Anti-interneuron antibodies in rapid-onset pediatric OCD: clinical generalization and target identification

ABSTRACT Obsessive-compulsive disorder affects 1-4% of children and produces substantial morbidity and disruption of normal development. In some children, rapid onset and a characteristic set of accompanying symptoms suggest a unique pathophysiology; this has been termed ‘Pediatric Acute-Onset Neuropsychiatric Syndrome’, or PANS. PANS onset often occurs following an infectious illness, suggesting that some cases of rapid-onset pediatric OCD are triggered by neuroinflammatory processes. Specifically, it has been proposed that certain infections can, in a susceptible child, lead to the production of autoantibodies that cross-react with brain epitopes and lead to basal ganglia inflammation, and thereby to OCD symptomatology. The targets of these autoantibodies remain unclear, and the hypothesis of autoimmunity as a cause of PANS is controversial. In recent work we used a novel approach to characterize the binding of antibodies from children with PANDAS (Pediatric Autoimmune Neuropsychiatric Disorder Associated with Streptococcus), a form of PANS in which symptoms are temporally associated with Streptococcal infection, to brain tissue. We have discovered, and multiply replicated, that IgG from children with PANDAS binds to specific interneurons within the caudate- putamen: the cholinergic interneurons, or CINs. CIN binding by PANDAS-associated IgG reduces their activity in two ex vivo assays. Both effects – IgG binding to CINs and reduction in their activity – are lost in children who exhibit symptomatic improvement following immune-modulating therapy. CIN pathology has been independently associated with Tourette syndrome, and we have found CIN disruption to lead to repetitive behavioral pathology in mice. These convergent findings give inherent plausibility to the novel hypothesis that antibody binding to and consequent inhibition of CINs contributes to the pathophysiology of PANDAS. Here we test and extend this analysis in critical ways. First, we will use the novel REAP screening methodology, based on yeast surface display of >3000 human proteins in their native conformation, to identify candidate antibody targets. We will validate these candidates by testing whether antibodies against them can bind to and inhibit CINs; pilot work has already identified three such candidates, and new data confirm CIN binding by one of them. Second, by examining sera from >350 children drawn from specialty treatment centers across the country, we will test whether CIN binding is specific to PANDAS or is seen more generally in PANS, or even in non-PANS OCD. Finally, we will culture individual B cells from selected children with PANDAS and PANS and will screen them for production of monoclonal antibodies against identified targets. This will, for the first time, allow us to identify and clone specific candidate autoantibodies associated with these conditions. Together, these innovative studies have the potential to fundamentally advance our understanding of PANS and PANDAS, paving the way for a pathophysiology-based nosology and for the development of new diagnostic tools and treatments.