Human obesity leads to an increase in respiratory demands. As obesity becomes more pronounced some individuals are unable to compensate, leading to elevated arterial carbon dioxide levels (PaCO2), alveolar hypoventilation, and increased cardiorespiratory morbidity and mortality (Pickwickian syndrome). The mechanisms that link obesity and hypoventilation are unknown, but thought to involve depression of central respiratory control mechanisms. Here we report that obese C57BL/6J-Lepob mice, which lack circulating leptin, also exhibit respiratory depression and elevated PaCO2 (> 10 mm Hg; p < 0. 0001). A role for leptin in restoring ventilation in these obese, mutant mice was investigated. Three days of leptin infusion (30 microg/d) markedly increased minute ventilation (V E) across all sleep/wake states, but particularly during rapid eye movement (REM) sleep when respiration was otherwise profoundly depressed. The effect of leptin was independent of food intake, weight, and CO2 production, indicating a reversal of hypoventilation by stimulation of central respiratory control centers. Furthermore, leptin replacement in mutant mice increased CO2 chemosensitivity during non-rapid eye movement (NREM) (4.0 +/- 0.5 to 5.6 +/- 0.4 ml/min/%CO2; p < 0.01) and REM (-0.1 +/- 0.5 to 3.0 +/- 0.8 ml/min/%CO2; p < 0.01) sleep. We also demonstrate in wild-type mice that ventilation is appropriately compensated when obesity is diet-induced and endogenous leptin levels are raised more than tenfold. These results suggest that leptin can prevent respiratory depression in obesity, but a deficiency in central nervous system (CNS) leptin levels or activity may induce hypoventilation and the Pickwickian syndrome in some obese subjects. O'Donnell CP, Schaub CD, Haines AS, Berkowitz DE, Tankersley CG, Schwartz AR, Smith PL. Leptin prevents respiratory depression in obesity.

Obese females are less predisposed to sleep-disordered breathing and have higher serum leptin levels than males of comparable body weight. Because leptin is a powerful respiratory stimulant, especially during sleep, we hypothesized that the elevated leptin level is necessary to maintain normal ventilatory control in obese females. We examined ventilatory control during sleep and wakefulness in male and female leptin-deficient obese C57BL/6J-Lep(ob) mice, wild-type C57BL/6J mice with dietary-induced obesity and high serum leptin levels, and normal weight wild-type C57BL/6J mice. Both male and female C57BL/6J-Lep(ob) mice had depressed hypercapnic ventilatory response (HCVR) in comparison with wild-type animals. In comparison with male C57BL/6J-Lep(ob) mice, female C57BL/6J-Lep(ob) mice had reduced HCVR and respiratory drive (a ratio of tidal volume to inspiratory time) both during non-rapid eye movement (NREM) sleep and wakefulness. In contrast, the HCVR did not differ between sexes in wild-type mice during NREM sleep and wakefulness, but was lower in females during REM sleep. Thus, leptin deficiency in female obesity is even more detrimental to hypercapnic ventilatory control during wakefulness and NREM sleep than in obese, leptin-deficient males.

The pattern of breathing during sleep could be a heritable trait. Our intent was to test this genetic hypothesis in inbred mouse strains known to vary in breathing patterns during wakefulness (Han F, Subramanian S, Dick TE, Dreshaj IA, and Strohl KP. J Appl Physiol 91: 1962-1970, 2001; Han F, Subramanian S, Price ER, Nadeau J, and Strohl KP, J Appl Physiol 92: 1133-1140, 2002) to determine whether such differences persisted into non-rapid eye movement (NREM) and rapid eye movement (REM) sleep. Measures assessed in C57BL/6J (B6; Jackson Laboratory) and two A/J strains (A/J Jackson and A/J Harlan) included ventilatory behavior [respiratory frequency, tidal volume, minute ventilation, mean inspiratory flow, and duty cycle (inspiratory time/total breath time)], and metabolism, as performed by the plethsmography method with animals instrumented to record EEG, electromyogram, and heart rate. In all strains, there were reductions in minute ventilation and CO2 production in NREM compared with wakefulness (P < 0.001) and a further reduction in REM compared with NREM (P < 0.001), but no state-by-stain interactions. Frequency showed strain (P < 0.0001) and state-by-strain interactions (P < 0.0001). The A/J Jackson did not change frequency in REM vs. NREM [141 +/- 15 (SD) vs. 139 +/- 14 breaths/min; P = 0.92], whereas, in the A/J Harlan, it was lower in REM vs. NREM (168 +/- 14 vs. 179 +/- 12 breaths/min; P = 0.0005), and, in the B6, it was higher in REM vs. NREM (209 +/- 12 vs. 188 +/- 13 breaths/min; P < 0.0001). Heart rate exhibited strain (P = 0.003), state (P < 0.0001), and state-by-strain interaction (P = 0.017) and was lower in NREM sleep in the A/J Harlan (P = 0.035) and B6 (P < 0.0001). We conclude that genetic background affects features of breathing during NREM and REM sleep, despite broad changes in state, metabolism, and heart rate.

Insulin-dependent diabetes mellitus (IDDM) can lead to ventilatory depression and decreased sensitivity to hypercapnia. We examined relationships between ventilation, plasma insulin, leptin, ketones, and blood glucose levels in two mouse models of IDDM: (1) streptozotocin-induced diabetes in C57BL/6J mice on a regular diet or with induced obesity from a high fat diet; and (2) spontaneous diabetes mellitus in NOD-Ltj mice. In both mouse models, IDDM resulted in depression of the hypercapnic ventilatory response (HCVR). This ventilatory depression was not associated with decreases in plasma insulin or leptin levels. There was, however, a strong association between the duration of hyperglycemia, the decline in HCVR, and increased glycosylation of the diaphragm. Hyperventilation was observed in only six of 14 C57BL/6J obese wild-type mice, despite a significant degree of diabetic ketoacidosis (DKA) in all 14 animals. In mice with DKA, there was a significant correlation between the increase in baseline minute ventilation (V E) and hyperleptinemia (r = 0.77, p < 0.01). In leptin-deficient C57BL/6J-Lep(ob) mice, low levels of both V E and ketones were observed. These results suggest that: (1) depression of the HCVR in IDDM is associated with hyperglycemia and glycosylation of the diaphragm; and (2) the hyperventilation of DKA is leptin dependent.