findings. The only other finding on microscopic exam was the presence of cytoplasmic basophilic 
granules in the livers of most animals. The significance of this finding is currently unclear, 
however, no inflammation, necrosis, apoptosis, or other abnormalities of the liver were noted. As 
mentioned above, we detected no evidence of biochemical liver injury in any liver function tests. We 
are planning to conduct additional studies with untreated animals to help assess the meaning of this 
histologic finding. 
A more detailed preliminary report is contained in Appendix E. 
In order to look for dissemination of the virus, a sensitive PCR based system that will allow 
detection of small amounts of message has been developed (See Figure 4 on the following page). 
Primers were designed to span a region of 18 base pairs on either side of a 500 basepair region of 
HSVTK gene. These primers were tested in PCR using the H5.010RSV7K plasmid as template (see 
lane 4) and produced the appropriate sized DNA (arrows). The sensitivity and specificity of the 
primers was tested by extracting mRNA from untransduced REN mesothelioma cells and REN cells 
transduced with varying MOI's of HS.OIORSVTK; performing reverse transcription, and using this 
cDNA for PCR. As shown in Figure 4, no HSV7K message was present in untransduced REN cells. In 
contrast, detectable message was present in cells transduced with as few as 0.001 pfu/cell. 
Although not truly quantitative, the amount of the 500 bp product was proportional to the MOI of 
virus. The utility of the primers in tissues was confirmed by doing RT-PCR on mRNA extracted 
from tumor tissue. When RNA from non-transduced tumors were examined no band at 500 bp was 
noted despite a positive control with actin (not shown). In contrast, a very strong signal was 
obtained from tumor tissue harvested from an animal injected with REN cells that had been 
preinfected with H5.010RSV7X six days earlier. 
These results demonstrate that mRNA for HSV TK can be detected from tumor tissues with a high 
degree of sensitivity and specificity. We next used this assay to analyze the organs of mice injected 
with the H5.010RSV77<. Mice with established tumors were injected with virus (6 x 10 9 pfu) 
intraperitoneally. Two days after injection, the mice were sacrificed and their tumors and organs 
removed. RNA was extracted for use in RT-PCR with mouse p-actin (as a positive control and HS MTK 
primers. These results are shown in Figure 5 on the next page. As expected, the primers amplified 
the correct sized band from REN tumor cells infected with HSV TK in vitro (Lane 1) and from the 
plasmid containing the HSVTK’ cDNA (Lane 2). No bands were seen when template was not added 
(Lane 3) or in tumor cells uninfected with HSV77< (Lane 4). HSVTK message was detected in a tumor 
nodule (Lane 12), and in smaller amounts in liver (Lane 8), and kidney (Lane 13) from infeced 
animals. This TK message may have been from the mesothelial cells surrounding the liver and 
kidney. In contrast, n& message was detected in any extra-abdominal organs including brain (Lane 
5), skeletal muscle (Lane 6), heart (Lanes 7 and 10), lung (Lane 11), and salivary gland (Lane 9). 
These results are consistent with our findings using p-gal staining, in which the only tissues where 
we have seen expression of transgene has been the normal mesothelium and liver after 
intraperitoneal injection of virus into animals with tumor. In these cases (using a LacZ marker 
adenoviral construct), we saw diffuse mesothelial staining and staining of scattered cells within the 
hepatic parenchyma. 
[ 402 ] 
Recombinant DNA Research, Volume 20 
