1. Hunger States Switch a Flip-Flop Memory Circuit via a Synaptic AMPK-Dependent Positive Feedback Loop 
Yunlei Yang, Deniz Atasoy, Helen H. Su, Scott M. Sternson 
Cell 
Volume 146, Issue 6, Pages (September 2011)
DOI: /j.cell
Copyright © 2011 Elsevier Inc. Terms and Conditions

2. Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

3. Figure 1
Deprivation-Induced Synaptic Plasticity in AGRP and POMC Neurons
(A) mEPSCs from fed and food-deprived (dep) mice.
(B) fmEPSC in AGRP neurons from deprived mice in the light period or from fed mice at the transition to the dark period (DP) were both significantly increased over fmEPSC from fed mice in the light period.
(C) Fluorescence micrograph of the arcuate nucleus from a Pomc-topazFP, Npy-sapphireFP double-transgenic mouse. POMC neurons (green) and AGRP neurons (blue) are intermingled.
(D) In POMC neurons, fmEPSC is decreased by food deprivation.
(E) Elevated spontaneous firing rate in AGRP neurons from deprived mice (n = 11) is reduced to the level of fed mice (n = 12) by CNQX.
∗∗∗p < Data are represented as mean ± SEM. See also Figure S1.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

4. Figure 2
Role of Calcium and Ghrelin in Synaptic Plasticity at AGRP Neurons
(A) Comparison of fmEPSC in AGRP neurons from fed and food-deprived (dep) mice with no treatment (nt, data from Figure 1B) or treated with BAPTA-AM, ryanodine, or ghrelin (left panel). The right panel shows treatments in the presence of CdCl2, which blocks VGCCs. All pairwise interactions were tested and p values were corrected with Holm's method. Significant differences are denoted by any interaction across the red dashed line; interactions on the same side of the line are not significant (p > 0.05). Left and right panels were analyzed separately.
(B) Caffeine increases fmEPSC (n = 8), which is blocked by ryanodine pretreatment (red, n = 3).
(C) Time course of ghrelin-mediated fmEPSC increase in an AGRP neuron.
(D) Ghrelin increases fmEPSC in AGRP neurons (n = 6), which is blocked by D-Lys3-GHRP6 (blue, n = 5) or ryanodine (red, n = 7).
(E) For fed mice, ghrelin injection increased fmEPSC relative to saline treatment.
(F) i.c.v. D-Lys3-GHRP6 (blue) during deprivation blocked the fmEPSC increase observed with i.c.v. saline (black).
∗∗∗p < Data are represented as mean ± SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

5. Figure 3
AMPK Signaling Mediates Deprivation- and Ghrelin-Induced Synaptic Activity
(A) AICAR increases fmEPSC in AGRP neurons from fed (n = 8) but not food-deprived (n = 8) mice.
(B) Cpd C reduces fmEPSC in AGRP neurons from deprived (n = 8) but not fed (n = 6) mice and blocks the ghrelin-mediated increase in fmEPSC.
(C) AGRP neuron dialysis with the AMPK activator ZMP (3 mM) in the patch pipette internal solution (int.) does not significantly increase fmEPSC relative to AGRP neurons recorded with standard internal solution (nt, data from Figure 1B).
(D) Targeted AMPK inhibition with Cpd C (int.) in AGRP neurons does not significantly (p > 0.05) change fmEPSC after neuron dialysis (15 min).
(E) AGRP neuron firing rate is increased by bath-applied AICAR after intracellular blockade of AMPK with Cpd C (int.).
(F) AGRP neuron firing is decreased by bath-applied Cpd C after dialysis with Cpd C (int.).
(G) Targeted AMPK inhibition with Cpd C (int.) does not block ghrelin-mediated fmEPSC increase.
(H) Ghrelin activates firing in AGRP neurons from fed (n = 8) but not deprived (n = 7) mice with postsynaptic AMPK blockade by Cpd C (int.).
(I) Glutamate receptor blockade prevents ghrelin activation of AGRP neuron firing.
(J) Paired-pulse ratio (PPR) in AGRP neurons from fed and food-deprived mice in 2 mM Ca2+.
(K) PPR in AGRP neurons from fed mice (0.5 mM external Ca2+) is reduced by ghrelin, and this is reversed by Cpd C. Average EPSC responses from two cells are also shown (inset).
(L and M) PPR in AGRP neurons from deprived mice (0.5 mM external Ca2+) is unaffected by ghrelin and increased by Cpd C. (M, inset) Average EPSC response from one cell is shown.
(N) Inhibition of CAMKK with STO-609 in AGRP neurons from fed mice blocks ghrelin-mediated but not AICAR-mediated increase of fmEPSC.
(O) AICAR does not significantly increase fmEPSC in the presence of ryanodine.
(P) 8-Br-cADP ribose blocks the ghrelin-mediated increase of fmEPSC.
(Q) Diagram of the signaling pathway supported by these experiments. Pointed and “T” arrows represent activation and inhibition, respectively.
n.s., p > 0.05, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < Data are represented as mean ± SEM. See also Figure S2.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

6. Figure 4
Synaptic Hysteresis Resulting from an AMPK-Dependent Positive Feedback Loop
(A) Ghrelin upregulation of fmEPSC shows hysteresis; fmEPSC is sustained after ghrelin washout and Ghsr1 blockade with SP∗. Duration of each transition is in parentheses (minutes).
(B) Caffeine increases fmEPSC, which remains elevated after washout. Synaptic activity returned to baseline after treatment with Cpd C, consistent with a positive feedback loop. Significant differences are denoted by any interaction across the red dashed line.
(C–F) Procedure for testing duration of persistent activity by transient exposure of brain slices to ghrelin or caffeine (5 min), transfer to a wash solution (10 min), and transfer again to a solution containing either aCSF alone or with D-Lys3-GHRP6 (3–5 hr). Slices were subsequently transferred to a recording chamber for electrophysiology.
(D and E) After transient ghrelin exposure and prolonged D-Lys3-GHRP6 incubation, fmEPSC remained elevated but was rapidly (10 min) reduced by (D) Cpd C or (E) STO-609. Control brain slices that were not treated (nt) with ghrelin but otherwise incubated as above (D) had low fmEPSC.
(F) After transient exposure to caffeine and prolonged incubation in caffeine-free aCSF (3–5 hr), fmEPSC was still elevated but was rapidly (10 min) reduced by Cpd C, consistent with the operation of a positive feedback loop.
n.s., p > 0.05, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < Data are represented as mean ± SEM. See also Figure S3.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

7. Figure 5
Persistent Synaptic Upregulation Is Reversed by Leptin-Mediated Opioid Release
(A) Synaptic activity before, during, and after food deprivation and refeeding (fed and deprived data from Figure 1C, 1 hr: n = 21, 24 hr: n = 27, 48 hr: n = 23). p values for multiple pairwise comparisons were adjusted with Holm's correction. The significant differences are denoted by any interaction across the red dashed line.
(B) AGRP neuron firing rate 24 hr after refeeding was still elevated, which was dependent on glutamatergic synaptic input.
(C) Injection of leptin in deprived mice reduced fmEPSC in AGRP neurons relative to saline injection.
(D) DAMGO reduced fmEPSC in AGRP neurons under VGCC block (CdCl2). The fmEPSC remained at this level during a 15 min wash but was increased by treatment with AICAR.
(E) fmEPSC in AGRP neurons from deprived mice treated with DAMGO is insensitive to Cpd C.
(F) NTX pretreatment of deprived mice blocks leptin-mediated reduction in fmEPSC observed with saline pretreatment.
(G) Coinjection of ghrelin and NTX, but not ghrelin alone, leads to elevated fmEPSC in brain slices prepared after 3 hr.
∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < Data are represented as mean ± SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

8. Figure 6
POMC Neurons Release an Opioid that Resets Persistent Synaptic Activity
(A) Epifluorescence micrograph of brain slice with POMC neurons expressing ChR2-tdtomato. Blue circles: photostimulation sites; ARC: arcuate nucleus; 3V: third ventricle; D: dorsal; V: ventral.
(B) ChR2-mediated photostimulation of POMC neurons in brain slices from deprived mice in the presence or absence of NTX. Photostimulation of POMC neurons reduces fmEPSC unless performed in the presence of NTX.
(C) Subset of neurons in (B) subjected to AICAR after photostimulation. POMC neuron photostimulation in the absence of NTX (n = 5) rendered fmEPSC in AGRP neurons sensitive to AICAR, whereas those photostimulated in the presence of NTX (n = 5) were insensitive to AICAR, indicating that a POMC neuron-derived opioid inactivates AMPK.
∗∗p < Data are represented as mean ± SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

9. Figure 7
SR Flip-Flop Model of a Neural Circuit with Synaptic Memory of Physiological State
(A) A core circuit in which AGRP neurons synaptically inhibit POMC neurons and are regulated by circulating hormones is controlled by ghrelin-responsive excitatory synapses. These synapses give this circuit a memory property based on an AMPK-dependent positive feedback loop (inset), which can be reversed by POMC neuron output, likely β-endorphin.
(B) A heuristic for the logic of this circuit is the SR flip-flop memory storage circuit. In the analogy with the neural circuit reported here, the set signal is ghrelin, which activates the green NOR gate, representing the conglomeration of AGRP neurons and their ghrelin-sensitive excitatory presynaptic terminals. The reset signal is leptin, which interacts with POMC neurons represented as the blue NOR gate. Notably, when R and S are both high, the circuit does not support memory, and this condition is consistent with the case of ghrelin treatment of fed mice where opioid signaling is sufficiently high to prevent persistent synaptic upregulation (Figure 5G).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

10. Figure S1
Deprivation-Induced Postsynaptic Properties in AGRP Neurons, Related to Figure 1
(A) Example recordings of evoked synaptic currents at +40 mV holding potential: Itotal (black), IAMPA-R (blue), and INMDA-R (orange, calculated by subtraction). Below, ratio of AMPA- and NMDA-receptor synaptic currents in AGRP neurons from fed (n = 9 cells) or food-deprived (dep) mice (n = 7 cells). Stimulus artifact has been removed from the data traces. Scale bar: 50 ms, 10 pA. Statistical analysis was with an unpaired t test.
(B) Example recordings and rectification index of glutamatergic synaptic currents [|(evoked current amplitude at +40 mV holding potential)/(evoked current amplitude at −60 mV holding potential)|] in AGRP neurons from fed (n = 7 cells) or deprived mice (n = 16 cells). Stimulus artifact has been removed from the data traces. Scale bar: 5 ms, 20 pA. Statistical analysis was with an unpaired t test.
(C) mEPSC amplitudes in AGRP neurons from fed mice (n = 60 cells), deprived mice (n = 62 cells), or fed mice at the start of the dark period (n = 26 cells). Note that the reduction in sample size for mEPSC amplitude (fed) relative to mEPSC frequency (fed) (Figure 1B) is because two AGRP neurons had fmEPSC = 0.
n.s., p > Data are represented as mean ± SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

11. Figure S2
Dose Responses and Pharmacological Control Experiments, Related to Figure 3
(A) fmEPSC dose-response curve for AICAR (19 μM, 57 μM, 100 μM, 167 μM, 500 μM, 1500 μM, n = 3–8 cells/concentration) in AGRP neurons from fed mice. Each concentration was tested on separate cells. A sigmoid fit to the data is also shown.
(B) fmEPSC dose-response curve for Cpd C (17 μM, 50 μM, 100 μM, 150 μM, 450 μM, n = 3–8 cells/concentration) in AGRP neurons from food deprived mice. Each concentration was tested on separate cells. A sigmoid fit to the data is also shown.
(C) Because AICAR can increase extracellular adenosine by acting as a competitive antagonist of the adenosine transporter, we tested the influence of adenosine receptors on AICAR upregulation of fmEPSC. In AGRP neurons from fed mice, the adenosine receptor antagonist, CGS15943 (either 1 μM, n = 7 or 10 μM, n = 3; concentrations are noted in parentheses [μM]), slightly increased fmEPSC, as expected from suppression of baseline adenosine receptor signaling. Subsequent addition of AICAR further increased fmEPSC, demonstrating that AICAR upregulates synaptic activity independently of adenosine receptor signaling. Statistical analysis was with paired t tests, using Holm's correction.
(D) Because AICAR might also directly or indirectly interact with purinergic (P2-type) receptors, we used the broad-spectrum P2 receptor antagonist, suramin, to block these receptors. In AGRP neurons from fed mice, suramin (either 5 μM, n = 8 or 15 μM, n = 5; concentrations are noted in parentheses [μM]), slightly increased fmEPSC. Subsequent addition of AICAR, further increased fmEPSC, indicating that AICAR upregulates synaptic activity independently of these receptors. Statistical analysis was with paired t tests, using Holm's correction.
(E) Because suramin does not act on P2Y6 receptors, which are expressed in the mouse brain, the P2Y6 receptor antagonist, MRS 2578 (15 μM), was applied to AGRP neurons from fed mice (n = 4), slightly increasing fmEPSC. Subsequent addition of AICAR further increased fmEPSC, indicating that AICAR upregulates synaptic activity independently of these receptors. Statistical analysis was with paired t tests, using Holm's correction for multiple comparisons.
(F) In AGRP neurons from fed mice, fmEPSC increased in response to caffeine and this was not significantly changed by subsequent treatment with Cpd C, indicating that Cpd C does not reduce fmEPSC nonspecifically (n = 5). Statistical analysis was with paired t tests, using Holm's correction.
(G) In AGRP neurons from fed mice, Cpd C (150 μM) did not block CGS15943 (1 μM)-mediated upregulation of fmEPSC (n = 3, paired t test p = 0.033), further indicating that Cpd C does not reduce fmEPSC nonspecifically.
(H) In AGRP neurons from fed mice, Cpd C (150 μM) blocked AICAR-mediated increase in fmEPSC (n = 3, paired t test, p = 0.81), consistent with AMPK as the functionally relevant molecular target for AICAR and Cpd C in the modulation of excitatory synaptic input at AGRP neurons.
(I) Validation of intracellular activity of ZMP dialyzed through a patch pipette (also see Experimental Procedures). In AGRP neurons from fed mice expressing the Ca2+ indicator, GCaMP3, dialysis of neurons with internal solution containing ZMP (3 mM, n = 6) significantly increased fluorescence response compared to dialysis with an internal solution lacking ZMP (n = 6), 15 min after establishing the whole-cell configuration (unpaired t test, p = 0.034). Decrease in fluorescence under control conditions was due to bleaching from fluorescence monitoring at time points between zero and 15 min, and this was also observed for other neurons in the imaging field that were not patched (data not shown).
(J) Ghrelin increases fmEPSC in AGRP neurons from fed mice dialyzed with a high concentration of BAPTA (10 mM). Statistical analysis was with paired t test.
(K) Partial antagonism of glutamate receptors with γ-DGG (2 mM) did not lead to a significant change in PPR within or between AGRP neurons from fed (n = 5) and food-deprived (n = 5) mice (two-way ANOVA with one factor repeated-measures; ± γ-DGG: F1,8 = 0.91, p = 0.37; fed/dep: F1,8 = 1.32, p = 0.28; interaction: F1,8 = 0.06, p = 0.82).
(L) Dose-response curves for inhibition of ghrelin (30 nM, black)-mediated upregulation of fmEPSC in AGRP neurons from fed mice by STO-609 (0.6 μM, 1.7 μM, 3 μM, 5 μM, 15 μM; n = 3–10 cells/concentration). Similarly, inhibition of AICAR (500 μM, red)-mediated upregulation of fmEPSC by STO-609 (5 μM, 15 μM, 50 μM, 135 μM; n = 5–10 cells/concentration) is shown. A window for selective CAMKK inhibition is denoted with blue shading. Each concentration was tested on separate cells. Sigmoid fits to the data are also shown.
(M) Further analysis of dose response for inhibition of AMPK activity by STO-609 as measured in brain slice lysates by phosphorylation of acetylCoA carboxylase (ACC), a downstream target of AMPK in many cell types. Western blots are shown from two representative experiments for phosphorylated ACC (pACC). Lysates were prepared from brain slices either not treated (nt) or treated with AICAR (A, 500 μM) or AICAR + STO-609 (A/S, STO-609 concentrations in parentheses [μM]). Actin loading controls are also shown.
(N) Analysis of western blots for pACC. Band intensity of pACC in each lane was normalized to actin intensity. Data are presented relative to slices not treated pharmacologically. STO-609 (3 μM, 15 μM) did not reverse AICAR-mediated phosphorylation of acetylCoA carboxylase (pACC), but a higher concentration (50 μM) did partially inhibit AICAR-mediated pACC phosphorylation (n = 4 brain slice lysates for each condition), similar to fmEPSC response in (L). ANOVA: F3,15 = 4.9, p =
n.s., p > 0.05, ∗p < 0.05, ∗∗p < Data are represented as mean ± SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

12. Figure S3
Role of Ghrelin-Mediated and Ligand-Independent Ghrelin Receptor Signaling for Sustained Elevation of fmEPSC, Related to Figure 4
Blockade of Ghsr1 with D-Lys3-GHRP6 (n = 4; paired t test, p = 0.86) or SP∗ (n = 6; paired t test, p = 0.13) does not significantly change fmEPSC in AGRP neurons from deprived mice (15 min treatment).
n.s., p > Data are represented as mean ± SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions