From Gut to Bone

Science Signaling  09 Dec 2008:
Vol. 1, Issue 49, pp. ec418
DOI: 10.1126/scisignal.149ec418

The signals that regulate bone density are clinically important because diseases affecting bone mass can arise through genetic mutations or as a consequence of the natural changes that occur during aging, such as osteoporosis associated with menopause in women. Yadav et al. provide evidence that serotonin synthesized by the duodenal enterochromaffin cells signals osteoblasts to inhibit bone formation and thereby decrease bone density. Mutations in the gene encoding low-density lipoprotein receptor–related protein 5 (LRP5) cause bone phenotypes in humans, with loss-of-function mutations causing a disease called osteoporosis pseudoglioma (OPPG) characterized by low bone mass and blindness, and gain-of-function mutations triggering diseases characterized by enhanced bone mass. Although LRP5 is best known as a coreceptor in the Wnt pathway, various lines of evidence indicate that Wnt signaling through β-catenin is not responsible for the regulation of bone by LRP5. Yadav et al. noted that the gene signatures of Lrp5 knockout mice (Lrp5−/−) and mice deficient in β-catenin in osteoblasts [β-Cat(ex3)osb−/−] had different gene signatures, although both mice exhibited decreased bone mass. The expression of the gene encoding the rate-limiting enzyme in serotonin synthesis in the periphery, tryptophan hydroxylase 1 (Tph1), was increased more than any other gene in the Lrp5−/− mice, whereas Tph1 expression was unchanged in the β-Cat(ex3)osb−/− mice. Blood serotonin concentrations were also elevated, and an increase in circulating serotonin was also noted in three patients with OPPG. Reducing circulating serotonin in the Lrp5−/− mice by reducing tryptophan ingestion or by pharmacological inhibition of its synthesis normalized the bone characteristics. To confirm that duodenal Lrp5 activity and serotonin production were responsible for the bone phenotypes, the authors created mice in which only gut or osteoblast Lrp5 or Tph1 was knocked out. Only the gut knockouts had altered circulating serotonin and, in the case of the gut-specific Lrp5 knockout, increased circulating serotonin and decreased bone mass and, in the case of the Tph1 knockout, reduced circulating serotonin concentration and increased bone mass. Serotonin acted through the Htr1b type of serotonin receptor on the osteoblasts; mice heterozygous for Htr1b showed increased bone mass. By generating mice lacking one allele of either the transcription factor ATF4 or CREB and combining them with the Lrp5−/− mice, the authors showed that serotonin appeared to act through CREB and not ATF4. Thus, serotonin released from the gut, the production of which is inhibited by LRP5, signals osteoblasts to inhibit bone formation. Thus, targeting circulating serotonin, which is completely separate from brain serotonin, may be an appealing target for bone density–related disorders (see commentary by Long).

V. K. Yadav. J.-H. Ryu, N. Suda, K. F. Tanaka, J. A. Gingrich, G. Schütz, F. H. Glorieux, C. Y. Chiang, J. D. Zajac, K. L. Insogna, J. J. Mann, R. Hen, P. Ducy, G. Karsenty, Lrp5 controls bone formation by inhibiting serotonin synthesis in the duodenum. Cell 135, 825–837 (2008). [PubMed]

F. Long, When the gut talks to bone. Cell 135, 795–796 (2008). [PubMed]