Immune rejection: the immune biology of allogeneic hematopoietic stem cell transplantation (from mice to humans) - Immune Biology of Allogeneic Hematopoietic Stem Cell Transplantation - page 101

In experiments designed to demonstrate that donor T cells can prevent
marrow graft rejection, strain combinations with bidirectional graft-ver-
sus-host and host-versus-graft mismatching for MHC-class I and -class II
antigens and minor histocompatibility antigens were used [112] . Recipients
were treated with 800 cGy TBI, the highest exposure that was not sufficient
to prevent rejection of T-cell-depleted grafts containing 5 × 10 6 marrow
cells. This cell dose translates to approximately 2.5 × 10 8 cells/kg recipi-
ent body weight, similar to the marrow cell dose used for HCT in humans.
Under these conditions, as few as 2.5 × 10 5 CD3 + donor T cells or 0.5 × 10 6
CD8 + donor T cells added to the T-cell-depleted marrow graft consistently
prevented rejection. Under these same conditions, 2.5 × 10 6 donor CD4 +
T cells did not prevent rejection, and higher doses of donor CD4 + T cells
caused early death due to GVHD.

To circumvent the problem of GVHD caused by donor CD4 + T cells, similar
experiments were repeated in a strain combination with bidirectional graft-
versus-host and host-versus-graft mismatching for MHC-class I antigens
and no mismatching for MHC-class II antigens or minor histocompatibility
antigens. With this strain combination, 750 cGy was the highest TBI expo-
sure that was not sufficient to prevent rejection of T-cell-depleted grafts.
Under these conditions, as few as 2.5 × 10 5 donor CD8 + T cells consistently
prevented rejection. In contrast, most recipients rejected grafts containing
2.5 × 10 5 donor CD4 + T cells, and even 12.5 × 10 5 donor CD4 + T cells were not
sufficient to prevent rejection in all recipients. These results showed that
the donor CD8 + T cells were at least fivefold more active than donor CD4 + T
cells in preventing graft rejection in this strain combination.

103

The key to understanding these results came from the fact that murine
T cells do not express MHC-class II antigens even after activation, whereas
in humans, MHC-class II antigens are express by activated T cells but not
by resting T cells. Based on the results of experiments with mice, it was
proposed that the ability of donor T cells to prevent rejection depended on
their ability to generate MHC-class I-specific or class I-restricted cytotoxic
T cells that recognize and eliminate or inactivate the recipient CD4 + and
CD8 + T cells responsible for causing rejection.

These results also suggested that graft failure would occur when donor
CD4 + T cells are added to T-cell-depleted grafts in strain combinations with
bidirectional class II mismatching. In this situation, the donor CD4 + T cells
would be expected to eliminate recipient marrow cells but would spare
recipient T cells, since they do not express MHC-class II antigens. In turn,
the spared recipient T cells would be expected to eliminate donor marrow
cells but would spare donor T cells for the same reason, thereby resulting in
mutually assured marrow destruction.

The role of donor T-cell cytotoxic effector function in preventing marrow
graft rejection was demonstrated in experiments with donor T cells defi-
cient in Fas ligand binding or in expression of granzyme B or perforin [113] .
Control experiments showed that approximately 1.0 × 10 5 wild-type donor
CD8 + T cells were sufficient to prevent rejection in 50% of the recipients.
Defective Fas-ligand-mediated cytotoxicity caused by the gld mutation had
only a minimal effect on the ability of donor CD8 + T cells to prevent rejec-
tion. Likewise, the absence of granzyme B had only a minimal effect on the