TH synthesis, Decreased leads to GABAergic interneurons, Decreased

TH levels influence the development of cortical GABAergic circuits (Friauf et al., 2008; Westerholz et al., 2010). In hypothyroid rats the expression of parvalbumin, the marker of a subpopulation of GABAergic neurons, is reduced (Gilbert et al., 2007). TH increase glutamic acid decarboxylase (GAD, GABA-synthesizing enzyme) activity in neonatal brain. Also GABA transaminase (GABA-T, GABA-degrading enzyme) activity appears to be increased by TH. Therefore, both GABA synthesis and degradation are increased by TH. This might reflect either the specific regulation of GABA levels, or general regulation of gene expression maintenance by TH, as commented by Shulga and Rivera, 2013. This strongly supports the link between the two KEs described in this indirect KER (decrease of TH synthesis leads to GABAergic interneuron decrease). It was also shown that low concentrations of T3 increase by non-genomic mechanism the depolarization-dependent release of GABA. GABA appears to provide negative feedback to thyroid endocrine axis, as TSH release is inhibited by GABA (Wiens and Trudeau, 2006).

TPO inhibitors (like PTU and MMI) decrease the volume of the granule cell layer of the dentate gyrus, the density of cells within the layer, and estimates of total granule cell number, as shown in hypothyroid rats (Madeira et al., 1991). Migration of granule cells from the proliferative zone to the granule cell layer (with different neuronal subtypes, including GABAergic neurons) is retarded by thyroid deficiency, as is dendritic arborization and synaptogenesis assessed by immunohistochemistry for the synaptic protein synaptophysin (Rami et al., 1986a, 1986b, Rami and Rabie, 1990).

- Sawano et al., 2013: This in vivo study investigated the effects methimazole (MMI, a TPO inhibitor) on the developing rat hippocampus, one of the brain regions most sensitive to TH status. MMI was administered at the concentration of 0.025% in drinking water to pregnant dams from gestational day 15 until 4 weeks postpartum. Looking at the pre- and post-synaptic components of the GABAergic system, the level of glutamic acid decarboxylase 65 (GAD65) protein was reduced to less than 50% of control in the hippocampus of hypothyroid rats, and recovered to control levels by daily thyroxine-replacement after birth. Reduction in GAD65 protein was correlated immunohistochemically with a 37% reduction in the number of GAD65⁺ cells, as well as a reduction in GAD65⁺ processes. In contrast, GAD67 was not affected by MMI treatment. A subpopulation of GABAergic neurons containing PV was also confirmed to be highly dependent on TH status (with a 33% reduction in total PV⁺ neurons compared with the control). Moreover, the physiologically occurring transient rise of KCC2 expression observed at PND 10 (followed by a large increase in KCC2 protein at PND 15) in the euthyroid hippocampus, was completely suppressed by MMI (~ 80% reduction in KCC2 protein at PND 15 vs control).

- Shiraki et al., 2012: this in vivo study compared the differential effects of MMI (0, 50, 200 ppm in the drinking water) comparing the developmental and adult-stage, in particular comparing pregnant rats treated from gestation day 10 to PND 21 (i.e., developmental hypothyroidism) and adult male rats treated from PND 46 through to PND 77 (i.e., adult-stage hypothyroidism). With regard to precursor granule cells, a sustained reduction of Pax6⁺ stem or early progenitor cells and a transient reduction of doublecortin⁺ late-stage progenitor cells were observed after developmental hypothyroidism with MMI at 50 and 200 ppm. These cells were unchanged by adult-stage hypothyroidism. The number of PV⁺ cells (a GABAergic interneuron subpopulation in the dentate hilus) was decreased (~ 60% reduction at PND 21, with 200 ppm MMI) and the number of calretinin⁺ cells was increased (~ 85% increase at PND 21, with 200 ppm MMI) after both developmental and adult-stage hypothyroidism.

- Gilbert et al., 2007: In this in vivo study pregnant rat dams were exposed to propylthiouracil (PTU, a TPO inhibitor, administered at 0, 3, 10 ppm in the drinking water, from gestational day 6 until PND 30). PTU decreased maternal serum T4 by ~ 50-75% and increased TSH. At weaning, T4 was reduced by approximately 70% in offspring in the low-dose group and fell below detectable levels in high-dose animals. PV⁺ cells were diminished in the hippocampus and neocortex of offspring sacrificed on PND 21 (~ 45% reduction in the cortex, and ~ 55% reduction in the dentate gyrus, with 3 ppm treatment), and altered staining persisted to adulthood despite the return of TH to control levels.

- Westerholz et al., 2010; 2013: In the developing cortex, spontaneous activity is characterized by synchronous bursts of action potentials in populations of glutamatergic and GABAergic neurons which propagate throughout developing neural networks. In these in vitro studies with cortical neurons (prepared from E16 rat cortex), synthesis of TH and T3, in particular, increased the density and growth of GABAergic neurons and accelerated the maturation of neural networks.

While some in vivo studies (Sawano et al., 2013; Shiraki et al., 2012) have shown a decrease of GABAergic cell populations upon induction of hypothyroidism, Saegusa and co-workers (Saegusa et al., 2010) reported about an increase of GABAergic interneurons. In Saegusa's study, rat dams were treated with either PTU or MMI in the drinking water, and male offspring were immunohistochemically examined on PND 20 and at the adult stage (i.e., 11-week-old). MMI and PTU caused in the offspring growth retardation, lasting into the adult stage. All exposure groups showed a sustained increase of GAD67⁺ cells in the adult stage, indicating an increase in GABAergic interneurons.

It should be noticed that in Saegusa et al., 2010 in vivo study, increase of GAD67⁺ cells was mainly observed in the adult stage (11-week-old rats) and analysis of GABAergic interneurons.  PV⁺ cells, which appear to be the GABAergic population most affected by TH dysregulation, was not evaluated. On the opposite, Sawano's and Shiraki's in vivo studies reported a decrease of GABAergic PV⁺ neurons at earlier stages, respectively on PND 15 and 21 induced by hypothyroidism (Sawano et al., 2013; Shiraki et al., 2012). Discrepancies in results are due to the fact that THs have effects on multiple components of the GABA system. For instance, in the developing brain, hypothyroidism generally decreases enzyme activities and GABA levels, whereas in adult brain, hypothyroidism generally increases enzyme activities and GABA levels.

There are also conflicting results on effects of long term changes in TH levels on GABA reuptake. Therefore, results variability from study to study is due to different experimental study designs, accounting for differences in brain development stages (PND vs adult), times of exposures to chemicals, and regional brain differences (Wiens and Trudeau, 2006).