Induction of acute graft-versus-host disease by T cells that do not respond to in vitro alloantigen stimulation

The mixed lymphocyte reaction (MLR) has been used extensively to measure alloreactive T cells. In clinical practice, a negative MLR of recipient T cells responding to donor cells does not necessarily mean that a donor-specific tolerance has been established. This discrepancy indicates that the presently used methods fail to demonstrate the full repertoire of alloreactive T cells. This could be the result of the fact that some alloreactive T cells do not respond in vitro but will mount a response towards alloantigens in vivo, or that some alloreactive T cells do not respond during the MLR but will respond later if the alloantigen stimulation remains. We therefore examined the non-proliferating T-cell population in a mouse primary alloreactive response. Spleen and lymph node cells derived from C57BL/6J (H-2(b)) mice were stained with carboxy-fluorescein diacetate succinimidyl ester and injected intravenously into C.B-17 severe combined immunodeficient (SCID) mice (H-2(d)). The donor cells were recovered 5 d after the injection. The non-proliferating T cells were sorted and were non-reactive to alloantigen stimulation in vitro. Nevertheless, these non-proliferating T cells could proliferate and cause acute graft-versus-host disease after being adoptively transferred to secondary recipients of SCID mice. These results suggest that there exists an alloreactive T-cell population that does not respond to in vitro alloantigen stimulation but can mount a delayed alloresponse in vivo.