Leptin Is a Hypoxia‐Inducible Gene

Although the long-term control of adipose mass is a major function of leptin, this hormone exerts a wide spectrum of biological effects. Among those, leptin stimulates angiogenesis, suggesting that adipose leptin production could be controlled by angiogenic stimuli such as hypoxia. This idea is supported by reports of increased leptinemia in humans after 20 hours under hypobaric hypoxia at high altitude ((1)) and in patients with sleep apnea ((2)). However, Yasumasu et al. reported recently that leptin secretion was suppressed by 50% when rat adipose cells were cultured under a 10% O2 atmosphere for 48 hours ((3)). The authors conclude that hypoxia inhibits adipose leptin production and that the effects observed in humans must be indirect. In sharp contrast, our recent data demonstrate that hypoxia directly increases leptin gene expression, leptin promoter activity, and leptin secretion in human PAZ6 adipose cells ((4)). Similar data were obtained previously in another leptin-producing cell type, the choriocarcinoma BeWo cells, by us ((5)) and an independent group ((6)). During the course of these studies, we tested the effect of hypoxia in primary rat adipocytes. Epididymal adipose cells were cultured in the presence of agents known to mimic hypoxia, cobalt chloride (CoCl2) or desferrioxamine (DFO), or under 6% O2. In response to CoCl2 or DFO, leptin mRNA and leptin release were increased by more than 2-fold, in keeping with the effect of these agents in PAZ6 cells ((4)). However, when the cells were cultured under 6% O2, leptin accumulation in the medium was not enhanced but rather decreased (0.81 ± 0.15 vs. 1.39 ± 0.20 ng/106 cells per 24 hours; n = 3, p > 0.05). In contrast to the conclusion of Yasumasu et al. ((3)), we suggest that reduction of leptin production in response to low O2 does not reflect an inhibitory effect hypoxia, but results from increased cell breakage. Indeed, it is well known that floating primary adipose cells are highly sensitive to breakage in culture ((7)). Decreased O2 availability is likely to represent a profound metabolic stress that could further compromise cell survival. Of note, as reported by Yasumasu ((3)), not only leptin production but also glycero-3-phosphate dehydrogenase activity, triglyceride content, and, to a lesser extent, glucose uptake are decreased in hypoxic adipose cells. Therefore, in the absence of an internal marker not affected by O2 depletion, there is no evidence in this study that reduced leptin production does not result from a general effect of hypoxia on cell integrity. The hypoxia-inducible factor-1 (HIF-1) transactivates a variety of genes with products that mediate adaptive responses to hypoxia. Inhibition of a prolyl-4-hydroxylase activity accounts for HIF-1 induction by O2 deprivation or by agents that eliminate (DFO) or compete (CoCl2) with iron ((8),(9)). In contrast to the pleiotropic effects of O2 depletion, CoCl2 and DFO represent more specific stimuli that primarily activate the HIF-1 pathway. Our data, which demonstrate a clear stimulatory effect of these agents in BeWo ((5)), PAZ6 ((4)), and primary adipose cells, provide experimental evidences that leptin is a new hypoxia-inducible gene.