Microglial dysfunction and genetic risk for neurodegenerative disease

Neurodegenerative disorders such as Alzheimer’s and Parkinson’s diseases are increasing in prevalence as world populations age. While tremendous progress has been made, our understanding of the mechanisms that underlie the development of these diseases remains far from complete. More troubling, despite the growing emotional and financial toll being taken by neurodegenerative disorders, existing treatment options are limited almost exclusively to those that help manage symptoms but that lack the ability to alter the progression of the disease (Liu et al., 2022). Large-scale genome-wide association studies have revolutionized our understanding of genetic contributions to human disease. While these studies have indicated a central role for genetics in most complex diseases, including those afflicting the central nervous system, delineating the cell types and specific cellular functions that are affected by disease-associated genes and mutations remains a major challenge for the field. Indeed, this knowledge is essential to achieve a deeper understanding of disease etiology and may ultimately hold the key to developing more effective treatments to combat neurodegeneration. One striking finding to emerge from these genetic studies has been the realization that microglia, the innate immune cells of the brain, play key roles in the neurodegenerative process (Hammond et al., 2019). It is known that in addition to the buildup of aggregated proteins and neuron degeneration itself, neuroinflammation is a hallmark of disease that is shared across the large majority of neurodegenerative conditions (Gao et al., 2023). Genetic evidence for the involvement of microglia is particularly clear in Alzheimer’s disease, where almost half of the prioritized risk genes identified so far are expressed more highly in microglia than in any other brain cell type (Penney et al., 2020). Thus, while microglia have long been appreciated to undergo changes in the neurodegenerating brain, genetic findings such as these argue strongly that microglial dysfunction can be a driver, rather than simply a consequence, of neurodegeneration in at least some patients. Mutations affecting the microglia-expressed protein Triggering receptor expressed on myeloid cells 2 (TREM2) have been linked to multiple neurodegenerative disorders (Zhou et al., 2019). Initially, TREM2 deletion mutations were found to cause polycystic lipomembranous osteodysplasia with sclerosing leukencephalopathy, a rare early-onset disease characterized by bone breaks, demyelination, neurodegeneration, and dementia. TREM2 rose to prominence, however, following the identification of heterozygous TREM2 point mutations as risk factors for the development of late-onset Alzheimer’s disease. Carriers of the TREM2 R47H/+ mutation, for instance, have a roughly 3-fold greater risk of developing Alzheimer’s disease than do non-carriers. Though the links have not been firmly established in all cases, TREM2 mutations have also been reported to be associated with the development of frontotemporal dementia, Parkinson’s disease, and amyotrophic lateral sclerosis (Zhou et al., 2019). Thus, identifying the specific cellular changes that arise due to disease-associated TREM2 mutations is likely to be instructive in pinpointing the microglial functions and dysfunctions that contribute to the development of neurodegenerative disease more generally. While considerable efforts have been made to this end using multiple model systems, there remains debate about the effects that different TREM2 mutations have on microglia (Hammond et al., 2019). This is particularly true for heterozygous point mutations such as that causing the TREM2 R47H/+ amino acid substitution. Indeed, there is still no consensus as to whether the R47H mutation causes TREM2 loss-of-function or if the effects are more complex (Zhang et al., 2023). This uncertainty is likely due in part to phenotypic variability that arises from cross species differences in cell and protein function. The mouse and human TREM2 proteins exhibit barely 50% amino acid identity, much less than the proteins encoded by most other Alzheimer’s disease-associated genes (Penney et al., 2020). Furthermore, microglia as a cell type exhibit considerably greater cross-species transcriptional and protein-coding differences than other brain cell types (Penney et al., 2020; Pembroke et al., 2021). These observations underscore the importance of studying human disease using a human model system, and in particular the value of studying microglia and microglial risk genes such as TREM2 using human cells (Penney et al., 2020). Rodent models have long been the workhorses of human disease research and have made tremendous contributions to our understanding of all major neurodegenerative diseases. These models possess complex brain circuitry and a full complement of cell types and sub-types whose interactions have been sculpted through a lifetime of development. Well-established disease models also allow for the assessment of disease pathologies and behavioral outcomes. However, cross-species differences in cell and protein function and the need for humanized and/or overexpressed proteins in most transgenic rodent models belie weaknesses that are thought to contribute to the poor translatability of therapeutics from animal studies to human patients (Penney et al., 2020; Pembroke et al., 2021; Liu et al., 2022). Human induced pluripotent stem cell (iPSC) technologies have developed rapidly in recent years, and now allow for iPSC differentiation to all of the main cell types of the brain, as well as for culturing these cells in co-culture and 3D culture systems of varying complexity (Penney et al., 2020). Further, gene editing techniques, and/or the use of patient-derived cells, facilitate the study of specific disease-associated mutations in human brain cells in vitro. Indeed, 3D culture models carrying disease-associated mutations can often recapitulate hallmark protein aggregation and other disease pathologies without the need for overexpression and on shorter timescales than rodent models. Stem cell systems also hold promise for personalized and regenerative medicine, as well as providing a platform for drug screening (Penney et al., 2020). Using these models, we and others have examined the effects of TREM2 R47H/+ mutations in human iPSC-derived microglia (Penney et al., 2024, and the references therein). Our recent study used CRISPR/Cas9 gene editing to generate TREM2 R47H/+ and isogenic control iPSC lines, followed by characterization of R47H/+-dependent effects in iPSC-derived microglia monocultures, neuron-microglia co-cultures, and xenotransplant mouse models (Figure 1; Penney et al., 2024).Figure 1: TREM2 R47H/+-dependent alterations in human microglia.Alterations of key microglial functions by the TREM2 R47H/+ mutation include (1) reduced uptake of disease-associated substrates such as amyloid-β peptides, myelin debris, and synaptic debris, (2) impaired injury sensing and response, (3) a pro-inflammatory gene expression signature and hyper-responsiveness to inflammatory challenge, and (4) upregulation of complement gene expression and other processes leading to increased synapse loss. Created with BioRender.com. ADP: Adenosine diphosphate; ATP: adenosine triphosphate; C1Q: complement component 1Q; C3: complement component 3.Gene expression profiling showed that the TREM2 R47H/+ mutation causes considerable transcriptional changes in microglia, with more than 1000 genes being differentially expressed between control and TREM2 R47H/+ cells. By far the most prominent transcriptional changes were those affecting immune activation and inflammation-related pathways, even though the cells were cultured without any type of exogenous challenge. Based on these observations, we next examined whether TREM2 R47H/+ microglia were hyper-responsive to inflammation-inducing agents. We found that following treatment with the cytokine interferon γ, or the bacterial cell wall component lipopolysaccharide, TREM2 R47H/+ microglia secreted significantly more of the inflammatory cytokines interleukin 6 and tumor necrosis factor α than did control microglia carrying wild-type TREM2 protein. At the transcriptional level, gene expression changes and cytokine gene induction following interferon γ and lipopolysaccharide treatments were also exaggerated in TREM2 R47H/+ microglia (Penney et al., 2024). Importantly, similar pro-inflammatory changes were found in microglia from human patients carrying TREM2 R47H/+ mutations, reinforcing the model system, as well as the pro-inflammatory effect of this mutation (Sayed et al., 2021). Mediating brain inflammatory responses is a key function of microglia, and as noted, microglia-mediated neuroinflammation is a pervasive feature present across neurodegenerative disorders (Hammond et al., 2019). In part, this is due to the pro-inflammatory nature of many proteins/peptides that form hallmark disease aggregates, such as amyloid beta (Aβ), the main constituent of senile plaques in Alzheimer’s disease, and alpha-synuclein, the primary component of Lewy bodies in Parkinson’s disease (Zhang et al., 2023). A number of other factors, including neuronal damage, myelin degradation, oxidative stress, and pro-inflammatory factors from additional brain and/or peripheral sources, can also lead to microglial activation during neurodegenerative disease. Given this array of inflammatory challenges faced by microglia, genetic changes that bias microglia towards hyper-activated or hyper-responsive states would be expected to pre-dispose individuals to the development of disease. Consistently, pro-inflammatory phenotypes have been commonly observed in microglia carrying mutations associated with increased neurodegenerative disease risk. In addition to the TREM2 R47H/+ mutation, the prevalent Alzheimer’s disease risk factor APO-ε4, and the key Parkinson’s disease mutation LRRK2 G2019S, to name just a few, have been shown to promote a pro- inflammatory state in microglia (Lin et al., 2018; Panagiotakopoulou et al., 2020; Zhang et al., 2023). Beyond initiating brain inflammatory responses due to protein aggregates, cellular damage, or exogenous challenges, a major microglial function is to detect and remove unwanted debris from the brain milieu (Gao et al., 2023). These sensing, uptake and degradation functions are important to maintain homeostasis of the central nervous system environment. We thus examined whether such processes were altered by the TREM2 R47H/+ mutation. Control and TREM2 R47H/+ microglia in culture were presented with recombinant fluorescently labeled Aβ, or synaptic or myelin debris isolated from mouse brains. Immunostaining demonstrated the internalization of each substrate by microglia of both genotypes. We then used flow cytometry to more carefully quantify microglial uptake of fluorescent Aβ, as well as synaptic and myelin debris that was labeled with the pH-sensitive dye pHrodo. These experiments demonstrated that TREM2 R47H/+ microglia exhibited an impaired ability to uptake each of these neurodegeneration-relevant substrates (Penney et al., 2024). Thus, in addition to biasing microglia to a pro-inflammatory phenotype, the TREM2 R47H/+ mutation also appears to impair the homeostatic debris removal functions of microglia for multiple major brain substrates. Responding to tissue injury is another key function of microglia in the brain. In a common in vivo injury model, microglia are monitored by live imaging and respond quickly and robustly to sites of laser-induced brain injury (Haynes et al., 2006). We sought to recapitulate this microglia injury response model in vitro using iPSC-based neuron-microglia co-cultures. By overlaying co-cultures with Matrigel prior to laser-induced injury, we were able to reliably detect and monitor microglial movement towards injury sites over time. These experiments revealed that microglia carrying TREM2 R47H/+ mutations have an impaired ability to sense and/or move toward laser-induced injury sites. While live-imaged TREM2 R47H/+ microglia also exhibited small but significant reductions in movement per se, the impairments in injury-induced movement were qualitatively much more severe (Penney et al., 2024). This observation suggested to us that the primary defect leading to reduced injury response by TREM2 R47H/+ microglia was likely a reduced ability to sense injury-related cellular damage. The best-characterized mechanism of acute injury sensing by microglia involves the release of ATP (adenosine triphosphate) by damaged neurons, its rapid conversion to adenosine diphosphate and subsequent sensing by the microglial purinergic receptor P2RY12 (Haynes et al., 2006). Consistent with impaired injury sensing via this mechanism, we observed that P2RY12 transcript and protein levels were both reduced in TREM2 R47H/+ microglia compared to controls (Penney et al., 2024). Thus, the R47H/+ mutation also appears to disrupt the homeostatic sensing and injury response functions of microglia through downregulated P2RY12 signaling. To gain further insight into the interactions between TREM2 R47H/+ microglia and other brain cell types, we then undertook xenotransplant experiments to introduce our iPSC-derived microglia into mouse brains. Following transplantation in the early post-natal period, adult mice were sacrificed to examine iPSC-microglia- and/or TREM2 R47H/+-dependent effects. We found no differences in the levels or organization of the myelin marker myelin basic protein in mouse hippocampi due to the presence of either control or TREM2 R47H/+ iPSC-microglia. In contrast, we did observe reduced hippocampal levels of the pre-synaptic marker synaptophysin in mice injected with TREM2 R47H/+ microglia compared to those that received either control microglia or injections of PBS only. To determine whether the reduced synaptophysin levels corresponded to fewer synapses in these mice, we performed co-staining for the pre- and post-synaptic markers vesicular glutamate transporter 1 and postsynaptic density protein 95, confirming that mice transplanted with TREM2 R47H/+ microglia did indeed exhibit a reduced density of co-localized synaptic puncta in the hippocampus (Penney et al., 2024). Synaptic pruning by microglia is an important developmental process that is impaired in mice lacking TREM2. Synaptic pruning can also be inappropriately activated in the adult brain in a number of pathological states, including neurodegeneration. The complement system is one major signaling axis involved in both developmental and pathological synaptic pruning (Hammond et al., 2019). Intriguingly, we found that multiple secreted complement components, including C1QA, C1QB, C1QC, C2, and C3, as well as the complement receptors CR1 and C1QR1, were all upregulated in microglia carrying the TREM2 R47H/+ mutation (Penney et al., 2024). In addition to this complement activation, the pro-inflammatory effect of the TREM2 R47H/+ mutation, and an upregulation of the purinergic receptor P2RY6 in R47H/+ microglia, also have the potential to promote synapse loss (Penney et al., 2024). Whether by one or a combination of these mechanisms, our observations indicate that microglia carrying the TREM2 R47H/+ mutation can promote synapse elimination in the mammalian brain, likely contributing to the disease process. While only based on one neurodegeneration-related mutation, these findings are instructive in focusing attention on key microglial processes that when perturbed can contribute to the development of neurodegenerative disease (Figure 1). The presence of the TREM2 R47H/+ mutation in human microglia appears to simultaneously impair homeostatic microglial functions, while also promoting toxic disease-associated functions such as inflammation and synaptic pruning. Thus, this work argues that TREM2 R47H/+ is not simply a loss-of-function mutation, but rather that it exerts both gain- and loss-of-function effects on human microglia (Penney et al., 2024). Furthermore, we believe that studies like this, using iPSC models to disentangle the effects of disease-associated mutations, are critical to understanding the mechanisms underlying disease development and may ultimately be key in developing more effective treatments for neurodegenerative disease. This work was supported by a Canada Research Chair award to JP. Presentation at a meeting: Alzheimer’s Association International Conference 2019, Los Angeles, CA, USA. C-Editors: Zhao M, Liu WJ, Qiu Y; T-Editor: Jia Y

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.