Leveraging neural plasticity for the treatment of amblyopia

Abstract

Amblyopia is a form of visual cortical impairment that arises from abnormal visual experience early in life. Most often, amblyopia is a unilateral visual impairment that can develop as a result of strabismus, anisometropia, or a combination of these conditions that result in discordant binocular experience. Characterized by reduced visual acuity and impaired binocular function, amblyopia places a substantial burden on the developing child. Although frontline treatment with glasses and patching can improve visual acuity, residual amblyopia remains for most children. Newer binocular-based therapies can elicit rapid recovery of visual acuity and may also improve stereoacuity in some children. Nevertheless, for both treatment modalities full recovery is elusive, recurrence of amblyopia is common, and improvements are negligible when treatment is administered at older ages. Insights derived from animal models about the factors that govern neural plasticity have been leveraged to develop innovative treatments for amblyopia. These novel therapies exhibit efficacy to promote recovery, and some are effective even at ages when conventional treatments fail to yield benefit. Approaches for enhancing visual system plasticity and promoting recovery from amblyopia include altering the balance between excitatory and inhibitory mechanisms, reversing the accumulation of proteins that inhibit plasticity, and harnessing the principles of metaplasticity. Although these therapies have exhibited promising results in animal models, their safety and ability to remediate amblyopia need to be evaluated in humans.

Keywords: Amblyopia; Critical period; Excitatory/inhibitory balance; Metaplasticity; Neuroplasticity; Suppression.

Figure 1.

Summary of visual outcomes following standard-of-care amblyopia treatment with glasses and patching from randomized clinical trials conducted by the Pediatric Eye Disease Investigator Group.^,,,,,– Percent of children achieving good visual acuity outcomes vs age at enrollment is plotted. For control children aged 3– 6 years, mean visual acuity is 0.1 logMAR and 95% have visual acuity ≤0.2 logMAR; for control children aged 7–17 years, mean visual acuity is 0.0 logMAR and 95% have visual acuities ≤ 0.1 logMAR.

Figure 2.

Proportion of suppressive units found in V1 and V2 of control and amblyopic monkeys.^,,

Figure 3.

Visual acuity improvement by number of weeks of amblyopia treatment: patching (top)^,,,,,,, or dichoptic treatment (bottom).^,,,,,, Only studies in which the children were treated with refractive correction alone until visual acuity stability was noted are included.

Figure 4.

The decline in plasticity potential with age involves an accumulation of factors that stabilize neural circuitry and inhibit plasticity. The loss of plasticity capacity with age is coincident with the emergence of molecules that act to consolidate neural connections for a lifetime. One factor is the development of perineuronal nets (A), extracellular matrix chondroitin sulfate proteoglycan containing structures that ensheathe the soma and dendrites of V1 neurons, and are highly enriched around PV inhibitory interneurons. This feature of the visual system is thought to attenuate plasticity by acting as a physical barrier to structural plasticity, altering neuronal excitability, and by containing molecules that inhibit neurite outgrowth. The accumulation of perineuronal nets and other factors that limit plasticity in adulthood progresses in accordance with a profile inverse to that of the critical period (B).

Figure 5.

Modification of synaptic connections mediates the loss of function in the amblyopic eye. In normal cats, inputs from the right and left eye express equal characteristics in V1 (A, top panel). This is reflected physiologically by similar VEPs measured from V1 following separate stimulation of the right and left eyes with grating stimuli phase-reversing at two hertz (A, bottom). After a period of monocular deprivation initiated early in postnatal development, deprived geniculocortical afferents branch less frequently and have fewer and smaller synaptic terminals (B top panel). These anatomical modifications are accompanied by a significant reduction in VEP amplitude elicited by stimulation of the deprived eye relative to those measured from the fellow eye (B, bottom panel). Graphs plot the normalized VEP power across the right (purple) and left (blue) eyes for gratings of variable spatial frequency and a control grey screen. Also plotted is the non-visual baseline power that was measured at an offset from the fundamental stimulation frequency (open circles).

Figure 6.

Dark exposure promotes anatomical, physiological and behavioral recovery from the effects of early monocular deprivation in cats. The monocular deprivation-induced atrophy of neuron soma size within layers of the LGN serving the deprived eye is reversed and restored to normal following 8 days of total darkness immersion (A). Within V1, the physiologically measured shift in ocular dominance away from the deprived eye recovers and is indistinguishable from normal when monocularly deprived cats are subjected to darkness for 10 days (B). Following a period of early monocular deprivation, the substantial impairment to spatial vision produced in the deprived eye recovers when cats are exposed to darkness for 10 days and then are provided binocular vision (C).

Figure 7.

Depiction of the BCM sliding threshold model for synaptic plasticity. Synaptic strength is modified by the postsynaptic response to input activity (A). When pre-synaptic activity is correlated with strong postsynaptic responses, the connection between them strengthens (LTP). For instance, when inputs from the two eyes consistently fire at the same time and activate a target neuron in V1, the synapses between them will strengthen and the V1 neuron becomes binocular. However, when post-synaptic responses do not strongly correlate with pre-synaptic activity the connection between them weakens (LTD). Using the example above, if activity of the V1 neuron is consistently out of sync with the activity of one input, that synapse will be weakened and eventually lost, rendering a monocular neuron. To avoid saturation of either LTP or LTD, the threshold to induce synaptic modification is proposed to slide according to the history of synaptic or cellular activity (B). When activity levels are reduced, for instance, the modification threshold shifts to the left making it easier to induce LTP. This capacity to modify the plasticity threshold is referred to as metaplasticity. Recovery from amblyopia that results from therapies such as darkness, patching, or binocular (dichoptic) therapy may promote recovery by reducing the level of cortical activity and engaging the mechanisms of metaplasticity.