Modeling Alzheimer's disease using human cell derived brain organoids and 3D models

Abstract

Age-related neurodegenerative diseases, like Alzheimer's disease (AD), are challenging diseases for those affected with no cure and limited treatment options. Functional, human derived brain tissues that represent the diverse genetic background and cellular subtypes contributing to sporadic AD (sAD) are limited. Human stem cell derived brain organoids recapitulate some features of human brain cytoarchitecture and AD-like pathology, providing a tool for illuminating the relationship between AD pathology and neural cell dysregulation leading to cognitive decline. In this review, we explore current strategies for implementing brain organoids in the study of AD as well as the challenges associated with investigating age-related brain diseases using organoid models.

Keywords: APOE; Alzheimer’s disease; aging; amyloid beta; brain organoids; cerebral organoids; hyperphosphorylated tau; neurodegenerative diseases.

Figure 1

Schematic diagram illustrating the generation of brain organoid models of AD. Brain organoids generated from patient or control donor somatic cells will retain the genetic background of the patient or donor. Most, if not all, age-related epigenetic signatures from the donor will be lost during the reprogramming of somatic cells to iPSCs. Genetic modifications associated with AD can be introduced into iPSCs or ESCs using techniques including lentivirus overexpression or CRISPR induced point mutations and gene knockouts. Genome edited PSCs can then be employed in the generation of embryoid bodies which could be patterned to obtain more directed, brain region specific organoids or unguided approaches can be implemented to yield more heterogeneous cerebral organoids. Following organoid cellular maturation, brain organoids could be treated with factors to simulate age-related events like BBB leakage, to induce AD pathology, to observe the effects of potential pharmaceuticals on AD pathology, or to inhibit mechanisms hypothesized to effect AD pathogenesis. These strategies have successfully established brain organoids with some AD pathological features, including Aβ aggregates, NFTs, increased apoptosis, and neuronal network dysregulation. fAD, familial Alzheimer’s disease; sAD, sporadic Alzheimer’s disease; DS, Down syndrome; iPSCs, induced pluripotent stem cells; ESCs, embryonic stem cells; Aβ, amyloid-β; BBB, blood brain barrier. Not to scale.

Figure 2

Schematic diagram outlining AD related cellular pathology and possible readouts to measure pathological effects. (A) As brain organoids mature, dendritic spines begin to form in neurons and functional synapses spontaneously fire. Aβ aggregates, increased p-tau and NFTs, decreased PSD95, and increased cleaved caspase-3 have all been observed in brain organoid models of AD. (B) Myelinating oligodendrocyte populations have been identified in unguided brain organoids after long-term culture or as early as 50 days using directed brain organoid protocols. It might be possible to observe the demyelination of axons and transcriptional or protein level regulation of this process in AD brain organoid models. (C) Astrocytes have been shown to increase inflammatory cytokine secretion in the presence of AD pathology. The reactivity of subpopulations of astrocytes in either unguided or astrocyte enriched brain organoid protocols can be studied in the presence of AD pathology. (D) Tools to observe cellular responses to AD pathology and the dysregulation of cellular processes in AD brain organoid models include scRNA-seq, ELISA, Western blot, flow cytometry, and multielectrode array technologies. Not to scale.

Figure 3

Healthy and diseased microglial processes implicated in AD which may be modeled in brain organoids. (A) Homeostatic microglia in a resting state in the healthy brain. Microglia participate in maintaining brain homeostasis, including synaptic pruning, surveillance of the microenvironment, and immune regulation. In response to stimuli or pathological conditions, homeostatic microglia transform into activated response microglia that have alterations in gene expression, morphology, and function. Microglial activation can promote an increase of cytokines such as tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL-1β), chemokine ligand 2 (CCL2), interleukin-8 (IL-8) and interleukin-6 (IL-6). (B) Neuronal synaptic pruning. Microglia perform synaptic pruning by engulfing synaptic components to eliminate unnecessary or pathogenic synapses throughout development and disease. This process is essential for optimizing brain function. However, in AD, synaptic pruning may become dysregulated and could lead to excessive pruning leading to impairments in neuronal communication and contributing to cognitive decline. (C) DAMs vs LDAMs. Disease-associated microglia (DAMs) refer to microglia characterized by an upregulation of genes involved in phagocytosis, lipid metabolism, and immune response. They are thought to play a role in the clearance of amyloid beta (Aβ) plaques in AD. Lipid droplet accumulating microglia (LDAMs), on the other hand, refer to activated microglia associated with aging, late-stage disease progression, and the presence of lipid droplets. LDAMs may be senescent-like and could contribute to neurodegeneration due to dystrophic morphology and impaired phagocytic machinery reducing their ability to clear Aβ plaques. While both populations may have an increase of lipids, further investigation needs to be done to understand which lipids are accumulating (triglycerides, cholesterol esters, free cholesterol, etc.) (D) TIMs and inflammation. Terminally inflamed microglia (TIMs) may represent an exhausted state for inflammatory microglia that could contribute to AD risk and pathology. TIMs are senescent-like and are a terminal state for activated microglia, shown to be more inflammatory and functionally impaired. Not to scale.