Hygromycin expression vector

[Jean-Paul Herman]

I've constructed recently some bicistronic retroviral vectors, 
containing a hygromycin resistance gene followed by an IRES sequence 
and, downstream of the IRES, the gene of interest (LacZ in one, the 
recombinase Cre in another). However, the producers (Psi2's), while 
keeping their hygromycin resistence (to up to 1.5 mg/ml!) , gradually 
lost the expression of the second gene and, correspondingly, the viral 
titer fell. Moreover, the same thing happened with cells infected with 
the virus and selected with hygromycin. Has anyone some insights as to 
what could go on here and if it a usual finding with that sort of 
vectors?

Jean-Paul



Jean-Paul Herman
ICNE, CNRS UMR 6544
Faculti de Midecine Nord
Marseille, France