Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jun 1;595(11):3267-3285.
doi: 10.1113/JP273484. Epub 2017 Mar 27.

Nutritional Status-Dependent Endocannabinoid Signalling Regulates the Integration of Rat Visceral Information

Affiliations
Free PMC article

Nutritional Status-Dependent Endocannabinoid Signalling Regulates the Integration of Rat Visceral Information

Abdessattar Khlaifia et al. J Physiol. .
Free PMC article

Abstract

Key points: Vagal sensory inputs transmit information from the viscera to brainstem neurones located in the nucleus tractus solitarii to set physiological parameters. These excitatory synapses exhibit a CB1 endocannabinoid-induced long-term depression (LTD) triggered by vagal fibre stimulation. We investigated the impact of nutritional status on long-term changes in this long-term synaptic plasticity. Food deprivation prevents LTD induction by disrupting CB1 receptor signalling. Short-term refeeding restores the capacity of vagal synapses to express LTD. Ghrelin and cholecystokinin, respectively released during fasting and refeeding, play a key role in the control of LTD via the activation of energy sensing pathways such as AMPK and the mTOR and ERK pathways.

Abstract: Communication form the viscera to the brain is essential to set physiological homoeostatic parameters but also to drive more complex behaviours such as mood, memory and emotional states. Here we investigated the impact of the nutritional status on long-term changes in excitatory synaptic transmission in the nucleus tractus solitarii, a neural hub integrating visceral signals. These excitatory synapses exhibit a CB1 endocannabinoid (eCB)-induced long-term depression (LTD) triggered by vagal fibre stimulation. Since eCB signalling is known to be an important component of homoeostatic regulation of the body and is regulated during various stressful conditions, we tested the hypothesis that food deprivation alters eCB signalling in central visceral afferent fibres. Food deprivation prevents eCB-LTD induction due to the absence of eCB signalling. This loss was reversed by blockade of ghrelin receptors. Activation of the cellular fuel sensor AMP-activated protein kinase or inhibition of the mechanistic target of rapamycin pathway abolished eCB-LTD in free-fed rats. Signals associated with energy surfeit, such as short-term refeeding, restore eCB-LTD induction, which in turn requires activation of cholecystokinin receptors and the extracellular signal-regulated kinase pathway. These data suggest a tight link between eCB-LTD in the NTS and nutritional status and shed light on the key role of eCB in the integration of visceral information.

Keywords: endocannabinoid; nutritional status; synaptic plasticity.

Figures

Figure 1
Figure 1. Long‐term depression of excitatory synapses depends on the nutritional status
A, normalized individual response amplitude to the first stimulation of visceral afferent fibres (TS stimulation, 0.05 Hz) recorded before and after LTD induction by LFS from a rat fed ad libitum (Fed). Insets are the average EPSC before and 25 min after LFS. Series resistance (Rs) is also plotted against time. B, rats fed ad libitum: normalized EPSC amplitude for 25 neurones before and after induction of LTD by LFS (black circles, each data point represents the amplitude of 3 consecutive EPSCs). Grey circles represent the normalized EPSC amplitude for 6 neurones recorded for 50 min without LFS. C and D, same representation as before for an NTS neuron recorded in a slice from a rat fasted for 24 h (C) and for 13 neurones (D). E and F, same representation as before for an NTS neuron recorded in a slice from a rat fasted for 24 h and then refed for 3 h (E) and for 10 neurones (F). Note that in fasted rats LTD is impaired. G, mean LTD for the control (Fed) group, the fasted group and the refed group. H, changes in CV before and after LFS in NTS neurones. Note that in the fed and refed groups CV is significantly increased suggesting a presynaptic site for LTD induction. For all figures, * P < 0.05, ** P < 0.01, *** P < 0.001.
Figure 2
Figure 2. eCB synaptic depression is modulated by the nutritional status
A, rats fed ad libitum: normalized EPSC amplitude for 12 neurones before and during ACEA application. B, fasted rats: normalized EPSC amplitude for 12 neurones before and during ACEA application. C, refed rats after fasting: normalized EPSC amplitude for 9 neurones before and during ACEA application. D, representative traces for average EPSCs recorded in NTS neurones for the three conditions before and at the end of ACEA application. E, mean ACEA‐induced depression for the control (Fed) group, the fasted group and the refed group. F, changes in CV before and during ACEA application in NTS neurones. Note that in the fed and refed groups CV is significantly increased suggesting a presynaptic site for ACEA‐induced depression.
Figure 3
Figure 3. Fasting does not induce a permanent depressive state at NTS synapses
A, normalized EPSC amplitude for 8 neurones before and after LFS in fasted rats injected with vehicle. B, normalized EPSC amplitude for 10 neurones before and after LFS in fasted rats injected with RU486. Note that LTD is still prevented by fasting. C, normalized EPSC amplitude for 6 neurones before and after AM251 application in fasted rats. CB1R blockade did not reveal an eCB endogenous tone. D, upper trace, EPSC of NTS neurones expressed a paired‐pulse depression when TS was stimulated twice with a 50 ms interval. Lower trace, superimposition of average EPSC (20 traces) recorded at +30 mv before and after subtraction of the fast AMPA component (see Methods). The grey trace represents the NMDA component of the EPSC. E, comparison of PPR, CV and the ratio NMDA to AMPA in fed and fasted conditions (n = 37 and n = 29 for PPR and CV for fed and fasted, respectively, and n = 11 and n = 10 for ratios for fed and fasted, respectively). F, comparison of frequency‐dependent depression of EPSCs in fed (n = 20) and fasted (n = 18) conditions. The inset shows a typical response to a train (50 Hz, 5 shocks) in an NTS neuron. G, example of miniature EPSC on two different time scales. The EPSC at the bottom is an average of 100 mEPSCs. H, comparison of mEPSCs frequency and amplitude in fed (n = 10) and fasted (n = 14) conditions. None of the parameters depicted here show a significant difference between groups.
Figure 4
Figure 4. Ghrelin is responsible for the loss of LFS‐LTD during fasting
A, normalized EPSC amplitude for 6 neurones before and after LFS in fasted rats pretreated with ghrelin antagonists. Note the recovery of LFS‐LTD. B and C, fed rats i.p. injected with ghrelin. B, normalized EPSC amplitude for 8 neurones before and after LFS. C, normalized EPSC amplitude for 8 neurones before and during ACEA application. D and E, slices from fed rats preincubated with ghrelin for 30 min. D, normalized EPSC amplitude for 7 neurones before and after LFS. E, normalized EPSC amplitude for 7 neurones before and during ACEA application. F and G, slices from fed rats preincubated with AICAR for 30 min. F, normalized EPSC amplitude for 7 neurones before and after LFS. G, normalized EPSC amplitude for 6 neurones before and during ACEA application. H, mean depression for ghrelin or AICAR group. Pre‐activation of ghrelin receptors or AICAR suppresses LFS‐ and eCB‐induced depression.
Figure 5
Figure 5. mTOR activation is necessary for LFS‐LTD and ACEA‐induced depression
A, normalized EPSC amplitude for 7 neurones before and during rapamycin application in fed rats. B and C, slices from fed rats pre‐treated with rapamycin. B, normalized EPSC amplitude for 10 neurones before and after LFS. C, normalized EPSC amplitude for 7 neurones before and during ACEA application. D, mean depression for all groups. Pre‐incubation with rapamycin suppresses eCB‐induced depression. E, normalized EPSC amplitude for 7 neurones before and during leucine application. F, normalized EPSC amplitude for 6 neurones before and during leucine application on slices pretreated with rapamycin. Leucine induces an EPSC depression which is blocked by rapamycin. G, leucine induces a significant increase in EPSC CV which is prevented by rapamycin. H, mean depression for the leucine groups.
Figure 6
Figure 6. CCK rescues eCB‐LTD
A, normalized EPSC amplitude for 12 neurones before and after induction of LTD by LFS in rats refed for 3 h and injected with vehicle. B, normalized EPSC amplitude for 10 neurones before and after induction of LTD by LFS in rats refed for 3 h and injected with devazepide. C, normalized EPSC amplitude for 8 neurones before and after induction of LTD by LFS in fasted rats injected with vehicle. D, normalized EPSC amplitude for 9 neurones before and after induction of LTD by LFS in fasted rats injected with CCK 3 h before killing. E, comparison of LTD amplitude for the different groups. CCK restores LFS‐LTD only in the fasted group injected 3 h before killing. Rescue of LTD by refeeding is prevented by injection of devazepide. F, normalized EPSC amplitude for 8 neurones before and during ACEA application in fasted rats injected with CCK 3 h before killing. G, normalized EPSC amplitude for 7 neurones before and during ACEA application in fasted rats first injected with MK‐801 then with CCK. H, comparison of ACEA‐induced depression for the different groups. CCK restores ACEA‐induced depression in the fasted group via activation of NMDAR.
Figure 7
Figure 7. Activation of MEK is necessary for eCB‐LTD rescue
A, normalized EPSC amplitude for 8 neurones before and after LFS in fasted rats injected with SL‐327 and CCK 3 h before killing. B, normalized EPSC amplitude for 10 neurones before and after LFS in fed rats injected with SL‐327. C, normalized EPSC amplitude for 6 neurones before and after LFS on slices treated with SL‐327 (10 μm) in fed rats. D, normalized EPSC amplitude for 10 neurones before and after LFS on slices treated with U0126 (10 μm) in fed rats. E and F, comparison of LTD amplitude for the different groups. Blockade of MEK by i.p. injection of SL‐327 prevents rescue of LTD by CCK but does not affect control LTD in fed rats. Direct blockade of MEK does not alter LFS‐LTD in fed rats.
Figure 8
Figure 8. Schematic diagram of LTD regulation
In the fed state, glutamate release by visceral afferent fibres activates NMDARs (either located in a neighbouring neuron or in the presynaptic terminal (Khlaifia et al. 2013) and triggers eCB release. eCB acting on CB1Rs (CBR) would reduce release probability of synaptic vesicles via mTOR activation. During fasting, ghrelin via its receptors (GHR) leads to AMPK activation that may inhibit mTOR and prevent LTD induction. Alternatively, decoupling G protein to CB1Rs may also reduce LTD induction. During refeeding, increased activation of vagal fibres by CCK would activate the NMDAR–ERK pathway that may on a slower time scale change the balance between AMPK and mTOR and therefore restore LTD induction. [Color figure can be viewed at wileyonlinelibrary.com]

Similar articles

See all similar articles

Cited by 2 articles

Publication types

MeSH terms

Feedback