Analysis of 11,430 recombinant protein production experiments reveals that protein yield is tunable by synonymous codon changes of translation initiation sites

Abstract

Recombinant protein production is a key process in generating proteins of interest in the pharmaceutical industry and biomedical research. However, about 50% of recombinant proteins fail to be expressed in a variety of host cells. Here we show that the accessibility of translation initiation sites modelled using the mRNA base-unpairing across the Boltzmann’s ensemble significantly outperforms alternative features. This approach accurately predicts the successes or failures of expression experiments, which utilised Escherichia coli cells to express 11,430 recombinant proteins from over 189 diverse species. On this basis, we develop TIsigner that uses simulated annealing to modify up to the first nine codons of mRNAs with synonymous substitutions. We show that accessibility captures the key propensity beyond the target region (initiation sites in this case), as a modest number of synonymous changes is sufficient to tune the recombinant protein expression levels. We build a stochastic simulation model and show that higher accessibility leads to higher protein production and slower cell growth, supporting the idea of protein cost, where cell growth is constrained by protein circuits during over expression.

Deciphering the determinants of recombinant protein yield across the human secretome

“…1b). This correlation pales in comparison to other studies which report numbers closer to 40% for endogenous genes in mammalian cells across various conditions [9][10][11][12] , likely due to the high mRNA expression achieved in the QMCF system. We conclude that adequate transgene mRNA is produced in these cells, and mRNA abundance is likely not the limiting factor.…”

Influence of the mRNA initial region on protein production: a case study using recombinant detoxified pneumolysin as a model

“…TIsigner revealed that the opening energy and expression score for PdT were 1.6 times higher and 2.7 times lower than those for His–TEV–PdT. The opening energy is inversely proportional to protein production levels ( Bhandari et al, 2021b ), and the His–TEV–PdT titer was 1.5-fold and 1.7-fold higher than for PdT, respectively, under C1 and C2 ( Table 3 ), which is in agreement with the difference observed in the opening energy. Furthermore, RNAfold revealed that the MFE of the mRNA structure of the pdt gene version 1 was 2.2 times lower than that of the his-tev-pdt gene and a predicted secondary structure with a long stem that includes the pdt start codon, which also corroborated with the other results obtained.…”

“…This observation indicates the relevance of having at least the two first codons free for ribosome binding. Moreover, although the pdt versions 1 and 2 have differences in the entire gene due to codon usage optimization of version 1, it was reported that changes in the first nine codons can achieve nearly optimum accessibility when compared to full-length modifications ( Bhandari et al, 2021b ). Therefore, codon differences in regions other than the 5′-end of the pdt version 2 are not likely to promote a significant reduction in opening energy and, consequently, should not increase protein production.…”

“…As shown in Figure 5 (bottom panel), the ribonucleotides related to the His-tag sequence seemed to have a major impact on this result since only the last two codons of the TEV cleavage site were considered in RNAfold analysis. The region from −30 to +30 was considered because this region was employed to calculate the MFE in previous studies ( Bhandari et al, 2021b ). The prediction of the entire mRNA structure would generate a different structure that would be much complex to analyze since we do not know for how long the complete mRNA remains stable until degradation starts.…”

Gene recoding by synonymous mutations creates promiscuous intragenic transcription initiation in mycobacteria

“…We provide evidence that bacterial promoters readily arise within open reading frames from synonymous mutations, and this lowers the previously held threshold for transcription initiation (and translation initiation), in agreement with a growing body of research showing that transcription initiation is a relatively promiscuous process in mycobacteria. In addition, codon context optimization has been reported to impact protein production to a greater extent than codon usage optimization alone in both prokaryotes and eukaryotes (47)(48)(49). Understanding the relationship between codon pairs and codon context more generally in thus necessary to improve recombinant protein research.…”