Developing an enhanced chimeric permuted intron-exon system for circular RNA therapeutics

Abstract

Rationale: Circular RNA (circRNA) therapeutics hold great promise as an iteration strategy in messenger RNA (mRNA) therapeutics due to their inherent stability and durable protein translation capability. Nevertheless, the efficiency of RNA circularization remains a significant constraint, particularly in establishing large-scale manufacturing processes for producing highly purified circRNAs. Hence, it is imperative to develop a universal and more efficient RNA circularization system when considering synthetic circRNAs as therapeutic agents with prospective clinical applications. Methods: We initially developed a chimeric RNA circularization system based on the original permuted intron-exon (PIE) and subsequently established a high-performance liquid chromatography (HPLC) method to obtain highly purified circRNAs. We then evaluated their translational ability and immunogenicity. The circRNAs expressing human papillomavirus (HPV) E7 peptide (43-62aa) and dimerized receptor binding domain (dRBD) from SARS-CoV-2 were encapsulated within lipid nanoparticles (LNPs) as vaccines, followed by an assessment of the in vivo efficacy through determination of antigen-specific T and B cell responses, respectively. Results: We have successfully developed a universal chimeric permuted intron-exon system (CPIE) through engineering of group I self-splicing introns derived from Anabaena pre-tRNA^Leu or T4 phage thymidylate (Td) synthase gene. Within CPIE, we have effectively enhanced RNA circularization efficiency. By utilizing size exclusion chromatography, circRNAs were effectively separated, which exhibit low immunogenicity and sustained potent protein expression property. In vivo data demonstrate that the constructed circRNA vaccines can elicit robust immune activation (B cell and/or T cell responses) against tumor or SARS-CoV-2 and its variants in mouse models. Conclusions: Overall, we provide an efficient and universal system to synthesize circRNA in vitro, which has extensive application prospect for circRNA therapeutics.

Keywords: chimeric permuted intron-exon system (CPIE); circular RNA (circRNA); mRNA therapeutics; ribozyme; size exclusion chromatography.

Figure 1

Design of the CPIE system for circRNA synthesis. A schematic diagram depicts the components of the CPIE system, along with its hypothetical splicing pathways. In pathway I, the PIE I module can facilitate the folding process of the PIE II module and expedite circularization. In pathway II, the PIE I module initially generates a circular intermediate RNA, followed by an intramolecular splicing reaction mediated by the PIE II module. The splicing sites (E1-E4), spacers and hybridization arms of the CPIE system are indicated.

Figure 2

Characterization of the CPIE system. (A) Schematic diagrams illustrating the engineered CPIEa and CPIEb systems and splicing procedure are presented. (B) Precursor RNA circularization within CPIEa and CPIEb systems are evaluated, separately. The bands corresponding to putative precursor RNA, intermediate RNA, circRNA, and free intron are indicated. (C) The efficiency of RNA circularization using the CPIEa and CPIEb systems is quantified by RT-qPCR. Data are shown as mean ± sd. Statistical analysis was carried out by using unpaired two-tailed t-tests (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 3

The purification and expression of circRNA. (A-B) HPLC chromatogram analysis was performed on the final circRNA products generated using CPIEa (A) and CPIEb (B) systems, respectively. The retention time corresponding to the collection of samples has been indicated. (C-D) Agarose gel electrophoresis results were obtained along with HPLC chromatograms of collected circRNA^EGFP produced through CPIEa (C) and CPIEb (D) systems, respectively. (E) RT-PCR was performed to target the putative splice junction, followed by Sanger sequencing for validation of the splicing sites of circRNAs generated by the CPIEa and CPIEb systems. A representative sequencing analysis is provided. (F-G) Fluorescence microscopy imaging was conducted at 24 h after transfecting HEK293F (F) and HEK293T (G) cells with circRNA^EGFP and m1ψ-modified linear mRNA expressing EGFP. (H-I) The mean fluorescence intensity (MFI) change in the HEK293F (H) and HEK293T (I) cells was measured starting from 24 h after transfection with either circRNA^EGFP or m1ψ-modified linear mRNA expressing EGFP, and this measurement continued for 5 days. All MFI data were normalized with their respective MFI values at 24 h.

Figure 4

Immunogenicity of the circRNA generated by CPIE systems. RT-qPCR analysis was performed to evaluate the relative expression of specific genes, RIG-1 (A), TNFα (B), IFNβ (C) and IL-6 (D) in A549 cells transfected with ePIE- and CPIE-generated circRNA^EGFP at 12 h post-transfection. The mRNA expression levels were normalized using the reference gene actin, and compared to those observed in mock-transfected cells. Notably, Poly(I:C) and unmodified linear mRNA^EGFP served as controls for this study. Data are shown as mean ± sd. Statistical analysis was carried out by using unpaired two-tailed t-tests (ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001).

Figure 5

Preparation of the LNP@circRNA^E7-D2GFP complex. (A) Schematic representation illustrating the components of the IVT template for circRNA^E7-D2GFP preparation. (B) Agarose gel electrophoresis result and HPLC chromatogram depicting the collected circRNA^E7-D2GFP generated through CPIEb system. (C) The expression of circRNA^E7-D2GFP was assessed in HEK293F cells using flow cytometry after transfection for 24 h. (D) Schematic depiction of the LNP@circRNA^E7-D2GFP complex formation. (E) Size distributions analysis conducted on the LNP@circRNA^E7-D2GFP complex. (F) Zeta potential evaluation performed on the LNP@circRNA^E7-D2GFP complex. (G) TEM image displaying the morphology of the LNP@circRNA^E7-D2GFP complex. Scale bar, 100 nm.

Figure 6

LNP@CircRNA^E7-D2GFP induces robust T cell responses and demonstrates efficacy as both prophylactic and therapeutic anti-tumor vaccines. (A) Timeline of the anti-tumor assay using LNP@circRNA^E7-D2GFP as a prophylactic vaccine in TC-1 tumor model is presented. (B) C57BL/6J mice (n = 5-6), which were immunized twice with circRNA^E7-D2GFP or vehicle control, were challenged with TC-1 cells (5 × 10⁵). (C) Timeline of the anti-tumor assay using LNP@circRNA^E7-D2GFP as a therapeutic vaccine in TC-1 tumor model is depicted. (D-E) TC-1 tumor-bearing C57BL/6J mice (n = 9) were treated twice with circRNA^E7-D2GFP or vehicle control as indicated. Tumor growth (D) and survival curves (E) are shown. (F-G) Seven days after second immunization, lymphocytes from the spleens were isolated and stimulated with E7 peptide (49-57aa). IFN-γ⁺CD8⁺ T cells were analyzed by flow cytometry, representative results (F) and quantification (G) of the data were shown. Data are shown as mean ± sd. P value was determined by log-rank test (E) or unpaired two-tailed t-tests (*P < 0.05, **P < 0.01, ***P < 0.001) (B, D and G).

Figure 7

CircRNA^dRBD vaccine induces potent B cell and T cell immune responses against SARS-CoV-2. (A) Schematic representation of the dRBD antigen is depicted. SP: signal peptide. (B) The agarose gel electrophoresis result and HPLC chromatogram illustrate the collected circRNA^E7-D2GFP generated through CPIEb system. (C) The Western blot result demonstrates the expression of circRNA^dRBD. (D) A timeline outlining the immunization and assessment process to evaluate the circRNA^dRBD vaccine is provided. BALB/c mice (n = 5-7) were immunized with circRNA^dRBD or vehicle control at a 2-week interval. Serum samples were collected 13 days after the first immunization and 14 days after the second immunization. (E-J) ELISA assays reveal specific total IgG titers against RBD from SARS-CoV-2 (E), as well as its delta (F), BA.2 (G), BA.4 (H), BQ.1 (I), and CH.1.1 (J) variants. (K) A radar map comparison showcases specific IgG titers against RBD of SARS-CoV-2 and its variants. (L-M) Seven days after second immunization, lymphocytes from the spleens were isolated and stimulated with dRBD fusion protein. IFN-γ⁺CD8⁺ T cells were analyzed by flow cytometry, representative results (l) and quantification (m) of the data were shown. Data are shown as mean ± sd. Statistical analysis was carried out by using unpaired two-tailed t-tests (ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001).