Human immunomodulatory ligand B7-1 mediates synaptic remodeling via the p75 neurotrophin receptor

Cell surface receptors, ligands, and adhesion molecules underlie development, circuit formation, and synaptic function of the central nervous system and represent important therapeutic targets for many neuropathologies. The functional contributions of interactions between cell surface proteins of neurons and nonneuronal cells have not been fully addressed. Using an unbiased protein-protein interaction screen, we showed that the human immunomodulatory ligand B7-1 (hB7-1) interacts with the p75 neurotrophin receptor (p75NTR) and that the B7-1:p75NTR interaction is a recent evolutionary adaptation present in humans and other primates, but absent in mice, rats, and other lower mammals. The surface of hB7-1 that engages p75NTR overlaps with the hB7-1 surface involved in CTLA-4/CD28 recognition, and these molecules directly compete for binding to p75NTR. Soluble or membrane-bound hB7-1 altered dendritic morphology of cultured hippocampal neurons, with loss of the postsynaptic protein PSD95 in a p75NTR-dependent manner. Abatacept, an FDA-approved therapeutic (CTLA-4-hFc fusion) inhibited these processes. In vivo injection of hB7-1 into the murine subiculum, a hippocampal region affected in Alzheimer's disease, resulted in p75NTR-dependent pruning of dendritic spines. Here, we report the biochemical interaction between B7-1 and p75NTR, describe biological effects on neuronal morphology, and identify a therapeutic opportunity for treatment of neuroinflammatory diseases.

Vagal afferent cholecystokinin receptor activation is required for glucagon-like peptide-1-induced satiation

Peripheral glucagon-like peptide-1 (GLP-1) and cholecystokinin (CCK) are secreted from enteroendocrine cells, and their plasma concentrations increase in response to eating. While the satiating effect of gut-derived CCK on food-intake control is well documented, the effect of peripheral GLP-1 is less clear. There is evidence that native GLP-1 can inhibit food intake only in the fed state but not in the fasting state. We therefore hypothesized that other gut peptides released during a meal might influence the subsequent effect of endogenous GLP-1 and investigated whether CCK could do so. We found that intraperitoneal injection of CCK in food-restricted mice inhibited food intake during the first 30-minute segment of a 1-hour session of ad libitum chow intake and that mice compensated by increasing their intake during the second half of the session. Importantly, this compensatory behaviour was abolished by an intraperitoneal injection of GLP-1 administered following an intraperitoneal injection of CCK and prior to the 1-hour session. In vivo activation of the free fatty acid 1 (FFA1) receptor with orally administered TAK875 increased plasma CCK concentration and, consistent with the effect of exogenous CCK, we found that prior oral administration of TAK875 increased the eating inhibitory effect of peripherally administered GLP-1. To examine the role of the vagus nerve in this effect, we utilized a saporin-based lesioning procedure to selectively ablate the CCK receptor-expressing gastrointestinal vagal afferent neurones (VANs). We found that the combined anorectic effect of TAK875 and GLP-1 was significantly attenuated in the absence of CCK receptor expressing VANs. Taken together, our results indicate that endogenous CCK interacts with GLP-1 to promote satiation and that activation of the FFA1 receptor can initiate this interaction by stimulating the release of CCK.