Blockade of phagocytosis and selective elimination of macrophages (m phi s) are generally accepted procedures for gaining knowledge about the function of m phi s in vivo. This study demonstrates that intravenous injection of gadolinium chloride (GdCl3) not only blocks phagocytosis by rat liver m phi s (Kupffer cells) but also selectively eliminates the large m phi s situated in the periportal zone of the liver acinus. Repopulation of m phi s starts at 4 days after injection. During repopulation, m phi s are less vulnerable to GdCl3. When repopulation is complete, the new m phi s show the same vulnerability as the original ones. Splenic m phi s are less vulnerable to GdCl3 because only some of the red pulp m phi s transiently disappear. The white pulp m phi s are not affected. Repopulation occurs sooner than in liver. These results indicate that administration of GdCl3 is a suitable approach to studying the in vivo function of large Kupffer cells.

The presence and distribution in the peri-insular region of extracellular matrix, and in particular basement membrane, was investigated in a comparative study comprising pancreata of rat, dog, pig, and man. Basement membrane markers, collagen type-IV and laminin, were determined immunohistochemically. Additional information pertaining to the structural relationships between endocrine and exocrine pancreas, in particular cell-to-cell and cell-to-matrix contacts, was obtained by electron microscopy. In pig, very little peri-insular capsule is present, and the structural integration of the porcine islet in the exocrine pancreas almost exclusively depends on cell-to-cell adhesion. In the canine pancreas, the islets are almost completely encapsulated with very little direct exocrine-to-endocrine cell-to-cell contact. In rat and man, the situation is intermediate with a tendency towards predominance of cell-to-matrix adhesion. The intra-insular adhesion mechanisms depend largely on cell-to-cell adhesion in all four species. The ultrastructural results suggest that collagenase preparations employed in islet isolation procedures should be of high purity as to preserve the protease-sensitive intra-islet cell-to-cell adhesion. Under these conditions, however, the endocrine-to-exocrine cell-to-cell contacts will be conserved also, resulting in an exocrine-tissue contamination of the islets of Langerhans. Consequently, additional steps for the effective removal of exocrine tissue and the purification of islets are required.

Human peripheral blood mononuclear cells (PBMCs) were incubated with large unilamellar vesicles (LUV) containing encapsulated dextran-magnetite particles (DMP). This resulted in an efficient incorporation of DMP. Electron microscopy revealed the presence of DMP in cells mainly in phagosomes and secondary lysosomes. DMP-labeled PBMCs showed a strong increase of the transverse relaxation rate (up to 16.6 s-1 for 5 x 10(7) cells/ml) and, accordingly, a great loss of signal intensity in MR imaging. The fraction of DMP containing PBMCs could be enriched by magnetic cell separation. The major population of the DMP containing cells proved to be monocytes. When PBMCs depleted of monocytes were used for labeling, DMP uptake was observed also in the peripheral blood lymphocytes. The labeling of PBMCs presented here may be used in future studies of selective MR imaging of in vivo cell migration in a variety of immunologically compromised tissue states, e.g., tumors, transplantations, and abscesses.

Human lymphocytes were labeled with biotinylated anti-lymphocyte-directed monoclonal antibodies, to which streptavidin and subsequently biotinylated dextran-magnetite particles were coupled. This labeling resulted in a strong and selective negative contrast enhancement of lymphocyte suspensions at 2.0 T, caused predominantly by the specific increase of R2 with a small but significant specific increase of R1. The R1 was found to decrease with increasing field strength. The immunolabeling procedure described here may be used for the selective signal depletion of target cells in MR imaging.

No abstract