Background: Ischemia-reperfusion injury (IRI) is a major clinical challenge during liver transplantation
that can lead to early allograft dysfunction (EAD), graft rejection, and even death
of the recipient. Induced lytic cell death (LCD) decreases functional cell mass and
triggers a harmful pro-inflammatory immune response. Rupture of the plasma membrane,
the final stage of several LCD pathways, could recently be shown to be an active and
highly regulated process mediated by the transmembrane protein Ninjurin-1 (NINJ1).
The role NINJ1 plays during hepatic IRI remains unknown.
Aims: To investigate the role of Ninjurin-1 activation on hepatic IRI.
Methods: A model of 70% segmental warm hepatic IRI was used in mice and rats with conventional
whole-body (Ninj1
-/-
) knockout and mice with hepatocyte-specific (Ninj1
fl/fl
Alb-Cre
+
) and myelomonocyte/Kupffer cell (KC)-specific (Ninj1
fl/fl
LysM-Cre
+
) knockout as well as their respective controls. The resulting injury was assessed
using serum (AST, ALT, LDH) and histopathological (Suzuki score, TUNEL staining) markers
for hepatic IRI. Treatment of WT mice with glycine or clone D1, an antagonizing antibody,
both known to preserve plasma membrane integrity in a NINJ1-dependent manner, was
used to assess pharmacological inhibition in vivo. To mimic IRI in vitro, we cultivated isolated and purified murine hepatocytes and KCs under hypoxia/reoxygenation
conditions (H/R) and quantified LCD. Non-denaturing gel electrophoresis (BN-PAGE)
was used to assess NINJ1 activation in liver samples from IRI- and sham-treated mice
and biopsies from human liver grafts (hLG) pre-implantation and post-reperfusion.
Results: NINJ1 deficiency in both mice and rats led to a significant reduction of all assessed
parameters for hepatic IRI. Accordingly, pharmacological inhibition protected mice
against IRI in vivo. Both hepatocytes and KCs undergo H/R-induced LCD contributing to hepatic IRI in vivo. Moreover, NINJ1 activation in post-reperfusion biopsies from hLG was markedly increased
in recipients with EAD (peak AST>5,000 U/l postoperatively) compared to recipients
with low AST levels (<200 U/l).
Conclusion: Across species, we demonstrate that NINJ1 mediates hepatic IRI, which can be prevented
by pharmacological targeting of NINJ1. Moreover, in hLG recipients, EAD was associated
with increased NINJ1 activation. Thus, we propose NINJ1 as a new target to treat hepatic
IRI and improve patient outcomes during liver transplantation.
