English

• Nucleic acid nanotechnology allows for the accurate assembly of nucleic acid molecules into nanostructures with customizable shapes, sizes and chemical compositions [1-5]. Given their programmable structure, ability for bottom-up self-assembly [6], these nanostructures have extensive applications in cellular protein translation [7], cell motility [8], and immune activation [9]. Ribonucleic acid (RNA) is an attractive material for constructing nanoparticles via nanotechnology [10]. The field of RNA nanotechnology has achieved rapid progress in the past decade. The first RNA nanostructures were proposed based on the three-dimensional organization of protein domains [11]. For the purpose of designing nanostructures that self-assemble in a controllable way, modular units of functional nucleic acids are employed to synthesize nano-scale supramolecular structures. Various geometric nano-scaffolds and RNA nanoparticles have been acquired via the methods of 'sticky' or 'dangling' ends [12,13], loop–receptor contact [14,15], branch extension [16-18], hand-in-hand [19-21], foot-to-foot [22,23] and RNA–protein complex interactions [24].

  In recent times, single-stranded RNA (ssRNA) origamis have garnered attention as a versatile platform for biomedical applications. These origami structures are nucleic acid polymers that self-fold into specified compact shapes in a manner similar to proteins [25]. They not only have significant scientific importance, but also possess key advantages in terms of practicality compared with current multicomponent self-assembly paradigms [26-28]. First, compared with multistranded nucleic acid nanostructures formed by self-assembly, ssRNA origamis formed by self-folding have greater potential for cloning, amplification and replication [29]. Second, the single-molecule folding process is uncorrelated with the concentration of reactant, allowing for a higher formation yield and more stable folding kinetics [25,28]. Third, single-stranded structures can be assembled into highly pure and homogeneous systems. In contrast, multistranded nucleic acid nanostructures usually consist of multiple components and often exhibit defects such as misassembled or missing strands [30]. Therefore, utilizing ssRNA origamis enables the large-scale production of stable, reliable, and morphologically uniform nanomaterials [25,28,31]. Furthermore, ssRNA origamis can also be equipped with a variety of structural or sequential modules to enhance specific biomedical functions, including nano-vaccine [32], gene editing [33], drug delivery [34], facilitating high-resolution structural characterization [35,36] and regulation of molecular interactions [36-38]. However, to our knowledge, there are no investigations utilizing these ssRNA origamis as modular units of functional nucleic acids to synthesize supramolecular structures with precisely controlled valence and explore their biological functions.

  In this report, we constructed a rod-like ssRNA origami, named Rod RNA-OG, using the cohesion of parallel crossovers (PXs). We then utilized Rod RNA-OG as a monomer to synthesize valence-programmed RNA structures in a one-pot manner in PBS buffer. We further demonstrated that Rod RNA-OG and its polyvalent assemblies are resistant to RNase degradation and can be efficiently internalized by macrophages for subsequent innate immune activation, even in the absence of external protective agents like polymers or lipids. These findings suggest that polyvalent RNA origamis could be a promising platform for immunotherapy applications.

  Naturally, most protein and RNA structures have no crossing number [39-41]. In contrast, designs derived from conventional scaffolded RNA origami often result in intricate knots with elevated crossing numbers, which may impede accurate folding. Referencing Yan's work, which utilized ssDNA or ssRNA to fold into complex shapes with simple topologies and a crossing number of zero [25], we designed an ssRNA origami unit (Fig. 1). To reduce the crossing count in the ssRNA origami structures, we employed PXs for interhelical cohesion. As illustrated in Fig. 1A, in these PX positions, RNA strands avoided passing through the central plane, marked by dashed lines in the model, thus reducing the complexity of knotting. The intermediate structure of the ssRNA origami included unpaired single-stranded regions and paired helical domains. While the design required multiple crossings of the RNA strand through the central plane, these occurred specifically within the helical domains and not the locking domains. Upon complete formation of all locking domains through base pair recognition, as shown in Fig. 1B, the intermediate would transform into the fully folded ssRNA origami structure.

  Figure 1

  

  Figure 1.  Design and synthesis of Rod RNA-OG. (A) Schematics of the PX designs. The central plane is marked by dashed lines in the model. (B) Schematics illustrating the complete formation of a locking domain through base pair recognition. (C-E) Models of double-stranded intermediate structures of Rod RNA-OG. Models of fully formed structures of Rod RNA-OG. (E, H) Pipeline-style models visualizing the folding pathway of the intermediate structure. The model is colored with a gradient, where the 5′ and 3′ ends of the ssRNA are represented in red and the middle portion is shown in blue. (I) Workflow of Rod RNA-OG synthesis. Linearized plasmid containing origami sequences is transcribed into ssRNA, and then self-folds into Rod RNA-OG after annealing in PBS buffer. (J, K) AFM images and the corresponding length, width and height of Rod RNA-OG.

  DownLoad: Full-Size Img PowerPoint

  Subsequently, we defined predetermined parameters based on PXs. We constructed a schematic diagram of a partially paired double-strand intermediate (Figs. 1C-E) and then went on to shape this double strand into predetermined forms (Figs. 1F-H). As shown in the three-dimensional (3D) models in Figs. 1C and F, white cylinders were utilized to represent continuous helical domains, whereas gray cylinders represented locking domains. In Figs. 1D and G, these helical and locking domains were respectively illustrated as rectangles and crosses as well. In Figs. 1E and H, we constructed pipeline-style models for visualizing the folding pathway of the intermediate structure. The model was colored using a gradient, where the 5′ and 3′ ends of the ssRNA were represented in red and the middle portion was shown in blue. As shown in Fig. 1H, the rod-like structure represented the fully folded Rod RNA-OG.

  The overall workflow for the synthesis of the Rod RNA-OG is shown in Fig. 1I. DNA plasmid containing T7 promoter, restriction sites and origami sequences was amplified in E. coli. After purification, the plasmid was linearized with BamHI-HF restriction enzyme and transcribed using T7 RNA polymerase. Thus, the RNA molecules transcribed in vitro can be easily produced in a large quantity with minimal effort and cost. The RNA molecules were then purified and assembled into Rod RNA-OG by annealing in 1× PBS buffer without any addition of divalent cations, gradually lowering the temperature from 65 ℃ to 25 ℃ at a rate of −1 ℃ per 30 min. According to the atomic force microscopy (AFM) results (Fig. 1J), the measured length, width and height of the Rod RNA-OG were ~42 nm, ~4 nm and ~2 nm respectively (Fig. 1K), consistent with our design.

  To synthesize polyvalent RNA origami structures with controlled valence, we precisely programmed the number and the “sticky end” of RNA strand units designed above and constructed the polyvalent RNA origamis named 2-Rod, 4-Rod and 6-Rod RNA-OGs which contain 2, 4 or 6 Rod RNA-OGs, respectively. As depicted in Fig. 2A, each type of ssRNA is distinguished by the varying sequences of sticky ends (represented by different colors) extending from the 5′ and 3′ ends, which hybridize to form polyvalent RNA origamis (Fig. 2B). The RNA molecules transcribed in vitro assembled in one pot to construct various types of polyvalent RNA origamis through the same annealing procedure as that of Rod RNA-OG.

  Figure 2

  

  Figure 2.  Synthesis of polyvalent RNA origamis. (A) ssRNA strands, transcribed from linearized plasmids containing origami sequences and sticky ends, assemble into polyvalent RNA origamis after annealing in PBS buffer. The models of 2-Rod, 4-Rod and 6-Rod RNA-OGs with increasing rod numbers from up to bottom are shown on the right. (B) Schematics of the linkers in the structures of 2-Rod, 4-Rod and 6-Rod RNA-OGs. (C) AFM images of polyvalent RNA origamis. Scale bar: 50 nm. (D) Histograms showing the distribution of the rod numbers of each designed polyvalent RNA origami.

  DownLoad: Full-Size Img PowerPoint

  To verify the synthesis of the valence-programmed polyvalent RNA origamis, we first used agarose gel electrophoresis to characterize the construction of polyvalent RNA origamis. As revealed by the gel shift assay (Fig. S1 in Supporting information), these formed polyvalent RNA origamis exhibited one clear band, indicating the successful assembly of each polyvalent RNA origamis. Then we used AFM to characterize the morphology of each polyvalent RNA origamis. As shown in Fig. 2C and Figs. S2A-D (Supporting information), the numbers of Rod RNA-OG in 2-Rod, 4-Rod and 6-Rod RNA-OGs were 2, 4 and 6 respectively, indicating that the structures of polyvalent RNA origamis were consistent with our design. According to the AFM results, we then evaluated the synthesis efficiency of various polyvalent RNA origamis (Fig. 2D). As expected, the synthesis rate gradually decreased as the valence of Rod RNA-OG increased. The 2-Rod RNA-OG exhibited a yield of 58.4%, whereas the purities of the target structures in the 4-Rod RNA-OG and 6-Rod RNA-OG designs were 23.4% and 12.2%, respectively. This variation in purity might be attributed to the high valence, which can lead to compromised connections due to steric hindrance.

  DNA nanostructures are vulnerable to the digestion of DNases, which renders them unstable in several biological applications [42]. To evaluate the enzymatic stability of polyvalent RNA origamis, we performed degradation assays under RNase A treatments in vitro. After being exposed to RNase A, the unfolded RNA strands and polyvalent RNA origamis were subjected to analysis through gel electrophoresis (Figs. 3A-D). The polyvalent RNA origamis demonstrated significant resistance to the degradation of RNase A, lasting over 24 h, while the unfolded ssRNA, bearing sequences identical to those of Rod RNA-OG, was highly sensitive to RNase A and degraded completely within 6 h (Fig. 3E). Other types of RNA origamis, such as lantern-shaped RNA origami [7], also exhibit greater biological stability compared with unfolded single-stranded RNA. However, they can only resist the digestion of fetal bovine serum (FBS) for 4 h. In contrast, the ssRNA origamis featuring PX motifs have been proven to be stable in FBS for at least 24 h [9,37]. As recent research has shown that the biostability of DNA nano-structures can be enhanced by PX motifs [43], we hypothesized that the excellent biostability of the RNA origamis might also be owing to their intrinsic PX motifs, which enabled the formation of a highly compact structure without internal nick positions, resulting in significantly enhanced RNA biostability without any additional chemical modifications. While other reported RNA nanostructures normally need chemical modifications [44,45] or delivery vehicles [46,47] for extended serum stability. Therefore, the polyvalent RNA origamis had an advantage over other RNA structures in terms of biological stability.

  Figure 3

  

  Figure 3.  Enzymatic stability of polyvalent RNA origamis. (A-D) Agarose gels demonstrating the stability of 1-Rod RNA-OG, 2-Rod RNA-OG, 4-Rod RNA-OG and 6-Rod RNA-OG under the treatment with RNase A (1 μg/mL, in 1× PBS buffer, pH 7.4) at 37 ℃ for 0.5, 6 and 24 h. (E) The fraction intact is calculated by the gray value ratio of the RNA-OG treated with RNase A to untreated RNA-OG through ImageJ.

  DownLoad: Full-Size Img PowerPoint

  Delivering ligands to specific cells and organelles is crucial for biological applications. However, many small nucleic acid therapeutics, including short single-stranded antisense oligos, have faced challenges in attaching to or penetrating cells because of their negative charge and small size. Consequently, they are either excluded from macrophage uptake, or repelled by the negatively charged cell membrane [48,49]. To assess the cellular uptake efficiency and biological function of various polyvalent RNA origamis, macrophage-like RAW 264.7 cells were treated with polyvalent RNA origamis at an equimolar concentration of Rod RNA-OG (Fig. 4A). As shown in Fig. 4B, the Alexa Fluor-647-labeled polyvalent RNA origamis were endocytosed by RAW 264.7 cells within a 1 h incubation, demonstrating the effective cellular uptake of polyvalent RNA origamis. We conducted a quantitative assessment of the fluorescence intensity of each RNA origami in cells (Fig. 4C). We observed that there was a reduction in internalization efficiency as the valence of Rod RNA-OG increased (mean fluorescence intensity: 1-Rod vs. 2-Rod: 1.589, 1-Rod vs. 4-Rod: 1.957, 1-Rod vs. 6-Rod: 3.043, all P < 0.05). This might be attributed to lower cell binding affinity resulting from increased electrostatic repulsion with the negatively charged cell membrane due to the growing amount of double-stranded RNA (dsRNA). Collectively, polyvalent RNA origamis can be efficiently internalized by macrophages, providing possibility for biological applications.

  Figure 4

  

  Figure 4.  Biological effects of polyvalent RNA origamis. (A) Schematic of the interaction between TLR3 and polyvalent RNA origamis in vivo. (B) Confocal fluorescence microscopy images depict RAW 264.7 cells treated with polyvalent RNA origamis, from left to right: 1-Rod, 2-Rod, 4-Rod and 6-Rod RNA-OGs, respectively. The scale bars in the zoom-in images represent 10 μm, while the ones in the zoom-out images are 50 μm. (C) Statistical distribution of fluorescence intensity for each group in (B). (D, E) The mRNA expression of Ifnb1 and Ccl5 after stimulation of macrophages by polyvalent RNA origamis. All data are shown as representative of at least three independent experiments. ****P < 0.0001, ****P < 0.001.

  DownLoad: Full-Size Img PowerPoint

  Nucleic acids, serving as natural ligands for multiple pattern-recognition receptors (PRRs), are integral components of the host's defense against microbial pathogens, including viruses. Host cells can sense invading nucleic acids through PRRs and initiate innate immune responses to eliminate these dangers [50]. The extracellular nucleic acids, once internalized into endosomes, are recognized by various Toll-like receptors (TLRs), including TLR3 [51], TLR7/8 [52,53], TLR9 [54] for dsRNA, ssRNA and cytosine-phosphate-guanine (CpG) DNA, respectively. Several RNA nanostructures have been identified as immunostimulants whose immunostimulatory effects depend on binding with CpG motifs [44] or forming complexes with a lipid-based moiety [55] rather than their inherent immunogenicity. Interestingly, the RNA origami nanostructures based on the intrinsic PX motifs have been identified as potent TLR3 agonists [9,37], and the degree of immune activation can be regulated by the edge length of the RNA origamis [37]. To investigate the innate immune activation characteristics of the polyvalent RNA origamis, we analyzed the levels of cytokines and chemokines associated with immune activation following macrophage stimulation using qRT-PCR. As shown in Fig. 4D, the RNA origamis significantly improved the expression level of Ifnb1 by 0.7–3.7 times compared with the control [56], which demonstrated that RNA origamis might activate macrophages via TLR3-TRIF signaling pathway [9]. Furthermore, the mRNA level of Ccl5 associated with the antiviral and antitumor functions of macrophages was also increased by 4.6–12.9 times in RNA origami groups (Fig. 4E), which was consistent with previous research [9,37]. Additionally, compared with the groups treated with polyvalent RNA origamis, the ssRNA origami unit induced a significantly stronger activation of the RAW 264.7 cells, and the mRNA levels of Ifnb1 and Ccl5 in each polyvalent RNA origami group showed slight alterations. Based on recent studies, multivalent ligands generally show a more powerful activating impact on receptors compared with single ligands. They can induce receptor aggregation and activation more effectively, consequently leading to stronger intracellular signal transduction and physiological responses [57-59]. This is precisely contrary to the results we have obtained. One possible reason for this outcome is that ssRNA origami units are more readily internalized by RAW 264.7 cells, thereby exerting a stronger immunoregulatory effect. Another possibility is that increasing the valence of each nanostructure might lead to the presence of “unused” ligands. Taken together, the polyvalent RNA origamis can independently activate the RAW 264.7 cell line as a TLR3 agonist, despite the absence of CpG motifs and lipid-based moieties.

  In summary, we have successfully constructed valence-programmed polyvalent RNA origami structures consisting of 2, 4 or 6 Rod RNA-OGs. These structures are not only economically scalable but also capable of self-assembling in a one-pot manner. Functionally, they exhibit remarkable resistance to RNase degradation and can be efficiently endocytosed by macrophages, leading to subsequent innate immune activation. This indicates that polyvalent RNA origamis hold promising potential as novel agonists for immunotherapy. Nevertheless, further studies are required to investigate the impacts of intrinsic characteristics of the ligands, such as topological features, on the innate immune system. Furthermore, the applications of polyvalent RNA origamis in other biological fields also hold great promise, such as drug delivery, biosensing and gene regulation. Additionally, interdisciplinary collaborations among materials science, biology, and medicine will be crucial for advancing the development and application of the polyvalent RNA origamis.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Yue Jin: Writing – original draft, Project administration, Data curation. Kun Dai: Validation, Methodology, Conceptualization. Lu Song: Validation, Conceptualization. Xiaolei Zuo: Validation, Supervision, Funding acquisition. Guangbao Yao: Writing – review & editing, Validation, Supervision. Min Li: Validation, Supervision, Funding acquisition.

  Acknowledgments

  This work is supported by the National Key Research and Development Program of China (Nos. 2021YFF1200300, 2020YFA0909000), National Natural Science Foundation of China (Nos. 22025404, 32471426), Innovative Research Group of High-Level Local Universities in Shanghai (No. SHSMU-ZLCX20212602), Natural Science Foundation of Shanghai (No. 23ZR1438700), Shanghai Municipal Health Commission (No. 2022JC027) and Shanghai Sailing Program (No. 22YF1424400).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110744.

  
  
• 1. [1]

     E. Benson, A. Mohammed, J. Gardell, et al., Nature 523 (2015) 441–444. doi: 10.1038/nature14586

  2. [2]

     H. Dietz, S.M. Douglas, W.M. Shih, Science 325 (2009) 725–730. doi: 10.1126/science.1174251

  3. [3]

     R. Veneziano, S. Ratanalert, K. Zhang, et al., Science 352 (2016) 1534. doi: 10.1126/science.aaf4388

  4. [4]

     F. Hong, F. Zhang, Y. Liu, H. Yan, Chem. Rev. 117 (2017) 12584–12640. doi: 10.1021/acs.chemrev.6b00825

  5. [5]

     F. Ricci, H. Dietz, Nat. Nanotechnol. 18 (2023) 541–542. doi: 10.1038/s41565-023-01362-x

  6. [6]

     P. Guo, Nat. Nanotechnol. 5 (2010) 833–842. doi: 10.1038/nnano.2010.231

  7. [7]

     M. Hu, C. Feng, Q. Yuan, et al., Nat. Commun. 14 (2023) 1307. doi: 10.1038/s41467-023-37020-y

  8. [8]

     Z. Ge, J. Liu, L. Guo, et al., J. Am. Chem. Soc. 142 (2020) 8800–8808. doi: 10.1021/jacs.0c01580

  9. [9]

     X. Qi, X. Liu, L. Matiski, et al., ACS Nano 14 (2020) 4727–4740. doi: 10.1021/acsnano.0c00602

  10. [10]

      P. Guo, C. Zhang, C. Chen, K. Garver, M. Trottier, Mol. Cell 2 (1998) 149–155. doi: 10.1016/S1097-2765(00)80124-0

  11. [11]

      E. Westhof, B. Masquida, L. Jaeger, Fold. Des. 1 (1996) R78–R88. doi: 10.1016/S1359-0278(96)00037-5

  12. [12]

      E.F. Khisamutdinov, D.L. Jasinski, P. Guo, ACS. Nano 8 (2014) 4771–4781. doi: 10.1021/nn5006254

  13. [13]

      K.A. Afonin, E. Bindewald, A.J. Yaghoubian, et al., Nat. Nanotechnol. 5 (2010) 676–682. doi: 10.1038/nnano.2010.160

  14. [14]

      L. Nasalean, S. Baudrey, N.B. Leontis, L. Jaeger, Nucleic. Acids. Res. 34 (2006) 1381–1392. doi: 10.1093/nar/gkl008

  15. [15]

      I.V. Novikova, B.H. Hassan, M.G. Mirzoyan, N.B. Leontis, Nucleic. Acids. Res. 39 (2011) 2903–2917. doi: 10.1093/nar/gkq1231

  16. [16]

      D. Shu, Y. Shu, F. Haque, S. Abdelmawla, P. Guo, Nat. Nanotechnol. 6 (2011) 658–667. doi: 10.1038/nnano.2011.105

  17. [17]

      L. Ponchon, F. Dardel, Nat. Methods 4 (2007) 571–576. doi: 10.1038/nmeth1058

  18. [18]

      I. Severcan, C. Geary, A. Chworos, et al., Nat. Chem. 2 (2010) 772–779. doi: 10.1038/nchem.733

  19. [19]

      C. Chen, C. Zhang, P. Guo, RNA 5 (1999) 805–818. doi: 10.1017/S1355838299990350

  20. [20]

      C. Geary, A. Chworos, L. Jaeger, Nucleic Acids Res. 39 (2011) 1066–1080. doi: 10.1093/nar/gkq748

  21. [21]

      W.W. Grabow, P. Zakrevsky, K.A. Afonin, et al., Nano Lett. 11 (2011) 878–887. doi: 10.1021/nl104271s

  22. [22]

      D. Shu, W.D. Moll, Z. Deng, C. Mao, P. Guo, Nano Lett. 4 (2004) 1717–1723. doi: 10.1021/nl0494497

  23. [23]

      Y. Shu, D. Shu, F. Haque, P. Guo, Nat. Protoc. 8 (2013) 1635–1659. doi: 10.1038/nprot.2013.097

  24. [24]

      H. Ohno, T. Kobayashi, R. Kabata, et al., Nat. Nanotechnol. 6 (2011) 116–120. doi: 10.1038/nnano.2010.268

  25. [25]

      D. Han, X. Qi, C. Myhrvold, et al., Science 358 (2017) eaao2648. doi: 10.1126/science.aao2648

  26. [26]

      W.M. Shih, J.D. Quispe, G.F. Joyce, Nature 427 (2004) 618–621. doi: 10.1038/nature02307

  27. [27]

      C. Lin, M. Xie, J.J. Chen, Y. Liu, H. Yan, Angew. Chem. Int. Ed. 45 (2006) 7537–7539. doi: 10.1002/anie.200602113

  28. [28]

      C. Geary, P.W. Rothemund, E.S. Andersen, Science 345 (2014) 799–804. doi: 10.1126/science.1253920

  29. [29]

      C. Lin, S. Rinker, X. Wang, et al., Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 17626–17631. doi: 10.1073/pnas.0805416105

  30. [30]

      C. Ducani, C. Kaul, M. Moche, W.M. Shih, B. Hogberg, Nat. Methods 10 (2013) 647–652. doi: 10.1038/nmeth.2503

  31. [31]

      C. Geary, G. Grossi, E.K.S. McRae, P.W.K. Rothemund, E.S. Andersen, Nat. Chem. 13 (2021) 549–558. doi: 10.1038/s41557-021-00679-1

  32. [32]

      T. Yip, X. Qi, H. Yan, Y. Chang, ACS. Nano 18 (2024) 4056–4067. doi: 10.1021/acsnano.3c07284

  33. [33]

      J.A. Kretzmann, A. Liedl, A. Monferrer, et al., Nat. Commun. 14 (2023) 1017. doi: 10.1038/s41467-023-36601-1

  34. [34]

      A. Krissanaprasit, C. Key, M. Fergione, et al., Adv. Mater. 31 (2019) e1808262. doi: 10.1002/adma.201808262

  35. [35]

      E.K.S. McRae, H.O. Rasmussen, J. Liu, et al., Nat. Nanotechnol. 18 (2023) 808–817. doi: 10.1038/s41565-023-01321-6

  36. [36]

      M.D.E. Jepsen, S.M. Sparvath, T.B. Nielsen, et al., Nat. Commun. 9 (2018) 18. doi: 10.1038/s41467-017-02435-x

  37. [37]

      K. Dai, Y. Xu, Y. Yang, et al., J. Am. Chem. Soc. 145 (2023) 17112–17124. doi: 10.1021/jacs.3c03477

  38. [38]

      N.Sampedro Vallina, E.K.S. McRae, B.K. Hansen, A. Boussebayle, E.S. Andersen, Nucleic. Acids. Res. 51 (2023) 4613–4624. doi: 10.1093/nar/gkad224

  39. [39]

      M.L. Mansfield, Nat. Struct. Biol. 1 (1994) 213–214. doi: 10.1038/nsb0494-213

  40. [40]

      W.R. Taylor, Nature 406 (2000) 916–919. doi: 10.1038/35022623

  41. [41]

      J.R. Wagner, J.S. Brunzelle, K.T. Forest, R.D. Vierstra, Nature 438 (2005) 325–331. doi: 10.1038/nature04118

  42. [42]

      J. Hahn, S.F. Wickham, W.M. Shih, S.D. Perrault, ACS. Nano 8 (2014) 8765–8775. doi: 10.1021/nn503513p

  43. [43]

      A.R. Chandrasekaran, J. Vilcapoma, P. Dey, et al., J. Am. Chem. Soc. 142 (2020) 6814–6821. doi: 10.1021/jacs.0c02211

  44. [44]

      S. Guo, H. Li, M. Ma, et al., Mol. Ther. Nucleic. Acids 9 (2017) 399–408. doi: 10.1016/j.omtn.2017.10.010

  45. [45]

      M. Shanmugasundaram, A. Senthilvelan, A.R. Kore, Chem. Rec. 22 (2022) e202200005. doi: 10.1002/tcr.202200005

  46. [46]

      E.L. Blanchard, D. Vanover, S.S. Bawage, et al., Nat. Biotechnol. 39 (2021) 717–726. doi: 10.1038/s41587-021-00822-w

  47. [47]

      A. Amiri, R. Bagherifar, E.Ansari Dezfouli, et al., J. Transl. Med. 20 (2022) 125. doi: 10.1007/s10799-021-00348-2

  48. [48]

      R.L. Juliano, X. Ming, K. Carver, B. Laing, Nucleic Acid Ther. 24 (2014) 101–113. doi: 10.1089/nat.2013.0463

  49. [49]

      R. Juliano, M.R. Alam, V. Dixit, H. Kang, Nucleic Acids Res. 36 (2008) 4158–4171. doi: 10.1093/nar/gkn342

  50. [50]

      M. Schlee, G. Hartmann, Nat. Rev. Immunol. 16 (2016) 566–580. doi: 10.1038/nri.2016.78

  51. [51]

      L. Alexopoulou, A.C. Holt, R. Medzhitov, R.A. Flavell, Nature 413 (2001) 732–738. doi: 10.1038/35099560

  52. [52]

      S.S. Diebold, T. Kaisho, H. Hemmi, S. Akira, C. Reis e Sousa, Science 303 (2004) 1529–1531. doi: 10.1126/science.1093616

  53. [53]

      F. Heil, H. Hemmi, H. Hochrein, et al., Science 303 (2004) 1526–1529. doi: 10.1126/science.1093620

  54. [54]

      H. Hemmi, O. Takeuchi, T. Kawai, et al., Nature 408 (2000) 740–745. doi: 10.1038/35047123

  55. [55]

      E. Hong, J.R. Halman, A.B. Shah, et al., Nano Lett. 18 (2018) 4309–4321. doi: 10.1021/acs.nanolett.8b01283

  56. [56]

      M. Matsumoto, M. Tatematsu, F. Nishikawa, et al., Nat. Commun. 6 (2015) 6280. doi: 10.1038/ncomms7280

  57. [57]

      R. Veneziano, T.J. Moyer, M.B. Stone, et al., Nat. Nanotechnol. 15 (2020) 716–723. doi: 10.1038/s41565-020-0719-0

  58. [58]

      Y.C. Zeng, O.J. Young, C.M. Wintersinger, et al., Nat. Nanotechnol. 19 (2024) 1055–1065. doi: 10.1038/s41565-024-01615-3

  59. [59]

      J. Spratt, J.M. Dias, C. Kolonelou, et al., Nat. Nanotechnol. 19 (2024) 237–245. doi: 10.1038/s41565-023-01507-y

• 

  Figure 1  Design and synthesis of Rod RNA-OG. (A) Schematics of the PX designs. The central plane is marked by dashed lines in the model. (B) Schematics illustrating the complete formation of a locking domain through base pair recognition. (C-E) Models of double-stranded intermediate structures of Rod RNA-OG. Models of fully formed structures of Rod RNA-OG. (E, H) Pipeline-style models visualizing the folding pathway of the intermediate structure. The model is colored with a gradient, where the 5′ and 3′ ends of the ssRNA are represented in red and the middle portion is shown in blue. (I) Workflow of Rod RNA-OG synthesis. Linearized plasmid containing origami sequences is transcribed into ssRNA, and then self-folds into Rod RNA-OG after annealing in PBS buffer. (J, K) AFM images and the corresponding length, width and height of Rod RNA-OG.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Synthesis of polyvalent RNA origamis. (A) ssRNA strands, transcribed from linearized plasmids containing origami sequences and sticky ends, assemble into polyvalent RNA origamis after annealing in PBS buffer. The models of 2-Rod, 4-Rod and 6-Rod RNA-OGs with increasing rod numbers from up to bottom are shown on the right. (B) Schematics of the linkers in the structures of 2-Rod, 4-Rod and 6-Rod RNA-OGs. (C) AFM images of polyvalent RNA origamis. Scale bar: 50 nm. (D) Histograms showing the distribution of the rod numbers of each designed polyvalent RNA origami.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Enzymatic stability of polyvalent RNA origamis. (A-D) Agarose gels demonstrating the stability of 1-Rod RNA-OG, 2-Rod RNA-OG, 4-Rod RNA-OG and 6-Rod RNA-OG under the treatment with RNase A (1 μg/mL, in 1× PBS buffer, pH 7.4) at 37 ℃ for 0.5, 6 and 24 h. (E) The fraction intact is calculated by the gray value ratio of the RNA-OG treated with RNase A to untreated RNA-OG through ImageJ.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Biological effects of polyvalent RNA origamis. (A) Schematic of the interaction between TLR3 and polyvalent RNA origamis in vivo. (B) Confocal fluorescence microscopy images depict RAW 264.7 cells treated with polyvalent RNA origamis, from left to right: 1-Rod, 2-Rod, 4-Rod and 6-Rod RNA-OGs, respectively. The scale bars in the zoom-in images represent 10 μm, while the ones in the zoom-out images are 50 μm. (C) Statistical distribution of fluorescence intensity for each group in (B). (D, E) The mRNA expression of Ifnb1 and Ccl5 after stimulation of macrophages by polyvalent RNA origamis. All data are shown as representative of at least three independent experiments. ****P < 0.0001, ****P < 0.001.

  下载: 全尺寸图片 幻灯片
•