C. HESDORFFER 4/93 
ability of MDR-expressing cells to pump out Rhodamine dye (94). 
It was implied that this higher level of MDR gives a meaningful 
selective advantage to stem cells when challenged with MDR- 
responsive drugs, and that transduction with an exogenous MDR 
gene would have little or no additional effect. We and others 
have shown that this is not correct: 1) Normal mice can be made 
severely leukopenic after treatment with daunomycin, while 
transgenic mice, whose bone marrow contains an exogenous human 
MDR gene regulated by a strong promoter, are unaffected (40,41); 
2) Cancer patients treated with MDR-responsive drugs have severe 
bone marrow toxicity (1-8) ; and 3) MDR-transduced bone marrow 
cells in live mice are preferentially protected from toxicity and 
enriched on exposure to taxol (31,32). 
If untransduced marrow stem cells were normally protected by 
their relatively higher levels of MDR, there should be little if 
any marrow toxicity induced by chemotherapy. This is clearly not 
the case. In addition, if transduction and expression of an 
exogenous MDR gene was not significant, then selection of MDR- 
transduced cells should not be possible. The data support the 
hypothesis that: 1) Although bone marrow stem cells have a 
relatively increased amount of MDR gene expression, this increase 
is not great enough to prevent bone marrow toxicity in mouse or 
human marrow exposed to MDR-responsive drugs; and 
2) Transduction and expression of MDR using an exogenous gene on 
a strong promoter results in resistance and selection of 
transduced murine marrow stem cells on exposure to MDR-responsive 
drugs, and, thus, is a rational approach to protecting human bone 
marrow cells from the toxicity of these drugs. This does not 
imply that MDR transduction is the only means of decreasing bone 
marrow toxicity in the treatment of patients with cancer 
undergoing chemotherapy with ABMT. Cytokines such as GM-CSF have 
been shown to shorten the time to marrow recovery (95,96); 
peripheral blood stem cell infusions are useful as well (97-100) . 
Combinations of these therapies with MDR transduction may be 
complementary, and more beneficial than any used alone to 
decrease or eliminate marrow toxicity. 
C. Human Marrow Transduction with the MDR Gene using a Safe 
Amphotrooic MDR Producer Line : 
In other experiments (101-107) , we have transduced human 
bone marrow cells obtained from marrow harvests for future ABMT 
with our highest titer (5 X 10 4 particles/ml) amphotropic 
producer line, A12M1 retrovirus (Figure 2). Supernatants from 
A12M1 have been used instead of coculture with MDR producer cells 
to avoid potential contamination of the bone marrow with the 
producer cells, an undesirable side-effect in clinical use. In 
one set of experiments, we have cultured Ficoll-separated 
[ 102 ] 
Recombinant DNA Research, Volume 18 
