Lateralized Expression of Cortical Perineuronal Nets during Maternal Experience is Dependent on MECP2

Surrogate Mecp2^WT (Wild type) mice exhibit hemisphere specific increase in PNN density in the primary somatosensory cortex (SS1). (A-C) Representative epifluorescent images of PNN expression in SS1 of naïve WT (NW) (A) as well as left (L) (B) and right (R) (C) hemispheres of surrogate WT (SW). Layers 1 through 6 are outlined. CP = caudate putamen and cc = corpus callosum. Arrows indicate examples of high-intensity PNNs analyzed for the study. Inset in A shows magnified PNN structures in the box. (D) Combined hemisphere analysis of the density of high-intensity PNNs was not significantly different between NW and SW (NW: n = 123 images; SW: n = 162; Mann-Whitney test, p > 0.05). (E) Separate analysis of left and right hemisphere revealed a significant increase of high-intensity PNNs in the right hemisphere of SW (SW-R; n = 81 images) compared to the left hemisphere of SW (SW-L; n = 81 images) (Kruskal-Wallis followed by Dunn’s test, *p < 0.05), while no significant difference was observed between hemispheres of NW (NW-L: n = 59 images; NW-R: n = 64 images; Mann-Whitney test, p > 0.05). (F,G) Size of SS1 was overall significantly larger in SW compared to NW (F) (NW: n = 123 images; SW: n = 162 images; Mann-Whitney test, *p < 0.05). Enlargement of SS1 in SW occurred in the left hemisphere (G) (NW-L: n = 59 images, NW-R: n = 64 images; SW-L: n = 81 images; SW-R: n = 81 images; Kruskal-Wallis followed by Dunn’s test, **p < 0.01). For D-G, n.s. = not significant. Different colors represent each of the five cohorts. Each open circle represents PNN density in an individual brain section.

Dynamic changes in PNN density across medial-lateral axis of both naïve and surrogate WT SS1. (A) Distribution of combined PNN density from left and right hemispheres of SS1 revealed an increase of PNN density in lateral sections compared to the medial sections of the same condition. Statistical analysis revealed significant differences between the combined two most medial regions and the combined two most lateral regions within conditions (grey p value denotes NW: n = 22 images for medial, 28 images for lateral; black p value denotes SW: n = 37 images for medial, 37 images for lateral). This medial-lateral comparison analysis method also applies to all subsequent figure panels. No significant difference in PNN density was found between conditions, represented by the light blue lines. Statistical analysis of the most medial region was a combined of 4 map numbers (NW: n = 40 images; SW: n = 62 images), middle region was a combined of 4 map numbers (NW: n = 32 images; SW: n = 46 images) and most lateral region was a combined of 5 map numbers (NW: n = 52 images; SW: n = 56 images). This sub-regional comparison method also applies to all subsequent figure panels. N = 5-24 images per map number. (B) Analysis of the right hemisphere in NW revealed that lateral sections had significantly higher density than the medial sections (NW-R, black p value; n = 14 images for medial, 15 images for lateral). In the left hemisphere, NW-L did not show any significant difference in medial-lateral axis (grey p value; n = 8 images for medial, 13 images for lateral). Across the subregions in the medial-lateral axis, there was no significant difference between the hemispheres in NW (blue lines; medial-NW: n = 18 images for L, 22 images for R; middle-NW: n = 15 images for L, 18 images for R; lateral-NW: n = 26 images per hemisphere). (C) A different pattern of hemisphere-specific differences was observed in SW. SW-R did not have significant differences between medial and lateral sections, while SW-L did (grey p value denotes SW-L: n = 20 images for medial, 17 images for lateral; black p value denotes SW-R: n = 17 images for medial, 20 images for lateral). While PNN density did not differ between SW-L and SW-R in middle and lateral regions, SW-R exhibited significantly higher PNN density in the medial regions than SW-L (medial-SW: n = 28 images for L, 34 images for R; middle-SW: n = 24 images for L, 22 images for R; lateral-SW: n = 30 images for L, 26 images for R). For A-C, lines represent the mean values. Each dot represents PNN density in an individual section. Mann-Whitney test: *p < 0.05, n.s. = not significant, 5 mice per condition. For B and C, n = 1–13 images per map number.

PNN density varies across subregions of WT SS1. (A–C) Analysis of both hemispheres for SS1 barrel field (S1BF, A), dysgranular zone (S1DZ, B) and forelimb (S1FL, C) revealed no significant differences between NW and SW (S1BF: NW – n = 131 images, SW - n = 168 images; S1DZ: NW − n = 22 images, SW – n = 35 images; S1FL: NW − n = 35 images, SW − n = 46 images; 5 mice per condition; Mann-Whitney test, p > 0.05, n.s. = not significant). (A′–C′) Analysis of sub regional SS1 by hemispheres revealed dynamic changes in PNN expression. In S1BF (A′), a significant increase of PNN expression was detected in the right hemisphere of SW (SW-R) compared to NW (NW-R) (NW-L: n = 65 images; NW-R: n = 66 images; SW-L: n = 85 images; SW-R: n = 83 images). In S1DZ (B′), while no significant difference was observed in PNN density between hemispheres in NW, there was a significant increase of PNN density in the right hemisphere (SW-R) compared to the left (SW-L) (NW-L: n = 8 images; NW-R: n = 14 images; SW-L: n = 18 images; SW-R: n = 17 images). A similar pattern of PNN plasticity was also detected in S1FL (C′), where PNN density was significantly increased in the right hemisphere than the left of SW. Interestingly, SW-L exhibited significantly lower PNN density compared to NW-L (Figure 4C′; NW-L: n = 13 images; NW-R: n = 22 images; SW-L: n = 21 images; SW-R: n = 25 images). For A′-C′, 5 mice per condition; Kruskal-Wallis followed by Dunn’s test, *p < 0.05, ***p < 0.001, n.s. = not significant.

Hemisphere-specific PNN density changes are not conserved in Mecp2^Het after maternal behavior. (A) Combined hemispheric analysis of PNN density of SS1 did not reveal significant changes between naïve Het (NH, n = 123 images) and surrogate Het (SH, n = 162 images) (Mann-Whitney test, p > 0.05). (B) Analysis of PNN density between hemispheres of SS1 revealed no significant difference between conditions or within hemispheres of naïve or surrogate Het (n = 54–82 images; Kruskal-Wallis followed by Dunn’s test, p > 0.05). (C) Analysis of SS1 area in both hemispheres revealed no significant changes after maternal learning (NH: n = 123 images; SH: n = 162 images; Mann-Whitney test, p > 0.05). (D) Area analysis by hemispheres reveal that left hemisphere of SH (SH-L) was significantly larger than the right hemisphere of SH (SH-R). This hemispheric area bias was absent in NH (n = 54–82 images; Kruskal-Wallis followed by Dunn’s test, **p < 0.01). n.s. = not significant. 5 mice per condition.

Overall patterns of PNN density across medial-lateral axis is preserved in Mecp2^Het. (A) Similar to NW-SW in Figure 3, NH and SH displayed significant increase of PNN density in lateral sections compared to the medial sections of the same condition (grey p value denotes NH: n = 23 images for medial, 27 images for lateral; black p value denotes SH: n = 36 images for medial, 25 images for lateral). Along the sub regions encompassing the medial-lateral axis, there was no significant difference between NH and SH, represented by the light blue lines (medial: n = 42 images for NH, 59 images for SH; middle: n = 37 images for NH, 55 images for SH; lateral: n = 44 images for NH, 48 images for SH). N = 5–22 images per map coordinate. (B) Within hemispheres of NH, statistical analysis of the lateral sections of the right (NH-R, black p value) and the left (NH-L, grey p value) showed significantly higher PNN density than their medial sections (NH-L: n = 9 images for medial, 16 images for lateral; NH-R: n = 14 images for medial, 11 images for lateral). There was no significant difference between hemispheres along the medial-lateral axis (medial: n = 23 images for L, 19 images for R; middle: n = 17 images for L, 20 images for R; lateral: n = 29 images for L, 15 images for R). (C) A different pattern of hemisphere-specific differences was observed in SH. SH-R did not display significant difference between medial and lateral sections, while SH-L did (grey p value denotes SH-L: n = 9 images in lateral, n = 20 images for medial; black p value denotes SH-R: n = 16 images for both lateral and medial). Statistical comparison between hemispheres revealed dynamic differences across medial-lateral axis, with SH-R had significantly higher PNN density than SH-L in medial and middle regions, while SH-L had significantly higher PNN density in the lateral regions (light blue lines; medial: n = 29 images for L, 30 images for R; middle: n = 31 images for L, 24 images for R; lateral: n = 20 images for L, 28 images for R). For A–C, lines represent mean values. Each dot represents PNN density in an individual section. Mann-Whitney test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n.s. = not significant, 5 mice per condition. For B–C, n = 1–12 images per map coordinate.

Left hemispheres of naïve and surrogate Mecp2^Het have higher PNN density in the most lateral sections, compared to the wildtype counterparts. (A) Combined hemispheres of NH had significantly higher PNN density than NW, in the middle and lateral sections. No significant difference was observed in the most medial sections (medial: n = 40 images for NW, 42 images for NH; middle: n = 32 images for NW, 37 images for NH; lateral: n = 52 images for NW, 44 images for NH). (B) Comparing left hemisphere PNN density between NW and NH, NH had significantly higher PNN density in the most lateral sections, with no significant differences in the more medial sections (medial: n = 18 images for NW-L, 23 images for NH-L; middle: n = 15 images for NW-L, 17 images for NH-L; lateral: n = 26 images for NW-L, 29 images for NH-L). (C) There were no significant differences in the right hemisphere between NW and NH, suggesting the PNN expression in the subregions of the lateral sections of the left hemisphere are particularly dysregulated in NH (medial: n = 22 images for NW-R, 19 images for NH-R; middle: n = 18 images for NW-R, 20 images for NH-R; lateral: n = 26 images for NW-R, 15 images for NH-R). (D) SH exhibited higher PNN density over SW only in the most lateral regions (medial: n = 62 images for SW, 59 images for SH; middle: n = 46 images for SW, 55 images for SH; lateral: n = 56 images for SW, 48 images for SH). (E) Similar to naïve conditions, only the lateral sections in left hemisphere of SH had significantly higher PNN density than the same regions of SW (medial: n = 28 images for SW-L, 29 images for SH-L; middle: n = 24 images for SW-L, 31 images for SH-L; lateral: n = 30 images for SW-L, 20 images for SH-L). (F) No significant differences between SW and SH were found in the right hemisphere (medial: n = 34 images for SW-R, 30 images for SH-R; middle: n = 22 images for SW-R, 24 images for SH-R; lateral: n = 26 images for SW-R, 28 images for SH-R). For A-F, lines represent the mean values. Each dot represents PNN density in an individual section. Mann-Whitney test, *p < 0.05, **p < 0.01, ****p < 0.0001, n.s. = not significant. 5 mice per condition. For A and D, n = 5-24 images per map coordinate. For B–C and E–F, n = 1–13 images per map coordinate.

Significant difference in S1 PNN density between SH and SW in Table 1 due to differential distribution. Histogram analysis showed that SW had more occurrences of 0 PNN density compared to SH, as shown by black dot (SW) above grey dot (SH). SH had more occurrences of 20–80 PNN density than SW, as shown by grey dots (SH) above black dots (SW). The average PNN density values of SW (n = 160 images) and SH (n = 152 images) are similar (Table 1) (5 mice per condition), but they are statistically different due to this differential distribution of data.

Principal component analysis of PNN expression segregates by conditions, lateral/medial axis and hemispheres. (A) The projection of each individual brain onto PC1 and 2. These PCs are defined by relative weights of the different brain regions – all of the weights for PC1 and 2 are shown in B. (Left) Individuals are colored by cohort with symbol shapes corresponding to their genotype and experience condition. (Right) Individuals are colored by K-means clustering assignment, showing that the primary separation is SH from the rest of the conditions. While PC1 tends to distinguish surrogates from naïve (in particular, surrogate het from all others), the PCA itself is an unbiased analysis that is not designed to calculate the differences in “NH vs. SH”. Rather, the PC1 represents the combination of factors that are the largest source of variation among all the samples. The PC1 itself is fully defined by the weights of the different brain sections. Inset shows the % variance explained by the different PCs, with PC1 explaining the most variance. (B) Weights for each brain region for principal component (PC) 1 (left panel) and 2 (right panel) from the analysis in (A). The map regions (corresponding lateral coordinates) with strongest positive and negative values contribute most strongly to the variation between individuals. The weights show a trend from medial to lateral, showing the differences in laterality are the largest source of variation in the data. (C) As in B, weights for each brain region for PC1 (left panel) and 2 (right panel) are shown, in this case for PCA on data in which all cohorts were averaged for each condition. (D) Conditions projected onto PC1 show a separation of NH from the rest while PC2 axis shows a separation of SH from SW.

Individual brains exhibit differential hemispheric bias in PNN density in specific subregions of SS1. (A–C) For each brain, PNN density from the left hemisphere was normalized to the right hemisphere. This left-right hemisphere normalization revealed varying patterns of cortical asymmetry. (A) In SS1, three NW brains exhibit left hemisphere bias of 2-fold. In SW, left hemisphere bias is seen in 2 brains (black, red). Two NH brains exhibit large fold differences favoring left hemisphere (red, blue circles). In SH, no brains display cortical asymmetry. Similar trends with larger fold differences are seen in S1BF (B) and S1ULp (C).

Hemispheric bias for MECP2 expression in individual NH brains. (A–E, A′–E′) Representative 20X magnification, tiled projection epifluorescent images showing MECP2 expression in left (A–E) and right (A′–E′) SS1 of NH from cohorts 1-5. R = rostral, C = caudal, D = dorsal and V = ventral. (A′′–E′′) Percentage cumulative frequency distribution of MECP2 intensity within left (grey) and right (black) SS1 of the corresponding NH cohorts. Cohort 1 (A′′) expressed more MECP2 in the right hemisphere than the left. Cohorts 2, 4–5 (B′′, D′′–E′′, respectively) expressed more MECP2 in the left hemisphere than the right, while cohort 3 (C′′) showed similar MECP2 expression in both hemispheres. A.U. = arbitrary intensity unit. N = 1 image per hemisphere.

Hemispheric bias for MECP2 expression in individual SH brains. (A–E, A′-E′) Representative 20X magnification, tiled projection epifluorescent images showing MECP2 expression in left (A–E) and right (A′-E′) SS1 of SH from cohorts 1-5. R = rostral, C = caudal, D = dorsal and V = ventral. (A′′-E′′) Percentage cumulative frequency distribution of MECP2 intensity within left (grey) and right (black) SS1 of the corresponding SH cohorts. Cohort 1 and 3 (A′′ and C′′, respectively) expressed more MECP2 in the right hemisphere than the left. Cohorts 2 and 5 (B′′ and E′′, respectively) expressed slightly more MECP2 in the left hemisphere than the right, while cohort 4 (D′′) showed similar MECP2 expression in both hemispheres. A.U. = arbitrary intensity unit. N = 1 image per hemisphere.

Hemispheric bias for MECP2 expression in individual NW brains. (A–E, A′–E′) Representative 20X magnification, tiled projection epifluorescent images showing MECP2 expression in left (A–E) and right (A′–E′) SS1 of NW from cohorts 1-5. R = rostral, C = caudal, D = dorsal and V = ventral. (A′′–E′′) Percentage cumulative frequency distribution of MECP2 intensity within left (grey) and right (black) SS1 of the corresponding NW cohorts. Cohort 1 and 4 (A′′ and D′′, respectively) expressed more MECP2 in the right hemisphere than the left. Cohorts 2, 3 and 5 (B′′, C′′ and E′′, respectively) expressed more MECP2 in the left hemisphere than the right. A.U. = arbitrary intensity unit. N = 1 image per hemisphere.

Hemispheric bias for MECP2 expression in individual SW brains. (A–E, A′–E′) Representative 20X magnification, tiled projection epifluorescent images showing MECP2 expression in left (A–E) and right (A′–E′) SS1 of SW from cohorts 1-5. R = rostral, C = caudal, D = dorsal and V = ventral. (A′′–E′′) Percentage cumulative frequency distribution of MECP2 intensity within left (grey) and right (black) SS1 of the corresponding SW cohorts. Cohort 1 and 2 (A′′ and B′′, respectively) expressed more MECP2 in the right hemisphere than the left. MECP2 expressions in Cohorts 3-5 (C′′-E′′) were similar between left and right hemispheres. A.U. = arbitrary intensity unit. N = 1 image per hemisphere.