Comparing development of synaptic proteins in rat visual, somatosensory, and frontal cortex

Abstract

Two theories have influenced our understanding of cortical development: the integrated network theory, where synaptic development is coordinated across areas; and the cascade theory, where the cortex develops in a wave-like manner from sensory to non-sensory areas. These different views on cortical development raise challenges for current studies aimed at comparing detailed maturation of the connectome among cortical areas. We have taken a different approach to compare synaptic development in rat visual, somatosensory, and frontal cortex by measuring expression of pre-synaptic (synapsin and synaptophysin) proteins that regulate vesicle cycling, and post-synaptic density (PSD-95 and Gephyrin) proteins that anchor excitatory or inhibitory (E-I) receptors. We also compared development of the balances between the pairs of pre- or post-synaptic proteins, and the overall pre- to post-synaptic balance, to address functional maturation and emergence of the E-I balance. We found that development of the individual proteins and the post-synaptic index overlapped among the three cortical areas, but the pre-synaptic index matured later in frontal cortex. Finally, we applied a neuroinformatics approach using principal component analysis and found that three components captured development of the synaptic proteins. The first component accounted for 64% of the variance in protein expression and reflected total protein expression, which overlapped among the three cortical areas. The second component was gephyrin and the E-I balance, it emerged as sequential waves starting in somatosensory, then frontal, and finally visual cortex. The third component was the balance between pre- and post-synaptic proteins, and this followed a different developmental trajectory in somatosensory cortex. Together, these results give the most support to an integrated network of synaptic development, but also highlight more complex patterns of development that vary in timing and end point among the cortical areas.

Keywords: E-I balance; PSD-95; cortical development; critical period; gephyrin; integrated network; synapsin; synaptophysin.

FIGURE 1

Developmental changes in GAPDH in rat visual (red, A), somatosensory (green, B), and frontal (black, C) cortex. Gray dots are results from all runs, and color dots are the average for each animal. Example bands are shown above the graphs. Exponential decay functions were fit (solid lines), and 95% confidence intervals (dotted lines) were added, adult levels are defined as 3τ. GAPDH reached adult levels in visual (A) at P50 (3τ = P49.9 ± 30.4; curve-fit R = 0.50; p < 0.001), somato - sensory (B) at P57 (3τ = P57.1 ± 26.9; curve-fit R = 0.69; p < 0.0001), and frontal (C) at P53 (3τ = P52.6 ± 24.3; curve-fit R = 0.69; p < 0.0001).

FIGURE 2

Developmental changes in synapsin (A,D,G), synaptophysin (B,E,H), and the pre-synaptic index (C,F,I) in rat visual (red), somatosensory (green) and frontal (black) cortex. Gray dots are results from all runs, and color dots are the average for each animal. Example bands for each protein and area are shown above the graphs. Exponential decay functions were fit (solid lines), and 95% confidence intervals (dotted lines) were added, adult levels are defined a 3τ. Synapsin reached adult levels in visual (A) at P37 (3τ = P37.2 ± 12.6; curve-fit R = 0.70; p < 0.0001), somatosensory (D) at P19 (3τ = P19.0 ± 14.9; curve-fit R = 0.45; p < 0.005), and frontal (G) at P38 (3τ = P37.6 ± 13.5; curve-fit R = 0.70; p < 0.0001). Synaptophysin reached adult levels in visual (B) at P45 (3τ = P44.8 ± 15.1; curve-fit R = 0.72; p < 0.0001), somatosensory (E) at P35 (3τ = P34.7 ± 22.2; curve-fit R = 0.64; p < 0.0001), and frontal (H) at P45 (3τ = P45.4 ± 14.2; curve-fit R = 0.75; p < 0.0001). Pre-synaptic index reached adult levels in visual (C) at P24 (3τ = P23.5 ± 4.2; curve-fit R = 0.88; p < 0.0001), somatosensory (F) at P20 (3τ = P20.2 ± 6.0; curve-fit R = 0.82; p < 0.0001), and frontal (I) at P37 (3τ = P36.8 ± 7.8; curve-fit R = 0.85; p < 0.0001).

FIGURE 3

Developmental changes in PSD-95 (A,D,G), gephyrin (B,E,H), and the post-synaptic index (C,F,I) in rat visual (red), somatosensory (green) and frontal (black) cortex. Gray dots are results from all runs, and color dots are the average for each. Example bands for each protein and area are shown above the graphs. An exponential decay function was fit (solid line) to PSD-95, a sigmoid function (solid line) was fit to the post-synaptic index, and 95% confidence intervals (dotted lines) were added. Adult levels are defined as 3τ, or the age when 87.5% of the asymptotic expression level was reached. Peak function was fit to gephyrin expression. PSD-95 reached adult levels in visual (A) at P62 (3τ = 61.9 ± 12.3; curve-fit R = 0.89; p < 0.0001), somatosensory (D) at P61 (3τ = P61.221.4; curve-fit R = 0.82; p < 0.0001), and frontal (G) at P49 (3τ = P49.1 ± 11.8; curve-fit R = 0.83; p < 0.0001). Gephyrin expression reached a peak in visual (B) at P29 (Peak = 29.3 ± 5.1; curve-fit R = 0.61 p < 0.0001), somatosensory (E) at P13 (Peak = 12.8 ± 5.1; curve-fit R = 0.48 p < 0.005), and frontal (H) at P20 (Peak = 20.4 ± 1.1; curve-fit R = 0.60 p < 0.0001). Post-synaptic index reached adult levels in visual (C) at P25 ( curve-fit R = 0.87; p < 0.0001, Adult Level = P24.6 ± 7.0), somatosensory (F) at P21 (curve-fit R = 0.98; p < 0.0001; Adult Level = P21.4 ± 2.2), and frontal (I) at P30 (curve-fit R = 0.95; p < 0.0001; Adult Level = P30.2 ± 4.9).

FIGURE 4

Time lines for maturation of protein expression and pre- and post-synaptic indices in frontal (black), somatosensory (green) and visual cortex (red) showing the age when adult levels (3τ) or peak in expression were reached (bright bars) and the standard error (light bars) around the 3τs or peak (light bars) (A) Adult levels of synapsin expression overlapped among the three cortical areas (Visual, 3τ = P37.2 ± 12.6; somatosensory cortex, 3τ = P19.0 ± 14.9; Frontal, 3τ = P37.6 ± 13.5). Adult levels of synaptophysin expression overlapped among the three cortical areas (Visual, 3τ = P44.8 ± 15.1; somatosensory, 3τ = P34.7 ± 22.2; Frontal, 3τ = P45.4 ± 14.2). Adult levels of PSD-95 expression overlapped among the three cortical areas (Visual, 3τ = 61.9 ± 12.3; somatosensory, 3τ = 61.2 ± 21.4; Frontal, 3τ = 49.1 ± 11.8). Peak in gephyrin expression was significantly later in Visual (Peak = 29.3 ± 5.1), than somatosensory (Peak = 12.8 ± 5.1, p < 0.05) and Frontal (Peak = 20.4 ± 1.1, p < 0.05) cortex. (B) The pre-synaptic index reached adult levels earlier in the sensory areas (Visual, 3τ = P23.5 ± 4.2, p = 0.06; somatosensory, 3τ = P20.2 ± 6.0, p < 0.05) than in frontal cortex (3τ = P36.7 ± 7.8). The post-synaptic index reached adult levels earlier in the sensory areas (Visual, 3τ = 46.4 ± 9.5, p < 0.05; somatosensory, 3τ = 43.5 ± 6.1, p < 0.005) than in frontal cortex (3τ = 75.6 ± 11.2).

FIGURE 5

Principal component analysis. (A) The percent variance captured by each component of the SVD analysis of protein expression in rat visual, somatosensory, and frontal cortex. The first three principal components represent the significant portion (99%) of the SVD. (B–D) The influence of each protein on the three principal components was reflected by the relative amplitude in the basis vectors. (E) Significant correlations (colored cells) between the three principal components and the combinations of proteins derived from the basis vectors. The color indicates the magnitude (represented by color intensity) and direction (green indicates positive, red indicates negative) of significant correlations (Bonferroni corrected, p < 0.0024).

FIGURE 6

Developmental changes in the three principal components in visual (red dots, solid line), somatosensory (green squares, dashed lines), and frontal (black diamonds, dotted line) cortex. (A) Principal component 1. Exponential decay functions were fit to the data for each areas, adult levels are defined as 3τ. Principal component one reached adult levels in visual at P45 (3τ = 44.7 ± 18.3; R = 0.85, p < 0.0001), somatosensory at P50 (3τ = 49.6 ± 14.8; R = 0.94, p < 0.0001), and frontal at P37 (3τ = 36.5 ± 19.0; R = 0.78, p < 0.001). (B) Principal component 2. Peak functions were fit to the data, and the timing of the peak was determined. Principal component two reached a peak in visual at P28 (Peak = 28.1 ± 0.6; R = 0.55, p < 0.05), somatosensory at P9 (Peak = 9.0 ± 1.2; R = 0.86, p < 0.0001), and frontal at P16 (Peak = 16.3 ± 1.9; R0.87, p < 0.0001). (C) Principal component 3. Exponential decay functions were fit to the data for visual, and frontal cortex, and adult levels were defined as 3τ. A sigmoid function was fit to the data for somatosensory cortex, and the timing of the inflection point was determined. Principal component three reached adult levels in visual at P26 (3τ = 25.5 ± 16.8; R = 0.69, p < 0.01), and frontal at P53 (3τ = 52.5 ± 28.2; R = 0.78, p < 0.001). The inflection point in somatosensory cortex occurred at P26 (Inflection Point = 25.7 ± 2.1; R = 0.87, p < 0.0001).

FIGURE 7

Developmental changes in total protein expression (A) and the pre-synaptic:post-synaptic Index in visual (red dots, solid line), somatosensory (green squares, dashed lines), and frontal (black diamonds, dotted line) cortex. (A) Exponential decay functions were fit to the data for each area, adult levels were defined as 3τ. Total protein expression reached adult levels in visual at P42 (3τ = 42.0 ± 17.6; R = 0.84, p < 0.0001), in somatosensory at P42(3τ = 41.6 ± 13.8; R = 0.92, p < 0.0001), and in frontal cortex at P33 (3τ = 32.9 ± 17.8; R = 0.77, p < 0.001). (B) Exponential decay functions were fit to the pre-synaptic:post-synaptic index in visual and frontal cortex, adult levels were defined as 3τ. Pre-synaptic:post-synaptic index reached adult levels in visual at P25 (3τ = 25.0 ± 15.0; R = 0.64, p < 0.05), and in frontal cortex at P86 (3τ = 86.2 ± 35.0; R = 0.89, p < 0.0001). A weighted average curve was fit to somatosensory cortex.