Prevention of murine graft-versus-host disease and bone marrow alloengraftment across the major histocompatibility barrier after donor graft preincubation with anti-LFA1 immunotoxin

We have investigated the effects of the vitro depletion of LFA1 positive cytolytic T lymphocytes, natural killer (NK) cells, and monocytes on the afferent phase of graft-versus-host disease (GVHD). Lethal GVHD was induced across the murine major histocompatibility complex by injecting C57BL/6 (H-2(b)) bone marrow (BM) cells (a source of stem cells) and splenocytes (S) (a source of T cells) into lethally irradiated B10.BR (H-2(k)) recipients. Because anti-LFA1 does not bind complement (C') effectively, we conjugated anti-LFA1 α chain monoclonal antibody (MoAb) to ricin toxin A chain (RTA) as a means of facilitating target cell elimination. A 2-hour preincubation of C57BL/6 bone marrow/spleen (BMS) with anti-LFA1-RTA in the presence of ammonium chloride (a potentiator of immunotoxin toxicity), but not a control immunotoxin (IT), reduced CTL activity by greater than 2 logs, significantly reduced NK cell activity, and prevented B10.BR mice from developing GVHD. Depletion of target cells by toxin-labeled-MoAb and not the blockade of the LFA1 molecule by the anti-LFA1 MoAb accounted for our results, because incubating cells with IT in the absence of a potentiator had no effect on GVHD prevention. In contrast, C57BL/6 recipients of C3H BMS grafts only partially benefitted from anti-LFA1-RTA preincubation, demonstrating that in this system, different cells not expressing LFA1 were involved in GVHD generation. The same findings observed with anti-LFA1-RTA preincubation were observed with preincubation with L-leucyl-L-leucine methyl ester, a chemical compound eliminating cytolytic cells, providing further support that GVHD induction in the C3H/HeJ into C57BL/6 system is not entirely mediated by classical cytolytic T cells. We next tested anti-LFA1-RTA in a model devised to measure its effect on alloengraftment (B10.BR recipients given lower doses of irradiation). Anti-LFA1-RTA BM preincubation selectively reduced alloengraftment in the model. This observation, combined with experiments showing that LFA1-RTA preincubation, but not anti-Thy1.2 + C' or control IT preincubation, reduced colony-forming unit-spleen formation, indicates that anti-LFA1 α chain IT may remove accessory cells or stem cells critical to engraftment. Still, anti-LFA1-RTA may be useful for clinical GVHD prevention when combined with positive selection techniques designed to enrich for stem cells.