The Dawn of the RNA Medicine Era: Can Circular RNA (circRNA) Surpass Its Predecessors?
Published on: July 31, 2024
Key features of circRNA include greater stability in the cytoplasm compared to mRNA and lower immunogenicity of unmodified transcripts. These advantages position circRNA to fully exploit the potential of RNA-based therapies.
RNA technology has made significant recent strides in biomedicine. The mRNA vaccine technology, for example, gained substantial attention during the global COVID-19 pandemic for its minimal side effects, low cost, and high efficacy. mRNA medicines have now established their therapeutic value across various diseases such as infections, cancers, and rare diseases. However, the instability and immunogenicity of RNA has limited its clinical use. The benefits of circRNA might address these mRNA drawbacks, further enhancing RNA’s clinical impact.
Researchers have devised several methods for synthesizing circRNA in vitro (Figure 1). In this issue, we will explore some prevalent techniques for in vitro circRNA synthesis.
Fig. 1: Timeline of important events in the discovery, research and development of circRNAs. The concept of circRNAs was proposed in 1976, and extensive research began in 2012 to confirm the abundance, function, and stability of circRNAs. doi.org/10.3389/fbioe.2021.78788)
The main approach for in vitro circRNA synthesis involves connecting the ends of linear RNA precursors to form a covalently closed ring (Figure 2). Linear RNA can be synthesized either chemically or enzymatically. Chemical methods can directly introduce a 5’-monophosphate during the synthesis process, aiding subsequent circularization. However, chemical methods are costly, yield low quantities of products, and are typically more suitable for generating small RNA molecules (<50-70 nt).
Fig. 2: Linear RNA precursors are produced by in vitro transcription, and linear RNA precursors are ligated to synthesize circRNA. (doi.org/10.3389/fbioe.2021.787881）
Enzymatic methods are generally preferred for producing longer linear RNA molecules due to their cost-effectiveness, scalability, and ability to generate higher quantities of RNA.
Enzymatic methods are typically carried out using in vitro transcription (IVT), which includes a DNA template, reaction buffer, and bacteriophage RNA polymerase. Commonly used polymerases come from bacteriophages such as T7, SP6, or T3. Among these, T7 RNA polymerase is frequently used in IVT processes due to its efficiency and reliability. IVT can generate longer RNA at lower cost compared to chemical synthesis methods, but the run-off nature of bacteriophage polymerases might sometimes result in incomplete RNA sequences. Some studies have enhanced transcription quality and reduced side effects through genetic engineering of these polymerases.
Several methods exist for circularizing linear RNA precursors in vitro, including chemical, enzymatic, and ribozyme ligation. Chemical ligation, using agents like BrCN or 1-ethyl-3(3’-dimethylaminopropyl) carbodiimide, links DNA-RNA hybrids but is inefficient and raises biosafety concerns. Moreover, it forms 2’,5’-phosphodiester bonds instead of the natural 3’,5’-phosphodiester bonds, making enzymatic synthesis methods more favorable
Enzymatic ligation involves T4 DNA ligase (t4 Dnl), T4 RNA ligase (T4 Rnl 1), and T4 RNA ligase 2 (T4 Rnl 2) from T4 bacteriophage (Figure 3).
It is worth noting that such ligation reactions require the linear RNA precursor to provide a 3'-OH on the acceptor substrate and a 5'-monophosphate on the donor substrate. If the linear RNA precursor is chemically synthesized, the 5'-monophosphate can be generated during the synthesis process or post-synthesis using ATP and T4 polynucleotide kinase. However, in IVT reactions, the synthesis of the linear RNA precursor typically starts from 5'-pppG, so the 5'-end must be dephosphorylated first, followed by the addition of a 5'-monophosphate using ATP and T4 polynucleotide kinase.
T4 Dnl can ligate double-stranded structures such as DNA/RNA hybrids (Figure 3A). This method needs the assistance of a cDNA bridge or template. . Generally, for high-quality ligation, the cDNA bridge should contain at least 10 nucleotides on either side of the ligation site. The advantage of this method is the increased precision of the ligation sites. However, it requires linear RNAs precursors to have minimal secondary structures and low “U” content in double-stranded regions.
T4 Rnl1 is a common RNA ligase that catalyzes the nucleophilic attack of a 3’-OH end on an activated 5’-end, forming a 5’,3’-phosphodiester bond to create circRNA (figure 3B). It shows nucleotide preferences and efficiently ligates short circRNA(6~8nt) and ssRNA, but its efficiency decreases with large RNA molecules or significant secondary structures. It may also cause oligomerization, which is a side effect that limits the amount of circRNA produced.
T4 Rnl2 also ligates RNA and shows higher activity with double-stranded RNA (Figure 3C). It can connect ssRNA ends with the help of RNA splints but it less efficient with long RNA segments and may have side effects.
Fig. 3 Enzyme-based methods. (Doi.org/10.3390/ IJMS232113247). (A) T4 DNL can be linked to the double-stranded structure of the double helix, such as the DNA/RNA hybrid. With the help of cDNA bridge, T4 Dnl enables precise RNA ligation.
(B) T4Rnl 1 catalyzes the nucleophilic attack of the 3′-OH terminus on the activated 5′-terminus, forming a covalent 5′,3′-phosphodiester bond. cDNA bridging prevents linear RNA precursors from folding into unsuitable structures. (C) T4Rnl2 is more suitable for linear RNA precursors with double-stranded junctions. With the help of RNA splints, T4Rnl2 can also achieve ligation of ssRNA ends.
Ribozymes, also known as ribonuclease enzymes, are special introns that self-spice to remove themselves from precursors and ligate adjacent exon into mature mRNA without external enzymes or protein factors. Group I introns are found mainly in mitochondria, chloroplasts, and some lower eukaryotes, with rare occurrences in bacteriophage. Group II introns are mostly present in organelles and bacteria.
In 1992, Puttaraju and Been first used the “Permuted Introns and Exons (PIE)” method to utilize the self-splicing activity of Group I introns for RNA circularization. PIE is currently one of the most common techniques for synthesizing circRNAs in vitro. The PIE method involves rearranging intron-exon constructs to form circular RNA. The gene of interest (GOI) can be integrated into exons, resulting in circRNA after transcription and splicing (Figure 4). RNA transcribed from the “PIE sequence” forms a ring after two ester exchange reactions. The first reaction releases the 5’ end intron, while the second reaction produces circRNA and releases the 3’ end intron.
Fig. 4. Principles of PIE method based on type I/II introns. The mechanism of type I intron or type II intron ribozyme is slightly different, but both can be applied to the PIE method. (a) The "PIE sequence" contains end-to-end fusion exons, with the 5′ half (gray) of the type I intron located at the tail of the exon and the 3′ half (black) located at the head of the exon, into which the gene of interest (GOI) can be integrated, allowing the synthesis of circular RNA.
(b) Group II introns can also be used for circRNA synthesis, involving reverse slicing between 5’ and 3’ splice sites of exons. This method can more accurately connect linear RNA precursors but forms 2’,5’-phosphodiester bonds instead of the natural 3’,5’-bonds.
Compared to T4 DNA/RNA ligase methods, the PIE method is more suitable for synthesizing large circRNAs with simpler reaction conditions and purification methods. For example, Orna Therapeutics employs the Group I intron self-splicing system for circRNA synthesis. However, the PIE method has limitations, such as variability in circRNA domains due to RNA secondary structures and potential differences between the final circRNA and the original linear RNA precursor sequence, which may affect functional validation. Group II intron PIE avoids some of these issues but introduces splicing scars that can cause immunogenicity.
To improve the PIE method, researchers have explored various modifications to enhance circularization efficiency. For instance, adding homologous arms to linear RNA precursors can stabilize introns and improve folding, significantly increasing circularization efficiency and extending lengths to 5kb. Optimized circRNA can produce large quantities of purifiable proteins. Curemed’s Clean-PIE system can identify optimal circularization sites within the coding region or IRES region, eliminating the need to introduce exogenous exonic sequences of E1 and E2. This approach also improves circularization efficiency. CirCode Biomed utilizes Group II intron PIE technology to ensures circularization efficiency while reducing immunogenicity.
HPR can generate circRNA through rolling circle reactions and self-splicing with circular single-stranded DNA template (Figure 5). Linear RNA precursors with HPR can adopt two selectable cleavage-active conformations, leading to the removal of the 3' and 5' ends. This process results in an intermediate structure with a 5'-OH group and a 2',3'-cyclic phosphate, ultimately generating the desired circRNA product.
Fig. 5. Strategies for circRNA synthesis by hairpin nucleases
This method is mainly used for producing small circRNAs and is influenced by conditions such as freezing and ion concentration. HPRs have notable drawbacks: 1) circRNA instability due to the dynamic balance in HPR-catalyzed cleavage and ligation, and 2) potential negative impacts of the HPR sequence on circRNA function.
RNA-based medicines show great promise as vaccines, therapeutics, and diagnostics. They can be manufactured quicker and more cost-effectively, and to-date have shown fewer clinical adverse effects than traditional medicines. Additionally, RNA-based medicines can be rapidly designed to be highly target specific, and they have very predictable pharmacodynamics/pharmacokinetics. Two of the most significant obstacles preventing broader clinical utility of these genomic medicines are RNA stability and delivery. circRNA is a very exciting RNA stability/durability innovation, leading to higher and longer-lasting RNA expression. Industrialization of circRNA manufacturing processes, building upon the strategies highlighted in this blog, will be critically important to further broadening the utility of RNA–based medicines.