Enhanced remyelination during late pregnancy: involvement of the GABAergic system

Pregnant women with MS experience fewer relapses, especially during the third trimester. In this study, we explore the cellular and molecular events that bring about the protective effect of late pregnancy on the course of de/remyelination in rats. Using cellular, molecular, and ultrastructural methods, we explored remyelination in response to a focal demyelination in the corpus callosum of late pregnant, virgin, and postpartum rats. We further explored the role of GABAA receptor (GABAAR) in the promyelinating effect observed during late pregnancy. Remyelination in response to a gliotoxin-induced demyelination in the corpus callosum was enhanced in late pregnant rats when compared to that seen in virgin and postpartum rats. This pregnancy-associated promyelinating effect was lost when either the GABAAR was blocked or when 5α-reductase, the rate limiting enzyme for the endogenous GABAAR activator allopregnanolone, was inhibited. Taken together, these data suggest that the pregnancy-associated pro-myelination operates, at least in part, through a GABAergic activated system.


Supplementary data 1: The extent of demyelination lesion is not different between virgin, pregnant, and postpartum animals at the peak demyelination 3 days postlysolecithin injection.
Images in (A), (B) and (C) show LFB staining of lysolecithin-injected corpora callosa of virgin, pregnant, and postpartum animals respectively. D) Bar graph shows that the demyelination lesion size was not significantly different between the three lysolecithin-injected groups (Lyso-Vir: n=9, Lyso-Preg: n=8, Lyso-PP: n=8, p>0.05). (E), (F) and (G) show immunofluorescent images of NF (green) and MBP (red) at the edge of the demyelination lesion in virgin, pregnant, and postpartum animals respectively. Arrowheads indicate unmyelinated axons while arrows indicate myelinated ones. There was no significant difference in either the percentage area covered by NF + fibers (H), the percentage area covered by MBP + fibers (I), the percentage of juxtaposed NF + and MBP + fibers (J), or the myelination index (K) between the three experimental groups (p>0.05). Data is presented as mean ± SEM. Scale bar in LFB staining = 200 µm. Scale bar in immunofluorescence images = 50 µm.

Supplementary data 3: Systemic Fin administration increases the demyelination lesion size in the corpus callosum of pregnant rats. (A) and (B)
show representative images of LFB staining in lysolecithin-injected corpus callosum of vehicle-treated and Fin-treated pregnant rats. C) Systemic administration of Fin resulted in a significantly larger demyelination lesion when compared to vehicletreated rats (Lyso-Vehicle: n=5, Lyso-Fin: n=7, p<0.05). Results are represented as mean ± SEM. Scale bar = 200µm.

Supplementary data 4: The expression of GABAARγ2 is upregulated in pregnant rats following demyelination in the corpus callosum.
A) Two isoforms of GABAARγ2 were detected at ~47 kDa and ~45 kDa. B) There was no significant difference in the expression level of the ~47 kDa isoform between the three experimental groups (Lyso-Vir: n=4, Lyso-Preg: n=4, Lyso-PP: n=4, p>0.05). C) Pregnant animals showed significantly higher expression of the ~45 kDa isoform compared to either virgin or postpartum animal groups (p<0.05). (D-F) GABAARγ2 (green) was immunodetected in combination with the OPC marker NG2 (red). A fraction of NG2 + cells coexpressed GABAARγ2 (arrows) in virgin (D), pregnant (E), and postpartum (F) animals. DAPI (blue) was used to visualize cell nuclei. (G-I) GABAARγ2 (green) was immunodetected with the microglial marker Iba1 (red). A fraction of Iba1 + cells co-expressed GABAARγ2 (arrows) in virgin (G), pregnant (H), and postpartum (I) animals. DAPI (blue) was used to visualize cell nuclei. Data is presented as mean ± SEM. Scale bar = 50µm.

Supplementary data 5: Experimental design.
Stereotaxic surgery was performed to inject lysolecithin into the corpus callosum to induce a local demyelination lesion. In experiment 1, this surgery was performed on virgin 2-3 months-old rats, pregnant rats at gestational day 13 (GD13), and postpartum rats at postpartum day 3 (PP3). Brains of these animals were collected at both day 3 post-surgery and day 7 post-surgery. Experiments 2 and 3 were performed on pregnant rats. Stereotaxic surgery was performed at GD13. Treatment with bicuculline (Bic) or finasteride (Fin) was administered daily from the day of the surgery until the day of sacrifice 7 days post-surgery.

Supplementary data 6: Lesion overview.
A) Immunofluorescent images showing an overview of the lysolecithin-injected corpus callosum (CC) 7 days post-lesion using DAPI, neurofilament (NF), and myelin basic protein (MBP). A series of images were aqcuired with 10x objective and stitched using the image J software (upper panel). The corpus callosum is outlined with the dashed lines. Lower panel shows a larger view of the demyelination lesion. The demyelination lesion is identified by the absence of MBP staining. Axonal loss at the centre of the lesion is evident as shown by the absence of NF staining. The centre of the lesion contains high density of DAPI. B) A series of immunoflourescent images of NF and MBP at 7 days post-lesion were acquired with 10x objective and stitched using the image J software. The two white boxes show an example of the sampled areas used for axonal integrity analysis at the edge of the lysolecithin-induced demyelination lesion. C) A series of immunoflourescent images of NG2 and PCNA at 7 days post-lesion were acquired with 10x objective and stitched using the image J software. The density of PCNA and NG2 is higher at the centre of the demyelination lesion. The white box shows an example of the area used for the evaluation of PCNA/NG2 density as well as PCNA/Olig2 density (see main manuscript). Scale bar = 50 µm.

Supplementary data 7: Microglial activation and astrocytosis in the corpus callosum following lysolecithin-induced demyelination.
Immunofluorescent images showing the distribution of the astrocytic marker GFAP (blue), and the microglial marker Iba1 (green) 7 days post-lesion. The staining was combined with MBP (red) to identify the demyelinated area of the corpus callosum. Activated astrocytes were found both at the center and the edges of the demyelination lesion. Activated microglia were mainly located at the center of the lesion. The white box shows an example of the area used for analysis of astrocytic and microglial activation. Scale bar = 50 µm.

Supplementary data 8: Lesion confirmation using toluidine blue.
Representative image showing toluidine blue staining in semi-thin sections obtained to confirm the presence of the demyelination lesion in the tissue collected for TEM. The dashed line separates the lesioned and non-lesioned corpus callosum. The non-lesioned area has multiple myelinated axons consistent with the appearance of white matter tracts (arrowheads). The lesion area has reduced number of these myelinated axons.