Multiple Roles for Nogo Receptor 1 in Visual System Plasticity

Abstract

During the developmental critical period for visual plasticity, discordant vision alters the responsiveness of neurons in visual cortex. The subsequent closure of the critical period not only consolidates neural function but also limits recovery of acuity from preceding abnormal visual experience. Despite species-specific differences in circuitry of the visual system, these characteristics are conserved. The nogo-66 receptor 1 (ngr1) is one of only a small number of genes identified thus far that is essential to closing the critical period. Mice lacking a functional ngr1 gene retain developmental visual plasticity as adults and their visual acuity spontaneously improves after prolonged visual deprivation. Experiments employing conditional mouse genetics have revealed that ngr1 restricts plasticity within distinct circuits for ocular dominance and visual acuity. However, the mechanisms by which NgR1 limits plasticity have not been elucidated, in part because the subcellular localization and signal transduction of the protein are only partially understood. Here we explore potential mechanisms for NgR1 function in relation to manipulations that reactivate visual plasticity in adults and propose lines of investigation to address relevant gaps in knowledge.

Keywords: amblyopia; critical period; dendritic spines; mouse; myelin; ocular dominance; visual acuity; visual cortex; visual plasticity.

Figure 1.

Aspects of visual system circuitry are conserved among mammals. Axons from projection neurons in the retina with receptive fields within the binocular zone (green) converge on the optic chiasm (blue and yellow lines) and then target the lateral geniculate nucleus (LGN) in each hemisphere. In mouse, the majority of axons cross the chiasm and the ipsilateral projection (blue dashed line) is smaller than in predatory mammals such as primates and cats. Within the LGN, axons from each eye innervate distinct domains. These domains comprise six layers in primate and three layers in cat, whereas the ipsilateral eye targets a smaller patch nestled within a larger domain of the contralateral eye in mouse. The axons from the neurons in LGN then project to primary visual cortex (V1).

Figure 2.

Aspects of visual system plasticity are conserved among mammals. (A) Monocular deprivation (MD) during the critical period disrupts eye dominance. Ocular dominance (OD) histograms plot the distribution of relative responsiveness of neurons in V1 to a visual stimulus presented independently to each eye. Neurons increasingly more responsive to the contralateral eye are categorized with lower numbers (3, 2, 1) while those with increasing preference for the ipsilateral eye are binned into higher number categories (5, 6, 7). Neurons with equal responsiveness to each eye are categorized as “4.” In primates and cats with normal vision (green bars and eye symbols below) this distribution is binocular. In mouse, normal vision is biased to the contralateral eye. Closing the contralateral eye for as briefly as a few days (purple bars and eye symbols below) shifts eye dominance toward the nondeprived ipsilateral eye. This plasticity is conserved between species although the magnitude of the OD shift varies. (B) Visual acuity increases during the critical period (green squares) and closing one eye during this maturation permanently impairs visual acuity. The resulting acuity following MD is similar to the acuity at the age of deprivation (black square), although these results are more variable in primate studies. In mice, deprivation for the duration of the critical period (long-term deprivation, LTMD) is required to impair acuity. (C) One model for how MD impairs visual performance in mammals. Discordant vision, such as deprivation or strabismus, first exaggerates eye dominance as in (A), diminishing responsiveness to the affected eye in visual cortex. This limited representation of the affected eye prevents the normal maturation of visual circuits subserving performance, such as acuity as in (B). After the critical period, these mechanisms of plasticity are no longer accessible. “Reactivating” developmental visual plasticity otherwise confined to the critical period is one strategy for rectifying eye dominance and potentially improving vision through the affected eye.

Figure 3.

The ngr1 functions in neurons within distinct circuits to limit ocular dominance (OD) plasticity and improvement in acuity. (A) Mice lacking ngr1 constitutively (ngr1−/−), or selectively in PV interneurons (ngr1 f/f;PV-Cre), retain developmental visual plasticity as adults. During the critical period, 4-days of MD (purple bars and eye symbols below) shifts the distribution of neuronal eye dominance (green bars and eye symbols below). This is not observed in adult WT mice. In contrast, in adult ngr1−/− mice or ngr1 f/f;PV-Cre mice, 4 days of MD continues to shift ocular dominance. (B) Adult ngr1−/− mice spontaneously recover visual acuity over 7 weeks following long-term deprivation (LTMD), but WT and ngr1 f/f;PV-Cre mice do not. (C) A comparison of the facets of visual plasticity present in different genotypes of mice. This genetic dissection of the expression of ngr1 reveals that OD plasticity is not sufficient to improve visual acuity.

Figure 4.

Several disparate extracellular ligands bind NgR1. Myelin-associated glycoprotein (MAG), oligodendrocyte-myelin glycoprotein (OMgp), and the Nogo-66 region of Nogo-A are ligands for NgR1. These proteins each bind the leucine-rich repeat (LRR) domain of NgR1. Several members of the family of chondroitin sulfate proteogylcans (CSPGs) also bind to NgR1. The sugar chains on these molecules interact with the stalk region of the receptor. As NgR1 is attached to the plasma membrane by a lipid anchor, NgR1 is proposed to transduce a signal from these ligands through one or more transmembrane “co-receptors” such as Lingo, TROY and p75, to activate the small GTPase RhoA. How this signal may limit anatomical plasticity and/or synaptic plasticity remains unclear.

Figure 5.

The ngr1 gene does not determine the set point for synaptic turnover in cerebral cortex. (A) One approach for investigating how specific genes may influence synaptic structural plasticity is in vivo 2-photon laser scanning microscopy (2plsm) through a cranial window, a small coverslip replacing a region of overlying skull. This approach permits repeated imaging of dendrites present in the most superficial layers of cortex, L1 and L2/3, in mice expressing green fluorescent protein (GFP+) in a sparse subset neurons. In this schematic, two GFP+ pyramidal neurons in L5 are shown with their apical dendrite branching in L1. The grey overlay emphasizes that only dendrites present in L1 and L2/3 are visible with this approach. (B) An example of apical dendrites in L1 from two pyramidal neurons. The yellow rectangle is the location of the dendrites presented in panel (C). At left, a segment of dendrite imaged four days apart. At right, a schematic of the locations of spines along the dendrite, including spines lost (red arrowheads) and spines gained (yellow arrowheads). (D) At left, a segment of axon imaged four days apart. At right, a schematic of the positions of boutons along the axon, including a bouton lost (red arrowhead). (E) The turnover ratio for dendritic spines ([percent gained + percent lost]/2) for wild-type (WT) and ngr1−/− mice across three consecutive imaging intervals as well as the average ± standard error of the mean (SEM). (F) The turnover ratio for axonal boutons ([percent gained + percent lost]/2) for WT and ngr1−/− mice across three consecutive imaging intervals as well as the average ± SEM.

Figure 6.

Potential mechanisms of NgR1 function in visual plasticity. (A) Ngr1 may affect the relative strength of excitatory and inhibitory neurotransmission (E/I balance) in visual cortex. Several manipulations that enhance visual plasticity after the critical period also may increase the E/I ratio. A specified threshold of inhibition is required to open the critical period (gray region), whereas elevated inhibition is associated with the close of the critical period. In normal mice (wild-type [WT]), this threshold to open the critical period is achieved in the third postnatal week and the critical period then closes approximately 2 weeks later (black line). Manipulations that alter E/I balance also affect the timing of the critical period. Treatments that elevate inhibition precociously, such as transgenic expression of brain-derived neurotrophic factor (BDNF) or administration of benzodiazepines (such as diazepam), cross this threshold to critical period plasticity earlier and the critical period subsequently closes sooner (black dashed line). Rearing animals in complete darkness (DR) prevents the opening of the critical period and delays maturation of cortical inhibition (lower black dashed arrow). Likewise, the critical period does not open in gad65−/− mice (red line). Distinct environmental (environmental enrichment [EE] and dark environment [DE]) and pharmacologic (3-mercaptopropionic acid [MPA] and fluoxetine) approaches for decreasing cortical inhibition also enhance adult visual plasticity (upper black dashed arrow). Ngr1−/− mice also exhibit a modest increase in E/I balance relative to WT mice (blue line). (B) NgR1 may close the critical period by transducing inhibitory signals from extracellular ligands that emerge with cortical maturation. NgR1 is a receptor for both multiple inhibitor factors associated with myelin membranes and chondroitin sulfate proteoglycans (CSPGs). The distributions of both cortical myelination (green line) and CSPGs (red line) increase as the critical period closes. Representative images of the distribution of CSGPs and myelin near the opening (P20) and closing (P40) of the critical period are provided at right. CSPGs are labeled with a lectin, Wisteria floribunda agglutinin (WFA), while myelin is stained with an antibody directed against myelin basic protein (MBP).