UU/UA dinucleotide frequency reduction in coding regions results in increased mRNA stability and protein expression

Abstract

UU and UA dinucleotides are rare in mammalian genes and may offer natural selection against endoribonuclease-mediated mRNA decay. This study hypothesized that reducing UU and UA (UW) dinucleotides in the mRNA-coding sequence, including the codons and the dicodon boundaries, may promote resistance to mRNA decay, thereby increasing protein production. Indeed, protein expression from UW-reduced coding regions of enhanced green fluorescent protein (EGFP), luciferase, interferon-α, and hepatitis B surface antigen (HBsAg) was higher when compared to the wild-type protein expression. The steady-state level of UW-reduced EGFP mRNA was higher and the mRNA half-life was also longer. Ectopic expression of the endoribonuclease, RNase L, did not reduce the wild type or UW-reduced mRNA. A mutant form of the mRNA decay-promoting protein, tristetraprolin (TTP/ZFP36), which has a point mutation in the zinc-finger domain (C124R), was used. The wild-type EGFP mRNA but not the UW-reduced mRNA responded to the dominant negative action of the C124R ZFP36/TTP mutant. The results indicate the efficacy of the described rational approach to formulate a general scheme for boosting recombinant protein production in mammalian cells.

Figure 1

Impact of UW dinucleotide frequency on protein expression. (a) The dicodon change algorithm eliminates UU or UA at the boundary by replacing the first amino acid codon that ends with U with a synonymous codon that ends with an S nucleotide. X is any amino acid. S (G or C), M (A or C), R (A or G) nucleotides are IUPAC symbols. (b) Dinucleotide frequency analysis in WT and UW reduced mRNA-coding region constructs. The Compseq program, an EMBOSS algorithm that calculates nucleotide composition in a sequence, was used http://mobyle.pasteur.fr/cgi-bin/portal.py#forms::compseq. The expected frequency of any dinucleotide is 1/16 (6.25%). Thus, the observed to expected ratio = % observed dinucleotide/6.25%. The observed dinucleotide (%) is shown on the columns. Several coding regions were designed based on the algorithm shown in figure. These were custom synthesized, and then subcloned into expression vectors. Two cell lines (c) HEK293 or (d) CHO1 were cotransfected with expression vectors containing wild-type (WT) coding region or UW-reduced coding region (UWr) of different genes and red fluorescent protein (RFP) plasmid at a ratio of 1/4. Protein expression at 18 hours after transfection was quantified by fluorescence (EGFP), chemiluminescence (luciferase), or enzyme-linked immunosorbent assay (ELISA) [interferon-α (IFN-α) and hepatitis B surface antigen (HBsAg)]. Data are presented as RFP-normalized fold increase which = (Expression values of UWr plasmid/WT plasmid) × (RFP normalization ratio). Data are mean ±SEM of three independent experiments.

Figure 2

Effect of sequence modification and normalization on DNA uptake and expression. (a) Representative image showing green (G), red (R; red fluorescent protein (RFP) plasmid normalization plasmid), and superimposed GR channels. (b) HEK293 cells were cotransfected with green fluorescent protein (GFP) (100 ng) and RFP (25 ng) expression plasmids. The fold ratio changes between the expression levels as a result of WT and UWr-coding region modification are shown (mean ± SEM) for three independent experiments. (c) HEK293 cells (3 × 10⁴ cells/well) in 96-well microplates were cotransfected with RFP expression plasmid and expression vectors containing either the WT-coding region or the UWr-coding region using low DNA input (10 ng/well) or high DNA input (75 ng/well).

Figure 3

mRNA stability and UW reduction. (a) Half-life mRNA determinations of the two enhanced green fluorescent protein (EGFP) mRNA sequence variants. HeLa Tet-off cell line was transfected with the indicated expression vectors, and transcriptional activity was blocked by doxycycline (1 µg/ml). RNA was extracted at the indicated points and subjected to reverse transcription (RT)-QPCR. The one-phase decay model was used to quantify the half-life for WT and UW-reduced (UWr) EGFP mRNAs; R² = 0.96 and 0.95, respectively. Data are average of two experiments (mean ± SEM); each with three replicates. (b) The steady state mRNA levels of WT and UW-reduced EGFP-coding region expression. (c) The EGFP variants were expressed from coding regions fused with a control growth hormone stable 3′UTR that lacks ARE or 3′ UTR that contains strong AU-rich elements (AREs) [from tumor necrosis factor (TNF)-α mRNA]. Data represent the expression ratio in comparison to WT EGFP (normalized to a ratio of 1). Data above are mean ± SEM of three replicates from at least two independent experiments.

Figure 4

Response of the enhanced green fluorescent protein (EGFP)-coding region variants to RNase L and dominant negative zinc-finger mutant, C124R tristetraprolin (TTP). HeLa Tet-off cell cells (3 × 10⁴ cells/well) in 96-well microplates were cotransfected with expression vectors containing (a) wild-type (WT) coding region or (b) UW-reduced coding region (UWr) for EGFP mRNA and each of RNase L or pcDNA3.1 control expression vector. Subsequently, the transcription from the cotransfected EGFP plasmid was blocked by doxycycline (1 µg/ml) for the indicated periods of time. Total RNA was extracted and subjected in reverse transcription (RT)-QPCR using TaqMan primers specific to WT and UWr EGFP mRNA (details in Materials and Methods). The one-phase exponential mRNA decay model was used to quantify the mRNA half-life for WT (R² = 0.96 and 0.99 for vector and RNase L curves) and UW-reduced EGFP mRNAs (R² = 0.9 and 0.85, vector and RNase L, respectively). The data are mean ± SEM of replicate from one representative experiment of two. Insets show Western blotting for transfected HA-tagged RNase L (HA-RL) expression using a specific antibody to HA tag. (c) Protein expression changes as a result of RNase L expression using fluorescence level ratio of green fluorescence to red fluorescence. Details of experiment are similar to those described in legend A and B. (d,e) HeLa Tet-off cells (3 × 10⁴ cells/well) in 96-well microplates were transfected for 18 hours with expression vectors containing (d) wild-type (WT) coding region or (e) UWr-coding region for EGFP in the presence of C124R expression vector or pcDNA3.1 control vector. Subsequently, the transcription was blocked by doxycycline (1 µg/ml) for the indicated periods of time. The one-phase exponential mRNA decay model was used to quantify the mRNA half life for WT (R² = 0.97 and 0.99 for vector and C124R, respectively) and UWr EGFP mRNAs (R² = 0.96 and 0.9.5 for vector and C124R, respectively). The data are mean ± SEM of three replicates from one representative experiment of two independent experiments performed. Insets show Western blotting for C124R expression. (f) Protein expression from WT EGFP-coding region, UWr EGFP-coding region, and UWr EGFP-coding region fused with tumor necrosis factor (TNF)-α ARE 3′UTR. Data is presented as fluorescence levels (mean ± SEM) of three independent experiments, each with three to four replicates. WT EGFP + pcDNA3.1 vector control was normalized to 1.0-fold ratio. The UWr EGFP and UWr EGFP+ARE data were normalized to UWr EGFP + pcDNA 3.1 control (fold ratio = 1.0).