Transient Hypothyroidism During Lactation Alters the Development of the Corpus Callosum in Rats. An in vivo Magnetic Resonance Image and Electron Microscopy Study

Abstract

Magnetic resonance imaging (MRI) data of children with late diagnosed congenital hypothyroidism and cognitive alterations such as abnormal verbal memory processing suggest altered telencephalic commissural connections. The corpus callosum (CC) is the major inter-hemispheric commissure that contra-laterally connects neocortical areas. However, in late diagnosed neonates with congenital hypothyroidism, the possible effect of early transient and chronic postnatal hypothyroidism still remains unknown. We have studied the development of the anterior, middle and posterior CC, using in vivo MRI and electron microscopy in hypothyroid and control male rats. Four groups of methimazole (MMI) treated rats were studied. One group, as a model for early transient hypothyroidism, was MMI-treated from postnatal day (P) 0 to P21; some of these rats were also treated with L-thyroxine (T4) from P15 to 21. Another group modeling chronic hypothyroid, were treated with MMI from P0 to 150 and from embryonic day 10 to P170. The results obtained from these groups were compared with same age control rats. The normalized T2 signal obtained using MRI was higher in MMI-treated rats and correlated with a low number and percentage of myelinated axons. The number and density of myelinated axons decreased in transient and chronic hypothyroid rats at P150. The g-ratio (inner to outer diameter ratio) and the estimated conduction velocity of myelinated axons were similar between MMI-treated and controls, but the conduction delay decreased in the posterior CC of MMI-treated rats compared to controls. These data show that early postnatal transient and chronic hypothyroidism alters CC maturation in a way that may affect the callosal transfer of information. These alterations cannot be reversed after delayed T4-treatment. Our data support the findings of neurocognitive delay in late T4-treated children with congenital hypothyroidism.

Keywords: attention deficit/hyperactivity disorder; autism; congenital hypothyroidism; iodine diet; neocortical development; psychiatric diseases; thyroid hormones.

FIGURE 1

Experimental groups and treatments. The bar chart shows the different treatments of the experimental groups studied. Chronic (MMI_E10 and MMI_P0) and transient (MMI_P0–21 and MMI_P0–21 + T4_P15–21) hypothyroid, and control (C) rats were treated (see color key) during their lifespan (white bars), time scale shown on the horizontal axis. The vertical lines within each lifespan indicate the age at which in vivo MRI scans were taken. All MMI pups were also treated with 1% KClO₄ up to P21, to additionally block thyroid function during fetal and lactating periods. Four pups (2 pups per litter per experimental group) were sacrificed at P150 for EM study. The last MRI scan was taken immediately prior to sacrifice at P150 in all groups except for C and MMI_E10 rats that were not processed for EM but scanned additionally at P170 (C and MMI rats) and P365 (C rats). Data and modified figure legend from Lucia et al. (2018).

FIGURE 2

MRI images of a C rat at P150. (A) Mid parasagittal T2w image showing the antero-posterior locations (white arrows) from which the anterior (B), middle (C) and posterior (D) coronal T2w images of were obtained. (B–D) Coronal images show the regions of interest (ROI) as the area between vertical white bars of the anterior, middle and posterior CC zones. Distances from Bregma are indicated in the lower-right corner, according to Paxinos et al. (2015). (E) Semithin toluidine blue-stained sections (1.5 μm thick) showing the structure of the CC at P150, and the main callosal regions: rostrum (RO), genu (GE), body (BO), and splenium (SP). Note that neither the isthmus nor the anterior and posterior zones of the body can be discerned, given that this region is mostly occupied by somatosensory axons in rodents. (F) Outlines show the anterior, middle and posterior zones comprising 30, 40, and 30 percent respectively of the total CC area. Dots give an approximate indication of where the EM micrographs and the T2w scans were taken (arrows in A). Horizontal bars (500 μm) at the top of the CC outline have the same length as the voxel z-thickness of T2w images of figures (B–D). Same scale for figures (A–D) and for (E,F).

FIGURE 3

Body weight and plasma concentration levels of thyroid hormone. (A) Changes in body weight with age for C and MMI rats. Note the arrested growth of chronic hypothyroid rats. (B,C) Bar charts show the total plasma concentrations of T4 (tT4) and T3 (tT3) at the ages indicated. TH plasma concentrations were recovered in transient hypothyroid rats at P50. Bars: mean ± SD. n.s., non-significant differences. Significant differences: *P < 0.001 (n = 8–11 rats per group). Data and figure legend from Lucia et al. (2018).

FIGURE 4

MRI images and T₂r of the anterior postnatal CC. (A) At all ages, T₂-w images of anterior CC (arrows and arrowheads) in MMI rats were less contrasted than in controls (arrowheads point to an undistinguishable anterior CC). The CC is hardly visible in MMI_P0 rats at P150 (arrow). Note the decreased contrast of CC in transient hypothyroid (MMI_P0–P21 + T4_P15–P21 and MMI_P0–P21) rats at P40. (B) Graph showing T₂r at postnatal ages. Bold symbols indicate that anterior CC was darker than the adjacent neuropil. In C rats, T₂r decreased rapidly from P8 to P40 and then more slowly. In MMI rats, T₂r values followed a similar trend but maintained higher values. (C) Bar charts show that T₂r was significantly higher in MMI than in C rats at all ages. At P40, differences between transient and MMI_P0 rats were not significant, but significant differences were seen between these groups and MMI_E10 and C rats (P < 0.001). At P75, significant T₂r differences (P < 0.001) were found between transient hypothyroid and both chronic (MMI_P0 and MMI_E10) and C rats. At P150, transient hypothyroid values were still significantly different to chronic hypothyroid (P < 0.001) and C (P < 0.05) rats. At all ages, significant differences between MMI_P0 and MMI_E10 (P < 0.001) rats were found. Bars: mean ± SD. n.s., non-significant differences. Significant differences: *P ≤ 0.05 and **P ≤ 0.001 (n = 8 rats per group). All figures at same scale.

FIGURE 5

MRI images and T₂r of the middle postnatal CC. (A) At all ages, T₂-w images of the middle (arrows and arrowheads) of MMI rats were less contrasted than in controls (arrowheads point to an undistinguishable middle CC). The CC is hardly visible in chronic hypothyroid (MMI_P0 and MMI_E10) rats at P40 and P60, respectively. Note the decreased contrast of the body in transient hypothyroid (MMI_P0–21 and MMI_P0–21 + T4_P15–21) rats at P40. (B) Graph showing T₂r at postnatal ages. Bold symbols indicate that anterior CC was darker than the adjacent neuropil. In C rats, T₂r decreased rapidly from P8 to P60 and then more slowly. In MMI rats, values followed a similar trend but maintained higher values. (C) Bar charts show that T₂r was significantly higher (P < 0.001) in chronic hypothyroid than in C rats at all ages. At P40, Differences between transient and MMI_P0 rats were not significant, but transient rats were significantly different (P < 0.001) to chronic hypothyroid and C rats. At P 75, differences (P < 0.001) were found between transient and chronic hypothyroid rats. The difference between MMI_P0–21 and C rats also significant (P < 0.05). At P150, differences between transient and C rats were not significant, however both transient and chronic hypothyroid were significantly different (P < 0.001) to C rats. At all ages, significant differences between MMI_P0 and MMI_E10 (P < 0.001) rats were found. Bars: mean ± SD. n.s., non-significant differences. Significant differences: *P ≤ 0.05 and **P ≤ 0.001 (n = 8 rats per group). All figures at same scale.

FIGURE 6

MRI images and T₂r of the posterior postnatal CC. (A) At all ages, T₂-w images of the posterior CC (arrows and arrowheads) of MMI rats were less contrasted than in controls (arrowheads point to an undistinguishable posterior CC). The CC is hardly visible in chronic hypothyroid (MMI_P0 and MMI_E10) rats at P150. Note the decreased contrast of the posterior CC in transient hypothyroid (MMI_P0–21 + T4_P15–P21 and MMI_P0–21) rats at P150. (B) Graph showing T₂r at postnatal ages. Bold symbols indicate that the posterior CC was darker than the adjacent neuropil. In C rats, T₂r decreased rapidly from P8 to P75 and then more slowly. In MMI rats, values followed a similar trend but maintained higher values. (C) Bar charts show that T₂r was significantly higher (P < 0.001) in chronic hypothyroid than in C rats at all ages. At P40, Differences between transient and MMI_P0 rats were not significant, but significantly different (P < 0.001) to chronic hypothyroid and C rats. At P75 and P150, significant differences (P < 0.001) were found between all groups of MMI treated rats and controls, except for a decrease in difference (P < 0.05) between untreated and T4-treated transient hypothyroid rats at P150. At all ages, significant differences between MMI_P0 and MMI_E10 (P < 0.001) rats were found. Bars: mean ± SD. n.s., non-significant differences. Significant differences: *P ≤ 0.05 and **P ≤ 0.001 (n = 8 rats per group). All figures at same scale.

FIGURE 7

EM photomicrographs of the CC ultrastructure in MMI and C rats at P150 in anterior (A,D,G,J,M), middle (B,E,H,K,N) and posterior (C,F,I,L,O) CC. Note the decreased posterior myelinated axon density, compared to anterior and middle CC in MMI and C rats. Compared to transient and C, the density of myelinated axons is significantly decreased in chronic MMI rats (J–O). No differences in the myelin thickness can be seen. All figures at same scale.

FIGURE 8

EM quantitative data of the CC at P150. (A–C) Bar charts show total, unmyelinated and myelinated axon number in the CC. Unmyelinated (D–F) and myelinated axon number (G–I), and myelinated axon percentage (J–L) in the anterior, middle and posterior CC are shown. (A) Differences in the number of MMI and C axons were not significant. (B) The difference in unmyelinated axon number between transient (MMI_P0–21 + T4_P15–P21 and MMI_P0–21) and MMI_P0 rats was not significant, transient and MMI_P0 numbers were significantly different (P < 0.05) to MMI_E0 chronic hypothyroid and C rats. (C) The number of myelinated axons of transient hypothyroid rats decreased significantly (P < 0.05) with respect to C rats and even more so in chronic hypothyroid rats (P < 0.001). (D,E) The unmyelinated axon number in anterior and middle CC in transient and MMI_P0 rats was not significantly different to C rats but increased significatively (P < 0.001) increased in MMI_E10 rats. (F) In the posterior CC, there was no significant difference in unmyelinated axon number between transient and hypothyroid rats which were both significantly higher (P < 0.001) compared to C rats. In contrast, myelinated axon number (G–I) and percentage (J–L) in the anterior, middle and posterior CC, decreased significantly in all MMI rats compared to controls, with the lowest values found in chronic hypothyroid rats. In middle CC (H,K), significant differences (P < 0.05) were found between untreated and T4-treated transient hypothyroid rats. Errors bars: SD. n.s., non-significant differences. Significant differences: *P ≤ 0.05 and **P ≤ 0.001 (n = 4 rats per group).

FIGURE 9

Regression functions between EM and T₂r values in the postnatal CC. (A,C,E) Graphs show T₂r (black lines) and myelinated axon number (red lines; data from Berbel et al., 1994) in anterior, middle and posterior CC in postnatal C rats at different ages. (B,D,F) Regression functions between myelinated axon number and T₂r. High fits were observed in the anterior (B; R² = 0.993), middle (D; R² = 0.858), and posterior (F; R² = 0.954) CC.

FIGURE 10

Estimated values for MMI treated rats. Estimated values were calculated using the paradigm published for work on the anterior commissure (Lucia et al., 2018). To assess the regression functions from Figures 9B,D,F, estimated values of myelinated axon number (white fill red circles) in the anterior (A), middle (C), and posterior (E) CC for MMI_E10 rats were obtained using T₂r values (solid red circle) and plotted Note that estimated values were similar to those previously published in EM studies (solid black diamond; data from Berbel et al., 1994). These regression functions were used to estimate myelinated axon number in the anterior (B), middle (D), and posterior (F) CC at postnatal ages in C, transient and chronic hypothyroid rats. At P150, estimated values of myelinated axon number of C rats (white fill diamond) were similar to published values (solid black diamond; data from Berbel et al., 1994). Estimated values for myelinated axon number in anterior CC decreased (B) in transient hypothyroid (on average, 23.2%), MMI_P0 (63.8%), and MMI_E10 (79.9%) rats compared to C. A similar decrease was found in middle (D) and posterior (F) CC. At P150, the estimated number of myelinated axons for transient and chronic hypothyroid rats was similar to the number of myelinated axons found in ME, except for MMI_P0 rats (filled black symbols; B,D,F). Est., estimated values.

FIGURE 11

Distribution of myelinated axon diameter and myelin thickness in the anterior CC at P150. (A,D,G,J,M) Histograms show unmyelinated and myelinated axon diameter distributions. Mean unmyelinated and myelinated axon diameter decreased (P < 0.001) in chronic hypothyroid rats compared to transient and C rats. (B,E,H,K,N) Histograms show myelin thickness distributions. The mean myelin thickness was similar between MMI and C rats. (C,F,I,L,O) Plots show the association between myelinated axon inner diameter and myelin thickness. Note the poor fit found between groups (R² range: 0.035–0.286), however the slope of the regression function was higher in C (3.9°) than in MMI (on average, 1.8°) rats. Unmyel, unmyelinated. Myel, myelinated. R², determination coefficients. n, number of axons and means (μ = mean ± SD) are indicated.

FIGURE 12

Distribution of myelinated axon diameter and myelin thickness in the middle CC at P150. (A,D,G,J,M) Histograms show unmyelinated and myelinated axon diameter distributions. Mean unmyelinated and myelinated axon diameter decreased (P < 0.001) in chronic hypothyroid rats compared to transient and C rats. (B,E,H,K,N) Histograms show myelin thickness distributions. The mean myelin thickness was similar between MMI and C rats. (C,F,I,L,O) Plots show the association between myelinated axon inner diameter and myelin thickness. Note the poor fit found between groups (R² range: 0.082–0.301), however the slope of the regression function was higher in C (4.1°) than in MMI (on average, 2.7°). Unmyel, unmyelinated. Myel, myelinated. R², determination coefficients. n, number of axons and means (μ = mean ± SD) are indicated.

FIGURE 13

Distribution of myelinated axon diameter and myelin thickness in the posterior CC at P150. (A,D,G,J,M) Histograms show unmyelinated and myelinated axon diameter distributions. Mean unmyelinated and myelinated axon diameter decreased (P < 0.001) in chronic hypothyroid rats compared to transient and C rats. (B,E,H,K,N) Histograms show myelin thickness distributions. The mean myelin thickness was similar between MMI and C rats. (C,F,I,L,O) Plots show the association between myelinated axon inner diameter and myelin thickness. Note the poor fit found between groups (R² range: 0.061–0.199), however the slope of the regression function was higher in C (3.8°) than in MMI (on average, 3.0°). Unmyel, unmyelinated. Myel, myelinated. μ = mean ± SD. R², determination coefficients. n, number of axons.

FIGURE 14

Association between axon inner diameter and g-ratio in the anterior, middle and posterior CC at P150. Plots show that in the anterior (A,D,G,J,M), middle (B,E,H,K,N) and posterior (C,F,I,L,O) CC, the g-ratio was lower (P < 0.01) in chronic hypothyroid than in transient hypothyroid and C rats. g-Ratio ranges were from 0.74 to 0.77 in anterior, from 0.73 to 0.77 in middle and from 0.72 to 0.75 in posterior CC. μ = mean ± SD. R², determination coefficients.

FIGURE 15

Distribution of myelinated axon conduction velocity of the CC in MMI and C rats at P150. (A,D,G,J,M) Histograms show myelinated axon conduction velocity distributions in anterior CC. Mean myelinated axon conduction velocity decreased in transient (P < 0.05) and chronic (P < 0.001) hypothyroid rats compared to C rats. (B,E,H,K,N) Histograms show myelinated axon conduction velocity distributions in middle CC. Mean myelinated axon conduction velocity decreased in chronic (P < 0.001) hypothyroid rats compared to transient and C rats. (C,F,I,L,O) Histograms show myelinated axon conduction velocity distributions in posterior CC. Mean myelinated axon conduction velocity decreased in transient and chronic (P < 0.01) hypothyroid rats compared to C rats. med, median. n, number of axons. μ = mean ± SD.