Development of T7 RNA polymerase (low dsRNA) with lower terminal transferase and RDRP activities through FADS and semi-rational design.

In recent years, mRNA-based drugs have garnered significant attention and attracted research focus. T7 RNAP is known for its simplicity and ability to catalyze RNA synthesis with high yield and fidelity in vitro. However, during the transcription process, T7 RNAP can produce by-products such as abortive short RNAs, prematurely terminated RNAs, and 3’-extended RNAs, which creates favorable conditions for the formation of dsRNA. Therefore, reducing the production of dsRNA has become an urgent need in practical applications.


Recently, a groundbreaking study titled "FADS and semi-rational design modified T7 RNA polymerase reduced dsRNA production, with lower terminal transferase and RDRP activities" was published online by Yeasen and Molefuture Team. Researchers successfully engineered T7 RNAP to significantly reduce the production of dsRNA byproducts during transcription, reducing it to 1.8% of the wild-type enzyme through a combination of directed evolution and semi-rational design.


FADS-based screening of T7 RNAP

For screening T7 RNAP, FADS(Fluorescence-activated droplet sorting) was chosen to meet the high-throughput requirements of protein evolution. To prevent premature fragmentation of cells, Researchers used a PDMS chip with dual aqueous phases, where lysozyme was mixed with the cells on the chip and exerted its activity within the droplets Figure 1. Researchers generated several random mutant libraries with a diversity ranging from 105 to106, After screening, 10 dominant variants were identified and designated as Mut1 to Mut10.  As shown in Figure 2E and Figure 2F, the dsRNA content of Mut1, Mut7, and Mut9 was significantly lower than that of the wild-type


Figure 1. Overview of the FADS




Figure 2. Random library screening and variants evaluation



Semi-rational design of T7 RNAP

Researchers conducted a semi-rational design exploration, created ten single-site saturated libraries and employed the traditional microtiter plate-based dual probesmethod for screening(Figure 3). The added 5'-end probe is employed for the detection of RNA yield.


Figure 3. Structure-guided semi-rational design for decreased dsRNA


After screening over 1000 mutants, Researchers identified six candidates: Mut11 (K180E), Mut12 (S228F), Mut13 (G238L), Mut14 (A70Q), Mut15 (A383H), and Mut16 (I743L). The total RNA yield of all six mutants did not significantly differ from the wild-type, indicating comparable overall transcription efficiency. Expectedly, the dsRNA content of Mut11 and Mut14 was significantly lower compared to the wild-type(Figure 4).


Figure 4. Microtiter plate screening and variant evaluation


Mut17 from DNA shuffling yields less dsRNA

To further optimize the T7 RNAP, the four variants with low dsRNA content while meeting production requirements were selected. Researchers then constructed a DNA shuffling library and screened with microtiter plates, identified a variant that exhibited lower dsRNA production compared to its parental T7 RNAP, named Mut17 (A70Q/F162S/K180E). They subsequently analyzed the transcriptional products of Mut17 and its parental variants (Mut14, Mut11, and Mut7). As shown in Figure 5, all four variants exhibited significant reductions in dsRNA production during transcription compared to the wild-type. Remarkably, Mut17 showed significantly lower dsRNA content of only 1.80% and as low as 0.007 ng/μg in the solvent system of screening.


Figure 5. dsRNA content of Mut17 and its parental mutations


The immunogenicity of the IVT product from the mutant is significantly lower than that of the wild type.

Next, We evaluated the immunogenicity of the IVT products in murine RAW264.7 cells. As shown in Figures 2A and 2B, IFN-β mRNA and protein levels were reduced in RAW264.7 cells transfected with mRNA produced by mutants compared to wild-type, indicating that mRNA synthesized by the wild-type T7 RNAP elicited the strongest immune response, while mRNA from the mutants showed a significantly reduced response. Specifically,the IFN-β mRNA of cells treated by Mut11 RNA was only 9.7% of the wild-type treated cells, and its protein was 12.93 pg/mL. Moreover, the expression of EGFP showed no significant differences in quantity or quality among mRNAs generated by different T7 RNAPs (Figure 6C), meeting the requirements for industrial applications.

Figure 6. IFN-β response and EGFP expression in mammalian cells


The mutant's RDRP and terminal transferase activities are much lower than those of the wild type.

To investigate the impact of mutations on reducing dsRNA, researchers conducted in silico studies on all these mutations. The result suggests a clear indication of reduced RDRP activity in the variants, as they show a decreased tendency to utilize RNA as a template. Similarly, this may have had an impact on the terminal transferase activity of T7 RNAP. As shown in Figure 7, under different concentrations, all 4 variants exhibited less than 50% of the fluorescence value compared to the wild-type.


Subsequently, a 3’-end heterogeneity assay was employed to characterize the terminal transfer activity of T7 RNAP. When focusing on the proportion of RNA with n>0, which reflects the terminal transferase activity, researchers found that these RNAs accounted for 82.94% in the wild-type, while the mutants exhibited the following percentages: Mut11 (70.29%), Mut7 (67.88%), Mut14 (63.31%), and Mut17 (55.62%) (Figure 8). Importantly, these percentages are highly consistent with the abundance of dsRNA in the products (Figure 5). These results indicate a significant reduction in terminal transferase activity of mutants, which, together with the decrease in RDRP activity, contributes to the decrease in dsRNA. Specifically, the terminal transferase activity may play a more crucial role, as evidenced by the lowest accumulation of n>0 RNAs observed in Mut17, which also exhibits the lowest dsRNA production (Figure 5 and Figure 8).


Figure 7. Relative RDRP activity


Figure 8. Heterogeneity in the 3’-end of RNA products






This study delivers a robust methodology for identifying mutants with reduced dsRNA content and successfully isolates multiple dsRNA mutants that have reduced RNA-dependent RNA polymerase and terminal transferase activity. It substantially bolsters the validation of safety and efficacy integral to mRNA therapeutics, underscoring its profound implications for advancing mRNA vaccines and gene therapy applications.


At the same time, the parent company Yeasen has also launched a T7 RNAP product that significantly reduces dsRNA content. Multiple partners have already tested the modified T7 RNAP mutants developed by Molefuture and the product has been highly recognized by customers. Our customers believe that the use of the new enzyme simplifies the purification process of mRNA and greatly reduces the immunogenicity of IVT products, demonstrates outstanding performance in multiple application scenarios, proving its wide potential application in the field of biopharmaceutical!



Ordering Information

The following are representative products offered by Yeasen. Additional sizes are available. Our products are highly optimized to work in concert, to help ensure superior performance and reproducibility. We can also provide customized services. If you’re interested in a product that isn’t shown, contact us and we’ll work with you to meet your needs.


Product name SKU Specifications
CleaScrip™ T7 RNA Polymerase (low ds RNA, 250 U/μL)  10628ES 10/100 KU
T7 High Yield RNA Synthesis Kit 10623ES 50/100/500 T
T7 RNA Polymerase GMP-grade (50 U/μL)  10624ES 5000/50000 U
T7 RNA Polymerase GMP-grade(250 U/μL)  10625ES 10/100 KU
10×Transcription Buffer 2 GMP-grade 10670ES 1/10 mL
Pyrophosphatase,Inorganic GMP-grade 0.1 U/μL)  10672ES 10/100/1000 U
Murine RNase inhibitor GMP-grade 10621ES 10/20/100 KU
BspQI GMP-grade  10664ES 500/2500 U
DNase I GMP-grade  10611ES 500/2000/10000 U
mRNA Vaccinia Capping Enzyme GMP-grade  10614ES 2000/10000/100000 U
mRNA Cap 2'-O-Methyltransferase GMP-grade  10612ES 2000/10000/50000 U
10×Capping buffer GMP-grade 10666ES 1/10 mL
S-adenosylmethionine (SAM)(32 mM) 10619ES 0.5/25/500 mL
Pseudouridine-5-triphosphate,trisodium salt solution (100 mM) 10650ES 20 μL/100 μL/1 mL
N1-Me-Pseudo UTP sodium solution(100 mM) 10651ES 20 μL/100 μL/1 mL
ATP Solution(100 mM) 10129ES 1/25/500 mL
CTP Solution(100 mM) 10130ES 1/25/500 mL
UTP Solution(100 mM) 10131ES 1/25/500 mL
GTP Solution(100 mM) 10132ES 1/25/500 mL
NTP Set Solution (ATP, CTP, UTP, GTP, 100 mM each) 10133ES 1 Set (4 vial)
Hieff NGS™ RNA Cleaner 12602ES 1/5/60/450 mL
ATP Tris Solution GMP-grade ( 100 mM) 10652ES 1/5/25/500 ml
CTP Tris Solution GMP-grade ( 100 mM) 10653ES 1/5/25/500 ml
GTP Tris Solution GMP-grade ( 100 mM) 10655ES 1/5/25/500 ml
Pseudo UTP Tris Solution GMP-grade ( 100 mM) 10656ES 1/5/25/500 ml
N1-Me-Pseudo UTP Tris Solution GMP-grade ( 100 mM) 10657ES 1/5/25/500 ml
ARCA (Anti Reverse Cap Analog)  10681ES 1/5/25/500 ml
Double-stranded RNA (dsRNA) ELISA kit  36717ES 48T/96T