A New Perspective on Synaptic Pruning · Soyon Hong

2026-06-09 · A faithful, transcript-grounded reading by PodLens

Source paper:https://pmc.ncbi.nlm.nih.gov/articles/PMC5479435/pdf/nihms863063.pdf

What This Paper Is About

This paper delves into the role and mechanisms of microglia-mediated synaptic pruning under physiological and pathological conditions. The authors Soyon Hong, Lasse Dissing-Olesen, and Beth Stevens propose that in the early stages of neurodegenerative diseases such as Alzheimer's disease (AD) (before the formation of amyloid plaques), the classical complement cascade pathway from development is aberrantly reactivated, leading to pathological synapse loss and thereby driving cognitive decline. The paper aims to shift the focus of AD research from late-stage plaque clearance to early-stage synaptic protection, elucidating the specific molecular pathways of the complement system and key molecules like Trem2 in microglia-mediated synapse elimination.

Paper Skeleton

• Problem Definition: Synapse loss is a key pathological feature in early AD, strongly correlated with cognitive decline, and precedes the accumulation of amyloid plaques (Aβ plaques). However, the contribution of microglia to synaptic degeneration in the very early, pre-plaque stages has been little explored in previous studies.
• Core Claim: The physiological synaptic pruning mechanism mediated by C1q and C3 complement protein tagging and the CR3 receptor in the developing brain is aberrantly reactivated in the adult and aging AD brain, triggering pathological synapse loss.
• Method of Argumentation: This review integrates multi-dimensional evidence from recent years, including dynamic microglial imaging, developmental biology, knockout mouse models, and human genome-wide association studies (GWAS), establishing a logical chain of argument from physiological microglial mechanisms to pathological activation in AD.
• Key Evidence:
• Dynamic Surveillance: Two-photon imaging confirms that in the healthy adult brain, microglia continuously monitor the synaptic microenvironment relying on their highly dynamic processes. (Dynamic microglia–synapse interactions in the healthy brain · "microglia in the healthy brain have highly dynamic processes")
• Complement Tagging: During development, classical complement proteins C1q and C3 are widely expressed and localized on subsets of immature synapses, tagging synapses to be pruned. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "classical complement cascade proteins, C1q and C3, are widely expressed")
• Genetic Association: GWAS association analyses of AD have identified key microglial and complement-related risk factors such as Trem2, CR1, ApoJ/Clusterin, and CD33. (Introduction · "Trem2, CR1, ApoJ/ Clusterin, CD33")
• Model Validation: In mouse models, knocking out C1qa reduced AD-related neuronal damage and gliosis, whereas knocking out C3 increased plaques and exacerbated late-stage degeneration. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "Genetic deletion of C1qa in an AD mouse model resulted in less plaque-related neuronal damage and gliosis")
• Boundaries & Limitations:
• The specific triggering events that cause dysfunction of the classical complement cascade in early AD (before plaques appear) remain unclear. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "it is not yet known whether complement is dysregulated early in the AD brain")
• The role of microglia in regulating synapse formation and elimination across different brain regions (such as the hippocampus), developmental periods, and behavioral contexts is highly context-dependent. (Dynamic microglia–synapse interactions in the healthy brain · "suggesting that microglia regulate synaptic formation and/or elimination depending on context")

Key Arguments List

1. Synapse loss is the strongest correlate of cognitive impairment in AD: Synapse loss in the hippocampus and association cortices correlates significantly more strongly with cognitive impairment than the load of amyloid plaques or neurofibrillary tangles, serving as a very early pathological hallmark that distinguishes AD from normal aging. - Anchor: AD: a disorder of degenerating selected synapses · "synapse loss in the hippocampus and association cortices serves as a much stronger correlate of cognitive impairment in AD" - Type: Fact
2. Soluble Aβ oligomers are the core synaptotoxic mediators: They can cause a significant reduction in synapse number before plaque formation and are the primary synaptotoxic active components. - Anchor: AD: a disorder of degenerating selected synapses · "soluble oligomers of Aβ, and not plaques per se" - Type: Claim
3. Physical barrier role of microglia on plaques: Microglia surrounding plaques appear to form a physical barrier, limiting plaque growth and potentially preventing the diffusion of synaptotoxic Aβ oligomers from the plaques. - Anchor: Microglia and their potential roles in Aβ plaque maintenance · "microglia that surround plaques appear to constitute a barrier" - Type: Claim
4. Dynamic surveillance of microglia under physiological conditions: In the healthy brain, microglia are not in a resting state but continuously monitor and sense the surrounding synaptic microenvironment through highly dynamic processes. - Anchor: Dynamic microglia–synapse interactions in the healthy brain · "microglia in the healthy brain have highly dynamic processes" - Type: Fact
5. Classical complement cascade localizes and tags synapses: In the healthy developing brain, C1q and C3 proteins localize on immature synapses, serving as tags to target synapses for elimination. - Anchor: Possible mechanisms of microglia–synapse dysfunction in the AD brain · "classical complement cascade proteins, C1q and C3, are widely expressed" - Type: Fact
6. Activity-dependent synaptic engulfment: Neuronal electrical activity regulates microglial engulfment of synapses, with microglia preferentially phagocytosing less active or inactive synaptic inputs. - Anchor: Dynamic microglia–synapse interactions in the healthy brain · "microglia preferentially phagocytosed less active inputs" - Type: Fact
7. Hypothesis of pathological activation of developmental synaptic pruning mechanisms in AD: The normal synaptic pruning pathway from developmental stages may be aberrantly upregulated in the early stages of AD and other neurodegenerative diseases, thereby mistakenly eliminating mature, healthy synaptic connections. - Anchor: Possible mechanisms of microglia–synapse dysfunction in the AD brain · "normal developmental pruning pathway becomes aberrantly upregulated" - Type: Prediction - Uncertainty: Deeper in vivo experiments are still needed to elucidate the exact triggers and pathological causal relationships of this aberrant upregulation.
8. Role of Trem2 in microglial survival and synaptic maturation maintenance: The Trem2 gene is crucial for the normal survival of microglia and the maintenance of synaptic connectivity; its dysfunction severely impairs synaptic function. - Anchor: Possible mechanisms of microglia–synapse dysfunction in the AD brain · "TREM2 may play an important role for maturation and maintenance" - Type: Claim
9. Model of aberrant neural network activity regulating regional vulnerability: The vulnerability of synapses and neural networks in specific brain regions is related to local electrical activity. Aberrant activity not only promotes Aβ production but also alters microglial synaptic pruning dynamics through activity-dependent mechanisms, jointly causing region-specific vulnerability. - Anchor: Role of microglia in region-specific network dysfunction · "aberrant neuronal activity through genetic or other causes" - Type: Prediction

Plain English Explanation

If you think of the brain as a massive city power grid, neurons are the wires, and the nodes where they connect to each other are called "synapses." In the early stages of brain development, an excess of connections is produced. To make the grid run more efficiently, the brain has a sophisticated set of "cleaners"—microglia. These cells are like smart patrol sweepers that constantly roam the healthy brain, selecting nodes based on their frequency of use: nodes that frequently have current passing through (active) are kept, while nodes with no current for a long time (inactive) are tagged with a yellow "eat me" label (namely, complement proteins C1q and C3) and then engulfed and cleared away by the sweepers.

What is this paper fighting against? In traditional AD research, scientists often focus their attention on the massive late-stage garbage dumps—"amyloid plaques"—and the reactive microglia hovering around them. However, the authors Soyon Hong, Lasse Dissing-Olesen, and Beth Stevens suggest we might have gotten our priorities wrong. They point out that before these giant plaques even form, synaptic nodes in the patient's brain have already begun to disappear in large numbers, which is the real culprit behind memory decline.

The authors propose a striking hypothesis: in early AD, soluble Aβ oligomers (a toxic protein) act like a destructive electrical disturbance, awakening the "synaptic pruning" program that was supposed to go into a low-profile mode after development ended. The cleaners, which were originally meant only for discarded wires, start frantically dismantling healthy, mature, and normal wire nodes due to the abnormal activation of the complement system.

Furthermore, the study found that genetic mutations like Trem2 disrupt the survival and recognition logic of microglia themselves, preventing them from functioning as healthy patrol cars. Therefore, to treat AD, the therapeutic focus must shift forward from "clearing plaques" to "how to protect synapses" and "how to prevent microglia from being abnormally activated in the early stages and mistakenly dismantling the normal power grid."

Glossary

• Microglia: Resident immune cells in the brain that play the roles of "patrol guards" and "cleaners," responsible for engulfing and clearing damaged neurons, redundant synapses, and pathological debris. (Abstract · "microglia prune developing synapses")
• Synaptic Pruning: The physiological process during brain development of eliminating redundant or non-functional synaptic connections to optimize neural network efficiency. (Abstract · "microglia prune developing synapses")
• Classical Complement Cascade: An immune defense system borrowed in the brain as a "synaptic localization tag," where C1q and C3 proteins can tag synapses that need to be cleared. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "classical complement cascade proteins, C1q and C3, are widely expressed")
• Soluble Oligomers of Aβ: The free polymeric form of amyloid protein, highly neurotoxic and synaptotoxic, which begins disrupting synaptic connections even before plaque formation. (AD: a disorder of degenerating selected synapses · "soluble oligomers of Aβ, and not plaques per se")
• Trem2: A receptor gene expressed on the surface of microglia, mutations in which greatly increase the risk of AD; it is responsible for regulating microglial survival, metabolism, and synaptic maintenance. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "TREM2 may play an important role for maturation and maintenance")

Before and After This Paper

• Before This Paper: The scientific community generally believed that the role of microglia in AD was primarily as secondary reactive cells responding to amyloid plaques in the mid-to-late stages of the disease. Research mostly revolved around how they cluster around plaques, whether they clear plaques, or whether they release inflammatory factors to exacerbate brain damage. Very little was known about how synapses are lost before plaques appear and whether microglia are involved. (AD: a disorder of degenerating selected synapses · "whether microglia play a role in these early, pre-plaque stages of pathogenesis and how they may impact synaptic function remains little explored")
• After This Paper: The paper successfully introduced the concept of synaptic pruning from brain development into the early pathogenesis of neurodegenerative diseases. This shift in perspective drove a large amount of early drug development targeting C1q and C3 complement inhibitors as well as Trem2 agonists, prompting the scientific community to start looking for interventions to protect synapses in the very early stages before plaques appear. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "An important question for future investigation is whether the normal developmental pruning pathway becomes aberrantly upregulated to mediate synapse loss by microglia in early stages of AD")

Most Worth-Reading Sections

1. On the dynamic process of synaptic surveillance in the healthy adult brain: This passage breaks the traditional bias that microglia are "resting/idle" in a disease-free state, describing an extremely active scene where they are constantly in brief contact with synapses. (Dynamic microglia–synapse interactions in the healthy brain · "microglia in the healthy brain have highly dynamic processes and continually survey their local microenvironment")
2. The mechanism of activity-dependent synaptic engulfment: This demonstrates how the brain selectively eliminates weaker connections based on neural electrical activity, which is key to understanding the physical mechanism of "use it or lose it" in neural networks. (Dynamic microglia–synapse interactions in the healthy brain · "Interestingly, neuronal activity regulated the microglial engulfment of synaptic input, where microglia preferentially phagocytosed less active inputs")
3. The proposal of the roles of C1q/C3 in normal aging and early AD: This section deduces in detail the chain of how abnormal upregulation of the complement system leads to synaptic degeneration in the hippocampus, serving as the core basis for the entire paper's hypothesis. (Possible mechanisms of microglia–synapse dysfunction in the AD brain · "Interestingly, C1q is highly upregulated and deposits on synapses with normal aging in human and mouse brains, particularly in the hippocampus, one of the brain regions most vulnerable to synapse loss in AD")

Resonances with past episodes

• Extension→ Active Forgetting and Neuropsychiatric Diseases · Jacob A. Berry
  The previous study pointed out that microenvironmental networks such as glial cells play a key regulatory role in mediating brain forgetting and memory elimination. This study extends on the microscopic mechanism, specifically elucidating how microglia specifically engulf and clear weaker synaptic connections by recognizing neuronal electrical activity.
  ThisDynamic microglia–synapse interactions in the healthy brain · "microglia preferentially phagocytosed less active inputs" Neuronal electrical activity regulates microglial engulfment of synapses, with microglia tending to preferentially clear weaker or inactive synaptic inputs.

  RelatedSUMMARY AND PERSPECTIVES · "heavily influenced by non-engram brain cells" Forgetting in the brain is not something that occurs in isolation within a single cell, but is highly controlled by microenvironmental networks such as non-engram cells, newly generated neurons, and glial cells.

A faithful reading and plain-language retelling of the paper, generated by PodLens.

This is one source-grounded reading, not a replacement for the original. Every point is anchored to its source, so you can check it yourself — and corrections are welcome.