We have developed a transfection vector for animal cells that contains long terminal repeat (LTR) sequences to promote expression. Plasmid p101/101, a derivative of plasmid pBR322 containing the complete Moloney murine sarcoma virus genome, was cut with restriction enzymes and religated so that both the 5′ and 3′ LTRs were retained and all but about 700 base pairs of the intervening viral sequences were removed. To test this vector, the
Escherichia coli
gene
gpt
was cloned into a unique
Pst
I site, between the two LTRs, with guanine and cytosine tailing, a method that can be generalized for insertion of any DNA segment into this vector. When DNA from recombinant plasmids in which the
gpt
gene was inserted in the same transcriptional polarity as the LTR sequences was transfected onto murine or rat fibroblast cultures, we obtained a high yield of Gpt
+
colonies. However, plasmid constructs with the
gpt
gene in the opposite polarity were virtually devoid of activity. With
gpt
in the proper orientation, restriction enzyme cuts within the LTRs or between the 5′ LTR and the
gpt
gene reduced transfection by more than 98%, whereas a cut between the
gpt
gene and the 3′ LTR gave an 80% reduction in activity. Thus, both 5′ and 3′ LTR sequences are essential for optimal
gpt
expression, although the 5′ LTR appears to play a more important role. When the LTR-
gpt
plasmid was transfected onto murine leukemia virus-infected mouse fibroblasts, we obtained evidence that RNA copies became pseudotyped into viral particles which could transfer the Gpt
+
phenotype into rodent cells with extremely high efficiency. This vector should prove useful for high-efficiency transduction of a variety of genes in mammalian cells.
Near quantitative synthesis of urea macrocycles enabled by bulky N-substituent
