Postnatal donor lymphocytes enhance prenatally-created chimerism at the risk of graft-versus-host disease

Abstract

The major barrier to clinical application of in utero hematopoietic stem cell transplantation is insufficient chimerism for phenotypic correction of target diseases or induction of graft tolerance. Postnatal donor lymphocyte infusion (DLI) may enhance donor cell levels so as to further facilitate tolerance induction. We created murine mixed chimeras in utero. Chimeras with <10% donor cells were subjected to postnatal DLI to evaluate the effects of DLI on chimerism augmentation and skin tolerance induction. Within one day after DLI, recipients experienced a transient peaking of donor chimerism, which could be as high as 20~40%. However, the transient chimerism peaking didn't benefit donor skin survivals despite immediate skin placement after DLI. In case of fruitful DLI, chimerism augmentation was usually observed after a latent period of 2~4 weeks. Otherwise, chimerism would return to around pre-DLI levels by days 7~14. Peripheral chimerism of >3% could be consistently boosted up to >10%, whereas chimerism of <0.2% hardly showed any significant enhancement. As for chimerism levels of 0.2~3%, chimerism augmentation up to >10% succeeded in 3(15%) of 20 recipients. Notably, chimerism augmentation by postnatal DLI was often associated with unexpected death or graft-versus-host disease (GVHD). In conclusion, transient chimerism augmentation by DLI played no role in facilitating graft tolerance. Substantial augmentation by DLI demanded a threshold chimerism level and posed a serious risk of GVHD to the recipients. It raised the concern about using postnatal DLI to broaden therapeutic horizons of in utero hematopoietic stem cell transplantation.

Keywords: Chimerism augmentation; donor lymphocyte infusion; graft-versus-host disease; in utero transplantation; tolerance induction.

Figure 1

Lung and liver involvement of GVHD after postnatal DLI. A representative recipient with low-level chimerism of 0.10% was subjected to postnatal DLI at 1 month old and had chimerism augmentation up to 12.94% at 2 months old. However, the mouse showed signs of GVHD. Histological examinations by hematoxylin-eosin staining showed mononuclear cell infiltration in the peribronchial area of the lung (upper left panel) and the portal area of the liver (lower left panel). The lung involvement of GVHD was destructive to airway epithelia (arrow, upper right panel) with epithelial shedding (arrowhead). The liver involvement of GVHD also damaged endothelia of bile ducts (arrow, lower right panel).

Figure 2

Multilineage expression of donor leukocytes following chimerism augmentation by postnatal DLI. A mixed chimera with peripheral chimerism of 4.01% at 1 month old had chimerism augmentation by postnatal DLI up to 26.2% at 4 months old. This augmentation displayed multilineage expression of donor H-2^b leukocytes.

Figure 3

Dynamics of donor cell levels within 1 month after postnatal DLI. 15 recipients (7, 6 and 2 respectively from groups “0.2~3%”, “0.01~0.19%” and “undetectable”) were examined for circulating donor cells levels within 30 days after postnatal DLI at the time points of 2 hours (2 H), 1, 2, 5, 7, 14 and 30 day (1~30 D).

Figure 4

Survival curves of donor skin grafts. H-2^q murine recipients in groups “0.2~3%”, “0.01~0.19%” and “undetectable” were subjected to H-2^b donor skin transplantation within one day (DLI D1) or 2 weeks (DLI W2) after postnatal DLI. Their skin graft survivals were compared with those from wild-type controls and their counterparts without DLI. DLI either on day 1 (DLI D1) or week 2 (DLI W2) did not show beneficial effect on donor skin survivals. Regardless of DLI, skin graft survivals were better in group “0.2~3%” than in groups “0.01~0.19%” (P<0.001~P=0.004), “undetectable” (P<0.001 for all) and wild-type controls (P<0.001 for both). Also, groups “0.01~0.19%” had better skin survivals than group “undetectable” (P<0.001~P=0.023). Skin survivals did not differed between group “undetectable” and wild-type controls.