There is an increasing understanding that overexpression of heterologous genes can impose a heavy metabolic burden on host cells, which limits the productivity and long-term stability of genetically-modified cells. However, few studies have investigated the mechanistic underpinnings of metabolic burden due to resource misallocation on protein overexpression. The degenerate genetic code provides an opportunity to allocate cellular tRNA and ribosomal resources optimally between host and heterologous protein expression. 18 of the 20 natural amino acids can be encoded with multiple codons, which can result in insignificant codon usage bias due to translational selection for codons with faster elongation times, or that are tuned to host tRNA supply. Through our research, we aim to improve the predictability and robustness of genetic engineering in microbes by systematically determining optimal codon bias schemes.
We investigate how the partitioning of microbial translational resources, specifically through the allocation of tRNA by incorporating dissimilar codon usage bias (CUB), can drastically alter the expression of proteins and reduce the burden on gene expression systems. Utilizing nearly identical fluorescent reporters (CFP and YFP), we assayed genetic resource competition both in vitro and in vivo using novel designs that isolate translation elongation from other variables. We find that alternative CUB designs can trans-regulate gene expression of competing heterologous and endogenous genes, yielding profitable or catastrophic design options. By isolating individual codons experimentally, we correlate specific codon usage patterns with genetic burden and derive novel coding schemes for multi-gene expression. This re-coding scheme, named CHI (codon health index) was tested to be beneficial for protein expression under a variety of growth conditions and with numerous proteins. This empirically derived coding scheme based on a new codon health index enables the designing of harmonious multi-gene expression systems while avoiding catastrophic cellular burden.
