Leptin

Leptin is a 167-amino-acid hormone of about 16 kilodaltons, secreted by adipocytes in rough proportion to fat mass and identified in 1994 through positional cloning of the mouse ob gene. It acts on the long isoform of the leptin receptor in the arcuate nucleus, signalling through JAK2 and STAT3 to stimulate POMC neurons and inhibit AgRP neurons. It is better described as a starvation signal than a satiety signal: concentrations fall sharply and early in energy deficit, and the response to that fall is much stronger than the response to a rise.

Congenital leptin deficiency is vanishingly rare and produces extreme hyperphagia from infancy that reverses on recombinant leptin, one of the cleanest hormone-replacement demonstrations in metabolic medicine. Metreleptin was approved in 2014 for generalised lipodystrophy, where fat tissue and therefore leptin are absent. In common obesity the opposite holds: leptin is already high, the hypothalamus responds poorly, and trials of recombinant leptin as a weight-loss drug failed. That failure is a load-bearing negative result, not a footnote.

Leptin explains why maintaining weight loss is biologically harder than achieving it. Fat mass falls, leptin falls faster still, and hunger and energy expenditure shift in the direction that restores the lost tissue. Small studies restoring leptin to pre-loss concentrations in weight-reduced people reversed part of that adaptation, which is mechanistically informative and has never become a therapy.

The commercial error is the phrase leptin resistance used as a diagnosis. It names an observation, not a measurable state with a threshold, and no supplement has shown it can be corrected. A serum leptin level in a person with obesity is predictable from their fat mass and adds nothing; the inference that raising leptin would suppress appetite is the exact hypothesis the trials refuted.