Controlling Gene Expression Using a Synthetic Biology Approach

Abstract

Gene expression is naturally regulated at multiple levels including transcription, translation and RNA processing. Cell biology experiments frequently require modifications of these regulatory systems, for example to express a heterologous protein, or to prevent expression of an endogenous gene. We have used the degeneracy of the genetic code to design genes that encode proteins with unchanged amino acid sequences but altered translational and RNA processing properties. We have performed a thorough statistical analysis of the codon distribution in all 4288 E. coli ORFs and analyzed the patterns of codon bias. The codon distribution pattern is distinctly different in the first 20 amino acids of ORFs. This could be attributed to the ribosome switching from initiation to elongation mode. Contrary to several other studies, we do not find codon-pair bias to deviate from what would be expected. By using the codon bias elucidated from this analysis we have designed and synthesized genes for fungal and plant cytochrome P450s that express well in E coli. RNA interference is a transforming technology that can shut down gene expression using an endogenous RNA processing pathway. However, as with antisense technology two decades ago, non-specific effects are becoming apparent. To control for such effects it is possible to redesign genes so that their mRNAs encode the wild type protein but share only 60–70% sequence identity with the original. We have found that such genes are resistant to both single RNAi species as well as wild type mRNA digested with DICER. Addition of RNAi-resistant genes should restore the wild type phenotype if the RNA interference is specific. By knocking out the endogenous gene and making mutations in added-back RNAi-resistant genes, the effects of amino acid changes on gene function can be easily tested in genetically intractable systems.