UAB 9405 
March 16, 1994 
PAGE - 4 
Recently, Shively’s group (19) has expressed the individual CEA domains in 
HeLa cells for epitope mapping purposes. 
It is clear that CEA can function as an effective tumor vaccine target in an animal 
model of CEA expressing murine colon carcinoma line (20). Vaccination with 
rV-CEA produced both humoral and cellular immune responses with resultant 
in vivo anti-tumor effects. It is pertinent that mice have the CEA gene family of 
molecules which share 50 to 60% homology with human CEA (16) and yet the 
mice suffered no toxic effects. This serum tumor marker is widely used in clinical 
medicine and CEA is expressed in >90% of colorectal tumors in man (also found 
in other tumor types like breast and non-small cell lung cancer). The small 
amounts of this glycoprotein in normal tissue supports the view that humans have 
tolerance to it. However, its immunogenicity in man is controversial with reports 
suggesting that colon and breast cancer patients have antibody to CEA (21,22) 
while other reports deny the presence of specific antibody (23,24). The issues of 
tolerance, cross-reacting antigens and immunogenicity will be serious concerns 
for this and almost any human tumor relevant antigen. 
2.3 Polynucleotide vaccines composed of plasmid DNA constructs of relevant 
genes when administered intramuscularly can induce gene expression in 
muscle with resultant host immune response and protective immunity. 
In 1990, myofiber cells in the mouse were shovm to express foreign genes that 
have been injected into muscle in the form of naked DNA without any cationic 
lipids, retroviruses or other special delivery systems (25). Such expression has 
also been demonstrated after injection of naked DNA into the skeletal muscle of 
fish (26), rats (27), cats, and rhesus monkeys (28). Expression in muscle has been 
demonstrated with DNA encoding reporter genes (25), dystrophin (29), and 
influenza nucleoprotein (30). The duration of gene expression in skeletal muscle 
using an RSV promoter driving a luciferase reporter exceeded 19 months post- 
injection, and the foreign plasmid DNA appeared to remain episomal (31). 
Injection of plasmid DNA into other murine tissues such as brain, liver, spleen, 
uterus, stomach, lung, or kidney did not produce expression (32). Pre-treatment 
of striated muscle with the local anesthetic bupivacaine significantly increases 
gene expression following naked plasmid DNA injection. This has been 
attributed to uptake and expression of the plasmid by a greater number of muscle 
fibers (33,34). 
Naked plasmid DNA encoding influenza A nucleoprotein (NP) delivered to mice 
by IM injection produced influenza NP-specific antibodies and cytolytic T cell 
response with protection from subsequent chedlenge with influenza A virus (30). 
Other examples of immunization using naked plasmid DNA include induction of 
antibody response to human growth hormone following administration of DNA- 
coated micro-projectiles into mouse skin (35); induction of neutralizing 
(hemagglutination inhibiting) antibodies by injection of DNA encoding influenza 
A hemagglutinin in mice, ferrets and rhesus monkeys (36) with protective 
immunity to challenge (37); antibody response to bovine herpes virus in mice and 
Recombinant DNA Research, Volume 19 [401] 
