Distinct Circuits for Recovery of Eye Dominance and Acuity in Murine Amblyopia

Abstract

Degrading vision by one eye during a developmental critical period yields enduring deficits in both eye dominance and visual acuity. A predominant model is that "reactivating" ocular dominance (OD) plasticity after the critical period is required to improve acuity in amblyopic adults. However, here we demonstrate that plasticity of eye dominance and acuity are independent and restricted by the nogo-66 receptor (ngr1) in distinct neuronal populations. Ngr1 mutant mice display greater excitatory synaptic input onto both inhibitory and excitatory neurons with restoration of normal vision. Deleting ngr1 in excitatory cortical neurons permits recovery of eye dominance but not acuity. Reciprocally, deleting ngr1 in thalamus is insufficient to rectify eye dominance but yields improvement of acuity to normal. Abolishing ngr1 expression in adult mice also promotes recovery of acuity. Together, these findings challenge the notion that mechanisms for OD plasticity contribute to the alterations in circuitry that restore acuity in amblyopia.

Keywords: amblyopia; eye dominance; leucine-rich repeat; plasticity; visual acuity.

Figure 1. Recovery normal eye dominance with 7 weeks of binocular vision following LTMD by ngr1(−/−) mice

(A) Schematic of the timeline for long-term monocular deprivation (LTMD) and period of binocular vision prior to electrophysiologic recording to assess ocular dominance (OD) in adult (P90-P120) ngr1(−/−) and WT mice. (B) CBI scores for non-deprived WT mice (n=8), and WT and ngr1(−/−) mice receiving LTMD (n=8, 11). Groups receiving LTMD are underlined. Each point represents the CBI for an individual animal, and the bars represent the average for each group with error bars for SEM. The gray box indicates the typical range of CBI scores for non-deprived mice. (C) Cumulative histograms of OD scores for groups reported in (B) (WT, 288 units; WT LTMD 248 units; ngr1(−/−) LTMD 336 units). (D) A comparison of recovery of eye dominance following increasing duration of binocular vision (BV) subsequent to LTMD. CBI scores for WT and ngr1(−/−) mice during LTMD and following 8 days of BV were previously published in [19] and are presented here for comparison to 7 weeks after re-opening the closed eye. See also Figure S1.

Figure 2. Recovery of eye dominance and acuity are preceded by elevated intracortical excitatory synaptic input onto L2/3 PV interneurons

(A) Parvalbumin-positive (PV) interneurons identified in both WT and ngr1(−/−) mice by tdTomato expression from a Cre-dependent reporter in combination with PV-Cre. Scale bar = 0.5mm (top), 50 m (bottom). (B) Example of a L2/3 PV interneuron recorded in binocular V1 in an acute slice; overlaid are 16 ×16 stimulation locations for laser-scanning photostimulation (LSPS) spanning pia to white matter. Scale bar = 250 m. (C) Representative traces from LSPS evoked EPSCs measured across 16 × 16 locations (75 m spacing) for a L2/3 PV interneuron. Direct somatic responses have been removed for clarity. A higher magnification trace is shown at right. (D) LSPS aggregate excitatory input maps pooled across PV interneurons for WT and ngr1(−/−) mice during LTMD or after 1d of binocular vision. Triangles indicate soma locations; n = number of cells is in parentheses for each group. Scale bar = 250 m. (E) Mean LSPS-evoked EPSC amplitude from neurons recorded in (D) binned into L2/3, L4, L5, and L6. The mean LSPS-evoked amplitudes are greater for both genotypes following 1d of vision, but the magnitude of the increase is more than double in ngr1(−/−) mice. Ngr1 limits increased excitatory synaptic input onto PV interneurons in visual cortex with restoration of binocular vision. See also Figure S2.

Figure 3. A sustained increase in excitatory synaptic input onto L4 pyramidal neurons precedes recovery from visual deprivation

(A) Representative traces of spontaneous activity in V1 from acute slices from WT and ngr1(−/−) mice during LTMD, and LTMD followed by 1 day (1d) and 4 weeks (4wks) of binocular vision. (B) The frequency of mEPSCs for WT LTMD (n=8), LTMD+1d (n=8) and LTMD+4wks (n=6) as well as ngr1(−/−) mice in these conditions (n=8, 8, and 6 respectively). While both WT and ngr1(−/−) mice display elevated frequency of mEPSCs with eye-opening, this increase is only sustained in ngr1(−/−) mice (C) Cumulative distributions of mEPSC amplitudes for WT and ngr1(−/−) mice during LTMD, and LTMD followed by 1d and 4wks of binocular vision. ngr1(−/−) exhibit larger synaptic events at 1d that are also evident at 4wks. See also Figure S3.

Figure 4. Recovery of eye dominance following LTMD is restricted in neocortex by ngr1

(A) Immunostaining for GFP in a coronal section from a ngr1(f/f) mouse. Positive staining for GFP indicates recombination of the ngr1(f/f) gene. The top right panel is an enlargement of V1, and the bottom right panel is an enlargement of thalamus from the same section as the wide field image on the left. Scale bars are equal to 225 m. (B)-(D) Immunostaining for GFP as in (A) for ngr1(f/f);CK-Cre, ngr1(f/f);Scnn1a-Cre, and ngr1(f/f);Olig-Cre mice. (E) CBI scores for ngr1(f/f) (flx) non-deprived mice (ND) (n=8), as well as flx (n=8), flx;CK-Cre (n=6), flx;Scnn1a-Cre (n=7), and flx;Olig-Cre (n=5) following LTMD and 6 weeks of binocular vision. Groups receiving long-term monocular deprivation (LTMD) are underlined. The grey box indicates the typical range of CBI values for non-deprived mice. Bars represent the average CBI score for each group with error bars for SEM. P-values for Kruskal-Wallis (KW) multiple comparison tests with Dunn’s correction to flx ND are presented above each column. (F) Cumulative histograms of ODI scores for flx ND (336 units), flx LTMD (326 units), flx;CK-Cre LTMD (240 units), flx;Scnn1a-Cre LTMD (296 units), flx;Olig-Cre LTMD (229 units). See also Figure S4 and S5.

Figure 5. Thalamus restricts recovery of visual acuity following LTMD

(A) Schematic of the timeline for long-term monocular deprivation (LTMD) and the period of binocular vision prior to the visual water task (VWT) to assess visual acuity in adult (P90) mice. The VWT proceeds in two phases: training and testing. (B) Performance of a representative ngr1(f/f) mouse that is non-deprived (black) or has received LTMD (grey). (C) Acuity of ND ngr1(f/f) mice (n=19) measured through one eye, and acuity following LTMD and 6 weeks of binocular vision for ngr1(f/f) (flx) (n=17), ngr1(f/f); CK-Cre (flx;CK-Cre) (n=10), ngr1(f/f); Scnn1a-Cre (flx;Scnn1a-Cre) (n=8), ngr1(f/f); Olig3-Cre (flx;Olig3-Cre) mice (n=7) measured through the previously deprived eye. Groups having received LTMD are underlined. See also Figure S6.

Figure 6. Visual acuity is restored following LTMD with deletion of ngr1 after the critical period

(A) Schematic of the timeline for tamoxifen (tmx) administration and LTMD prior to the period of binocular vision that precedes the visual water task (VWT) to assess visual acuity in adult (P90) mice. The VWT proceeds in two phases: training and testing. Following administration of tamoxifen (tmx), the global expression of NgR1 is nearly completely absent after 7 days. (B) Representative immunoblots confirm the recombination of ngr1 in ngr1(f/f); ER-Cre (flx;ER-Cre) mice following testing on the VWT. (C) Acuity following LTMD and 6 weeks of binocular vision for ngr1(f/f) (flx) (n=13), and ngr1(f/f);ER-Cre (flx;ER-Cre) (n=7) mice measured through the previously deprived eye. Groups having received LTMD and treatment with tamoxifen are underlined.