IRES-dependent second gene expression is significantly lower than cap-dependent first gene expression in a bicistronic vector

Mol Ther. 2000 Apr;1(4):376-82. doi: 10.1006/mthe.2000.0050.


The internal ribosome entry site (IRES) has been widely used to coexpress heterologous gene products by a message from a single promoter. However, little is known about the efficiency of IRES-dependent second gene expression in comparison with that of first gene expression. This study was undertaken to characterize the relative expression of IRES-dependent second gene in a bicistronic vector, which was derived from the 5' untranslated regions of the encephalomyocarditis virus (EMCV). IRES-dependent second gene expression was compared with cap-dependent first gene expression in several cultured cell lines and in mouse liver in vivo. The expression of the IRES-dependent second gene ranged from 6 to 100% (in most cases between 20 and 50%) that of the first gene. Second gene expression in a plasmid without the IRES was 0.1-0.8% (with some exceptions) that of the first gene. These findings have important implications for the use of IRES, i.e., care should be taken regarding the decreased capacity of IRES-dependent downstream gene expression as well as in determining which gene should be positioned as the first or second gene in a bicistronic vector.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • 5' Untranslated Regions
  • Alkaline Phosphatase / genetics
  • Animals
  • CHO Cells
  • Chloramphenicol O-Acetyltransferase / genetics
  • Cricetinae
  • Encephalomyocarditis virus / genetics
  • Gene Expression*
  • Gene Transfer Techniques
  • Genes / genetics
  • Genetic Vectors*
  • HeLa Cells
  • Humans
  • L Cells
  • Liver / metabolism
  • Luciferases / genetics
  • Mice
  • Plasmids / genetics
  • RNA Caps / genetics


  • 5' Untranslated Regions
  • RNA Caps
  • Luciferases
  • Chloramphenicol O-Acetyltransferase
  • Alkaline Phosphatase