Predictive models of glucose control: roles for glucose-sensing neurones

C Kosse, A Gonzalez, D Burdakov (Corresponding Author)

Research output: Contribution to journalReview article

13 Citations (Scopus)
4 Downloads (Pure)

Abstract

The brain can be viewed as a sophisticated control module for stabilizing blood glucose. A review of classical behavioural evidence indicates that central circuits add predictive (feedforward/anticipatory) control to the reactive (feedback/compensatory) control by peripheral organs. The brain/cephalic control is constructed and engaged, via associative learning, by sensory cues predicting energy intake or expenditure (e.g. sight, smell, taste, sound). This allows rapidly measurable sensory information (rather than slowly generated internal feedback signals, e.g. digested nutrients) to control food selection, glucose supply for fight-or-flight responses or preparedness for digestion/absorption. Predictive control is therefore useful for preventing large glucose fluctuations. We review emerging roles in predictive control of two classes of widely projecting hypothalamic neurones, orexin/hypocretin (ORX) and melanin-concentrating hormone (MCH) cells. Evidence is cited that ORX neurones (i) are activated by sensory cues (e.g. taste, sound), (ii) drive hepatic production, and muscle uptake, of glucose, via sympathetic nerves, (iii) stimulate wakefulness and exploration via global brain projections and (iv) are glucose-inhibited. MCH neurones are (i) glucose-excited, (ii) innervate learning and reward centres to promote synaptic plasticity, learning and memory and (iii) are critical for learning associations useful for predictive control (e.g. using taste to predict nutrient value of food). This evidence is unified into a model for predictive glucose control. During associative learning, inputs from some glucose-excited neurones may promote connections between the 'fast' senses and reward circuits, constructing neural shortcuts for efficient action selection. In turn, glucose-inhibited neurones may engage locomotion/exploration and coordinate the required fuel supply. Feedback inhibition of the latter neurones by glucose would ensure that glucose fluxes they stimulate (from liver, into muscle) are balanced. Estimating nutrient challenges from indirect sensory cues may become more difficult when the cues become complex and variable (e.g. like human foods today). Consequent errors of predictive glucose control may contribute to obesity and diabetes.

Original languageEnglish
Pages (from-to)7-18
Number of pages12
JournalActa Physiologica
Volume213
Issue number1
Early online date28 Aug 2014
DOIs
Publication statusPublished - Jan 2015

Fingerprint

Neurons
Glucose
Cues
Food
Learning
Reward
Brain
Association Learning
Food Preferences
Muscles
Neuronal Plasticity
Smell
Wakefulness
Liver
Locomotion
Energy Intake
Energy Metabolism
Blood Glucose
Digestion
Obesity

Keywords

  • Animals
  • Brain
  • Glucose
  • Humans
  • Hypothalamic Hormones
  • Intracellular Signaling Peptides and Proteins
  • Learning
  • Melanins
  • Neurons
  • Neuropeptides
  • Orexins
  • Pituitary Hormones
  • Journal Article
  • Research Support, Non-U.S. Gov't
  • Review

Cite this

Predictive models of glucose control : roles for glucose-sensing neurones. / Kosse, C; Gonzalez, A; Burdakov, D (Corresponding Author).

In: Acta Physiologica, Vol. 213, No. 1, 01.2015, p. 7-18.

Research output: Contribution to journalReview article

@article{e87ebcbdc2de402c8b5c73aa21e391f6,
title = "Predictive models of glucose control: roles for glucose-sensing neurones",
abstract = "The brain can be viewed as a sophisticated control module for stabilizing blood glucose. A review of classical behavioural evidence indicates that central circuits add predictive (feedforward/anticipatory) control to the reactive (feedback/compensatory) control by peripheral organs. The brain/cephalic control is constructed and engaged, via associative learning, by sensory cues predicting energy intake or expenditure (e.g. sight, smell, taste, sound). This allows rapidly measurable sensory information (rather than slowly generated internal feedback signals, e.g. digested nutrients) to control food selection, glucose supply for fight-or-flight responses or preparedness for digestion/absorption. Predictive control is therefore useful for preventing large glucose fluctuations. We review emerging roles in predictive control of two classes of widely projecting hypothalamic neurones, orexin/hypocretin (ORX) and melanin-concentrating hormone (MCH) cells. Evidence is cited that ORX neurones (i) are activated by sensory cues (e.g. taste, sound), (ii) drive hepatic production, and muscle uptake, of glucose, via sympathetic nerves, (iii) stimulate wakefulness and exploration via global brain projections and (iv) are glucose-inhibited. MCH neurones are (i) glucose-excited, (ii) innervate learning and reward centres to promote synaptic plasticity, learning and memory and (iii) are critical for learning associations useful for predictive control (e.g. using taste to predict nutrient value of food). This evidence is unified into a model for predictive glucose control. During associative learning, inputs from some glucose-excited neurones may promote connections between the 'fast' senses and reward circuits, constructing neural shortcuts for efficient action selection. In turn, glucose-inhibited neurones may engage locomotion/exploration and coordinate the required fuel supply. Feedback inhibition of the latter neurones by glucose would ensure that glucose fluxes they stimulate (from liver, into muscle) are balanced. Estimating nutrient challenges from indirect sensory cues may become more difficult when the cues become complex and variable (e.g. like human foods today). Consequent errors of predictive glucose control may contribute to obesity and diabetes.",
keywords = "Animals, Brain, Glucose, Humans, Hypothalamic Hormones, Intracellular Signaling Peptides and Proteins, Learning, Melanins, Neurons, Neuropeptides, Orexins, Pituitary Hormones, Journal Article, Research Support, Non-U.S. Gov't, Review",
author = "C Kosse and A Gonzalez and D Burdakov",
note = "{\circledC} 2014 The Authors. Acta Physiologica published by John Wiley & Sons Ltd on behalf of Scandinavian Physiological Society.",
year = "2015",
month = "1",
doi = "10.1111/apha.12360",
language = "English",
volume = "213",
pages = "7--18",
journal = "Acta Physiologica",
issn = "1748-1708",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Predictive models of glucose control

T2 - roles for glucose-sensing neurones

AU - Kosse, C

AU - Gonzalez, A

AU - Burdakov, D

N1 - © 2014 The Authors. Acta Physiologica published by John Wiley & Sons Ltd on behalf of Scandinavian Physiological Society.

PY - 2015/1

Y1 - 2015/1

N2 - The brain can be viewed as a sophisticated control module for stabilizing blood glucose. A review of classical behavioural evidence indicates that central circuits add predictive (feedforward/anticipatory) control to the reactive (feedback/compensatory) control by peripheral organs. The brain/cephalic control is constructed and engaged, via associative learning, by sensory cues predicting energy intake or expenditure (e.g. sight, smell, taste, sound). This allows rapidly measurable sensory information (rather than slowly generated internal feedback signals, e.g. digested nutrients) to control food selection, glucose supply for fight-or-flight responses or preparedness for digestion/absorption. Predictive control is therefore useful for preventing large glucose fluctuations. We review emerging roles in predictive control of two classes of widely projecting hypothalamic neurones, orexin/hypocretin (ORX) and melanin-concentrating hormone (MCH) cells. Evidence is cited that ORX neurones (i) are activated by sensory cues (e.g. taste, sound), (ii) drive hepatic production, and muscle uptake, of glucose, via sympathetic nerves, (iii) stimulate wakefulness and exploration via global brain projections and (iv) are glucose-inhibited. MCH neurones are (i) glucose-excited, (ii) innervate learning and reward centres to promote synaptic plasticity, learning and memory and (iii) are critical for learning associations useful for predictive control (e.g. using taste to predict nutrient value of food). This evidence is unified into a model for predictive glucose control. During associative learning, inputs from some glucose-excited neurones may promote connections between the 'fast' senses and reward circuits, constructing neural shortcuts for efficient action selection. In turn, glucose-inhibited neurones may engage locomotion/exploration and coordinate the required fuel supply. Feedback inhibition of the latter neurones by glucose would ensure that glucose fluxes they stimulate (from liver, into muscle) are balanced. Estimating nutrient challenges from indirect sensory cues may become more difficult when the cues become complex and variable (e.g. like human foods today). Consequent errors of predictive glucose control may contribute to obesity and diabetes.

AB - The brain can be viewed as a sophisticated control module for stabilizing blood glucose. A review of classical behavioural evidence indicates that central circuits add predictive (feedforward/anticipatory) control to the reactive (feedback/compensatory) control by peripheral organs. The brain/cephalic control is constructed and engaged, via associative learning, by sensory cues predicting energy intake or expenditure (e.g. sight, smell, taste, sound). This allows rapidly measurable sensory information (rather than slowly generated internal feedback signals, e.g. digested nutrients) to control food selection, glucose supply for fight-or-flight responses or preparedness for digestion/absorption. Predictive control is therefore useful for preventing large glucose fluctuations. We review emerging roles in predictive control of two classes of widely projecting hypothalamic neurones, orexin/hypocretin (ORX) and melanin-concentrating hormone (MCH) cells. Evidence is cited that ORX neurones (i) are activated by sensory cues (e.g. taste, sound), (ii) drive hepatic production, and muscle uptake, of glucose, via sympathetic nerves, (iii) stimulate wakefulness and exploration via global brain projections and (iv) are glucose-inhibited. MCH neurones are (i) glucose-excited, (ii) innervate learning and reward centres to promote synaptic plasticity, learning and memory and (iii) are critical for learning associations useful for predictive control (e.g. using taste to predict nutrient value of food). This evidence is unified into a model for predictive glucose control. During associative learning, inputs from some glucose-excited neurones may promote connections between the 'fast' senses and reward circuits, constructing neural shortcuts for efficient action selection. In turn, glucose-inhibited neurones may engage locomotion/exploration and coordinate the required fuel supply. Feedback inhibition of the latter neurones by glucose would ensure that glucose fluxes they stimulate (from liver, into muscle) are balanced. Estimating nutrient challenges from indirect sensory cues may become more difficult when the cues become complex and variable (e.g. like human foods today). Consequent errors of predictive glucose control may contribute to obesity and diabetes.

KW - Animals

KW - Brain

KW - Glucose

KW - Humans

KW - Hypothalamic Hormones

KW - Intracellular Signaling Peptides and Proteins

KW - Learning

KW - Melanins

KW - Neurons

KW - Neuropeptides

KW - Orexins

KW - Pituitary Hormones

KW - Journal Article

KW - Research Support, Non-U.S. Gov't

KW - Review

U2 - 10.1111/apha.12360

DO - 10.1111/apha.12360

M3 - Review article

VL - 213

SP - 7

EP - 18

JO - Acta Physiologica

JF - Acta Physiologica

SN - 1748-1708

IS - 1

ER -