Leptin and insulin regulate food intake and appetite through actions in the hypothalamus. Both hormones reduce the firing rate of a subpopulation of hypothalamic neurones, an action attributed to K_ATP channel activation (Spanswick et al. 1997, 2000). We have previously reported that insulin activation of hypothalamic K_ATP channels is dependent upon phosphoinositide 3-kinase (PI 3-kinase; Spanswick et al. 2000). Consequently, we have examined whether PI 3-kinase also mediates leptin activation of these channels.

Recordings were made from neurones acutely isolated from Sprague-Dawley rats anaesthetized with enflurane (4 %) prior to decapitation (Spanswick et al. 1997). For cell-attached recordings, electrodes contained a high (140 mM) K⁺ solution, and neurones were bathed in 135 Na⁺-containing saline. Inside-out recordings were made in symmetrical 140 K⁺ at -40 mV. Mean channel activity was determined as N_f P_o (where N_f is the number of functional channels and P_o the open state probability) over a 120 s period. Means ± S.E.M. are given. Statistical analysis used Student’s t test.

Leptin (10 nM), present in the electrode solution during cell-attached recordings, increased channel activity in 64 % of neurones (n = 16/25). N_f P_o 1-2 min following cell-attached formation was 0.12 ± 0.03 and increased to 0.43 ± 0.06 after 10-20 min. In control experiments, N_f P_o was unchanged over 20 min (n = 9). Tolbutamide (200 µM) inhibited leptin-induced channel activity by 53 ± 12 % (n = 3; P < 0.05), indicating the channel was K_ATP. In inside-out patches, N_f P_o was reversibly inhibited by 63 ± 10 % on addition of 3 mM MgATP (n = 3; P < 0.05) and the single-channel conductance was 156 ± 15 pS (n = 3), consistent with K_ATP in rat glucose-responsive (GR) neurones.

The actions of the PI 3-kinase inhibitors LY294002 and wortmannin were examined on K_ATP activity during cell-attached recordings. Leptin increased N_f P_o from 0.21 ± 0.10 to 0.68 ± 0.28, which on application of 10 nM wortmannin in the continuous presence of leptin, was reduced to 0.33 ± 0.13 (n = 3; P < 0.01). In separate experiments leptin increased N_f P_o from 0.14 ± 0.03 to 0.39 ± 0.02 and 10 µM LY294002 reduced this to 0.19 ± 0.04 (n = 4; P < 0.01). Neither drug effects K_ATP channels directly (Spanswick et al. 2000).

These data indicate that leptin-induced activation of hypothalamic K_ATP channels is mediated through PI 3-kinase and thus provides evidence for potential cross-talk between the insulin and leptin pathways in GR neurones.This work was supported by The Wellcome Trust and Biovitrum.

Spanswick, D., Smith, M.A., Groppi, V.E., Logan, S.D. & Ashford, M.L.J. (1997). Nature 390, 521-525.

Spanswick, D., Smith, M.A., Mirshamsi, S., Routh, V.H. & Ashford, M.L.J. (2000). Nature Neurosci. 3, 757-758.