Optimizing Large-scale Circular RNA Production Process with KACTUS Engineered RNase R

Circular RNA: A type of therapeutic drug modality. 

Circular RNA (circRNA) is a class of covalently closed, circular non-coding RNA molecules formed through back-splicing of precursor mRNA. The 5' and 3' ends are covalently linked, resulting in a structure without free termini. This unique structure grants circRNA high resistance to exonucleases and significantly greater stability compared to linear RNA, allowing it to persist longer in cells. Theoretically, under an internal ribosome entry site (IRES)-driven protein expression mechanism, circRNA can exert therapeutic effects even at lower dosage levels. These characteristics have attracted substantial interest and investment from researchers and companies, including Orna Therapeutics, Laronde (now Sail Biomedicines), Circle Pharma, CirCode Bio, and Transcripta Bio. Notably, both Transcripta Bio and Ringcode Bio have projects that have entered the clinical stage. 

Challenges in Large-Scale circRNA Production

However, the efficient large-scale production of high-purity circRNA remains challenging due to the difficulty of achieving higher circularization efficiency based on current methods such as intron self-circularization or in vitro ligation via T4 RNA Ligase I or II. Consequently, removing uncyclized linear RNA and open-circle RNA has become an indispensable step in the manufacturing process.

In large-scale circRNA manufacturing, using exonuclease to selectively digest linear RNA is the typical approach for CirRNA purification. Alexandra Sophie et al.compared the performance of current digestive enzymes in circRNA enrichment [2]. These results showed that these enzymes cannot fully deplete linear RNAs and also non-specifically digested circRNAs. This was due to 1) the existence of enzyme-sensitive sequences in certain circRNAs or 2) the secondary structure of linear RNA with strong enzyme resistance. Simply increasing enzyme concentration or extending digestion time may exacerbate the non-specific digestion of CirRNA and further reduce circRNA recovery rates. Therefore, optimizing the digestion condition in a case-by-case scenario is critical to balance circRNA recovery purity and purity.  

Figure 1. Enrichment of circular polyribonucleotide using two different combinations of 5' and 3' exonucleases. Source: Patent US20240240218A1

Optimizing RNase R Digestion for circRNA Purification

We leveraged KACTUS’ in-house engineered RNase R (Cat. No. RNR-EE001) to explore the conditions for optimal linear RNA digestion including enzyme concentration, reaction time, and temperature. Given that RNA samples with different sequences exhibit varying sensitivities to RNase R, an initial screening test is conducted before targeted optimizations. 

Initial screening reaction setup

We titrated the Rnase R as below (Table 1). The concentrations of purified RNAs were measured by Nanodrop, together with E-Gel electrophoresis for purity and recovery rate analysis. We did grayscale quantification on three major bands representing circRNA, precursors & intermediates, and nicked RNAs, the primary products in the system.  

Purity of circRNA = circRNA / (circRNA + Precursor & Intermediate + Nicked RNA)%.

Component	1	2	3	4	5	6	NC
RNase R (Unit)	2000	1000	500	340	250	125	0
10x Reaction buffer (μl)	100	100	100	100	100	100	100
Cyclized RNA Sample (μg)	500	500	500	500	500	500	500
RNase-free Water	Up to 1mL
Incubate at 37°C for 60min. After reaction, add 0.5M EDTA 50μL to terminate the reaction. Purify RNA in 5 minutes.

Table 1. Reaction conditions used for measuring RNase R activity across different RNA sequences.

Two distinct circRNA samples were selected to investigate the difference in RNase R reactivity across varied RNA sequences. Under the same reaction condition, Sample 1 (1200nt) had a lower recovery rate than Sample 2 (1600nt) (Figure 2), indicating that it is more sensitive to RNase R. Under the low enzyme concentration condition (Condition 6), the purity of Sample 1 increased from the original 44% to 65%, but the recovery rate remained suboptimal. For Sample 2, the purity increased from 28% to 56%, with a recovery rate of 30%. However, two undigested product bands were observed, indicating insufficient digestion.

Figure 2. E-gel analysis purity and recovery rate for Sample 1 (1200nt)  and Sample 2 (1600nt) in RNase R titration study, 1hr incubation at 37℃. 

Impact of enzyme concentration and incubation time in circRNA digestion

Two experimental scenarios were designed to further improve digestion efficiency. Experiments were conducted in parallel and all circRNA samples were freshly prepared to avoid the impact of repeated freeze-thaw cycles: 

1. Reducing enzyme concentration while extending reaction time (Table 2, Figure 3).

2. Increasing enzyme concentration while shortening reaction time (Table 3, Figure 4).

Component	Cyclized RNA sample 1					Cyclized RNA sample 2
	1	2	3	4	NC	1	2	3	NC
RNase R (Unit)	125	60	30	15	0	60	30	15	0
10x Reaction buffer (μL)	100	100	100	100	100	100	100	100	100
Cyclized RNA Sample (μg)	500	500	500	500	500	500	500	500	500
RNase-free Water	Up to 1mL
Incubate at 37°C for 120min. After reaction, add 0.5M EDTA 50μL to terminate the reaction. Purify RNA in 5 minutes.

Table 2. Reaction conditions for reduced enzyme concentration and extended reaction time (2 hours).

Component	Cyclized RNA sample 1							Cyclized RNA sample 2
	NC	1	2	3	4	5	6	NC	1	2	3	4	5	6
RNase R (Unit)	0	2000	1000	500	340	250	125	0	2000	1000	500	340	250	125
10x Reaction buffer (μl)	100	100	100	100	100	100	100	100	100	100	100	100	100	100
Cyclized RNA Sample (μg)	500	500	500	500	500	500	500	500	500	500	500	500	500	500
RNase- free Water	Up to 1mL
Incubate at 37°C for 30min. After reaction, add 0.5M EDTA 50μL to terminate the reaction. Purify RNA in 5 minutes.

Table 3. Reaction conditions for increased enzyme concentration and shortened reaction time (30 minutes). 

Under the lowest concentration (Condition 4, Figure 3), Sample 1 achieved a 20% recovery rate and 70% purity, indicating improved purity but a a reduced recovery rate compared to the initial screening study. For Sample 2, the recovery rate and purity reached 21% and 64%, respectively, a better balance between purity and recovery rate compared to the first round.

Figure 3. Purity and Recovery Rate of Sample 1 and Sample 2 after 2-hour incubation with lower RNase R concentrations.

A significant increase in non-specific digestion was observed in the presence of excessive RNase R, even with a digestion time of just 30 minutes (Figure 4). Under the three highest RNase R concentrations, the recovery rate was extremely low for both circRNA samples. Under the lowest RNase R concentration (Condition 6, Figure 5), both samples achieved ~70% purity. Notably, Sample 2, which had a low initial circularization efficiency of only 25%, showed a significant improvement in purity. However, the recovery rate for both samples was only around 10%, which presents a critical challenge for industrial-scale production.

Figure 4. Purity and Recovery Rate of  Sample 1 and Sample 2 after 30-minute incubation with high RNase R concentration.

Taken together, these data suggested that either lower-dose RNase R with extended reaction time or higher-dose RNase R with shortened reaction time can improve circRNA purity. However, the lower dose-longer incubation is a preferred scenario for a more balanced purity and recovery rate.

Effects of System Volume and Temperature in circRNA Purification

Next, we explored how the volume of the reaction system (Table 4, Figure 5) and the temperature (Table 5, Figure 6) could influence digestion outcomes, using Sample 1 as an example with low enzyme dose, long incubation time strategy:

Component	Cyclized RNA sample 1
	1	2	3	4	1	2	3	4	1	2	3	4
RNase R (Unit)	125	62.5	30	15	125	62.5	30	15	125	62.5	30	15
10x Reaction Buffer (μL)	100	100	100	100	100	100	100	100	100	100	100	100
Cyclized RNA Sample (μg)	500	500	500	500	500	500	500	500	500	500	500	500
RNase-free Water	up to 500μL				up to 1mL				up to 2mL
Incubate at 37°C for 120min. After reaction, add 0.5M EDTA 50μL to terminate the reaction. Purify RNA in 5 minutes.

Table 4. Reaction conditions following low-enzyme dose and long reaction time strategy. 

It was shown that, when maintaining the same input amounts of each component, increasing system volume did not significantly affect the final product purity while the recovery rate was significantly reduced (Figure 5). This may be related to our precipitation method using lithium chloride. In the actual circRNA production circumstances, the optimal reaction system concentration should be selected based on the purification method used.

Figure 5. Purity and Recovery Rate of Sample 1 under different system volumes under lower RNase R dosage.

Finally, the temperature study was conducted at 37°C, 39°C, and 41°C (Table 5). The results indicate that under the low enzyme dose condition, moderately increasing the reaction temperature helped improve the purity without significantly compromising the recovery rate (Figure 6).

Component	Cyclized RNA sample 1
	1	2	3	1	2	3	1	2	3
RNase R (Unit)	60	30	15	60	30	15	60	30	15
10x Reaction buffer (μL)	100	100	100	100	100	100	100	100	100
Cyclized RNA Sample (μg)	500	500	500	500	500	500	500	500	500
RNase-free Water	up to 500μL
Condition	Incubate at 37°C for 120min			Incubate at 39°C for 120min			Incubate at 41°C for 120min
After reaction, add 0.5M EDTA 50μL to terminate the reaction. Purify RNA in 5 minutes.

Table 5. Reaction conditions for low enzyme dose, long reaction time, and high reaction system concentration strategy. 

Figure 6. Purity and Recovery Rate of Sample 1 at 37°C, 39°C, and 41°C under lower RNase R dosage.

Optimizing circRNA Purification Process using KACTUS engineered RNase R with high linear RNA specificity.

Overall, strategies such as using a lower enzyme dose with a longer reaction time, increasing reaction system concentration, and moderately raising reaction temperature can help balance purity and recovery rate. Given that circRNA sequence and structure significantly impact digestion efficiency, we recommend that users optimize reaction conditions for each specific RNA sequence across these dimensions. When necessary, Design of Experiments (DOE) can be applied to streamline optimization efforts, reducing experimental workload while pre-enriching circRNA to ease the burden on downstream purification processes.

The wildtype RNase R typically exhibits higher nonspecific digesting activity. To overcome this issue, KACTUS developed the off-the-shelf RNase R enzyme for linear RNA digestion. This enzyme has been engineered for high specificity to linear RNA without digesting circRNA. To view this product, click here.

KACTUS has developed a full GMP-Grade and GMP-Ready enzyme portfolio for RNA production. To learn more about these products from a team member, please contact us. 

References

1. Qu L, Yi Z, Shen Y, Lin L, Chen F, Xu Y, Wu Z, Tang H, Zhang X, Tian F, Wang C, Xiao X, Dong X, Guo L, Lu S, Yang C, Tang C, Yang Y, Yu W, Wang J, Zhou Y, Huang Q, Yisimayi A, Liu S, Huang W, Cao Y, Wang Y, Zhou Z, Peng X, Wang J, Xie XS, Wei W. Circular RNA vaccines against SARS-CoV-2 and emerging variants. Cell. 2022 May 12;185(10):1728-1744.e16. doi: 10.1016/j.cell.2022.03.044. Epub 2022 Apr 1. PMID: 35460644; PMCID: PMC8971115. 

2. De Boer, A. S., Hobert, E. M., & Depeter, J. A. (2024). Methods of Enriching for Circular Polyribonucleotides (Patent No. US20240240218A1). U.S. Patent and Trademark Office. https://patents.google.com/patent/US20240240218A1/ko