Abstract
Market authorization of a pesticide or biocide in Europe requires the hazard assessment of endocrine-disrupting properties, including the thyroid modality. Substances inducing thyroid histopathological and/or thyroid hormone effects in rodent studies need to be further investigated to rule out whether the substance can be considered as an endocrine disruptor for the thyroid pathway, including neurodevelopmental impact in pups and its relevance in humans. Histopathological assessment for identifying reliable biomarkers for assessing neurodevelopmental effects is an important aspect of this testing scheme in rats. Periventricular heterotopia in the corpus callosum and persistence of the external granular layer in the cerebellum have been proposed as potential histopathological biomarkers in the brain. The correlation in the cochlea for hearing impairment seen in rat pups derived from hypothyroid dams is another potential biomarker. Herein, we provide a brief overview of the histopathological endpoints. The technical challenges in correctly identifying these changes during brain development and their significance in detecting the impact of maternal hypothyroidism in rodents are discussed. This mini review is part of a scientific presentation by Dr Gangadharan during the developmental neurotoxicity (DNT) session at the 21st ESTP’s Annual Congress (2024).
Keywords
Introduction
Neurodevelopmental risks associated with thyroid hormone (TH) imbalance during pregnancy are a hot topic in the regulatory environment of agrochemicals in Europe. The human relevance of data obtained from neurodevelopmental rodent studies is considered too sensitive and often scientifically challenging. The thresholds of TH changes observed in rodent studies and the differences between the roles of direct and indirect thyroid toxicants are important. Recently, a Thyroid Task force between the European Center for Ecotoxicology and Toxicology of Chemicals (ECETOC) and Crop Life Europe (CLE) proposed a tiered testing and Assessment Scheme, the thyroid function-neurodevelopmental toxicity-testing and assessment Scheme (Thyroid-NDT-TAS), to identify TH disruptors and their potential to induce neurodevelopmental effects in the progeny. 24 In rodent studies assessing the neurodevelopmental effect following experimentally induced maternal hypothyroidism, histopathological assessment is instrumental in identifying the effect; however, the identification of a specific or reliable biomarker for neurodevelopment is not straightforward. Decreased maternal TH is not necessarily associated with impaired neurodevelopment in the offspring, whereas lower TH brain concentrations play a significant role. This mini review provides a brief overview of the histopathological endpoints used to detect the impact of experimentally induced maternal hypothyroidism in rodents.
Animal Models to Study Maternal Hypothyroidism in Rodents
Recently, several animal models of maternal hypothyroidism have been reviewed. 11 Experimental models to recapitulate maternal hypothyroidism include thyroidectomy, 3 restriction of iodine, 23 and exposure to thyrotoxicants targeting thyroperoxidase (TPO), such as propylthiouracil (PTU) or methimazole (MMI).14,36 The PTU also inhibits deiodinase 1 (DIO1). In rodent studies assessing neurodevelopmental endpoints (eg, OECD TG 426), dams are exposed to the test substance from gestational day 6 (GD6) to post-natal day 21 (PND21), and the required measurements are conducted at PND21 and PND60/70. Measurements at other time points at birth (GD21/PND0), PND4, etc, were also conducted as needed. In these studies, histopathological analyses of the thyroid and brain were conducted. Histopathology of the thyroid was used to ascertain its correlation with hormonal changes.
Review of Histopathological Endpoints Concerning Neurodevelopment in the Context of Maternal Hypothyroidism in Rat
Assessing the consequences of thyroid-disrupting chemicals (TDCs) is challenging considering the complexities of TH synthesis and its action in the maternal and fetal compartments. 31 Thus, the research community in this area is redefining or proposing new methods to capture the potential of agrochemicals that may induce neurodevelopmental effects before they are released into the market.12,45 Histopathological examination of the brain and characterization of the induced lesions are key areas in this context and are considered the gold standards. There are a few histopathological endpoints, such as periventricular heterotopia in the corpus callosum, persistence of the external granular layer (EGL) in the cerebellum, and delayed maturation of the cochlea in the ear, used to diagnose the impact of maternal hypothyroidism in pups.
Periventricular Heterotopia (Figure 1)
Heterotopia is a common manifestation of neurodevelopmental toxicity and has been reported in the cerebral cortex or hippocampus 19 . Heterotopia is defined as a cluster of neurons in an abnormal location due to improper migration of developing neurons during neurodevelopment. In the context of maternal hypothyroidism, reported heterotopia is referred to as periventricular heterotopia32,34,35 or subcortical band heterotopia.14,42 Bilateral cellular malformations, heterotopia in the corpus callosum in both hemispheres of the brain of pups at PND23, induced by maternal hypothyroidism by treating their mothers from GD6 to PND30 with graded levels of PTU, was first reported by Goodman and Gilbert in 2007. 16 Heterotopia primarily consist of neurons and show a dose-dependent increase in size with a decrease in TH levels. Using BrdU immunohistochemistry, they also reported that neurons present in heterotopia were born between GD17 and GD19. In their study, 50-micron vibratome sectioning of the brain at the hippocampal region was employed to detect heterotopia. Further studies with detailed characterization of the lesion and experiments to formulate the molecular pathogenesis were conducted by the same group.14,32 -34 Dose-dependent increases in the volume of heterotopia with a clear presence of heterotopia and a minor reduction in maternal serum T4 (<15%) induced by PTU when the pups were weaned were demonstrated. Using immunohistochemical methods, they also showed that heterotopia consists of neurons, oligodendrocytes, astrocytes, and microglia. Heterotopia was defined as permanent damage if not treated with thyroxine before its development, and the offspring displayed increased sensitivity to seizures. The TH-dependent heterotopia formation in the neonatal brain does not worsen under conditions of low dietary iron intake. 44 A cross-fostering study by the same group confirmed that heterotopic formation requires gestational hormone deficiency (GD6 to PND2). 32 They found that only five days of maternal exposure to 10 ppm by PTU on GD19 and PND2 was both sufficient and necessary to induce a permanent periventricular heterotopia in 100% of analyzed pups. 34 This study also reported decreased sonic hedgehog (Shh) expression, abnormal cell adhesion, and altered radial glial morphology, thereby suggesting its role in the development of this lesion in a periventricular location. Further research using laser capture microdissection in the ventricular zone and RNA sequence analysis revealed 358 differentially expressed genes in hypothyroid neonates compared to controls. 33 Pathway analysis revealed that processes such as the maintenance of the extracellular matrix and cytoskeleton, cell adhesion, and cell migration were seriously affected. This suggests that multiple components of cell junctions in the ventricular zone are involved in the pathogenesis of this lesion.

Periventricular heterotopia in the cerebrum (Panel A). From a control PND21 pup to demonstrate the comparable location at the level of corpus callosum (Level 4 as per Garman et al (2016) [Panel B]). From a PND21 pup from a dam treated with PTU at 10 ppm during gestation day 6 (GD6) to lactation day 21 (LD21). Note the circled area with a collection of ectopic neurons H&E (Panels A and B). Original objective magnifications = 20× (Panels A and B).
The PTU-based experimental model was used to test the potential of other environmental chemicals that were able to reduce serum T4 levels, such as perfluoroalkyl perfluorohexane sulfonate (PFHxS) (50 mg/kg/d) and the antimicrobial triclosan (3000 mg/kg/d), and did not show brain morphological changes (heterotopia) or neurobehavioral changes, as seen in pups born out of PTU-treated rats. 13 Although both the chemicals reduced serum T4 levels in this model, there was no increase in serum thyroid-stimulating hormone (TSH) levels. Brain tissue T4 was reduced in newborns but recovered in PND6 pup brain. These observations suggest that a mere reduction in serum T4 levels alone may not be sufficient to induce neurodevelopmental effects in the brain.
Other research groups42,25,26,29,38 have conducted similar studies using PTU alone or in comparison with other TPO inhibitors or indirect thyroid toxicants. Shibutani et al compared the ability of MMI to PTU in inducing neurodevelopmental effects in rats with that of PTU by inducing maternal/perinatal hypothyroidism. The dose regime ranged from 3 to 10 ppm PTU and 200 ppm MMI in drinking water from GD10 to PND20. Although other neurodevelopmental defects were observed, the incidence of periventricular heterotopia in their study was low. They could not observe any heterotopia in the brain at PND20 in rats administered 3 or 12 ppm PTU, and only in 2/5 pups administered 200 ppm MMI. The reported reduction in the incidence of heterotopia may be related to differences in the histological processing of the brain in this study compared to the technique employed by the Environmental Protection Agency (EPA) research group. Minami et al conducted a modified Comparative Thyroid Assay (CTA) to induce hypothyroidism using either 10 ppm PTU or 1000 ppm sodium phenobarbital. Brain TH levels and histopathology were assessed in addition to the original endpoints required for CTA. As described by Garman et al, 10 only three levels of the brain were examined for the histopathological examination of the brain. The incidence of heterotopia in the first examination using a single section of the brain was lower than expected in the PTU group (3/31 PND21 pups). The second examination using step sections from the middle region of the brain, including the hippocampus, revealed an increased number of heterotopias. This study shows that step-sectioning of the brain in the hippocampal region is necessary for successful and accurate detection of heterotopic formation in the brain. In addition to PTU and MMI, treatment with an environmental chemical, amitrole, at 50 mg/kg in maternal rats was shown to induce heterotopia formation in all pups at PND16. 36 This study also evaluated the potential of other TPO inhibitors, 2-Mercaptobenzimidazole (MBI) and cyanamide, to induce heterotopia in the same experiment and reported that they were incapable of producing heterotopia. In this experiment, brains were immersion fixed in 10 % formalin and were subsequently blocked with coronal cuts anterior and posterior to the hippocampus. The blocks were sectioned at 10 micron, and every third section was collected. More recently, it has been reported that the environmental pollutant PFHxS reduces serum thyroxine levels but does not alter thyroid activity in the post-natal brain. 30 They compared the development of periventricular heterotopic formation with PTU.
The possibility of using heterotopia as a biomarker for maternal hypothyroidism in a mouse model was evaluated using dietary PTU at 1500 ppm. 37 This study concluded that heterotopia formation in mice is not a useful histopathological endpoint for detecting neurodevelopmental endpoints in mouse pups following perinatally induced hypothyroidism in pregnant mothers using the histopathological techniques described in rat studies. Similar studies have also reported disruptions in the layering of cells in the neocortex and hippocampus or the development of radial glial cells in the post-natal neocortex after mild and transient reductions in the maternal TH induced by methimazole.2,21
Persistence of External Granular Layer in Cerebellum in the Rat (Figure 2)
The cerebellum is known for its contribution to coordination, posture, balance, and learning. Recent studies have also shown its contribution to functions such as cognition, emotion, and language 43 . Granule cells in the cerebellum originate from superficially located external granule cells (EGLs) that are visible in the neonatal brain. 1 The EGL is the zone of granule cell progenitors that migrate to the core of the cerebellum using Bergmann glia astrocytes as a guide. The TH plays a significant role in cerebellar development by influencing neuronal differentiation, axonal elongation, synaptogenesis, dendritic arborization, and cellular migration. 20 The development of the rodent cerebellum is a post-natal event, and in the hypothyroid cerebellum, the disappearance of the EGL is delayed due to the delay in its proliferation and migration of granule cells to populate the granule cell layer of the cerebellum (Koibuchi and Chin 2000). A mutant mouse model with a mutation at TRα1 showed defective post-natal cerebellum development, which included inhibition of inward movement of cells from the EGL. 8 The same group further clarified using CRE/loxP-mediated conditional expression approach that the effect of this specific TRα1 mutation primarily alters the differentiation of Purkinje cells and Bergmann glia, thus indirectly affecting the global development of cerebellum, notably the inward migration and terminal differentiation of the granule cell precursors from the EGL. 7

Persistence of external granular layer in the cerebellum (Panel A). From a control PND21 pup to demonstrate the comparable location at the level of cerebellum (Level 7 as per Garman et al (2016) [Panel B]). From a PND21 pup from a dam treated with PTU at 10 ppm during GD6 to LD21. Note the arrows with persistence of external granular layer in Panel B and comparable location in Panel A. H&E (Panels A and B). Original objective magnifications = 20× (Panels A and B).
First-generation rats fed a severely iodine-deficient diet showed delayed disappearance of the cerebellar EGL along with other features of delayed brain development. 23 The EGL thickness in the control group peaked at 9 days of age and disappeared by 20 days of age. In the experimental group, the largest EGL thickness occurred at 10 days of age and disappeared by 25 days of age. A TPO inhibitor, methimazole, was used to induce hypothyroidism in pregnant rats, and its effect on cerebellar development in pups was studied. 17 The prominent anomaly reported in this study was in the internal granular layer of the cerebellum, with excess bulges or branching and the formation of excess sublobules, although normal lobulation was maintained. Impaired motor coordination has been reported as a functional correlate of this defect. In this study, the EGL survived longer in the treatment group than in the control group. Increased severity and incidence of persistent EGL in the cerebellum at PND21 pups were reported in PTU-treated pups compared to phenobarbital-treated pups. 26 The PND21 pups born to pregnant rats fed 1000 ppm NaPB from GD6 to PND21 had 5/14 persistent EGL compared to 16/16 in pregnant rats fed PTU at 10 ppm. For PND21 pups in the PTU group, serum T4 levels were significantly reduced in both sexes (81% and 88% decrease in males and females, respectively), with a significant increase in serum TSH levels. On PND21, the brain T3 and T4 levels were significantly reduced.
Delayed Development of the Cochlea (Figure 3)
The relationships among deaf mutism, cretinism, and goiters are well known 40 . Gene knockout studies in mice have shown that TH influences the maturation of cochlear tissues during auditory development. 28 THRB-deficient mice are deaf and exhibit multiple cochlear defects. 9 In mice lacking Dio2 (an enzyme involved in converting T4 to T3), the Greater Epithelial Ridge (GER, a transient epithelial cell structure located medially to the sensory hair cells) remodeling in the cochlea was delayed, and the mice had permanent auditory deficits. 27 Deafness and loss of cochlear cells in the absence of TH transporters have been reported previously. 41 The GER cells secrete several important sulphonated glycoproteins that contribute to tectorial membrane formation.22,39 In euthyroid rodents, GER regresses during the first two weeks of post-natal development to create an inner sulcus that allows proper suspension of the tectorial membrane. 18 The mechanism involved in the regression of GER is not completely understood, and the delayed disappearance of GER is related to hearing function deficits. Cochlear remodeling is a precisely ordered regulatory process, and T3 provides important molecular signaling support for these remodeling events. As caspase 3-impaired mice also have abnormal persistence of GER, it is considered that an apoptosis-mediated mechanism takes part in the remodeling of GER. 28

Delayed remodeling of greater epithelial ridge (GER) in the cochlea (Panel A). From a control PND21 pup to demonstrate the comparable location at the level of cochlea (Panel B). From a PND21 pup from a dam treated with PTU at 10 ppm during GD6 to LD 21. Note the arrows with delayed remodeling of GER in Panel B and comparable location in Panel A. H&E (Panels A and B). Original objective magnifications = 20× (Panels A and B).
Pregnant mice treated with PTU in drinking water and the development of the cochlea in pups were studied at various sequential post-natal time points, and it was concluded that, in the inner ear, the remodeling of the GER, also known as Kolleker’s organ, requires TH (Deol 1976). 6 Perinatal exposure of dams at 5 or 25 ppm PTU in drinking water from GD18 to GD21 resulted in dose-dependent defective hearing function in rats. 15 The TDC, which can lower serum T4 concentrations at 14 or 21 days of post-natal age correlated with hearing loss in adult rats. 4 The TDC’s investigated in this study were Aroclor 1254, PTU, dioxins, PCB 126, and DE-71. The structural defect responsible for defective hearing function in pups following gestational/lactational exposure to Aroclor A1254 has been reported to be the loss of outer hair cells in the upper-middle and apical turns of the cochlea. 5
Conclusion
Identifying and establishing reliable histopathological endpoints are crucial for assessing the neurodevelopmental risks associated with TH imbalance during pregnancy. This review discusses three histopathological endpoints in rodent studies that relate maternal hypothyroidism to the neurodevelopmental effects in pups. Establishing a clear protocol for histological processing of the brain and ear is crucial for achieving this objective.
The majority of mechanistic in vivo studies have been conducted with PTU, and it is important to further investigate which histopathological endpoints are related to a decrease in fetal/pup TH levels and which might be more specific to PTU (via direct interaction with neural progenitor cells [NPCs] or other pathways that PTU might impact). As a histopathological endpoint, the majority of studies focused only on periventricular heterotopia, and it is worth investigating other histopathological endpoints discussed in this mini review, such as the persistence of EGL in the cerebellum and delayed development of the cochlea, to establish a set of reliable histopathological endpoints.
Footnotes
Acknowledgements
Author Contributions
All authors contributed to writing the manuscript. Wrote the manuscript (BG) and performed a critical review (FS). The opinions expressed in this article are solely those of the authors and should not be construed as representing the views or policies of their employers.
Authors’ Note
This document has been reviewed in accordance with policies of the authors’ institution of employment. The opinions expressed in this paper solely represent those of the authors and should not be construed as official views or policies of the authors’ institution.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
