Generation and characterization of novel stromal specific antibodies.

Rheumatoid synovial fibroblasts were used as an immunogen to produce monoclonal antibodies selected for their reactivity with stromal cell antigens. Mice were immunised with low passage whole cell preparations and the subsequent hybridomas were screened by immunohistochemistry on rheumatoid synovium and tonsil sections. The aim was to identify those antibodies that recognised antigens that were restricted to stromal cells and were not expressed on CD45 positive leucocytes. A significant number of antibodies detected antigen that identified endothelial cells. These antibodies were further characterised to determine whether the vessels identified by these antibodies were vascular or lymphatic. From five fusions clones were identified with predominant reactivity with: 1) fibroblasts and endothelial cells; or 2) broad stromal elements (fibroblast, endothelium, epithelium, follicular dendritic cells). A fibroblast-specific antibody that did not also identify vessels was not generated. Examples of each reactivity pattern are discussed.

Report on antibodies submitted to the stromal cell section of HLDA8.

The paradigm for tissue specific homing of leukocytes is the "area code" hypothesis, which predicts that a specific combination of adhesive interactions and chemokine signals from the endothelium directs leukocyte migration into specific tissue sites. This area code hypothesis has been supported by studies from previous HLDA workshops where endothelial specific cell antigens have been studied. Similarly, a clear haematopoietic "stem cell code" comprising the chemokine SDF-1 (CXCL12) and the adhesion receptor VCAM-1 (CD106) has been shown to contribute to the stem cell niche within bone marrow [K. Tokoyoda, T. Egawa, T. Sugiyama, B.I. Chai, T. Nagasawa, Cellular niches controlling B lymphocyte behaviour within bone marrow during development, Immunity 20 (2004) 707-718]. HLDA 7 included a section devoted to stem cell antigens, which began to define additional antigens important in these processes. During the course of HLDA 8 we have extended these observations to determine whether a more global stromal address code defined by fibroblasts, exists in variety of different tissues [G. Parsonage, A.D. Filer, O. Haworth, G.B. Nash, G.E. Rainger, M. Salmon, C.D. Buckley, A stromal area postcode defined by fibroblasts, Trends Immunol. 26 (2005) 150-156]. The stromal cell section in HLDA 8 was designed to complement the malignant cell, endothelial cell, and stem cell/progenitor cell sections. Seven new CD numbers were assigned to antibodies included in this section at the HLDA 8 Workshop meeting held during December 2004.

A panel of monoclonal antibodies recognizing GPI-anchored ADP-ribosyltransferase ART4, the carrier of the Dombrock blood group antigens.

ART4 (CD297) is a member of the family of toxin-related ADP-ribosyltransferases (ARTs) and is the carrier of the Dombrock blood group alloantigens (Do). Two mouse monoclonal antibodies (MIMA-52 and MIMA-53), and two rat monoclonal antibodies (N0NI-B4 and NONI-B63) were obtained following immunization of mice with human Do/ART4-transfected cells and of rats with human Do/ART4 cDNA, respectively. All four mAbs recognize Do/ART4-transfected Jurkat cells but not untransfected cells by FACS analysis. Staining of Do/ART4-transfected cells by these mAbs was reduced following treatment of cells with PI-PLC, confirming that Do/ART4 is anchored in the cell membrane by linkage to glycosylphosphatidylinositol as predicted from its amino acid sequence. The four mAbs did not react with Gy(a-) (Dombrock null) erythrocytes but agglutinated other red blood cells. By flow cytometric analysis, all mAbs reacted prominently with erythrocytes, and weakly with peripheral blood monocytes and splenic macrophages, but not with B-lymphocytes or T-lymphocytes. The mAbs reacted weakly also with human umbilical vein endothelial cells and the basophilic leukemia KU-812. Immunohistology revealed staining of epithelia and endothelia on sections of tonsils. In FACS analyses NONI-B4 competed with MIMA-52 for binding to Do/ART4-transfected cells and erythrocytes, whereas NONI-B63 competed with MIMA-53. Neither of the mAbs reacted with mouse ART4-transfected cells, but NONI-B63 and MIMA-53 did react with a mouse/human ART4 chimera, indicating that the epitope recognized by these mAbs lies in the C-terminal half of the protein.