Contribution
Introduction
When considering protein expression, promoter strength is often the first factor that comes to mind since it directly influences transcription initiation. The impact of 3'-untranslated regions (3'-UTR) of genes in this context is rarely considered as relevant as the former. However, the role of terminators in cells is not limited only to the regulation of transcription completion.
The terminator sequence can also directly affect the expression levels of the upstream encoded protein, primarily by influencing the amount of mRNA available for translation. It has already been established that the terminator affects the abundance of the corresponding mRNA, which directly correlates with the respective protein expression level. The terminator region can physically block access to nucleases resulting in the longer mRNA half-life. The stability of the mRNA's 3'-UTR is the key factor determining the strength of the terminator sequence [1].
Many plasmid vectors commonly used in laboratories include only a ‘standard’ set of terminators, and protein expression is typically assumed to depend primarily on promoter strength.
In this work, we characterized three S. cerevisiae terminators for their ability to affect protein expression, providing researchers with data to make more informed choices when selecting a terminator.
To investigate the role of terminators in regulating protein expression, we tested their potential as gene expression control elements. We constructed three yeast strains, each carrying the Venus-encoding sequence under the medium-strength pRPL18b promoter [2], followed by one of three terminators: DIT1, GIC1, or CYC1 in the 3′-UTR. The resulting protein expression levels were compared by measuring fluorescence intensity using flow cytometry.
Terminators
The DIT1 and GIC1 terminator sequences were chosen based on the article [3]. The sequence of CYC1 terminator was obtained from the Saccharomyces Genome Database (SGD) for the strain W303 (the length of the CYC1 terminator based on the information available in the literature [4]. Forward and reverse primers for the terminator amplification were designed to allow further use of the yeast MoClo toolkit [2] for plasmid construction. Terminators were amplified by PCR from yeast genomic DNA.
Table 1. Terminators used for the characterization
| Terminator | Length (bp) |
|---|---|
| DIT1 | 250 |
| GIC1 | 401 |
| CYC1 | 248 |
Plasmid Construction
After amplification and gel purification, the terminators were assembled in Golden Gate reaction into a pYTK001 plasmid vector as a type 4 custom part using Esp3I restriction enzyme [2]. In the second level Golden Gate assembly (Eco31I restriction enzyme was used), pYTK001-terminator parts plasmids were used to assemble transcriptional units (Table2).
Table 2. Transcriptional unit plasmids created after second level Golden Gate assembly
| Transcriptional unit | Terminator | Length (bp) | Comments |
|---|---|---|---|
| pRPl18b_Venus_tDIT1 | tDIT1 | 1916 | |
| pRPl18b_Venus_tCYC1 | tCYC1 | 1914 | |
| pRPl18b_Venus_tGIC1 | tGIC1 | 2067 | |
| pALD6_mRuby2_tSSA1 | tSSA1 | 6009 | Internal Control |
Each of the three transcriptional unit plasmids contained medium strength pRPl18b promoter [2], Venus-encoding sequence, and one of the terminators: DIT1, CYC1, or GIC1.
As it can be seen from Table 2, an additional transcriptional unit plasmid carrying the mRuby2 gene was constructed (explained below). At the next level assembly, transcriptional unit plasmid were used to create three multigene cassette vectors: each of which contained the same pALD6_mRuby2_tSSA1 transcriptional unit (internal control) and one of the pRPl18b_Venus transcriptional units (Venus with one of the terminators of interest) (Table 3).
Many factors can affect fluorescence intensity in living cells, such as the number of plasmids integrated in the genome, cellular state, cell-to-cell variability, etc. In order to minimize the effect of other factors on Venus fluorescence, apart from terminator strength, the pALD6_mRuby2_tSSA1 was constructed. Since the same pALD6_mRuby2_tSSA1 transcriptional unit was present in every multigene assembly, it was used to normalise the fluorescent signal of Venus to signal obtained for mRuby2 from the same strain. This would allow us to minimize the influence of all other factors on Venus efficiency, except for the terminator type.
All MoClo kit plasmid vectors used in the assembly are listed in Table 3.
Table 3. MoClo kit plasmid vectors used for transcriptional unit assembly.
| TU_tDIT1 | TU_tCYC1 | TU_GIC1 | Internal control |
|---|---|---|---|
| pYTK002 | pYTK002 | pYTK002 | pYTK003 |
| pYTK017 | pYTK017 | pYTK017 | pYTK018 |
| pYTK033 | pYTK033 | pYTK033 | pYTK034 |
| pYTK001_tDIT1 | pYTK001_tCYC1 | pYTK001_tGIC1 | pYTK052 |
| pYTK067 | pYTK067 | pYTK067 | pYTK072 |
| pYTK074 | pYTK074 | pYTK074 | pYTK074 |
| pYTK086 | pYTK086 | pYTK086 | pYTK086 |
| pYTK089 | pYTK089 | pYTK089 | pYTK089 |
| pYTK092 | pYTK092 | pYTK092 | pYTK092 |
Yeast Strain Construction
After linearization of the plasmids in URA3 selection marker, all three multigene cassette plasmids were transformed into the DOM-90 yeast strain (MATa {leu2-3,112 trp1-1 can1-100 ura3-1 ade2-1 his3-11,15 bar1::hisG} [phi+]) to get three yeast strains. Additionally two control yeast strains were created by separately transforming the pALD6_mRuby2_tDIT1 and pRPl18b_Venus_tDIT1 transcriptional units (the strongest terminator tDIT1 was chosen for the positive control plasmid) to get reference strains for each fluorescent protein.
Successful transformations were first confirmed with colony PCR and fluorescent microscopy, and then quantified using flow cytometry.
Flow cytometry and fluorescence detection
Yeast strains were grown in 3mL of complete synthetic media (CSM) with glucose 2% (m/v) for three hours. Next, 200 uL of the culture was pipetted into a 96-well plate (3 technical replicates from each colony). Samples were read using Attune™ NxT Flow Cytometer (Thermo Fisher Scientific). To measure Venus and mRuby2 signals, the blue laser I (488 nm) and yellow laser II (561 nm) were selected, and voltage was set to 470. A bandpass filter (530/30 nm) for blue laser I, and a (615/20 nm) bandpass filter for yellow laser II were selected. Cells were gated out on the FSC-A/SSC-A plot. The median values of the signals were used to create barplots.
Results and discussion
From the bar plots it can be seen that the highest reporter expression levels were achieved for the DIT1 terminator (Figure 1A, B). Strains carrying Venus with GIC1 terminator showed much lower levels of fluorescence intensities in comparison to DIT1 terminator (Figure 1 A,C), while Venus_Cyc1 strains showed the lowest Venus fluorescence (Figure 1A, C).
Figure 1. Fluorescence signal intensity of Venus protein measured with three different terminators. Values are shown after subtracting the background fluorescence and normalisation to the internal control (mRuby2 fluorescence). The experiment was repeated twice.
- A. Venus fluorescence signal in all strains with different terminators.
- B. Median values of Venus signal in the strains with DIT1 terminator (three colonies are shown).
- C. Median values of Venus signal in strains with CYC1 and GIC1 terminators.