HEPATOCELLULAR TRANSPLANTATION AND TARGETING GENETIC MARKERS TO HEPATIC CELLS 
The major impediment to performing these experiments in a controlled and verifiable 
manner is the aforementioned inability to identify transplanted cells in the host 
following transplantation. In our murine experiments, the use of transgenic marker 
genes enabled an unequivocal demonstration of engraftment and function of transplanted 
cells. Clinical experience suggests that the clinical phenotype reflects many factors 
in addition to the efficiency of engraftment. We do not believe that the short-term 
prognosis of hepatic failure (days to weeks) is sufficiently predictable to serve as a 
indicator of the success or failure of hepatic engraftment. Hepatic failure, 
complications of hepatic failure, complications of surgery, or death may ensue even if 
engraftment is technically successful. Conversely, clinical improvement may occur in 
the absence of engraftment due to the normal course of the disease or coincident medical 
management. We have implemented a clinical protocol at TCH which is designed to address 
the sensitivity and predictive value of clinical and laboratory methods in assessing low 
level hepatic function. This study will assess the utility of conventional measures of 
hepatic function and dysfunction, the recently described MEGX test (Oellerich et al, 
1987, 1989, 1990), and loading tests for liver specific enzymes phenylalanine 
hydroxylase and ornithine transcarboxylase in extreme liver failure prior to 
transplantation in an attempt to ascertain whether these tests could provide a method 
for assessing engraftment and function of hepatocytes in the future (Ferry et al. 
Clinical course and management of Hepatic Transplantation) . 
We believe that the presence of a genetic marker for transplanted hepatocytes would 
greatly enhance our ability to demonstrate the success or failure of engraftment. This 
is particularly important in these initial trials of HCT in which the basic procedures 
for harvesting, manipulation, and surgically implanting these cells under clinically 
applicable conditions will need to be evaluated. 
There may be natural polymorphism between donor and host cells which may be used to 
detect the transplanted cells. Polymorphism in serum proteins may be used to 
differentiate the products of the host and transplanted hepatocytes (Ritzman and 
Johnson, 1983), though their relative infrequency and the frequent use of transfused 
plasma and blood products may make this approach unfeasible. We are assessing our 
ability to detect these polymorphism in our ongoing clinical study of OLT. Genetic 
(DNA) polymorphism between the donor and host cells are ubiquitous and may be used to 
identify the presence of donor cells. The high information content of VNTR segments is 
sufficient to predict that the donor and host will be different and this method has been 
used to follow chimeric engraftment in bone marrow transplantation (Gasparini et al, 
1989; Gatti et al, 1989). This technique may indeed be useful in demonstrating the 
presence of the engrafted cells (the applicability of this technique will be assessed 
in the present study) , but could not enable identification of individual cells in 
histological sections. The sensitivity of this technique is also inherently limited by 
the "stutter" which occurs at low frequency when amplifying repetitive sequences. This 
stutter may preclude unambiguous identification of cells at frequencies of 1-3%. The 
most sensitive method is to use PCR to identify conventional restriction fragment length 
polymorphism which are present and homozygous in the recipient, but absent or 
heterozygous in the donor. Such polymorphism can be detected by digesting genomic DNA 
with the polymorphic enzjmie and then performing PCR across the polymorphic site. The 
[786] 
Recombinant DNA Research, Volume 14 
