any one of three melanocortin receptors (MC1R, MC3R or MC4R), but not in cells lacking these receptors; this finding supports the idea that LCN2 binds to melanocortin receptors. Of these receptors, MC4R is expressed in the brainstem and hypothalamus and has been linked to feeding behaviour5. The affinity of LCN2 for MC4R binding was similar to that of α-MSH, and LCN2 could compete with α-MSH for binding to MC4R, despite the fact that the two molecules have no obvious similarities.

Further proof that LCN2 promotes anorexia by activating MC4R (Fig. 1) came from Mosialou and colleagues' demonstration that LCN2 bound to slices of hypothalamus in which MC4R is known to reside, but not to slices from mice lacking MC4R. Most importantly, they showed that LCN2 had no biological effects on food intake or glucose metabolism in mice lacking MC4R.

Figure 1: A bone–brain axis modulates hunger.

Mosialou et al.1 report that, after mice eat a meal, absorbed nutrients are sensed by bone-forming cells called osteoblasts, which respond by releasing the glycoprotein lipocalin 2 (LCN2). LCN2 enters the bloodstream and circulates around the body, passing into the brain's hypothalamus. Here, the glycoprotein binds to melanocortin 4 receptor (MC4R) proteins on neurons. Neuronal activation by MC4R binding induces a signalling pathway that leads to loss of appetite.

People with mutations in MC4R are often obese6, and the authors showed that some of these people have elevated levels of LCN2 in their blood compared with weight-matched people without MC4R mutations. This result suggests that signalling from the brain to the bones controls LCN2 production in an attempt to establish homeostasis.

Although Mosialou et al. concentrated their efforts on the binding of LCN2 to MC4R in the hypothalamus, these receptors are also abundant on the vagus nerve, which projects from most internal organs to the hindbrain7, where it can activate a neural circuit that promotes anorexia8. These vagal MC4Rs are more accessible to circulating hormones than are hypothalamic receptors, because they do not lie behind the blood–brain barrier. As such, they may be involved in the everyday appetite suppression induced by LCN2 after a meal.

It is well known9 that sepsis, a condition caused by bacteria, or experimentally induced by injecting rodents with bacterial lipopolysaccharide molecules, produces profound anorexia. LCN2 is robustly induced in many cells by this condition; hence, it may also contribute to the anorexia caused by sepsis2.

Overall, it is remarkable that hormones can still be discovered. Moreover, it is good to be reminded Footnote 1that unexpected functions of well-known proteins can be elucidated by astute observations and critical experiments.


  1. 1.

    See all news & views


  1. 1

    Mosialou, I. et al. Nature 543, 385–390 (2017).

    CAS  Article  ADS  PubMed  PubMed Central  Google Scholar 

  2. 2

    Ferreira, A. C. et al. Prog. Neurobiol. 131, 120–136 (2015).

    CAS  Article  PubMed  Google Scholar 

  3. 3

    Flo, T. H. et al. Nature 432, 917–921 (2004).

    CAS  Article  ADS  PubMed  Google Scholar 

  4. 4

    Flower, D. R. FEBS Lett. 354, 7–11 (1994).

    CAS  Article  PubMed  Google Scholar 

  5. 5

    Cone, R. D. Endocrine Rev. 27, 736–749 (2006).

    CAS  Article  Google Scholar 

  6. 6

    Farooqi, I. S. & O'Rahilly, S. Nature Clin. Pract. Endocrinol. Metab. 4, 569–577 (2008).

    CAS  Article  Google Scholar 

  7. 7

    Campos, C. A., Shiina, H. & Ritter, R. C. J. Neurosci. 34, 12636–12645 (2014).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  8. 8

    Roman, C. W., Derkach, V. A. & Palmiter, R. D. Nature Commun. 7, 11905 (2016).

    CAS  Article  ADS  Google Scholar 

  9. 9

    Liu, Y. et al. Endocrinology 157, 2380–2392 (2016).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Richard D. Palmiter.

Related links

Related links

Related links in Nature Research

Physiology: Microbial signals to the brain control weight

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Palmiter, R. Bone-derived hormone suppresses appetite. Nature 543, 320–321 (2017).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.