Abstract
Semliki Forest virus (SFV) vectors are widely used in neurobiological studies because they efficiently infect neurons. As with any viral vector, they possess a limited cloning capacity, so infection with different SFV vectors may be required to introduce multiple transgenes into individual cells. However, this approach is limited by superinfection exclusion. The authors examined marker expression in baby hamster kidney cells, mouse cortical neurons, and rat hippocampal neurons using different fluorophore-encoding vectors that are based on the wild-type SFV4 strain and on the less cytopathic SFV4(PD) mutant, which carries two point mutations in nonstructural protein 2. For every fluorophore tested, SFV4(PD) gave higher (up to 22-fold) expression compared to SFV4. In infections using two and three different vectors, SFV4 caused relatively few multifluorescent baby hamster kidney cells when applied at 0-s, 15-min, or 2-h intervals. In contrast, SFV4(PD) permitted significantly enhanced marker coexpression, resulting in 46% doubly and 21% triply fluorescent baby hamster kidney cells, and 67% to 78% doubly fluorescent cortical and hippocampal neurons. At 15-min or 2-h addition intervals, SFV4(PD) still permitted 23% to 36% doubly fluorescent baby hamster kidney cells. The increased efficiency of SFV4(PD) in coexpressing separate markers from different viral particles suggests that mutations in nonstructural protein 2 affect alphaviral superinfection exclusion. The results demonstrate that SFV4(PD) is well-suited to coexpress multiple proteins in neuronal and non-neuronal cells. This capability is particularly valuable to express the various components of heteromeric protein complexes, especially when the individual cDNAs cannot be combined into single SFV particles.
This paper is dedicated to Drs. James H. and Ellen G. Strauss in vivid remembrance of their common ornithological excursions during Markus U Ehrengruber's time at Caltech—with a look onto alphavirsues from the avian side.
The authors thank Eric Sung for designing the sequencing primers, Joyce Iping for technical help with the preparation of SFV vectors and cortical cell cultures, Sheri L. McKinney for supplying dispersed hippocampal neurons, and Drs. K. George Chandy and George A. Gutman for access to the multiwell plate reader. The authors are grateful to Dr. Sondra Schlesinger for critically reviewing the manuscript. This work was supported by grants from the NIH (NS48336), the McKnight Endowment Fund for Neuroscience, and the National Multiple Sclerosis Society. M.U.E. acknowledges Starbucks Co. for providing a comfortable environment in which to write this paper.