Eye opening differentially modulates inhibitory synaptic transmission in the developing visual cortex

Abstract

Eye opening, a natural and timed event during animal development, influences cortical circuit assembly and maturation; yet, little is known about its precise effect on inhibitory synaptic connections. Here, we show that coinciding with eye opening, the strength of unitary inhibitory postsynaptic currents (uIPSCs) from somatostatin-expressing interneurons (Sst-INs) to nearby excitatory neurons, but not interneurons, sharply decreases in layer 2/3 of the mouse visual cortex. In contrast, the strength of uIPSCs from fast-spiking interneurons (FS-INs) to excitatory neurons significantly increases during eye opening. More importantly, these developmental changes can be prevented by dark rearing or binocular lid suture, and reproduced by the artificial opening of sutured lids. Mechanistically, this differential maturation of synaptic transmission is accompanied by a significant change in the postsynaptic quantal size. Together, our study reveals a differential regulation in GABAergic circuits in the cortex driven by eye opening may be crucial for cortical maturation and function.

Keywords: development; eye opening; fast-spiking interneurons; mouse; neuroscience; somatostatin-expressing interneurons; synaptic transmission.

Figure 1.. Development of synaptic transmission from Sst-INs onto PCs in layer 2/3 of the visual cortex.

(A) Schema of a quadruple whole-cell recording from an Sst-IN (red) and three PCs (blue) in layer 2/3. (B) Representative fluorescent (tdTomato, Sst-INs; Alexa 488, recorded neurons), IR-DIC, and merged images of a quadruple recording of an Sst-IN and three PCs. The dashed lines indicate the border between layer 1 and layer 2/3. Scale bar, 20 μm. (C) Left, representative traces showing synaptic transmission from an Sst-IN to two PCs recorded at P11. The red and blue lines indicate the averaged traces. Scale bars: 50 mV (vertical, red), 25 pA (vertical, blue), and 20 ms (horizontal). Right, morphological reconstruction of the Sst-IN. Scale bar: 80 µm. (D) Left, representative traces showing synaptic transmission from an Sst-IN to two PCs recorded at P15. Scale bars: 50 mV (vertical, red), 25 pA (vertical, blue), and 20 ms (horizontal). Right, morphological reconstruction of the Sst-IN. Scale bar: 80 µm. (E) The probability of synaptic connection from Sst-INs to PCs at P5–20. Data label indicates the number of pairs in each group. A total of 391 pairs were recorded from 82 mice. (F) Quantification of the peak amplitude of uIPSCs from Sst-INs to PCs at different postnatal ages. (G–H) Quantification of the 10–90% rise time (G) and half-width (H) of uIPSCs at P7–20. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 1—source data 1. Error bars indicate mean ±SEM. *p<0.05; **p<0.01; ***p<0.001; n.s., p>0.05.

Figure 1—figure supplement 1.. Morphological and electrophysiological properties of tdTomato⁺ neurons in Sst-tdTomato mice.

(A) Top panel, the morphologies of reconstructed non-FS tdTomato⁺ neurons in layer 2/3 of visual cortex. Bottom panel, corresponding traces of voltage responses to 500 ms current pulse step injections recorded in the current-clamp mode. (B) Some tdTomato⁺ cells in Sst-tdTomato line expressed PV. Arrowheads indicated PV⁺/tdTomato⁺ cells. (C) Top panel, the morphologies of reconstructed fast-spiking tdTomato⁺ cells. These cells are basket-like interneurons, with dense axonal arborization in layer 2/3. Bottom panel, corresponding membrane voltage responses of cells to current injections.

Figure 1—figure supplement 2.. The timing of eye opening in mice.

Quantification of mice undergoing eye opening in each litter (n = 10) from P13 to P15. There were no mice with opened eyes at P13. At P14, 77.9% ± 7.4% of mice had opened eyes; at P15, 100% of mice had opened eyes.

Figure 1—figure supplement 3.. Dendritic morphology of PCs does not change significantly before and after eye opening.

(A) The morphological identity of recorded PCs. Top panel, the morphologies of reconstructed PCs. Bottom panel, corresponding membrane responses of PCs to current injections. (B–G) Quantification of the end number of apical dendrites (B), the end number of basal dendrites (C), the node number of apical dendrites (D), the node number of basal dendrites (E), the total length of apical dendrites (F), and the total length of basal dendrites (G) between P12–13 (before eye opening) and P14–15 (after eye opening). (H–I) Sholl analysis of dendritic length per 40 μm radial unit distance from the soma. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 1—source data 1. n.s., p>0.05.

Figure 1—figure supplement 4.. Synaptic responses from Sst-INs to PCs recorded with a cesium-based intracellular solution.

(A) Representative traces showing synaptic transmission from Sst-INs to PCs recorded at P12 and P15, respectively. Postsynaptic responses were recorded with a cesium-based intracellular solution containing 60 mM Cl^-. (B) Summary of connection probability. (C) Quantification of the peak amplitude of Sst-IN→PC uIPSCs. Detailed statistical analysis, detailed data and exact sample numbers are presented in Figure 1—source data 1. *p<0.05; **p<0.01; ***p<0.001; n.s., p>0.05.

Figure 2.. Development of synaptic transmission from Sst-INs onto PCs in the prefrontal Cg1/2 area.

(A) Representative traces of synaptic transmission from an Sst-IN to a PC in layer 2/3 in the prefrontal Cg1/2 area at P13 and P14. Inset schema indicates paired patch recording of an Sst-IN and a PC. Scale bars: 50 mV (vertical, red), 20 pA (vertical, blue), and 20 ms (horizontal). (B) Histogram of the connection probability from P9 to P20. Data label indicates the number of pairs in each group. (C) Summary of the peak amplitude of uIPSCs from P9 to P20. Detailed statistical analysis, detailed data, and number of experiments are presented in Figure 2—source data 1. *p<0.05; **p<0.01; ***p<0.001; n.s., p>0.05.

Figure 3.. Strength of synaptic transmission from FS-INs onto PCs significantly increases in layer 2/3 of the visual cortex during eye opening.

(A) Fluorescent image of a visual coronal section from Sst-tdTomato::Lhx6-EGFP line. TdTomato, Sst-INs; EGFP, Lhx6-EGFP cells. Arrowheads indicate EGFP⁺/tdTomato^- cells. Scale bar, 100 μm. (B) Schema of a quadruple whole-cell recording from an FS-IN (EGFP⁺/tdTomato^-) and three PCs in layer 2/3. (C) Two examples of connection from an FS-IN to a PC at P13 and P15. Left panels, membrane potential responses of recorded FS-INs and PCs to current injections. Middle panels, synaptic transmission from FS-INs to PCs. Right panels, the reconstructed morphology of recorded FS-INs. (D) The connection probability from FS-INs to PCs did not change from P12 to P18. Data label indicates the number of pairs in each group. (E) The peak amplitude of FS-IN→PC uIPSCs at P14–15 and P16–18 was significantly larger than that at P12–13. (F–G) Quantification of the 10–90% rise time (F) and half-width (G) of uIPSCs at P12–18. Detailed statistical analysis, detailed data and number of experiments are presented in Figure 3—source data 1. *p<0.05; **p<0.01; n.s., p>0.05.

Figure 3—figure supplement 1.. Morphologies of reconstructed FS-INs and corresponding evoked membrane responses.

Top panel, the morphologic reconstructions of FS-INs. Bottom panel, the membrane responses of FS-INs to current injections.

Figure 3—figure supplement 2.. Synaptic transmission from FS-INs onto PCs increases in Cg1/2 area during eye opening.

(A) Two examples of connection from an FS-IN onto a PC at P13 and P15 in Cg1/2 area. (B) Summary of the connection probability from FS-INs onto PCs. (C) The peak amplitude of FS-IN→PC uIPSCs significantly increased during eye opening. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 3—source data 1. *p<0.05; **p<0.01.

Figure 4.. Development of synaptic transmission from Sst-INs to FS-INs and Htr3a-INs.

(A) Representative evoked responses from an Sst-IN to an FS-IN in layer 2/3 at P13 and P14, respectively. Inset panel, schematic of a paired recording from an Sst-IN and an FS-IN. Scale bars: 50 mV (vertical, red), 25 pA (vertical, green), and 20 ms (horizontal). (B) Summary of connection probability from Sst-INs to FS-INs at different postnatal ages. A total of 68 pairs were recorded from 11 mice. Data label indicates the number of pairs in each group. (C) The peak amplitude of uIPSCs from Sst-INs to FS-INs was unchanged from P12 to P18. (D) Fluorescent image of a visual coronal section from Sst-tdTomato::Htr3a-EGFP line. TdTomato, Sst-INs; EGFP, Htr3a-INs. Scale bar, 50 μm. (E) Schematic of a paired recording from an Sst-IN (red) and an Htr3a-IN (gray) in layer 2/3. (F) Traces showing representative synaptic transmission from an Sst-IN and Htr3a-IN at P11 and P15. Scale bars: 40 mV (vertical, red), 20 pA (vertical, black), and 20 ms (horizontal). (G) Summary of connection probability from Sst-INs to Htr3a-INs at different postnatal ages. A total of 126 pairs were recorded from 17 mice. Data label indicates the number of pairs in each group. (H) The peak amplitude of uIPSCs from Sst-INs to Htr3a-INs did not significantly change from P9 to P17. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 4—source data 1. Error bars indicate mean ±SEM. n.s., p>0.05.

Figure 4—figure supplement 1.. Synaptic transmission from FS-INs onto FS-INs does not change during eye opening.

(A) Representative traces of FS-IN→FS-IN uIPSCs at P13 and P15. (B) The probability of chemical connections between FS-INs at different postnatal ages. (C) Summarized results of the peak amplitude of FS-IN→FS-IN uIPSCs. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 4—source data 1. n.s., p>0.05.

Figure 5.. Differential inhibition by Sst-INs to various target cell types before and after eye opening.

(A) Schematic of a triple recording from an Sst-IN, a PC, and an FS-IN. (B) Representative evoked responses from an Sst-IN simultaneously onto a PC and an FS-IN at P13 and P15. Scale bars: 50 mV (vertical, red), 50 pA (vertical, black), and 5 ms (horizontal). (C) uIPSC amplitude evoked by Sst-INs was not significantly different between PCs and FS-INs at P12–13. (D) uIPSC amplitude evoked by Sst-INs was significantly smaller in PCs than in FS-INs at P14–15. (E) The logarithm of the ratio between uIPSC amplitude in FS-INs and PCs at P12–13 and P14–15. (F) Schematic of a triple recording from an Sst-IN a PC and an Htr3a-IN. (G) uIPSC amplitude evoked by Sst-INs was not significantly different between PCs and Htr3a-INs at P12–13. (H) uIPSC amplitude evoked by Sst-INs was significantly smaller in PCs than in Htr3a-INs at P14–15. (I) The logarithm of the ratio between uIPSC amplitude in Htr3a-INs and PCs at P12–13 and P14–15. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 5—source data 1. Error bars indicate mean ±SEM. *p<0.05; n.s., p>0.05.

Figure 6.. Eye opening modulates the strength of synaptic transmission from Sst-INs to PCs and from FS-INs to PCs.

(A) Schematic schedule of eyelid suturing, eyelid reopening, and recording of synaptic connection. (B) Summary of Sst-IN→PC connection probability in visual cortex (VC). (C) Quantification of the peak amplitude of Sst-IN→PC uIPSCs in VC. (D) Summary of Sst-IN→PC connection probability in Cg1/2. (E) Quantification of the peak amplitude of Sst-IN→PC uIPSCs in Cg1/2. (F) Summary of FS-IN→PC connection probability in VC. (G) Quantification of the peak amplitude of FS-IN→PC uIPSCs in VC. (H) Connection probability from FS-INs to PCs in the Cg1/2 area from continuously sutured mice and reopened mice. (I) Summary of the peak amplitude of FS-IN→PC uIPSCs in Cg1/2. Data label indicates the number of pairs in each group. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 6—source data 1. Error bars indicate mean ±SEM. *p<0.05; **p<0.01; n.s., p>0.05.

Figure 6—figure supplement 1.. Dark rearing prevents the developmental decrease of synaptic transmission from Sst-INs to PCs.

(A) Schematic schedule of dark rearing and electrophysiological recording. (B) Representative evoked responses from an Sst-IN to a PC at P12 and P15 in dark-reared mice. Insert, schematic of paired recording of an Sst-IN (red) and a PC (blue). (C) The occurrence of connection from Sst-INs to PCs in dark-reared mice was not significantly different between P12–13 and P14–15. (D) The peak amplitude of uIPSCs did not change significantly from P12–13 to P14–15 in dark-reared mice. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 6—source data 1. n.s., p>0.05.

Figure 6—figure supplement 2.. Representative traces of synaptic transmission (related to Figure 6).

Top panels indicate the direction of synaptic transmission.

Figure 6—figure supplement 3.. The strength of Sst-IN→FS-IN synaptic transmission does not change in continuously sutured mice during the time of eye opening.

(A) Schema of a paired recording from an Sst-IN to an FS-IN in continuously sutured mice at P12–15. (B) Representative traces of synaptic transmission from an Sst-IN to an FS-IN at P13 and P15 in continuously sutured mice. (C) Quantification of the connection probability. (D) Quantification of the uIPSC peak amplitude. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 6—source data 1. n.s., p>0.05.

Figure 6—figure supplement 4.. Suturing does not change the strength of FS-IN→FS-IN uIPSCs.

(A) Schematic of paired recording of two FS-INs from sutured mice. (B) Representative traces of synaptic connections between two FS-INs at P13 and P15. (C) The connection probability between FS-INs at P12-13 and P14-15 from sutured mice. (D) Summarized results of averaged amplitude of FS-IN→FS-IN uIPSCs from sutured mice. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 6—source data 1. n.s., p>0.05.

Figure 7.. Postsynaptic mechanisms underlying the changes of synaptic transmission from Sst-INs to PCs and from FS-INs to PCs.

(A) Amplitude-scaled overlay of paired-pulse ratio (PPR) responses in Sst-IN→PC connections at P13 and P15. Red, P13; blue, P15. Scale bars: 20 pA (vertical red), 10 pA (vertical, blue), 50 mV (vertical, black), and 20 ms (horizontal). Four presynaptic action potentials were evoked at 20 Hz. (B) The normalized peak amplitude of Sst-IN→PC uIPSCs showed short-term depression, and no significant difference in PPR was found between P12–13 (red) and P14–15 (blue) mice. (C) The coefficient of variation (C.V.) in Sst-IN→PC connections did not change from P12–13 to P14–15. (D) The failure rate in Sst-IN→PC connections did not change from P12–13 to P14–15. (E) Representative uIPSC responses from an Sst-IN to a PC evoked by a train of 3 presynaptic action potentials at 20 Hz under three different external Ca²⁺/Mg²⁺ concentrations. The postsynaptic cells were recorded with Cs-based and high Cl^- intracellular solution. Scale bars: 100 pA (vertical) and 20 ms (horizontal). (F) The parabola plot of the variance and mean of the peak amplitude of Sst-IN→PC uIPSCs in (E) under three different external Ca²⁺/Mg²⁺ concentrations. Red dots, 1 mM Ca²⁺/3 mM Mg²⁺; green dots, 2 mM Ca²⁺/2 mM Mg²⁺; blue dots, 3.7 mM Ca²⁺/0.3 mM Mg²⁺. (G) The number of release sites in Sst-IN→PC connections did not change from P12–13 to P14–15. (H) The quantal size in Sst-IN→PC connections significantly decreased from P12–13 to P14–15. (I) Amplitude-scaled overlay of paired-pulse ratio (PPR) responses in FS-IN→PC connections at P13 and P15. Red, P13; blue, P15. Scale bars: 100 pA (vertical red and blue), 50 mV (vertical, black), and 20 ms (horizontal). (J) PPR in FS-IN→PC connections was similar between P12–13 (red) and P14–15 (blue) mice. (K) The coefficient of variation (C.V.) in FS-IN→PC connections was unchanged from P12–13 to P14–15. (L) The failure rate in FS-IN→PC connections did not change from P12–13 to P14–15. (M) Representative uIPSC responses from an FS-IN to a PC evoked by a train of 3 presynaptic action potentials at 20 Hz in different external Ca²⁺/Mg²⁺ concentrations. Scale bars: 200 pA (vertical) and 20 ms (horizontal). (N) The parabola plot of the variance and mean of uIPSC amplitude in (M) at different external Ca²⁺/Mg2⁺ concentrations. Red dots, 1 mM Ca²⁺/3 mM Mg²⁺; green dots, 2 mM Ca²⁺/2 mM Mg²⁺; blue dots, 3.7 mM Ca²⁺/0.3 mM Mg²⁺. (O) The number of release sites in FS-IN→PC connections did not change from P12–13 to P14–15. (P) The quantal size in FS-IN→PC connections significantly increased from P12–13 to P14–15. Detailed statistical analysis, detailed data, and exact sample numbers are presented in Figure 7—source data 1. Error bars indicate mean ±SEM. *p<0.05; ***p<0.001; n.s., p>0.05.

Figure 8.. A model depicting how eye opening differentially regulates inhibitory synaptic transmission in developing layer 2/3 of the visual cortex.

The development of inhibitory synaptic transmission before (A) and after eye opening (B). Sst-INs (red) primarily innervate distal dendrites of PCs (blue), while FS-INs (green) mainly target and inhibit somatic and perisomatic regions of PCs. Eye opening decreases the inhibition from Sst-INs to PCs. Decreased Sst-IN→PC inhibition after eye opening is predicted to enhance the effect of visual input (stronger signal input depicted by the thicker black arrow) onto excitatory neurons in the visual cortex by facilitating dendritic events. In contrast, synaptic inputs from FS-INs to PCs increase after eye opening. The increase of FS-IN→PC inhibition is speculated to involve in a homeostatic rebalancing of inhibition. The inhibition from Sst-INs to FS-INs remains unchanged during the time of eye opening.