Abstract
Mounting evidence challenges the view that antibodies reach the brain solely via vascular leaks. Neurons, astrocytes and microglia appear able to assemble restricted and context-dependent immunoglobulin repertoires via cryptic V(D)J recombination, splice-and-link RNA editing and retroelement-assisted rearrangements. These endogenous antibodies have been associated with complement-mediated pruning, receptor trafficking and astrocyte-neuron metabolic coupling and may change after injury. However, causal evidence from in vivo, cell-type-specific gain- or loss-of-function models remains sparse. Accordingly, we synthesise current evidence and outline testable, conservative hypotheses about when and how neural immunoglobulins might influence CNS function and disease.