C1q and immunoglobulins mediate activity-dependent synapse loss in the adult brain, 2026, Crowley et al.

[SNT Gatchaman]

C1q and immunoglobulins mediate activity-dependent synapse loss in the adult brain

Spoiler: Authors

Gerard Crowley; Minjung Kim; Nathanael O’Neill; Emir Turkes; Fateme Ghasemi; Luca Giudice; Sebastiaan De Schepper; Benjy J Y Tan; Benito Maffei; Laís S S Ferreira; Julie Rebejac; Javier Rueda-Carrasco; Margarita Toneva; John Christian Fajardo; Judy Z Ge; Zhengyue Grace Yang; Paula Korhonen; Phillip Muckett; Damaris Bennett; Camille Paoletti; Tammie T M Sow; David A Posner; Annerieke Sierksma; Dimitra Sokolova; Viktoras Konstantellos; Leen Ali; Kiavash Movahedi; Andrew F MacAskill; Victor L J Tybulewicz; Tarja Malm; Gabriele Lignani; Menna R Clatworthy; Soyon Hong

INTRODUCTION
Synaptic elimination is a fundamental process that can engage microglia and the classical complement cascade, yet the mechanisms that specify which synapses are targeted by complement component 1q (C1q), the initiating component of this cascade, remain incompletely understood. Neuronal activity shapes synaptic connectivity, but whether it interfaces with complement pathways to regulate synapse loss in the adult brain, including in disorders such as Alzheimer’s disease (AD), is unclear. In peripheral tissues, antigen-bound immunoglobulins can recruit C1q to targets, raising the possibility that adaptive immune mechanisms contribute to complement-dependent synaptic elimination in the central nervous system.

RATIONALE
We tested whether neuronal activity specifies complement-dependent synaptic elimination in the adult brain. Further, guided by unbiased spatial transcriptomic analyses, we investigated whether adaptive immune components contribute under conditions of altered circuit activity. To address this, we manipulated neuronal activity in defined hippocampal circuits in adult wild-type mice and examined the impact on synapses and neuroimmune interactions, with additional validation in an AD mouse model.

RESULTS
Inducing neuronal hyperactivity in adult wild-type mice, using chemogenetic and pharmacological approaches, led to region- and input-specific loss of synaptic terminals in a C1q-dependent manner. Conversely, reducing neuronal hyperactivity in an AD mouse model decreased C1q deposition and partially restored synaptic density, supporting a link between activity state and complement-associated synapse loss in both physiological and disease-relevant contexts.

Spatial transcriptomic analyses in the chemogenetically activated wild-type brains identified activity-dependent accumulation of B lymphocyte lineage cells, including antibody-secreting cells, in leptomeningeal niches adjacent to regions of elevated activity. B cell receptor sequencing and photoconversion experiments supported their origin in the dura and localization to activity-modulated hippocampal regions. At the synaptic level, super-resolution imaging demonstrated association of immunoglobulin M (IgM) with C1q-localized synaptic terminals. Genetic perturbation approaches further indicated that secreted IgM and antigen specificity contribute to activity-dependent C1q deposition and synapse loss.

CONCLUSION
These findings identify neuronal activity as a key regulator of complement-dependent synaptic elimination in the adult brain. In addition, they support a contribution of adaptive immune components in the hyperactivity-induced wild-type brain, whereby B lymphocyte lineage cells and IgM participate in C1q-associated synaptic loss. Together, this work supports a model in which activity state interfaces with innate immune mechanisms to mediate synaptic elimination, with adaptive immune components providing an additional modulatory layer.


Web | DOI | PDF | Science | Paywall

 

[SNT Gatchaman]

EDITOR’S SUMMARY
Loss of synapses is a major hallmark of Alzheimer’s disease (AD) and other neurodegenerative disorders. In AD-relevant disease models, deposition of C1q, the central protein in the classical complement cascade, contributes to synapse loss. Crowley et al. showed that neuronal hyperactivity triggers C1q protein deposition at synaptic targeting sites accompanied by accumulation of B cells, resulting in the loss of synaptic components (see the Perspective by Miarka and Prinz). In an AD model already displaying neuronal hyperactivity, dampening the hyperactivity lowered C1q deposition and partially restored loss of synaptic components. Intriguingly, B cell–derived immunoglobulin M participated in activity-dependent synapse loss. These findings provide evidence of neuronal activity as a regulator of C1q-mediated synapse loss in the adult brain and introduce immunoglobulins as key players in this process.

 

[SNT Gatchaman]

Commentary.

Antibodies sculpt adult brain circuits
Lauritz Miarka; Marco Prinz

During development, the mammalian brain forges more neuronal connections, or synapses, than needed. Excess synapses are tagged by the complement system (1), triggering their elimination by microglia, the tissue-resident macrophages (2). The remaining synapses are strengthened and neuronal circuits mature. During neurodegeneration, this developmental program is reactivated and leads to pathological synapse loss and cognitive decline (3). Whether similar mechanisms shape circuits in the adult brain in the absence of disease is not known. On page 166 of this issue, Crowley et al. (4) reframe developmental, complement-mediated synaptic pruning as a constitutive sensor of neuronal hyperactivity in the adult brain. Neuronal hyperactivity recruits antibody-secreting plasma cells to the hippocampus. Locally produced antibodies [immunoglobulins (Igs)] then facilitate complement deposition on synapses as a cue for engulfment by microglia. The finding reveals an unexpected interface between resident innate immunity of the central nervous system (CNS) and peripheral adaptive immunity in synaptic pruning.

Web | DOI | PDF | Science | Paywall

 

[Jonathan Edwards]

I would worry that what they have found is a model artefact. There are a number of elements of the story that do not really add up in terms of what we think we now about human brains and B cells.