Fig. 8.
Renal pathology following the transfer of AQP-4296–304-specific T cells in combination with AQP-4 antibody containing NMO-IgG. In the intact kidney (a–c), AQP-4 positive collecting ducts (brown) are essentially absent in the cortex, appear at the cortico-medullary junction (a), and are found in high numbers in the renal papilla (a). Intact collecting duct cells show a basolateral staining pattern (b, c), which is much stronger, when the commercial anti-AQP-4 antibody was used (b brown) than it was, when anti-AQP-4 antibody containing NMO-IgG was utilized (c red). d–e Peri-ductal inflammation at the renal cortico-medullary junction following transfer of AQP-4296–304-specific T cells in combination with AQP-4 antibody containing NMO-IgG. Consecutive sections of the cortico-medullary junction, stained with HE (d) and commercial anti-AQP-4 (e brown). Inflammation was not seen in the renal papilla, although collecting ducts at this site expressed high levels of AQP-4 (f brown). g Flow cytometry analysis revealed that binding of AQP-4 antibody containing NMO-IgG (gray) to the surface of collecting duct cells (MDCK cells) was below the limit of detection in the absence of IFN-γ. In the presence of IFN-γ, the availability of the antibody-reactive AQP-4 epitope increased, and AQP-4 antibody containing NMO-IgG could readily bind. In both cases, binding of Subcuvia (red) served as negative control