Anti-GluA3 antibodies in frontotemporal dementia: effects on glutamatergic neurotransmission and synaptic failure.

Despite the great effort of the scientific community in the field, the pathogenesis of frontotemporal dementia (FTD) remains elusive. Recently, a role for autoimmunity and altered glutamatergic neurotransmission in triggering disease onset has been put forward. We reported the presence of autoantibodies recognizing the GluA3 subunit of α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors in about 25% of FTD cases. In this study, we evaluated the mechanisms involved in anti-GluA3 autoimmunity, through molecular/neurochemical analyses conducted on patients' brain specimens with frontotemporal lobar degeneration-tau neuropathology. We then corroborated these results in vivo in FTD patients with transcranial magnetic stimulation and glutamate, D-serine, and L-serine dosages in the cerebrospinal fluid and serum. We observed that GluA3 autoantibodies affect glutamatergic neurotransmission, decreasing glutamate release and altering GluA3-containing α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptor levels. These alterations were accompanied by changes of scaffolding proteins involved in receptor synaptic retention/internalization. The above results were confirmed by transcranial magnetic stimulation, suggesting a significant impairment of indirect measures of glutamatergic neurotransmission in FTD patients compared with controls, with further add-on harmful effect in those FTD patients with anti-GluA3 antibodies. Finally, FTD patients showed a significant increase of glutamate, D-serine, and L-serine levels in the cerebrospinal fluid.