To establish the process by which transplanted cells integrate into the liver parenchyma, we used dipeptidyl peptidase IV-deficient F344 rats as hosts. On intrasplenic injection, transplanted hepatocytes immediately entered liver sinusoids, along with attenuation of portal vein radicles on angiography. However, a large fraction of transplanted cells (Ͼ70%) was rapidly cleared from portal spaces by phagocyte/macrophage responses. On the other hand, transplanted hepatocytes entering the hepatic sinusoids showed superior survival. These cells translocated from sinusoids into liver plates between 16 and 20 hours after transplantation, during which electron microscopy showed disruption of the sinusoidal endothelium. Interestingly, production of vascular endothelial growth factor was observed in hepatocytes before endothelial disruptions. Portal hypertension and angiographic changes resulting from cell transplantation resolved promptly. Integration of transplanted hepatocytes in the liver parenchyma required cell membrane regenesis, with hybrid gap junctions and bile canaliculi forming over 3 to 7 days after cell transplantation. We propose that strategies to deposit cells into distal hepatic sinusoids, to disrupt sinusoidal endothelium for facilitating cell entry into liver plates, and to accelerate cell integrations into liver parenchyma will advance applications of hepatocyte transplantation. (HEPATOLOGY 1999;29:509-519.)

No abstract

No abstract

The feasibility of liver repopulation with hepatocytes has been shown, although clinical applications demand significant hepatic replacement. To show whether portal vascular bed in large animals could accomodate a greater cell number, we analyzed liver repopulation in syngeneic Fischer 344 rats deficient in dipeptidyl peptidase IV. This system allowed localization of transplanted normal hepatocytes in liver or various ectopic sites, as well as dual studies for analysis of gene expression. Interestingly, the product of a dipeptidyl peptidase IV substrate inactivated bile canalicular adenosine triphosphatase (ATPase) activity in normal but not in dipeptidyl peptidase IV-deficient rats, which allowed localization of dipeptidyl peptidase IV-deficient hepatocytes in normal rat liver for additional reversed transplantation systems. Further studies with genetically marked cells showed that because of the size difference between hepatocytes and portal vein radicles, intrasplenically transplanted cells were distributed in periportal areas (zone 1) in mice, whereas in larger animals (rats or rabbits) cells were also distributed downstream to midlobular (zone 2) or perivenous (zone 3) areas. Transplantation of an escalating number of hepatocytes showed that adult rats tolerated intrasplenic injection of a large cell number in single sessions (up to 1 X 10(8), approximately 10% to 15% of the host hepatocyte mass). Morphometric analysis of recipient livers showed survival of a significantly greater cell number with incorporation in host liver plates. At 4 weeks, transplantation of 2 x 10(7) hepatocytes into adult rats led to a survival of 1.4 +/- 1.0 x 10(6) transplanted cells/cm3 liver, whereas after transplantation of 5 x 10(7) cells or 7.5 x 10(7) cells, the number of surviving transplanted cells in the liver significantly increased to 4.1 +/- 1.4 x 10(6) transplanted cells/cm3 liver (mean, 2.9-fold; P<.003) and 5.5 +/- 1.3 x 10(6) transplanted cells/cm3 liver (mean, 3.9-fold; P<.003), respectively. When cells were injected in greater numbers, transplanted hepatocytes retained normal function and produced more serum albumin or hepatitis B surface antigen in deficient hosts. These data indicate the feasibility in larger animals of significant liver repopulation with hepatocyte transplantation. Use of dipeptidyl peptidase IV-deficient rats should help further analysis of mechanisms in liver repopulation.