Neurodegenerative Diseases: Pathogenesis and Preclinical Models for Translational Drug Discovery

Abstract

The fourth session of the 2024 European Society of Toxicologic Pathology (ESTP) Congress brought together lectures focused on the use of in vitro and in vivo models to investigate neurodegenerative diseases. Four presentations highlighted various aspects of neurodegenerative diseases including dementia, immune-mediated conditions, and neuromuscular disorders. The session began with an overview of animal models of dementia underscoring their critical role in understanding disease pathogenesis and supporting the development of effective therapeutic drugs. Subsequent presentations investigated immunological self-tolerance in autoimmune neurodegenerative diseases, such as multiple sclerosis and Guillain-Barré syndrome, and the application of in vitro models to study neuromuscular diseases such as amyotrophic lateral sclerosis. The final presentation examined cannabinoid-based therapeutic options for treating neurodegenerative diseases, highlighting their potential in neuroprotection and neurorepair. This session provided valuable insights into the latest research and advancements in neurodegenerative disease modeling and therapy, offering promising directions for improved modeling and therapeutic strategies.

Introduction

Neurodegenerative diseases represent a significant challenge in the medical field due to their complex pathogenesis, diverse clinical manifestations, and limited therapeutic options. The current paper summarizes the principles and practices discussed during the fourth session of the ESTP Congress, which provided an integrated perspective on innovative models and therapeutic approaches for neurodegeneration. Presentations spanned from traditional animal models and cutting-edge in vitro systems to emerging therapeutic modalities, including cannabinoids. By bridging these fields, the session highlighted opportunities to improve translational research and accelerate the development of effective treatments.

Bridging the Gap: Understanding Dementia Through Animal Models

Dr. Laura Fusaro from Denali Therapeutics delivered an insightful presentation on the critical role of animal models in advancing our understanding of dementia pathogenesis and supporting the development of effective therapies. Dementia encompasses a diverse group of neurodegenerative disorders characterized by progressive cognitive decline that interferes with daily functioning. Common forms include Alzheimer’s disease (AD), vascular dementia, frontotemporal dementia (FTD), and dementia with Lewy bodies. These disorders are underpinned by overlapping pathological features such as neurodegeneration, protein misfolding, and chronic neuroinflammation. ²²

AD: Pathology and Modeling

AD is the most common form of dementia, accounting for approximately 60 to 80% of cases. While sporadic Alzheimer’s predominates, early-onset familial Alzheimer’s is strongly linked to APP, PSEN1, and PSEN2 mutations, which increase amyloid-β (Aβ) production. In contrast, late-onset Alzheimer’s is associated with the APOEε4 allele and modifiable risk factors such as hypertension, diabetes, and diet. ⁷ The widely accepted Aβ cascade hypothesis posits that Alzheimer’s pathology originates from an imbalance between Aβ production and clearance, leading to extracellular Aβ plaque accumulation, neurofibrillary tangle (NFT) formation, neuronal loss, and cortical atrophy (Figures 1 and 2).^11,21,46

Figure 1.

Alzheimer senile plaques. Immunohistochemistry of affected Alzheimer’s tissues using antibodies directed against Aβ peptides demonstrates the presence of both diffuse (A) and dense core (B) senile plaques. These dense core plaques are often associated with neuritic elements that can stain filamentous tau and correlate with disease severity. Neuritic AD plaques are readily observed using Bielschowsky silver staining (C) or Thioflavin S staining (D). These stains can also label neurofibrillary tangles (NFTs) as shown by the arrowheads. The scale bars are 40 μm. Reproduced with permission from DeTure & Dickson, 2019 (Mol. Neurodegener.) under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0).

Figure 2.

Neurofibrillary tangles. Neurofibrillary tangles develop from intracellular pre-tangles containing misfolded tau and small tau aggregates to mature NFTs containing bundles of cross-linked tau filaments before the neuron dies and an extracellular ghost tangle (asterisk) remains. Silver staining (A) and Thioflavin S (B) capture many mature tangles (arrows) and some pre-tangles (arrowheads) along with amyloid plaques and tau neuropil threads. Development of NFTs from the pre-tangles is more easily visualized using tau immunohistochemistry (C, D). This allows the mis-localized somal tau to be distinguished readily from the bundles of paired helical filaments (PHFs) in NFTs in addition to the neuropil threads that can also be pronounced (d). The scale bars are 40 μm. Reproduced with permission from DeTure & Dickson, 2019 (Mol. Neurodegener.) under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0).

Several transgenic mouse models have been developed to replicate aspects of AD. For example, the TG2576 model, which overexpresses human APPswe, effectively models amyloid deposition and cognitive deficits; however, it lacks tau pathology and neurodegeneration.^17,35 Similarly, the 5XFAD model, incorporating multiple familial AD mutations, displays rapid amyloid pathology and neuroinflammation but also fails to reproduce tau pathology and neuronal loss.^26,35 The 3xTg model, which combines amyloid and tau pathologies with cognitive impairments, provides a more integrated approach, but it is limited by overexpression artifacts and the absence of significant vascular pathology. ²⁷

Due to their genetic and physiological similarities to humans, non-human primates (NHPs) offer unique advantages as models for AD. While NHPs can develop amyloid plaques and vascular amyloidosis with age, they resist tau pathology and neuronal loss. ⁴⁶ Experimental approaches, such as inducing tauopathy by delivering tau via viral vectors into the entorhinal cortex, have demonstrated hippocampal atrophy, tau propagation, microglial activation, and cognitive decline, further emphasizing their value in modeling specific aspects of AD. ³

FTD: Insights From GRN Models

FTD is a leading cause of early-onset dementia, typically presenting between the ages of 45 and 65 years. Unlike AD, behavioral changes, personality shifts, and language deficits often precede memory impairments. Familial FTD, which comprises 30% to 50% of cases, is associated with mutations in MAPT, GRN, and C9orf72, while sporadic forms also occur. Pathologically, FTD subtypes are distinguished by intracellular inclusions of specific proteins. FTLD-Tau features tau inclusions in cortical neurons, while FTLD-TDP is characterized by TDP-43 inclusions, often linked to GRN mutations. ³⁰

Loss-of-function mutations in GRN result in progranulin deficiency, leading to lysosomal dysfunction, neuroinflammation, and neuronal loss.^23,24 The GRN knockout mouse model replicates many phenotypes of human GRN-related FTD, including lipofuscinosis, TDP-43 aggregation, and cognitive deficits. However, it fails to capture the typical heterozygous loss of function found in human disease. These models have been invaluable for elucidating FTD mechanisms and testing therapeutic strategies.

Emerging Therapies and Challenges

Recent advancements in AD therapeutics focus on targeting Aβ pathology. Monoclonal antibodies such as aducanumab, lecanemab, and donanemab have effectively reduced amyloid plaques. ⁹ However, these treatments are associated with amyloid-related imaging abnormalities (ARIAs), including cerebral edema and microhemorrhages, particularly in APOEε4 homozygous carriers. ⁶ Preclinical studies using models like 5XFAD have provided critical insights into the mechanisms underlying ARIA, including vascular amyloid deposition and altered vascular integrity. ³²

Despite progress, no disease-modifying therapies currently exist for FTD. GRN-targeted approaches, tau-focused treatments, and neuroinflammation modulators remain active areas of investigation. Advances in imaging techniques and biomarkers, such as neurofilament light chain (NfL), ²⁴ are facilitating earlier diagnosis and enabling the monitoring of disease progression and therapeutic efficacy.

Bridging the Gap in Translational Research

Animal models are essential for understanding the complexities of dementia but are limited in replicating the full spectrum of human disease. Choosing suitable models depends on the specific research question, guided by factors such as disease relevance, genetic similarity, and ethical considerations. Dr. Fusaro emphasized the need to refine existing models and incorporate humanized approaches, like organoids and multi-omics technologies, to improve translatability.

In conclusion, although no single model can fully capture the complexity of human dementia, integrating innovative technologies and interdisciplinary collaboration continues to drive progress in understanding disease mechanisms and developing effective therapies. Addressing the unmet needs of patients requires sustained efforts to refine animal models and translate findings into clinical breakthroughs.

Autoimmune Neurodegenerative Diseases: Shifting Paradigms and Emerging Opportunities

Dr. Dinesh S. Bangari introduced the concept of immunological self-tolerance and the mechanisms underlying its breach in autoimmune diseases. Most of the presentation was focused on discussing shifting paradigms in the pathophysiology and animal models of two illustrative autoimmune neurodegenerative diseases (ANDs): multiple sclerosis (MS) and Guillain-Barré syndrome (GBS). The talk concluded with a brief discussion on the critical role of nonclinical pathologists in the discovery and development of medicines targeting ANDs.

ANDs arise from the loss of immunological tolerance to self-antigens in the central (CNS) and/or peripheral nervous system (PNS). ⁴³ They are triggered by genetic, epigenetic, and environmental factors, including infections (via mechanisms like molecular mimicry, epitope spreading, bystander activation, superantigens), microbiota, tumors, vitamin D deficiency, smoking, air pollution, stress, toxicants, chemotherapy, immunotherapy, and hormones. ⁵

Of the over 80 autoimmune diseases affecting about 4.5% of the global population, around 30 impact the nervous system.^15,36 ANDs can be categorized based on pathomechanisms: (1) spontaneous (e.g., MS), (2) post-infectious (e.g., GBS), (3) paraneoplastic (e.g., anti-NMDA encephalitis), and (4) iatrogenic (e.g., checkpoint inhibitor-induced neuropathy). Clinically, ANDs are marked by progressive organ dysfunction, disability, heterogeneous presentations, and varying prevalence by sex, ethnicity, and geography.

MS, the most common demyelinating AND, affects over 2.9 million people worldwide.^19,44 Key pathological features include inflammation, demyelination, and neurodegeneration. MS etiology is unclear, but factors like Epstein-Barr Virus (EBV) infection are implicated. Two models of MS pathogenesis exist: the CNS intrinsic “inside-out” model and the CNS extrinsic “outside-in” model. In the CNS intrinsic “inside-out” model, the initial event is hypothesized to take place in the CNS, which leads to the release of CNS antigens to the periphery either by drainage to the lymph nodes or active carriage by antigen-presenting cells. In the CNS extrinsic “outside-in” model, the initial event takes place outside the CNS (e.g., in the context of a systemic infection) and leads to an aberrant immune response against the CNS. Clinically, MS presents as relapsing-remitting, primary progressive, or secondary progressive forms.

GBS affects the PNS, presenting as acute paralytic neuropathy with limb weakness and hypoflexia or areflexia.^4,45 Diagnostic hallmarks include aberrant nerve conduction, albuminocytologic dissociation, and anti-ganglioside antibodies. GBS is often post-infectious, linked to Campylobacter jejuni enteritis, other infections, vaccination, and checkpoint inhibitor therapy. Pathological phenotypes include acute inflammatory demyelinating polyneuropathy (AIDP), acute motor axonal neuropathy (AMAN), and Miller-Fisher syndrome.

Animal models for MS, GBS, and other ANDs replicate disease features like inflammation, demyelination, and neuronal degeneration.^25,33,37,38 MS models include experimental autoimmune encephalomyelitis (EAE), cuprizone (Cz)-induced demyelination, lysolecithin-induced myelin loss, and Theiler’s murine encephalopathy virus. GBS models include experimental autoimmune neuritis (EAN) and C. jejuni lipopolysaccharide and GM1 passive transfer models.

EAE models emphasize inflammation, while Cz models highlight demyelination. Disease induction and progression in these models depend on species, sex, strain, and other factors. Histopathologic assessments use conventional and specialized methods, including plasma neurofilament light levels as a biomarker; neurofilament light has also been credentialed in the EAE mouse model using an approved MS drug, glatiramer acetate. ¹

Several considerations for optimal use of the Cz model were discussed. For example, CD-1 stock mice are highly resistant to disease induction, impact of diet formulation (pellets vs powder vs oral gavage) on disease induction, regional variation in lesion severity as evidenced by relative resistance of hippocampal fornix region to demyelination as opposed to severe loss of myelin in the midline of corpus callosum. Such preanalytical factors must be borne in mind when investigations using these models are planned.

Neurological immune-related adverse events (iRAEs) represent autoimmune-like diseases associated with the use of monoclonal antibodies, targeted cancer therapeutics such as immune checkpoint inhibitors, tumor necrosis factor alpha (TNF-α) blockers, and so on. ²⁰

During nonclinical safety assessments, iRAEs can manifest in animal subjects both clinically and histologically. Pathologists, therefore, play a critical role in understanding these neurotoxicologic liabilities. For animal model characterization, pathology labs use routine histopathology and immunohistochemistry (IHC) markers for demyelination (MBP, MOG, PLP, citrullinated MBP), axonal degeneration (Smi32, neurofilament light), immune cells (CD3, CD45R, CD19, FoxP3), microglial cells (Iba-1), and oligodendrocytes (olig2, GSTpi). Additionally, in vivo multimodal imaging technologies (e.g., MRI, microCT, molecular ultrasound, positron emission tomography), spatial transcriptomics, and new approach methodologies, including patient-derived organoids, can be leveraged by modern pathology laboratories to investigate ANDs and support drug-development efforts.

Using Advanced Human in Vitro Models to Study Neurodegenerative Disease

Prof. Dr. Pasterkamp presented different advanced in vitro models to investigate and study neurodegenerative disease with an emphasis on amyotrophic lateral sclerosis (ALS). ALS is a fatal adult-onset neurodegenerative disease characterized by the progressive loss of upper and lower motor neurons resulting in muscle weakness and atrophy. Treatment options for ALS patients are limited and more effective therapies are needed. However, the development of such therapeutic strategies requires a better understanding of the pathogenic mechanisms underlying ALS. ⁴¹ Several different animal models have been generated to model diverse aspects of ALS pathogenesis and these have been invaluable in unveiling disease pathways. ³¹ However, such approach is mostly feasible for forms of ALS where disease-causing genetic mutations are known. Furthermore, animal models have not been very successful in predicting the efficacy of experimental drugs in a human setting for ALS patients.

With the development of induced pluripotent stem cell (iPSC) technology it has become possible to use patient cells to generate desired cell types and tissues. This approach is now widely used for the generation of human in vitro models that model specific tissues or organs, such as the brain. ²⁹ These models are used to define the pathogenic events that cause motor neuron degeneration and ALS and for the development of therapeutic approaches. In combination with molecular cell biological approaches (e.g. scRNAseq, CRISPR, spatial omics), 3D microscopy, and microfluidics, different cell types that are relevant for ALS pathogenesis, or neuromuscular disease in general, can be generated and assessed, including motor neurons, glial cells, and skeletal muscle. These cells can be combined in microfluidics devices to reconstitute the neuromuscular system and study neuronal function in relation to functional outputs, e.g. muscle strength or activity using calcium imaging or muscle cell contraction (Figure 3). This setup also allows the interrogation of specific subcellular compartments such as motor neuron axons. Approaches here include live cell imaging of intracellular cargo and RNA sequencing of axons (axon-seq). ¹⁰

Figure 3.

Overview of different iPSC-derived models for studying neuromuscular and neurodegenerative disease. iPSCs can be used to generate many different neural and non-neural cell types and cultured in 2D. Furthermore, protocols are available to generate neural organoids representing different parts of the brain and spinal cord but also skeletal muscle. Finally, cell and organoid types can be combined in microfluidic platforms for reconstituting the neuromuscular system.

iPSCs can also be a starting point for the generation of more complex tissues or models, such as organoids. In the recent years, organoid models have been developed to model different parts of the nervous system, including motor cortex and spinal cord. ⁴² For example, brain organoid models were developed in which microglia develop innately, alongside neurons and other glial cells. ²⁸ Microglia are the immune cells of the brain and an important cell type in the pathogenesis of most neurodegenerative diseases, including ALS. Importantly, these models show established pathological hallmarks of ALS, as well as pathogenic changes and can therefore be used to further dissect disease mechanisms and to identify therapeutic targets. Brain organoids derived from ALS patient iPSCs show various phenotypes including cell death, changes in cell type composition and synaptic connectivity. To examine the latter phenotype further, air-liquid interphase cortical organoid (ALI-CO) slice culture were developed to perform electrophysiological recordings, that indeed confirm changes in neuronal activity. ⁴⁰ ALI-CO also allow the seeding of cell types, such as microglia, at the desired density and genotype. Finally, brain organoids, or iPSC-derived models in general, can not only be used to study disease mechanisms in symptomatic patients but also offer the opportunity to generate models before symptom onset, when specific gene defects can be detected early on. This approach could help in testing drug efficacy before symptom onset and facilitate early intervention.

Update on Cannabinoid-Based Therapeutic Options for Treating Neurodegenerative Disorders

Prof. Dr. Fernández-Ruiz introduced and discussed the therapeutic potential of cannabinoids in various neurodegenerative disease. Cannabinoids form a singular family of plant-derived natural compounds and synthetic derivatives able to exert multiple biological actions in the human body, which derive from their ability to mimic, block or modulate the action of various endogenous signaling lipids that form part of the so-called endogenous cannabinoid system. ¹⁴ Many of these actions have been found to have potential therapeutic applications in human pathologies. The regulation of cell homeostasis, integrity and survival in different tissues, in particular, in neural cells, is one of the different biological actions of endocannabinoids that is attracting more interest, as it explains why the pharmacological modulation of different elements of the endocannabinoid system (eCB) may afford benefits in pathologies related to brain damage, in particular in chronic progressive neurodegenerative disorders. ¹² These beneficial effects appear to be facilitated by the location of those endocannabinoid elements modulated by plant-derived or synthetic cannabinoids (e.g., CB₁, CB₂, GPR55, PPAR-γ receptors, endocannabinoid inactivating enzymes) in cellular substrates (e.g., neurons, astrocytes, microglial cells, neural progenitor cells, blood-brain barrier) that are important in the control of neural cell survival,^8,12 as summarized in Figure 4.

Figure 4.

Distribution of elements of the endocannabinoid system in key cell and tissue substrates for the control of neural cell survival; eCB = endocannabinoid system.

This fact allows these endogenous compounds to be active in the preservation, rescue, repair and/or replacement of neurons and other neural cells against the numerous insults that contribute to potentially damage these cells in neurodegenerative disorders. This lecture has been an attempt to update the most recent and relevant experimental evidence, mainly obtained at the preclinical level, supporting that different endogenous, plant-derived or synthetic cannabinoids may behave as neuroprotective and neurorepair agents in AD, ² Parkinson’s disease, ³⁹ ALS, ³⁴ Huntington’s disease, ¹⁸ spinocerebellar ataxias, ¹⁶ and also in accidental brain damage (e.g., stroke, brain trauma, spinal injury). ¹³ Examples of this preclinical development in ALS were presented in the lecture and are summarized in Figure 5.

Figure 5.

Molecular and cellular mechanisms underlying the neuroprotective effects of different cannabinoid compounds in amyotrophic lateral sclerosis following data obtained in experimental models of this disease (adapted from ref. 34); eCB = endocannabinoid system.

The major conclusion derived from this lecture was that the experimental evidence collected so far has already reached the necessary biomedical relevance to move toward a prompt clinical validation of cannabinoid-based medicines for these disorders. In this sense, cannabinoid-based clinical trials have already been initiated in some of these disorders or will be initiated soon in others such as ALS using, for example, cannabinoids that selectively activate the CB₂ receptor.

Conclusions

This session highlighted significant advancements in understanding neurodegenerative diseases, showcasing innovative models and emerging therapies, and underscoring the importance of interdisciplinary collaboration. Continued efforts to bridge preclinical findings with clinical applications remain critical to addressing the unmet needs of patients with neurodegenerative diseases.