Local inflammatory response around diffusion chambers containing xenografts: Nonspecific destruction of tissues and decreased local vascularization

Immunoisolation of xenogeneic pancreatic islets within membrane-bound devices has been proposed as an approach to cure diabetes. We examined the local response to implanted xenografts and allografts in comparison with isografts in diffusion chambers with 0.4-μm pore membranes when implanted into epididymal fat pads of rats. These membranes prevented host cell entry into the device but did not prevent passage of large molecules such as IgG and IgM. Well-differentiated allogeneic tissues (Sprague-Dawley rat embryonic lung implanted into Lewis rats) survived for 1 year when implanted in intact devices, but similar tissues were destroyed within 3 weeks when implanted within devices with holes poked in the membrane to allow host cell contact. In contrast, xenografts (CF1 mouse embryonic lung implanted into Lewis rats) were destroyed within 3 weeks even when implanted in devices with intact membranes. The death of the xenogeneic tissues was accompanied by a severe local accumulation of inflammatory cells and a decrease in local vascularization. When isogeneic tissues (Lewis rat embryonic lung implanted in Lewis rats) were mixed with xenogeneic tissues, a local inflammatory response occurred and both iso- and xenogeneic tissues were destroyed within 5 weeks. These results suggest the possibility that xenografts are killed by local accumulation of inflammatory cells, perhaps mediated by the release of antigens from the tissues within the device and presentation by an indirect pathway. The observation that the local response to xenografts is sufficient to kill isografts complicates issues of immunoprotection, suggesting that successful immunoisolation will require membranes that not only provide protection of the encapsulated tissues from the host immune system but also have properties that diminish the release of xenogeneic antigens.