Response variability in Attention-Deficit/Hyperactivity Disorder: a neuronal and glial energetics hypothesis

Abstract

Background: Current concepts of Attention-Deficit/Hyperactivity Disorder (ADHD) emphasize the role of higher-order cognitive functions and reinforcement processes attributed to structural and biochemical anomalies in cortical and limbic neural networks innervated by the monoamines, dopamine, noradrenaline and serotonin. However, these explanations do not account for the ubiquitous findings in ADHD of intra-individual performance variability, particularly on tasks that require continual responses to rapid, externally-paced stimuli. Nor do they consider attention as a temporal process dependent upon a continuous energy supply for efficient and consistent function. A consideration of this feature of intra-individual response variability, which is not unique to ADHD but is also found in other disorders, leads to a new perspective on the causes and potential remedies of specific aspects of ADHD.

The hypothesis: We propose that in ADHD, astrocyte function is insufficient, particularly in terms of its formation and supply of lactate. This insufficiency has implications both for performance and development: H1) In rapidly firing neurons there is deficient ATP production, slow restoration of ionic gradients across neuronal membranes and delayed neuronal firing; H2) In oligodendrocytes insufficient lactate supply impairs fatty acid synthesis and myelination of axons during development. These effects occur over vastly different time scales: those due to deficient ATP (H1) occur over milliseconds, whereas those due to deficient myelination (H2) occur over months and years. Collectively the neural outcomes of impaired astrocytic release of lactate manifest behaviourally as inefficient and inconsistent performance (variable response times across the lifespan, especially during activities that require sustained speeded responses and complex information processing).

Testing the hypothesis: Multi-level and multi-method approaches are required. These include: 1) Use of dynamic strategies to evaluate cognitive performance under conditions that vary in duration, complexity, speed, and reinforcement; 2) Use of sensitive neuroimaging techniques such as diffusion tensor imaging, magnetic resonance spectroscopy, electroencephalography or magnetoencephalopathy to quantify developmental changes in myelination in ADHD as a potential basis for the delayed maturation of brain function and coordination, and 3) Investigation of the prevalence of genetic markers for factors that regulate energy metabolism (lactate, glutamate, glucose transporters, glycogen synthase, glycogen phosphorylase, glycolytic enzymes), release of glutamate from synaptic terminals and glutamate-stimulated lactate production (SNAP25, glutamate receptors, adenosine receptors, neurexins, intracellular Ca2+), as well as astrocyte function (alpha1, alpha2 and beta-adrenoceptors, dopamine D1 receptors) and myelin synthesis (lactate transporter, Lingo-1, Quaking homolog, leukemia inhibitory factor, and Transferrin).

Implications of the hypothesis: The hypothesis extends existing theories of ADHD by proposing a physiological basis for specific aspects of the ADHD phenotype - namely frequent, transient and impairing fluctuations in functioning, particularly during performance of speeded, effortful tasks. The immediate effects of deficient ATP production and slow restoration of ionic gradients across membranes of rapidly firing neurons have implications for daily functioning: For individuals with ADHD, performance efficacy would be enhanced if repetitive and lengthy effortful tasks were segmented to reduce concurrent demands for speed and accuracy of response (introduction of breaks into lengthy/effortful activities such as examinations, motorway driving, assembly-line production). Also, variations in task or modality and the use of self- rather than system-paced schedules would be helpful. This would enable energetic demands to be distributed to alternate neural resources, and energy reserves to be re-established. Longer-term effects may manifest as reduction in regional brain volumes since brain areas with the highest energy demand will be most affected by a restricted energy supply and may be reduced in size. Novel forms of therapeutic agent and delivery system could be based on factors that regulate energy production and myelin synthesis. Since the phenomena and our proposed basis for it are not unique to ADHD but also manifests in other disorders, the implications of our hypotheses may be relevant to understanding and remediating these other conditions as well.

Figure 1

Means (open symbols) and standard deviations (filled symbols) of reaction times for a group of boys with ADHD (squares, n = 17) and age-matched control subjects (circles, n = 18) in a 4-choice reaction time task. The data are from Leth-Steensen et al. [52]. The lines through the data derive from a series latency mechanism described in the text.

Figure 2

A scheme illustrating a glutamatergic neuron (left) a glial cell (astrocyte) and a small blood vessel (right) and the major components contributing to hypotheses 1 and 2 (H1 and H2). Neural activity triggers release of the neurotransmitter glutamate that is taken up into the astrocyte (via GLAST and GLT-1 transporters), and stimulates the breakdown of glycogen, the uptake of glucose, and glycolysis, to produce lactate. Rapid neuronal firing is sustained by the energy provided by the astrocyte-neuron lactate shuttle. Energy demands are high during rapid (burst) and maintained rates of neuronal firing. H1: At times of increased neuronal demand, deficient lactate results in decreased neuronal conversion of lactate to acetyl CoA, decreased ATP formation, deficient ATPase function, delayed restoration of ion gradients, elevated extracellular K⁺, deficient Na⁺-dependent transport of glutamate into astrocytes that is required to drive glycolysis and lactate release by the astrocytes. The result is that situationally appropriate firing rates are achieved only episodically. Methylphenidate treatment results in an increase of the extracellular levels of the catecholamines, NA (and DA) that stimulate glycolysis and release of lactate from the astrocytes. This is followed by glycogen replenishment, thereby correcting the energy deficiency, and restoring appropriate firing rates. H2: A deficient supply of lactate for oligodendrocytes in the developing nervous system slows and reduces the synthesis of fatty acids required for the synthesis of myelin. Poorly myelinated axons would transmit action potentials more slowly, accounting for inefficient integration (coherence) between brain regions and for slow reaction times. A number of neurotransmitter receptors present on astrocytes are not illustrated (e.g. muscarinic, α₂, DA D3, D4, D5 and receptors for several neuropeptides).