Development of delivery strategies for CRISPR‐Cas9 genome editing

Abstract

The clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR‐related protein 9 (Cas9) genome editing system has attracted much attention due to its powerful genome editing capacity. However, CRISPR‐Cas9 components are easily degraded by acids, enzymes, and other substances in the body fluids after entering the organism, thus efficiently delivering the CRISPR‐Cas9 system into targeted organs or cells has been a central theme for promoting the application of CRISPR‐Cas9 technology. Although several physical methods and viral vectors have been developed for CRISPR‐Cas9 delivery, their clinical application still suffers from disadvantages, such as the risks of mutagenesis, cell damage, and poor specificity. As an alternative, non‐viral nanocarriers hold great promise for circumventing these challenges. Furthermore, with aim to realize more efficient and precise genome editing and reduce the undesirable side effects, stimuli‐responsive nanocarriers are designed for the spatiotemporal CRISPR‐Cas9 delivery in responsive to various stimuli. In this review, we will summarize the recent progress in delivery strategies for CRISPR‐Cas9 genome editing. The mechanisms and advantages of these strategies were reviewed, providing a comprehensive review of the rational design of materials and techniques for efficient and precise genome editing. At last, the potential challenges of current CRISPR‐Cas9 delivery are discussed.

INTRODUCTION

The clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-related protein 9 (Cas9) genome editing system was reported for gene editing in 2012 by Emmanuelle Charpentier and Jennifer Doudna, and driving CRISPR-Cas9 technology to a new research boom. Therefore, Emmanuelle Charpentier and Jennifer Doudna were awarded the 2020 Nobel Prize in Chemistry. In 2013, Zhang Feng's team implemented gene editing of eukaryotic cells using CRISPR-Cas9 technology for the first time. As research progressed, the CRISPR-Cas9 system has been employed in multifarious fields, offering a powerful tool for scientific research and treatment of various diseases.

CRISPR-Cas9 system contains two critical components: Cas9 nuclease and single guide RNA (sgRNA). In the presence of protospacer adjacent motif sequences (5′-NGG for widely used SpCas9), Cas9 nuclease could be directed by sgRNA to any targeted genes through complementary base pairing, forming double-strand breaks (DSBs) at specific location. DSBs are mainly repaired through non-homologous end joining (NHEJ), homology-directed repair (HDR), homology-mediated end joining (HMEJ), and microhomology-mediated end joining (MMEJ) pathways. A repair mechanism based on the NHEJ pathway would generate random insertion or deletion at the breaks to induce gene mutation. By providing the DNA templates containing sequences homologous to DSBs, specific sequences can be inserted at the targeted loci via HMEJ, MMEJ, and HDR pathways. Furthermore, CRISPR interference (CRISPRi) technology has also been developed in recent years. By fusing the inactivated nuclease (dead Cas9, dCas9) with transcription activators or repressors, the CRISPR-dCas9 system can achieve accurate and efficient control of gene expression without cutting targeted DNA under the guidance of sgRNA, providing a natural and effective method to precisely control endogenous gene expression.

Compared to conventional genome editing tools that recognize the targeted sequence via the specific interaction between DNA and proteins, including zinc finger nucleases (ZFNs) and transcription activator-like nucleases (TALENs), CRISPR-Cas9 system can re-target new targeted sequences or simultaneously target multiple sequences for genome editing by changing or adding sgRNA sequences (about 100 base pairs), which is much simpler than ZFNs and TALEN genome editing systems that require sophisticated design and synthesis of a bulky guiding protein (zinc finger DNA-binding domain and transcription activator-like effector). In addition, CRISPR-Cas9 system holds a higher precision in genome editing compared to other conventional tools. While CRISPR-Cas9 technology has shown great application potential in biology research and medicine, several practical problems and technical challenges remain to be solved before applying it to clinical practice. At the molecular level, there are two main problems to be overcome. Firstly, efficient and accurate cutting and repair sites have been designed to improve the specificity of genome editing and reduce the probability of off-targeting. Secondly, reasonably choose repair pathway according to different experimental purposes. Although the HDR-mediated CRISPR-Cas9 genome editing has made great progress, the instability of the recombination process remains a great challenge for the application of HDR-mediated CRISPR-Cas9 genome editing. To solve these problems, it is necessary to develop more advanced CRISPR techniques. For example, homology-dependent targeted integration strategies have been developed as an alternative for CRISPR-Cas9-based gene repair. This method allows stable gene knock-in at both dividing and non-dividing cells in vitro and in vivo. Overall, there is no doubt that CRISPR-Cas9 technology has greater potential in various applications with the continuous advances in the CRISPR-Cas9 system.

On the other hand, an essential prerequisite for the CRISPR-Cas9 system to exert its function is efficiently delivering the CRISPR-Cas9 system into targeted cells, tissues or organs. The CRISPR-Cas9 system can be delivered in the form of either plasmid DNA (pDNA) encoding both Cas9 nuclease and sgRNA, CRISPR mRNA and sgRNA, or Cas9 ribonucleoprotein (RNP, compound of Cas9 nuclease and sgRNA). Each form has its advantages and disadvantages. The pDNA form is convenient, low cost, and high stability. However, the larger size (more than 10 kb) would significantly increase the difficulty of CRISPR-Cas9 system package and delivery, which make transfection inefficient. Furthermore, pDNA usually leads to long-term expression of Cas9 nuclease in cells, which may lead to high off-target effects and strong immune responses. The advantage of the mRNA form is that it can express Cas9 nuclease more quickly than pDNA in cells, while its inherent instability may reduce the overall genome editing efficiency. Cas9 RNP is the most straightforward form without transcription and translation, and shows excellent stability, quick action, low immunogenicity and off-target effects despite its high cost and relatively short genome editing duration. It is widely known that pDNA, mRNA, sgRNA, and Cas9 RNP hold strong negative charges, which results in scarcely cellular internalization of CRISPR-Cas9 components because of the electrostatic repulsion with the negatively charged cell membrane. Therefore, developing strategies that can efficiently deliver CRISPR-Cas9 components into targeted cells have been a cutting-edge theme for promoting the application of CRISPR-Cas9 technology.

Viral vectors, such as adenovirus (AV), adeno-associated virus (AAV), and lentivirus (LV), have been developed for the intracellularly delivery of CRISPR-Cas9 system with high gene editing efficiency. However, viral vectors have inherent shortcomings, including the risks of mutagenesis, restriction of insertion size, and immunogenicity, which seriously limit the application of viral vectors. Additionally, viral vectors can induce long-term genome editing in cells, which may increase off-target effects. As an alternative, non-viral nanocarriers hold great promise to circumvent the challenges mentioned by viral vectors mentioned. To date, multiple polymeric nanoparticles, liposomes, and inorganic nanoparticles have been reported for the delivery of the CRISPR-Cas9 system. However, these non-viral nanocarriers generally display much lower genome editing efficiency compared to viral vectors. The major challenge is that these non-viral nanocarriers need to overcome multiple physiological and intracellular barriers. Physiological barriers include non-specific protein absorption, rapid blood clearance, and poor cell-targeting ability; intracellular barriers mainly refer to inefficient endo/lysosome escape and cytosol release. Therefore, with aim to realize efficient and precise genome editing and reduce undesirable side effects, the next-generation non-viral nanocarriers (stimuli-responsive nanocarriers) are designed for the spatiotemporal CRISPR-Cas9 delivery in responsive to various stimuli. These stimuli include exogenous signals (e.g., pH, redox, enzyme, adenosine triphosphate (ATP), microRNA, and hypoxia) and external stimuli (e.g., photo, magnetism, and ultrasound). In this review, we summarized the development of delivery strategies for CRISPR-Cas9 genome editing, including physical methods, viral vectors, non-viral nanocarriers, and stimuli-responsive nanocarriers (Figure ). We further discussed the advantages of these strategies and materials, providing a comprehensive review of the rational design of materials and techniques for efficient and precise genome editing.

[IMAGE OMITTED. SEE PDF]

PHYSICAL METHODS FOR CRISPR-CAS9 DELIVERY

Microinjection

The microinjection method is commonly applied in mammalian oocyte and early embryo studies, and exogenous gene fragments were directly injected into the targeted cell through a microinjection device (Figure , tube tip diameter of [0.1–0.5 μm]). The microinjection technique has also been employed to prepare mouse models of human diseases. For example, Yan et al. injected the complexes of Cas9 mRNA and sgRNA into the cytoplasm of rabbit embryos by microinjection, successfully constructed the single-gene and multi-gene knockout rabbit model. Marcela et al. microinjected the complexes of Cas9 mRNA and sgRNA into oocytes to inhibit the formation of pancreas in sheep. Through intra-fertilized injection of Streptococcus pasteurianus Cas9 mRNA and sgRNA, TYR gene mutant mice were successfully prepared to mimic a human oculocutaneous albinism model. Furthermore, no off-target effects were detected at potential off-target sites in mice. The delivery of exogenous gene fragments into microalgae cells is difficult due to a pliable membrane on the Euglena gracilis, Chen et al. proposed a microinjection method to introduce Cas9 mRNA into E. gracilis and knocked out the CRTP1 gene with acceptable genome editing efficiency (16.7%). The resulting CRTP1 mutant provides a good candidate strain for studying carotenoid metabolism in E. gracilis. The microinjection technique is good at directly and rapidly introducing gene fragments into single cells. However, it also requires sophisticated instruments and professional person to operate, and each experiment can only process hundreds of cells at most, so it is impractical for in vivo applications that need to edit millions of cells at the same time.

[IMAGE OMITTED. SEE PDF]

Electroporation

Electroporation is a cell transfection tool that instantly increases the permeability of the cell membrane by applying a high-intensity electric field to form tiny pores outside the cell membrane. Electroporation technology is applicable to the particles with sizes of several to dozens of nanometers, and has been broadly applied to deliver CRISPR-Cas9 components. For example, Latella et al. delivered pDNA encoding Cas9 nuclease and sgRNA to disrupt the human rhodopsin gene in a mouse model by electroporation. Xie and colleagues delivered CRISPR-Cas9 system into induced pluripotent stem cells (iPSCs) of β-thalassemia patients, and thus successfully corrected β gene mutation. Wan et al. also loaded Cas9 RNP into exosomes derived from hepatic stellate cells (HSCs) using electroporation technique and named this system as Exosome^RNP (Figure ). Exosome^RNP holds the good biocompatibility of exosomes and can specifically target the liver. Exosome^RNP also exhibited powerful therapeutic potential in mouse models of acute liver injury, chronic liver fibrosis, and liver cancer by targeting p53 up-regulated apoptosis regulator (PUMA), cyclin E1 (CCNE1), and K(lysine) acetyltransferase 5 (KAT5). Recently, Kutter and colleagues found that large pDNA can enter cells along with the flow of small pDNA without getting entangled in two or more open membrane pores. As a result, co-transfection of large CRISPR pDNA (15 kb) together with a small empty pDNA (3 kb) via electroporation can significantly increase the transfection efficiency and reduce cell death. Furthermore, during the process of delivering CRISPR systems, the damage of phospholipid bilayer caused by transient voltage may be accompanied by abnormal apoptosis and decrease in cell activity. To minimize the side effects of electroporation, researchers should optimize the parameters of voltage and current of this method.

Hydrodynamic injection

Hydrodynamic injection is another method to achieve transgene expression in mice by rapidly injecting a large volume of macromolecule solution into the liver via the tail vein. A large volume of macromolecule solution in the liver raises the intrahepatic pressure, resulting in temporary pores on the hepatocyte membrane and subsequent entry of gene cargoes. In addition to liver, other tissues (e.g., muscle, kidney, spleen, and brain) can allow the entry of gene cargoes after receiving huge fluid pressure. Therefore, hydrodynamic injection was also broadly employed for the delivery of the CRISPR-Cas9 system. Chiu et al. constructed a mouse model of hepatocellular carcinoma with Trp53 deletion and overexpression of c-Myc by delivering a transposon system expressing c-Myc and a CRISPR-Cas9 system targeting Trp53 into hepatocyte through the hydrodynamic injection. Yin et al. successfully corrected the mutation of Fah locus in the mouse model of hereditary tyrosinemia through the hydrodynamic injection of CRISPR-Cas9 components. He and colleagues also successfully constructed a rat nonalcoholic fatty liver model by knocking out the Pten gene in rat liver through the hydrodynamic injection of pDNA encoding CRISPR-Cas9 system (Figure ). Overall, hydrodynamic injection is superior to other physical methods due to its simplicity, high efficiency, and versatility. However, this method has its disadvantages, such as high blood pressure, impaired cardiovascular function, and the risk of animal death.

Induced transduction by osmocytosis and propanebetaine

Due to the challenge in directly delivering proteins into cells, Diego et al. developed a method called induced transduction by osmocytosis and propanebetaine (iTOP) to induce efficient transduction of proteins into various primary cells in 2015, mainly through the high osmotic pressure of NaCl and the transduction of propanebetaine (Figure ). Kholosy and colleagues described an optimized iTOP method for effective CRISPR-Cas9 gene editing in various human cells. It is found that this method achieved 85% and 70% genome editing efficiency in ARPE-19 and HEK293 cells, respectively. Compared to commonly used methods, including electroporation and lipofection, iTOP-mediated delivery of CRISPR-Cas9 exhibited a higher gene editing efficiency and lower decrease in cell viability. Although iTOP has shown great promise for the intracellular delivery of CRISPR-Cas9 system, its application in vivo still faces many challenges due to the use of highly concentrated salt solutions.

Mechanical cell deformation

Rapid mechanical deformation of cells can induce transient membrane disruption. When cells pass through a channel that is smaller than the diameter of cells, the shear force caused by the physical contraction of cells could produce tiny pores and then facilitate the passive diffusion of macromolecules around the cells into the cytosol. The pores disappear after recovering the membrane deformation. This strategy can transport almost any macromolecules into almost any type of cells. For the first time, Han et al. further optimized the membrane deformation method to co-deliver sgRNA and Cas9 nuclease into various cell lines (Figure ), such as adherent and non-adherent cells, refractory lymphomas, and embryonic stem (ES) cells, achieved successfully gene editing without obvious decrease in cell viability. In a later study, Han et al. optimized the delivery chip to facilitate the delivery efficiency of Cas9 RNP into human primary CD4⁺ T cells and thus successfully edited the programmed cell death protein 1 (PD-1), which would promote the T-cell-mediated cancer immunotherapy. They concluded that this microfluidic cell deformation delivery method is effective for fragile cells, which are unsuitable for electroporation. It also avoids the risk of infection and insertion of viral vectors. In another study, Ma et al. designed a new delivery microfluidic chip (Nano-Blade Chip, NB-Chip), which can rapidly and efficiently deliver the CRISPR-Cas9 system into hard-to-transfect human HSCs, providing an effective approach to treat genetic diseases associated with HSCs. It was found that the successful disruption of the P42 isoform in C/EBPα can promote HSCs to enter the bone marrow for proliferation, indicating that NB-Chip has excellent therapeutic potential in HSC diseases.

VIRAL VECTORS FOR CRISPR-CAS9 DELIVERY

Adenovirus

AVs are a group of double-stranded linear DNA viral particles with a diameter of 70–100 nm and have an icosahedral capsid. The genome of AV is about 36 kb in length, with 103 bp inverted terminal repeats (ITRs) at each end of the genome, which are involved in DNA replication. The loading capacity of AVs are approximately 8.5 kb, twice that of AAVs. AV vectors with E2a and E4 gene knockout are termed as the second generation AV vectors, which have a higher loading capacity up to 14 kb and weaker immune responses compared to first-generation AV vectors. However, its application is seriously limited by the difficulty of virus package and low virus titer. The third-generation AV vectors also called gutted AV vectors are devoid of almost all coding viral regions and only keep the ITRs and packaging signal sequences. Its loading capacity can be increased up to 35 kb. Due to the absence of the viral packaging gene, third-generation AV vectors require a helper virus that carries all coding regions. Third-generation AV vectors hold high loading capacity and low immunogenicity, but are difficult to mass-produce and have the risk of contamination by helper viruses. Marta and colleagues delivered the CRISPR-Cas9 system into a humanized mouse model of recessive dystrophic epidermolysis bullosa using AV vectors and successfully corrected the mutated COL7A1 gene that encodes the type VII collagen, resulting in significant recovery in areas of skin damage. This finding provides a new solution for treating recessive dystrophic herpetic epidermolysis bullosa. The CRISPR-Cas9 delivery using AV vectors can also be utilized to treat other rare diseases. For instance, in another study on hemophilia B, Stephens and colleagues used AV vectors as the carrier of CRISPR-Cas9 system to treat mouse model of hemophilia B by correcting the Factor IX (FIX) gene (mutation that causes hemophilia B). The results found that FIX mutations in all mice were corrected without error.

Adeno-associated virus

AAVs are a group of small single-stranded linear DNA virus particles without an envelope on their surface, and their genome length is approximately 5 kb. Their genome contains three genes: Rep (replication), Cap (capsid) and AAP (assembly). AAVs do not encode polymerases, thus AAV vectors rely on other viruses for infection and replication. ITRs are indispensable for the replication and package of AAVs. The length of ITRs at both ends is 0.3 kb, the cargo-size limitation of AAVs is about 4.7 kb. To date, AAV vectors have been extensively studied in gene therapy. The advantages of AAV vectors include low risk of pathogenicity and immunogenicity. The major disadvantage of AAV vectors is insufficient loading capacity. AAVs have also been employed as delivery vehicle for CRISPR-Cas9 gene editing. For example, Yoon et al. delivered the CRISPR-Cas9 system into the mouse pre-implant embryos using recombinant AAV, successfully constructed a mutant mouse model. Yao and colleagues also reported an AAV-CPP.16, a novel variant of AAV that binds cell-penetrating peptides (CPPs), which can efficiently trans-blood-brain barrier (BBB) to treat brain tumors and genetic diseases of central nervous system (CNS), providing a promising approach for the clinical treatment of CNS. Alessio et al. developed two kinds of AAV vectors to simultaneously deliver CRISPR-Cas9 system and HDR-based DNA templates targeting immunoglobulin genes, and effectively modified B cells to produce anti-HIV neutralizing antibodies in vivo.

Lentivirus

LVs, derived from HIV, are a kind of enveloped RNA virus with a diameter of 80–120 nm and an icosahedron shell. Similar to AVs, LVs have a loading capacity of about 8 kb and can infect both dividing and non-dividing cells. Compared to AAVs, LVs hold a larger loading capacity, and thus can encode differently-sized Cas9 nucleases or several sgRNAs for multiplex genome editing. Currently, LV vectors have been broadly applied for the delivery of CRISPR-Cas9 components. To solve the endogenous immune recognition of Cas9 nuclease, Dubrot et al. designed a LV vector that can specifically knock out CRISPR antigen from tumor cells. This work provides a promising way to promote the application of CRISPR-Cas9 system and avoid the undesired immune recognition of gene editing cells. Heckl and colleagues designed a LV to deliver pDNA encoding CRISPR-Cas9 system and a fluorescent protein to construct a mouse model of acute myeloid leukemia. In another study, Lee et al. reported a hepatitis C virus/E1E2 envelope glycoprotein pseudotyped LV to deliver CRISPR-Cas9 system into the hepatic tumors in vivo, resulted in successful disruption of kinesin spindle protein to inhibit the tumor growth. It was also found that E1E2-pseudotyping is more suitable for in vivo CRISPR-Cas9 delivery than commonly used vesicular stomatitis virus-pseudotyping due to excellent serum stability, low immunogenicity, and cell-specific targeting capacity. Benefiting for the integration capacity of LVs, LVs were also employed to construct gene libraries to explore mechanisms of disease. For example, LVs were used to deliver 73,000 sgRNAs into two cell lines that had been transfected to express Cas9 nuclease. The CRISPR-Cas9-mediated genomic screening using LV vectors was also employed to explore the mechanisms for West-Nile-virus-induced cell death. In general, LV vectors can integrate exogenous genes into the genome of host cells for prolonged expression with low immunogenicity, and thus are commonly used in clinical and animal experiments.

NON-VIRAL NANOCARRIERS FOR CRISPR-CAS9 DELIVERY

Polymeric nanoparticles

Polyethylenimine

Polyethylenimine (PEI) is one of the most commonly used polymer gene carriers, abundant amino groups endow PEI with a high charge density, which contributes to gene compression and lysosomal escape. Nowadays, high molecular weight branched PEI (Mw = 25 kDa) has become the gold standard for nucleic acid delivery. Although high charge density is helpful to improve the transfection efficiency, it also increases the cytotoxicity of the polymers, which limits its application in vivo. To reduce its cytotoxicity, researchers tried to modify the surface of PEI. For example, Zhang and colleagues developed a PEI-β-cyclodextrin (PC) for CRISPR delivery. Compared to PEI with high molecular weight, the cytotoxicity of PC is significantly lower. Zhao et al. developed an optical switch controlled CRISPR-Cas9 system, which can permanently block the PD-1/PD-L1 pathway through CRISPR-Cas9 genome disruption (Figure ). This system was constructed using a photoactivated self-degradation PEI derivative and pX330/sgPD-L1 that expresses Cas9 nuclease and sgRNA targeting PD-L1. Under light irradiation, PEI derivative was degraded and then released pX330/sgPD-L1 into the cytoplasm of cancer cells. In another study, PEI, CRISPR-Cas9 pDNA, and magnetic nanoparticles (MNPs) formed CRISPR-Cas9-PEI-MNPs complexes, which can bind to residues on the surface of cells and then enter the cells via endocytosis, thereby releasing CRISPR-Cas9 system into cytoplasm. Overall, PEI is a potential cationic polymer for the efficient delivery of gene fragments and macromolecular proteins.

[IMAGE OMITTED. SEE PDF]

Dendrimers

Dendrimers are a class of synthetic polymers with hyperbranched structure and high-density surface functional groups. These polymers have been widely employed as delivery carriers for drugs, nucleic acids, and proteins. For example, Chen and colleagues modified the surface of polyamidoamine (PAMAM) with high-density phenylboronic acid (PBA) for Cas9 RNP delivery (Figure ). These PBA groups can bind with amino groups and imidazolyl groups on proteins through nitrogen borate ester, while the residual amino groups on PAMAM would combine with anionic groups on proteins via the electrostatic interaction. Therefore, PBA-modified PAMAM can complex with proteins with different isoelectric points to form uniform nanoparticles, as well as Cas9 RNP, and then effectively delivered Cas9 RNP into various cell lines with a higher genome editing efficiency than commercialized CRISPR^MAX. Although the net charge of Cas9 RNP is negative, there are still some charged residues in Cas9 protein. This strategy provided a potential strategy to bind both negative charges of sgRNA and positive charged residues in Cas9 protein. Additionally, more specially designed delivery systems should be developed. In another study, 6-O-α-(4-O-α-D-glucuronosyl)-D-glucosyl-β-cyclodextrin was modified onto the third generation PAMAM for Cas9 RNP delivery (Figure ). After in situ intraventricular administration, this synthetic dendrimer achieved effective genome editing in human neuroblastoma (SH-SY5Y) cells and mouse brain tissue.

Chitosan

Chitosan is a kind of biocompatible and biodegradable cationic polysaccharide, which is usually employed as a polymer carrier for various drugs. Chitosan can also be adapted to deliver the CRISPR-Cas9 system via the forms of pDNA and RNP. Zhang et al. found that the complexes of PEGylated chitosan and CRISPR-Cas9 pDNA exhibited the optimal genome editing efficiency, when the N/P ratios of complexes formed by pDNA and PEGylated chitosan with low molecular or medium molecular were set as 20 or 5, respectively. However, free chitosan can not effectively encapsulate RNP and deliver it into cells. In recent work, red fluorescent protein (RFP) with negative charge was first complexed with chitosan from RFP@CS nanocomposites. Then, Cas9 protein fused with 20 glutamate residues at N-terminus and donor DNA templates were further complexed with RFP@CS to form nanoparticles. Such nanoparticles achieved a similar HDR efficiency to CRISPR^MAX in HEK 293T cells (12.5 ± 3.0%). In this study, RFP can also be employed as a fluorescent probe to monitor the delivery of RNP in cells.

Lipids

Lipids are the most commonly used materials for transferring exogenous nucleic acids into cells. Lipids are composed of a polar head group, a hydrophobic tail, and connecting part between two structures. Early studies showed that the cellular internalization of lipid polyplexes was achieved through direct membrane fusion, but now it is more likely to believe that this process mainly depends on endocytosis. After cellular internalization, lipids would disrupt the endosomal membrane, leading to flip-flop reorganization of phospholipids and eventual release of payloads into the cytoplasm. To date, lipids have been successfully employed for the delivery of small-molecule drugs and nucleic acids, as well as CRISPR-Cas9 pDNA and Cas9 mRNA. For example, Zhang et al. developed polyethylene glycol phospholipid-modified cationic lipid nanoparticles (PLNP) to condense and encapsulate a CRISPR-Cas9 pDNA (targeting polo-like kinase 1 (PLK-1)) and achieved 47.4% transfection efficiency in A375 cells in vitro. After intratumoral injection into melanoma tumor-bearing mice, PLNP/pDNA complex achieved significant downregulation of PLK-1 and suppression of the tumor growth (>67%) in vivo. Guo et al. reported a noncationic, deformable, and tumor-targeted nanolipogel system (tNLG) for tumor-specific CRISPR-Cas9 genome editing (Figure ). Compared to cationic lipids, tNLG was made of zwitterionic and anionic lipid, which can encapsulate CRISPR-Cas9 pDNA to form a core-shell nanostructure independent on electrostatic interaction, avoiding the potential cytotoxicity of cationic materials. By surface-grafting of ICAM1 antibody, tNLG can selectively recognize and bind triple-negative breast cancer (TNBC) cells to knock out Lipocalin 2 (>81%), and significantly inhibit the growth of TNBC (>77%).

[IMAGE OMITTED. SEE PDF]

For Cas9 mRNA form, Miller et al. synthetized a zwitterionic amino lipids (ZALs) nanoparticles (ZNP) to co-deliver long RNAs including Cas9 mRNA and sgRNA in vitro and in vivo (Figure ). ZNP enabled permanent genome editing with an indefinitely sustained 95% decrease in protein expression (luciferase). ZNP-based mRNA delivery achieved high protein expression (mCherry) at low doses in vitro (<600 p.m.) and in vivo (1 mg kg⁻¹). With the systemic administration of Cas9 mRNA and sgRNA against LoxP in genetically engineered mice, ZNP induced expression of floxed tdTomato in the liver, kidneys, and lungs. To achieve selective tissue-targeting delivery of CRISPR-Cas9 system, Siegwart and colleagues reported a novel selective organ targeting strategy (SORT), which allows the targeted delivery of various gene cargoes, including mRNA, Cas9 mRNA and sgRNA, and Cas9 RNP into lung, spleen, and liver of mice after the systemic administration. The SORT nanoparticle is composed of ionizable cationic lipids, amphipathic phospholipids, cholesterol, PEGylated lipids, and SORT lipids. The results found that SORT nanoparticles with anionic SORT lipids can specifically accumulate in the spleen. With the increasing number of cationic SORT lipids, SORT nanoparticle would gradually accumulate from liver to spleen, and then to the lung, respectively. In addition, SORT nanoparticle with ionizable cationic SORT lipids displayed effective liver accumulation but without tissue-targeting capability. SORT nanoparticle with zwitterionic cationic SORT lipids exhibited weaker tissue-targeting capability compared to that with cationic SORT lipids.

Although Cas9 nuclease is highly positive, the net charge of Cas9 RNP (compound of Cas9 and sgRNA) is negative. Thus, some commercial cationic lipids, including Lipofectamine 2000, Lipofectamine 3000, and Lipofectamine RNAi^MAX, can be complex with Cas9 RNP for effective genome editing. Furthermore, Zuris et al. fused Cas9 protein with negatively charged GFP (recombinant Cas9) to achieve protein delivery using commercial cationic transfection reagents. However, it should be noted that the recombinant Cas9 may reduce Cas9 activity, resulting in decreased genome editing efficiency. Considering the effectiveness of cationic lipids in CRISPR-Cas9 delivery, Thermo Fisher Company developed a new transfection reagent, called CRISPR^MAX, for Cas9 RNP delivery. By optimizing the transfection condition, the genome editing efficiency of CRISPR^MAX in human iPSC, mouse ES cells, and HEK 293FT cells reached 55%, 75%, and 85%, respectively. To maximize the delivery efficiency of Cas9 RNP, Xu et al. designed a lipid library to screen effective lipid structure for intracellular delivery of Cas9 RNP. A series of bioreducible lipids were synthetized via Michael addition of primary or secondary amines and an acrylate that containing a disulfide bond and a 14-carbon hydrophobic tail. To demonstrate the potential of these lipids in Cas9 RNP delivery, cholesterol, C16-PEG₂₀₀₀-ceramide, and 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) were employed to stabilize the liposome. By loading a Cas9 RNP targeting GFP, three lipids in this library induced more than 70% EGFP disruption in EGFP-expressing HEK cells. In another study, Li and colleagues reported a library of chalcogen containing lipid nanoparticles for the intracellular delivery of Cas9 RNP. The chalcogen-containing lipids were synthesized through reacting lipophilic tails containing O, S and Se ethers (O17O, O17S and O17Se) with compound bearing amines. The results demonstrated that lipids with the O17Se tail showed the highest GFP knockout and low cytotoxicity profiles. In addition to cationic lipids mentioned above, noncationic lipids modified with a nitrilotriacetic acid (NTA) group were designed to deliver His-tagged proteins (Figure ). Nickel ions (Ni²⁺) were added to enhance the interaction between His-tagged proteins and NTA-modified lipids. Noncationic lipid nanoparticles incorporated NTA-modified lipids can efficiently deliver His-tagged Cas9 RNP into mammalian cells.

DNA nanostructures

As an endogenous substance, DNA has good biocompatibility. DNA can also be programmed to form specific DNA nanostructures through its excellent molecular recognition function. In recent years, self-assembled DNA nanostructures have been an attractive approach for drug delivery. A yarn-like DNA nanoclew was synthesized by rolling circle amplification (RCA), and its palindrome sequences were partially complementary to sgRNA in Cas9 RNP (Figure ). After complexation of DNA nanoclew with RNP, the cationic polymer PEI was further coated on the surface of the nanoclew, which can promote the cell uptake and endosomal escape. DNA nanoclews (NC-12) with 12 nucleotides complementary to sgRNA have higher genome editing efficiency than NC-0 and NC-23, which could be attributed to the balance between Cas9 RNP binding and intracellular release. Intratumoral injection of NC-12/RNP/PEI complexes can induce 25% gene knockout in surrounding U2OS cells. In later work, a hepatocyte-targeted charge reversal polymer layer (Gal-PEI-DM) was introduced to the surface of NC/RNP/PEI for targeted delivery of Cas12a/crRNA RNP (Figure ). This Gal-PEI-DM can revert the overall charge of nano-assembly from negative charge to positive charge in acidic endosome microenvironment for the effective endosomal escape. A general strategy for the construction of a double-stranded DNA-RNP hybrid nanostructure by folding double-stranded DNA was developed. In this study, dCas9 and dCas12a were fused through a flexible and stimuli-responsive peptide linker. The covalently bivalent dCas9-12a RNP can recognize targeted sequences in the double-stranded DNA scaffold with guide RNAs and form double-stranded DNA-RNP nanostructures. This hybrid nanostructure can protect genetic information in the folded state and induce efficient gene transcription in the unfolded state. Lin-Shiao et al. described a DNA nanostructure encoding an intact human gene and a fluorescent protein as compact templates for CRISPR-mediated HDR. This DNA nanostructure was designed to include Cas9 RNP binding sites to promote shuttling into the nucleus, resulting in lower toxicity and higher insertion efficiency than unstructured double-stranded DNA templates in human primary cells. In general, DNA nanostructures have been widely employed in CRISPR delivery, whereas the in vivo stability of DNA nanostructures should be improved.

[IMAGE OMITTED. SEE PDF]

Inorganic nanomaterials

Metal-organic frameworks

Metal-organic frameworks (MOFs) are a class of organic-inorganic hybrid crystalline porous materials, which were made of inorganic metal ions and organic molecules, and valued for its tunable porous structure and good biocompatibility. Zeolitic imidazole frameworks (ZIFs) are a subtype of MOFs. The payloads loaded by ZIFs can be released under the stimulation of pH or ATP. ZIF-8 synthetized by Zn²⁺ and 2-methylimidazole was first reported to deliver Cas9 RNP. When ZIF-8-Cas9 RNP complexes (CC-ZIFs) were internalized by cells, excellent protonation capability of ZIF-8 in an acidic environment would facilitate endosomal escape of CC-ZIFs, and release payloads into cytosol (Figure ). However, the Cas9 RNP loading efficiency of ZIF-8 was only about 17%. In order to increase the loading efficiency of ZIF-8, Wang et al. reported a silica-metal-organic framework hybrid nanoparticle (SMOF NP) composed of both silica and ZIF-8 with a higher loading efficiency (90%). Furthermore, various functional groups can be easily incorporated into silica component on the surface of SMOF NP. Besides, ZIF-90 made of Zn²⁺ and imidazole-2-carboxaldehyde has been proved to possess a high protein loading efficiency (>90%). In addition, in the presence of ATP (2 mM), ZIF-90 can be degraded to release protein due to the competitive coordination between ATP and Zn²⁺. In vitro studies found that ZIF-90 could deliver various proteins into the cytoplasm regardless of the size and molecular weight of the protein. When loading Cas9 RNP, ZIF-90 enabled the effective intracellular delivery of Cas9 RNP and disrupted the GFP expression in HeLa cells with high efficiency (35%).

[IMAGE OMITTED. SEE PDF]

Mesoporous silica nanoparticles

In recent years, mesoporous silica nanoparticles (MSNs) has received widespread attention in the delivery of various drugs due to its tunable morphology and pore diameter, easily functional surface, and large surface area. Such advanced nanostructures also have great prospect in delivering CRISPR-Cas9 system. For example, Xu et al. reported a polymer-coated MSN for the intracellular delivery of CRISPR-Cas9 pDNA (Figure ). This MSN was first functionalized with a fluorescent dye (Cy5.5) and nuclear localization sequence (NLS), and further coated with a poly(dimethyldiallylammonium chloride) (PDDA) through microfluidic nanoprecipitation. The PDDA endows MSNs with enhanced stability and prevents premature release. Benefiting from the pH-responsive charge of PDDA, MSNs achieved enhanced cellular uptake and endosomal escape. Thereafter, the NLS in the MSN surface would facilitate the nuclear transport of CRISPR-Cas9 pDNA, enabling the effective GFP-tag knockout of the PXN genomic sequence in U2OS cells. Similar to previous work, Noureddine et al. developed a monosized lipid-coated mesoporous silica nanoparticle (LC-MSN) delivery vehicle, which enables loading both CRISPR components (RNP or pDNA) and efficient release within cancer cells. The RNP-loaded LC-MSN exhibited 10% genome disruption in both in vitro reporter cancer cell lines and in vivo Ai9-tdTomato reporter mouse model.

Black phosphorus nanosheets

Black phosphorus nanosheets (BPs) are a class of two-dimensional materials with puckered honeycomb surfaces, which provide an ultra-high specific surface area for the loading of Cas9 RNP. The BPs exhibit excellent biocompatibility and can be degraded into phosphite/phosphate ions without toxicity. Based on these features of BPs, some BPs-based carriers have been developed for the CRISPR delivery. For example, Zhou et al. reported a biodegradable BP for the delivery of Cas9 RNP with three NLS at the C terminus (Cas9N3) (Figure ). In vitro studies found that Cas9N3-BPs can be efficiently internalized through the membrane penetration and endocytosis pathways, followed by successful endosomal escape and cytosolic release of RNP with the degradation of BPs. Compared to other delivery platforms, Cas9N3-BPs enables effective genome editing at a relatively low dose, providing a universal cytosolic delivery strategy for CRISPR-Cas9 system and other biomacromolecules. In another similar study, Yue et al. developed a polylysine functionalized black phosphorus (PBP) for the delivery of Cas13a/crRNA. After entering the cytoplasm, PBP/Cas13a/crRNA complexes would escape from endosome with the degradation of PBP, and eventually down-regulate Mcl-1 at the transcriptional level with an efficiency of 58.64%. Mcl-1 amplification has been demonstrated as a key factor for breast cancer cell survival. PBP/Cas13a/crRNA targeting Mcl-1 showed exciting tumor suppression efficacy (65.16%) after the intratumoral injection. In general, BPs have great potential for the delivery of the CRISPR-Cas9 system in gene therapy.

Gold nanoparticles

Gold nanoparticles (GNPs) are a nanomaterial with good biocompatibility and low toxicity, offering a promising platform for the delivery of multiple biomacromolecules. Thiol-containing molecules can be easily grafted onto GNPs through gold-thiol bonds. For example, Rotello et al. developed several arginine-functionalized GNPs (ArgNPs) for the cytosolic delivery of protein and nucleic acid. To enhance the loading efficiency of Cas9 RNP by ArgNPs, an anionic glutamate tag (E-tag) was fused to the N-terminus of Cas9 protein (Cas9En) to increase the charge density of Cas9 RNP. The number of E-tag affected the size and cellular internalization of ArgNPs-RNP nanoassemblies, Cas9 modified with 15 or 20 repeated glutamate units exhibited the highly efficient direct cytoplasmic delivery of RNP (~90%). ArgNPs-RNP nanoassemblies achieved about 30% genome disruption of both AAVS1 and PTEN genes in HeLa cells. Due to simultaneous delivery of Cas9, sgRNA, and donor DNA, in vivo gene editing via HDR remains a challenge. Lee and colleagues developed a GNPs-based delivery vehicle (CRISPR-Gold) for simultaneous delivery of Cas9 RNP and donor DNA. To obtain CRISPR-Gold, GNPs were first modified with thiol-oligonucleotides, followed by hybridization with donor DNA. Cas9 RNP was introduced onto the surface of particles via the binding affinity between Cas9 RNP and DNA. Eventually, a silica layer was used to enhance the negative charge density and then complexed with the cationic endosomal disruptive polymer PAsp(DET) to prepare CRISPR-Gold. In vitro studies found that CRISPR-Gold can effectively enter cells via the caveolae/raft-dependent endocytosis. CRISPR-Gold showed less toxic than lipofectamine or nucleofection methods and induced an HDR efficiency between 3% and 4% in several cell lines, such as human ES cells, human iPSC, and bone-marrow-derived dendritic cells. As a result, CRISPR-Gold successfully corrected the DNA mutation associated with Duchenne muscular dystrophy in mice after the local injection with minimal off-target effects. In another study, Lee et al. reported a similar CRISPR-Gold without donor DNA. By delivering Cas9 RNP targeting Cpf1, CRISPR-Gold can edit multiple brain cells without detectable toxicity. CRISPR-Gold targeting the metabotropic glutamate receptor 5 (mGluR5) effectively reduced mGluR5 levels in the striatum after intracranial injection and rescued mice from exaggerated repetitive behaviors. Shahbazi et al. also developed a similar nanoformulation (AuNP/CRISPR) by layer-by-layer conjugation of CRISPR components (Cas9, sgRNA, and donor DNA) on the surface of GNPs to achieve HDR, and demonstrated safety delivery of multiple CRISPR components in primary human blood progenitors. In addition to GNPs, gold nanowires (AuNWs), another type of gold-based nanomaterials, were designed for active and direct intracellular delivery of Cas9 RNP (Figure ). In this study, Cas9 RNP was immobilized onto the surface of AuNWs via a degradable disulfide bond. Under an ultrasound field, Cas9-sgRNA@AnNWs can be rapidly internalized by tumor cells, and then released Cas9 RNP into cytoplasm due to the high concentration glutathione (GSH) in tumor cells.

Calcium phosphate nanoparticles

Calcium phosphate nanoparticles are also usually used as non-viral gene vectors because of good biocompatibility and strong binding affinity with nucleic acids. For example, Liu et al. reported a dual-targeting delivery system based on polymer/inorganic hybrid nanoparticles. CRISPR pDNA was firstly encapsulated in an inorganic core made of protamine sulfate, calcium carbonate, and calcium phosphate through co-precipitation. Thereafter, functionalized carboxymethyl chitosan was further introduced to the surface of inorganic core to improve cell uptake and nuclear transport efficiency. This dual-targeting delivery system delivered CRISPR pDNA into the nucleus of human breast cancer cell and reduced the expression of CDK 11% to 10%. Additionally, Cas9 RNP can also be in situ mineralized by calcium phosphate under physiological conditions. Calcium phosphate mineralization significantly increased the stability of Cas9 RNP and cellular internalization without decrease in bioactivity. The mineralized Cas9 RNP nanoparticles were efficiently delivered into protoplast cells, and achieved 20% genomic editing for Magnaporthe oryzae compared to RNPs alone.

Cell-penetrating peptides

CPPs are a class of short peptides, which can deliver a variety of biomacromolecules into cells, such as RNA, DNA, and proteins. Compared to synthetic delivery reagents, CPPs hold less cytotoxicity, excellent biocompatibility, biodegradability, and even can be directly fused with proteins without decrease in bioactivity. Therefore, CPPs are other potential materials that can directly deliver various forms of CRISPR-Cas9 system into cells. For example, as one of the commonly used CPPs, octaarginine (R8) can enhance the transfection efficiency of nucleic acid drugs. Li and colleagues encapsulated paclitaxel (PTX) and pDNA encoding Cas9 and hypoxia-inducible factor-1α (HIF-1α)-targeting sgRNA with a R8-modified cationic liposome (R8-dGR-Lip). Compared to PEGylated liposomes, R8-dGR-Lip exhibited improved cell targeting and deeper tumor penetration. Considering the importance of HIF-1α in inhibiting tumor metastasis, R8-dGR-Lip disrupted the expression of HIF-1α and its downstream proteins, resulting in superior antimetastatic effects in pancreatic cancer. In addition, HIF-1α disruption also enhanced the cytotoxicity of PTX on BxPC-3 cell lines. In another study, CPPs were directly conjugated to recombinant Cas9 protein via a thioether bond, and further complexed with CPPs/sgRNA to form positively charged nanoparticles. This nanoparticle achieved efficient genome disruption with less off-target effects compared to pDNA transfection in various human cells, including ES cells, dermal fibroblasts, HEK 293T cells, HeLa cells, and embryonic carcinoma cells. Similarly, Yin et al. Developed a generic method based on genetic fusion of supercharged polypeptide (SCP) for the intracellular protein delivery (Figure ). They found that SCP can bind to the nuclear import protein importin β1, and thus enables efficient delivery of proteins with various charges and sizes into the cell nucleus. In addition, SCP exhibits negligible hemolytic activity, good serum stability, and low immunogenicity in vivo. Cas9 RNP (targeting CCR5 gene) delivery mediated by SCP resulted in 15.2% indel efficiency in HeLa cells. In later work, by incorporating dithiocyclopeptide linker containing a matrix metalloproteinase 2 (MMP-2) sensitive sequence and an intramolecular disulfide bond between Cas9 and SCP, higher genome editing efficiency in tumor cells was observed. What's more, Cas9 nuclease was fused with both NLS and low-molecular-weight protamine, achieved 43.9% indels of KRAS gene in A549 cells. This system also disrupted both PD-L1 and PD-L2 in cancer cells, and thus induced robust anti-tumor immune responses in vivo. CPPs can also achieve Cas9 RNP delivery through non-covalent interactions. For instance, Shen et al. designed an amphipathic α-helical peptide called endo-porter (EP), which can complex with Cas9 RNP through electrostatic interaction. Upon “proton-sponge effects,” EP-induced efficient GFP gene deletion in multiple cells isolated from GFP transgenic mice.

[IMAGE OMITTED. SEE PDF]

STIMULI-RESPONSIVE NANOCARRIERS FOR CRISPR-CAS9 DELIVERY

Although non-viral delivery systems have been designed to deliver various forms of CRISPR-Cas9 system (e.g., CRISPR pDNA, Cas9 mRNA and sgRNA, and Cas9 RNP), and displayed comparable genome editing efficiency, low immunogenicity, and potential of mass manufacturing. Despite these advantages in CRISPR-Cas9 delivery, some challenges remain to be addressed. These challenges include the potential off-target effects and the suboptimal therapeutic efficiency, which are mainly caused by the undesired accumulation of CRISPR-Cas9 components in non-targeted tissues. With the rapid advances in materials science and life science, intelligent stimuli-responsive nanocarriers that can respond to various internal (e.g., pH, enzymes, ATP, glucose, redox, and hypoxia) or external stimuli (e.g., photo, ultrasound, and magnetism), have been broadly developed for the spatiotemporal delivery of the CRISPR-Cas9 system.

Internal stimuli-responsive nanocarriers for CRISPR-Cas9 delivery

pH-responsive CRISPR-Cas9 delivery

pH values display obvious heterogeneity in different physiological microenvironments. For instance, endosomes/lysosomes (pH 4.0–6.5) and the extracellular environment of tumors or inflammatory tissues (pH 6.5–6.8) show a significant acidic environment. However, the cytosol and normal physiological environment is near neutral (about pH 7.4). Therefore, these pH differences have been broadly explored for the intelligent responsive delivery of various drugs as well as CRISPR-Cas9 system. For instance, Tu et al. developed a novel pH-responsive nanoparticle for the co-delivery of CRISPR-Cas9 pDNA targeting Cdk5 and PTX. This nanocarrier was achieved through the self-assemble of poly-(ethyleneimine)-poly(lactic-co-glycolic acid) and a PEG corona covalently linked via acid cleavable bonds. When entering acidic tumor microenvironment, pDNA would be released with the detachment of PEG corona. The successful delivery of the CRISPR-Cas9 system could effectively disrupt the Cdk5 gene, followed by down-regulating the expression of PD-L1 to block the immune checkpoint pathway and restore antitumor immunity. Moreover, PTX can induce the ICD of tumor cells, achieving effective activation of DCs, suppression of Tregs, and the polarization of M2 to M1 in tumor microenvironment. As a result, this nanocarrier could convert cold tumor into hot tumor, and then effectively inhibited tumor growth and prolonged overall survival of tumor-bearing mice. Similarly, our group reported a multistage delivery nanoparticle (MDNP) with a core-shell structure, in which the core was formed by PBA-modified PEI (PEI-PBA) and pDNA encoding CRISPR-dCas9 system, whereas the responsive shell was made of 2, 3-dimethylmaleic anhydride (DMMA)-modified poly(ethylene glycol)-b-polylysine (mPEG₁₁₃-b-PLys₁₀₀/DMMA) (Figure ). In order to overcome the multiple physiological barriers, MDNP was designed to exhibit compatible surface properties in different physiological environments. During the blood circulation, MDNP can maintain the core-shell structure, which can avoid the undesirable protein adsorption and immune clearance. As entering tumor acidic microenvironment, the polymer shell could rapidly detach from the cationic core due to the electrostatic repulsion. The exposure of PBA-modified polyplex core enhances the tumor accumulation and cellular internalization of CRISPR-dCas9 system. With the systemic administration of MDNP, we successfully recovered the expression of endogenous miR-524 and thus inhibited the MDA-MB-231 tumor growth in mice. Owing to abundant amino groups, PEI in these works effectively triggers endosomal escape via “proton sponge effects” and thus is widely developed for CRISPR-Cas9 delivery. Furthermore, PEI coated nanoparticles, such as GNPs, DNA nanoclews, and graphene oxide, exhibited enhanced endosomal escape. In later work, we also reported a nano-Cas9 ribonucleoprotein system (nanoRNP) that can load any combination of sgRNAs to overcome the tumor heterogeneous in cancer treatment (Figure ). NanoRNP was prepared by mixing Cas9 RNP and a pH-responsive graft polymer, 2, 5-dihydro-2, 5-dioxofuran-3-acetic acid (CA)-bridged polylysine-g-poly(ethylene glycol) (PLys₁₀₀-CA-mPEG₇₇). Similar to previous work, pH-responsive polymer endows nanoRNP with prolonged blood circulation, improved cellular internalization, and enhanced tumor accumulation. By loading a combination of sgRNAs targeting STAT3 and RUNX1, nanoRNP significantly suppressed the growth of tumor model with two cell subpopulations whose proliferation is sensitive to STAT3 or RUNX1 disruption, respectively. Recently, Xie et al. also reported a simple, safe, and efficient pH-responsive polymer nanoparticle that can deliver Cas9 RNP alone (NHEJ-NP) or Cas9 RNP together with donor DNA (HDR-NP). After intravenous, intratracheal, or intramuscular injection, NHEJ-NP effectively disrupted the targeted gene in mouse liver, lung, or skeletal muscle, respectively. Intramuscular injection of HDR-NP recovered the muscle strength of the Duchenne muscular dystrophy mouse. In another study, Zhen et al. designed a novel pH-sensitive cationic nano-liposome to selectively deliver the CRISPR-Cas9 system into tumor tissues for the treatment of cervical cancer. This nano-liposome can remain stable in normal tissues, whereas it decomposes in a weakly acidic tumor microenvironment, thereby achieving efficient tumor targeting and genome disruption.

[IMAGE OMITTED. SEE PDF]

In addition to polymer and liposome materials, other pH-responsive materials were designed for CRISPR-Cas9 delivery. For example, the aforementioned ZIF-8 can efficiently encapsulate Cas9 RNP via the coordination bonds, but release them in response to an acidic condition (pH = 5). In another study, Wang et al. also developed a pH-responsive SMOF NP. The SMOF NP was made of silica and ZIF, enables efficient delivery of various drugs, including intracellular delivery of a variety of payloads, including hydrophilic small molecule drugs, nucleic acids, as well as RNP and RNP together with donor DNA. The superior pH-responsive release and endosomal escape capability of SMOF NPs endow its exciting drug delivery efficiency. Moreover, the surface of SMOF NPs can also be modified with PEG and various targeted ligands. By subretinal injection of SMOF NP loading Cas9 RNP, efficient in vivo genome editing of retinal pigment epithelial cells was achieved. Li et al. developed a proton-activated DNA nanosystem (H-DNC) for the co-delivery of Cas9 RNP and DNAzyme. To prepare H-DNC, ultra-long DNA chains, which contains the sgRNA recognition sequence, DNAzyme sequence and HhaI enzyme cleavage site were firstly synthesized via RCA. Then, the cofactor DNAzyme Mn²⁺ was added to condense the long DNA chains to form DNA cage. Eventually, an acid-degradable polymer-encapsulated Hhal enzyme was assembled on the surface of DNA cage to form a pH-responsive DNA nanosystem. This nanosystem was then intravenously injected into the mice and entered the tumor cells via a lysosome-mediated endocytosis pathway. With the degradation of polymer on the surface of the Hhal enzyme, the released Hhal enzyme could recognize and cut off the cleavage sites in long DNA chains to further release Cas9 RNP and DNAzyme, thus achieving effective gene therapy for breast cancer.

Redox-responsive CRISPR-Cas9 delivery

It is well known that the redox potentials in intracellular and extracellular environments are significantly different. For example, it is found that the intracellular concentration (0.5–10 mM) of GSH is about three orders of magnitude higher than that of the extracellular environment (2–20 μM). Therefore, the use of GSH-sensitive disulfide bonds in intracellular drug delivery efficiently enables release of the payloads after cellular internalization. In an earlier study, Liu et al. designed a lipid nanoparticle (BAMEA-O16B) integrated with disulfide bonds for the intracellular delivery of Cas9 mRNA and sgRNA (Figure ). After entering the tumor cell, high intracellular concentration of GSH breaks the disulfide bonds, resulting in rapid and efficient release of Cas9 mRNA and sgRNA. In vitro experiments showed that BAMEA-O16B effectively disrupted the expression of GFP (with efficiency up to 90%) in human embryonic kidney cells. Furthermore, BAMEA-O16B can efficiently accumulate in the hepatocytes after the intravenous injection. For the effective malignant tumor treatment, our group designed a virus-like nanoparticle (VLN) to co-deliver Cas9 RNP and small molecule drugs (Figure ). Small molecule drugs were firstly encapsulated into surface-thiolated MSNs and then locked through conjugating Cas9 RNP via disulfide bonds. These MSNs loaded small molecule drugs and Cas9 RNP was further coated with a lipid layer to prolong the blood circulation and avoid enzymatic degradation in physiological environment. After cellular internalization, VLN releases Cas9 RNP and small molecule drugs in response to the intracellular reduction microenvironment, leading to the synergistic regulation of multiple cancer-associated pathways. By loading Cas9 RNP targeting PD-L1 and axitinib, a small molecule inhibitor of tyrosine kinase, VLN achieved effective blockade of PD-1/PD-L1 pathway and reinvigoration of exhausted T cells.

Recently, Gong and colleagues developed multiple redox-responsive nanocarriers containing disulfide bonds for CRISPR-Cas9 delivery. For example, they designed a customizable nanocapsule for the delivery of Cas9 RNP with controlled stoichiometry. This nanocapsule was composed of Cas9 RNP and a biodegradable polymer network, which was formed through in situ free-radical polymerization of GSH-degradable crosslinker and several functional monomers, including cationic monomer, anionic monomer, imidazole monomer, PEGylated monomer, and ligand-containing monomer. The performance of the nanocapsules can be optimized by changing the ratios of various monomers and crosslinkers, thus offering customizable size (25 nm), excellent stability, successful endosomal escape, efficient intracellular release of Cas9 RNP, and an easily modified surface. As a result, the optimized nanocapsule induced robust gene editing in murine retinal pigment epithelium tissue and skeletal muscle after local injection. In a later study, Zou et al. designed a unique nanocapsule for the noninvasive brain delivery of the CRISPR-Cas9 system. Similarly, CRISPR-Cas9 nanocapsules were constructed by coating Cas9 RNP with a GSH-degradable polymer shell containing a double-action ligand, which can promote the BBB penetration and tumor cell targeting. They demonstrated the effective in vivo glioblastoma targeting and 38.1% PLK1 genome editing efficiency, ultimately achieving an effective and safe gene therapy for malignant gliomas.

In addition to the aforementioned reductive GSH, the level of reactive oxygen species (ROS) in cancer cells and inflammatory tissues is usually higher than that in healthy cells. This difference in ROS level can also be utilized to design stimuli-responsive nanocarriers. For example, Yan et al. developed a CRISPR-Cas9 prodrug nanosystem (NanoProCas9) that can specifically activate genome editing in response to inflammatory environment. Poly(β-amino ester) (PBAE) was firstly employed to condense a pDNA encoding destabilized Cas9 (dsCas9) with dihydrofolate reductase domains. Then, PBAE/pDNA complex was further coated with macrophage membrane to achieve the targeted delivery of CRISPR-Cas9 system into inflammatory tissue. Finally, trimethoprim, a ROS-responsive precursory molecule containing a stabilizer for dsCas9, was introduced into the macrophage membrane via lipid fusion. In normal environment, NanoProCas9-induced dsCas9 would be degraded by ubiquitin-dependent proteasome. However, dsCas9 can be stabilized for robust genome editing by trimethoprim, which is activated in an inflammatory environment with high-level ROS. More importantly, NanoProCas9 may lose genome editing capability when the ROS level is normal after therapy, providing a safer strategy for the in vivo application of the CRISPR-Cas9 system. In another study, Zhao et al. reported an HSP70-promoter-driven CRISPR-Cas9 delivery system (denoted as F-PC/pHCP). F-PC/pHCP was composed of a fluorinated dendrimer containing chlorin e6 (Ce6) and an HSP70-promoter-driven pDNA encoding CRISPR-Cas9 system. Under 660 nm laser irradiation, F-PC/pHCP would generate ROS to activate the HSP70 promoter, thus inducing specific PD-L1 disruption to prevent the immune escape of tumor cells. In addition, ROS generated by irradiation with F-PC/pHCP also triggered the ICD of tumor cells to reprogram the immunosuppressive tumor microenvironment. As mentioned above, BPs in this system can be degraded in response to intracellular ROS, and then release the encapsulated CRISPR-Cas9 system to realize genome editing.

Enzyme-responsive CRISPR-Cas9 delivery

The expression levels of certain enzymes are varied in pathological conditions, and this property has been utilized to design a series of enzyme-responsive drug delivery systems. For example, the expression levels of hyaluronidase (HAase) and MMPs in tumor microenvironment are generally higher than normal tissues, and controlled release of drugs could be achieved by using enzyme-sensitive peptides or substrates to construct delivery system. In an aforementioned study, Yin et al. developed a SCP-mediated Cas9 RNP delivery system (Cas9-LS) by incorporating dithiocyclopeptide linker containing an MMP-2 sensitive polypeptide and an intramolecular disulfide bond between Cas9 and SCP. When entering tumor microenvironment, dithiocyclopeptide linker can be completely broken by the high levels of MMP-2 and GSH, resulting in cytoplasmic release of Cas9 and gRNA for efficient genome editing. In another study, Zhang et al. designed a peptide nucleic acid (PNA) containing an MMP-2-cleavable peptide sequence (Gly-Pro-Leu-Gly-Val-Arg-Gly) to control the activity of the CRISPR-Cas13a system. This strategy can reduce the cleavage activity of CRISPR-Cas13a system by inhibiting the binding of crRNA to CRISPR-Cas13a system. However, the high levels of MMP-2 can cleave the peptide sequence and thus prevent PNA from binding to crRNA. Subsequently, crRNA would bind to the CRISPR-Cas13a system in the presence of an RNA activator, thereby exerting its DNA cleavage capability.

Yang and colleagues reported a programmable unlocking nano-matryoshka-CRISPR system (PUN@Cas-PT) to disrupt both PD-L1 and protein tyrosine phosphatase N2 (PTPN2) for improved cancer immunotherapy (Figure ). PUN@Cas-PT was composed of a ROS-responsive core and dual enzyme-responsive (MMPs and HAase) corona. Reasonably designed structure endows PUN@Cas-PT with improved circulation stability, enhanced tumor retention and internalization, effective endosomal escape, and cytoplasmic release of CRISPR-Cas9 system. This PUN@Cas-PT can overcome multiple biological barriers and strictly limit the activation of the CRISPR-Cas9 system in tumor tissues, which generally overexpressed both MMPs and HAase, and showed a significant high level of ROS. The successful activation of CRISPR-Cas9 enabled efficient disruption of PD-L1 and PTPN2, and thus eliciting robust adaptive antitumor immunity and long-term immune memory with excellent biocompatibility. Based on the elevated level of HAase in tumor tissues, Li et al. also developed a multifunctional nuclear-targeting “core-shell” artificial virus (RRPHC), which can deliver CRISPR-Cas9 pDNA into the nucleus without NLS. RRPHC comprises a core of fluorinated polymer (PF33)/pDNA and a multifunctional shell of hyaluronic acid (HA) derivatives. When reaching the tumor tissues, the HA shell was degraded by HAase overexpressed in tumor microenvironment, thereby exposing the cationic core to facilitate the cell uptake and endosomal escape. Excitingly, RRPHC exhibited a higher transfection efficiency than commercialized reagents in many types of cell lines, such as HEK 293, HCT 116, SW 480, and B16F10 cells. When loading a CRISPR-Cas9 system targeting MTH1, RRPHC induced MTH1 disruption with efficiency up to 44% in SKOV3 cells and thus effectively inhibited tumor growth in vivo. Cai and colleagues developed a controllable CRISPR-Cas9 system (eiCRISPR) for cell-selective genome editing (Figure ). This eiCRISPR is composed of Cas9 nuclease, self-blocked inactive sgRNA (bsgRNA), and chemically caged DNAzyme, which could activate CRISPR-Cas9 in a controllable manner. They inhibit DNAzyme activity via chemical modification, while reactivating DNAzyme to cut off the blocking region of bsgRNA in the presence of certain cancer cell-overexpressed enzyme, and thus achieving selective genome editing in cancer cells. In vivo results demonstrated that eiCRISPR delivered by biodegradable liposomes enabled effective genome disruption of human papillomavirus 18 E6, inhibiting the tumor growth in a tumor-bearing xenograft.

[IMAGE OMITTED. SEE PDF]

ATP-responsive CRISPR-Cas9 delivery

As the most direct energy source in the organism, ATP content in the intracellular environment (1–10 mM) is about 1000 times higher than that in the extracellular environment (<5 μM). Taking advantage of this feature, ATP can be exploited to design the spatiotemporal delivery of the CRISPR-Cas9 system. In an aforementioned study, Yang et al. found that ZIF-90 can efficiently encapsulate Cas9 nuclease and deliver it into the cytoplasm via “proton sponge effects” (Figure ). Moreover, in the presence of ATP (2 mM), ZIF-90 can be degraded to selective-release payloads due to the competitive coordination between ATP and Zn²⁺ (Figure ).

MicroRNA-responsive CRISPR-Cas9 delivery

MicroRNAs (miRNAs) are class of small non-coding RNAs, which plays an important role in the post-transcriptional regulation of gene expression in plants, animals, and viruses. The miRNA activities varied in different cells, and thus can be used as an endogenous stimulus for designing smart-responsive CRISPR-Cas9 genome editing. For example, Hirosawa et al. developed a synthetic mRNA that can respond to miRNA (miR-Cas9 switch), thereby controlling the genome editing activity of Cas9 based on specific miRNA expressed in targeted cells. This mRNA was constructed by inserting a miRNA-complementary sequence into the 5′-UTR of the mRNA encoding Cas9. The developed two switches can specifically respond to miR-21-5p or miR-302a-5p to activate CRISPR-Cas9 genome editing in HeLa and human iPSCs, respectively. In a later study, Hirosawa et al. also designed a miRNA-responsive AcrllA4 switch with a similar strategy to control the expression of AcrllA4, which is a Cas9 inhibitor. When combined with Cas9- or dCas9-VPR-guided RNA complexes, the AcrllA4 switch would induce genome disruption or activation depending on miRNAs specifically expressed in targeted cells, preventing off-target effects caused by random cleavage at undesired sites. In another study, Wang et al. designed a miRNA sensor to detect the miRNA activity at cell levels using a miRNA-responsive sgRNA and dCas9-VPR that can activate the expression of a RFP. By designing sgRNAs targeting endogenous genes, this strategy can further manipulate the expression of endogenous genes and mutate specific DNA sequences guided with specific miRNAs and siRNAs in targeted cells. Overall, the miRNA-responsive CRISPR system can achieve a more precise regulation of endogenous genes or genome disruption in specific types of cells according to particular miRNAs in different compartments.

Hypoxia-responsive CRISPR-Cas9 delivery

Hypoxia is a typical pathological feature of many diseases, such as solid tumors. This pathological state of local hypoxia would disrupt the redox level of cells, leading to increased reductive stress. In this regard, Li et al. developed a CRISPR-Cas9 nanocomposite (APACPs) depended on hypoxic-responsive gold nanorods. The on-demand release of CRISPR-Cas9 was realized by conjugating Cas9 RNP onto gold nanorods via oxygen-sensitive azo bonds, and further coated with PEI for rapid endosomal escape. In the hypoxic tumor microenvironment, the azo group in APACPs was selectively cleaved by the overexpressed of reductase, achieving the on-demand release of Cas9 RNP and subsequent genome disruption of HSP90α. Disruption of HSP90α can relieve the thermal resistance of tumor cells, and thus APACPs enabled efficient in vivo and in vitro tumor ablation through low-temperature hyperthermia. APACPs not only realized the on-demand release of CRISPR-Cas9 system in tumor microenvironment, but also effectively avoided the thermal damage of normal organs caused by hyperthermia, providing an efficient and safe strategy for low-temperature hyperthermia. By inserting hypoxia-responsive element sequences into the promoter of targeted genes, Davis et al. developed a lipid nanoparticle-based therapeutic system that enveloped both human herpes simplex virus thymidine kinase suicide gene and CRISPR-Cas9 system to achieve cancer-specific gene expression. This strategy shows high transgene expression in cells exposed to hypoxic condition, equivalent to the efficiency of traditional cytomegalovirus and cellobiohydrolase promoters.

External stimuli-responsive nanocarriers for CRISPR-Cas9 delivery

Photo-responsive CRISPR-Cas9 delivery

Photo-responsive CRISPR-Cas9 delivery and activation provides a non-invasive and superior spatiotemporal solution. Upon rational design, materials could experience certain physical and chemical changes under laser irradiation, including photothermal effects based on photothermic agents, ROS generated by photosensitizers, and upconversion effects based on upconverting luminescent materials. For example, Wang et al. reported a lipid/AuNPs-based genome editing delivery vehicle (LACP). To construct LACP, AuNPs were first modified with TAT to efficiently condense CRISPR-Cas9 pDNA (CP) to from AuNPs/CP via the electrostatic interaction, and further coated with a lipid layer made of DOTAP, DOPE, cholesterol, and PEG₂₀₀₀-DSPE. Under laser irradiation, LACP can release TAT/pDNA complexes into the cytosol with the cleavage of Au-S bonds induced by laser triggered thermo-effects of AuNPs. Thereafter, the CP would enter the nucleus under the guidance of TAT, achieving effective genome editing. By carrying a CP targeting PLK-1 gene, a master regulator of mitosis overexpressed in multiple cancer cells, LACP effectively disrupted the expression of PLK1 under laser irradiation (24 mW cm⁻², 20 min), thus inhibiting the growth of A375 cells both in vitro and in vivo. To obtain a higher penetration depth, Li et al. designed a semiconducting polymer brush (SPPF) that can respond to the light in the second near-infrared window (NIR-II) for the controlled activation of the CRISPR-Cas9 system (Figure ). This SPPF can complex with CRISPR pDNA via the electrostatic and supramolecular interaction. Dexamethasone (Dex) that can dilate the nuclear pore by binding with the nuclear glucocorticoid receptor, was also encapsulated in the hydrophobic core of the SPPF nanoparticle to enhance genome editing efficiency. Under laser irradiation of NIR-II (808 nm), SPPF could promote the endosomal escape and release of payloads through photothermal conversion. Furthermore, SPPF can also be utilized to monitor the in vivo distribution of the CRISPR-Cas9 system, thus guiding the laser irradiation in real time. Yin and colleagues constructed a CRISPR-Cas9 2D gene-editing nanosystem (silicene Cas9) by loading Cas9 RNP onto the PEI-modified 2D silicene nanosheets through physical adsorption and π-stacking interaction. Because of high photothermal-conversion efficiency, superior loading capability, good biocompatibility and biodegradability, silicene nanosheets have attracted enormous attention in drug delivery. The periodic atomic grooves on the surface of 2D silicene can also be adapted to load Cas9 nuclease and sgRNA. After the cellular internalization, the “proton sponge effects” of PEI and NIR-II light (1064 nm)-mediated photonic hyperthermia effects promoted the rapid endosomal escape of silicene Cas9 and subsequent release of Cas9 RNP. By disrupting TXNDC5 (with a genome editing efficiency up to 47.68%), an important regulatory factor involved in interfering with photothermal therapy efficiency, silicene Cas9 significantly improved the photothermal hyperthermia effects on tumor.

[IMAGE OMITTED. SEE PDF]

In addition to photothermal effects, laser irradiation can generate ROS in the presence of photosensitizers. The generated ROS can be utilized to induce endosomal membrane disruption for effective endosomal escape and cleave ROS-sensitive bonds for the controlled release of drugs. For example, Deng et al. developed a dual NIR and GSH-responsive nanoparticles (T-CC-NPs) for the controlled release of Cas9 RNP and photosensitizer in the tumor cell cytoplasm (Figure ). To construct T-CC-NPs, anionic micelle self-assembled of nitrilotriacetic acid-disulfanediyldipropionate-polyethyleneglycol-b-polycaprolactone (NTA-SS-PEG-PCL) was firstly prepared to encapsulate Ce6 into hydrophobic core and bind Cas9 RNP tagged with histidine on the surface, and further coated with iRGD-modified copolymer. The photosensitizer (Ce6) can generate ROS under 671 nm irradiation to promote the endosomal escape of T-CC-NPs. Subsequently, GSH in the cytosol breaks the disulfide bonds to release Cas9 RNP. By loading a Cas9 RNP targeting the antioxidant regulator Nrf2, T-CC-NPs reduced Nrf2 expression in tumor cells under NIR irradiation, thus enhancing tumor cell sensitivity to ROS for effective photodynamic therapy. Zhao et al. developed a photo-switched CRISPR-Cas9 system to permanently block PD-1/PD-L1 pathway for effective cancer immunotherapy. Photo-switched CRISPR-Cas9 system was formed with photoactivated self-degradable PEI derivatives (PPCe) and a pDNA encoding Cas9 nuclease and sgRNA targeting PD-L1. Under 660 nm light irradiation, the PPCe can promote the endosomal escape via the photochemical internalization (PCI) effects, and then release pDNA in the cytoplasm with the cleavage of ROS-sensitive thioketal bonds. Consequently, photo-switched CRISPR-Cas9 system significantly enhanced the intratumoral infiltration of CD8⁺ T cells, achieving effective activation of T cell-mediated antitumor immunity.

Upconverting nanoparticles (UCNPs) can convert long-wavelength light (NIR light) into short-wavelength light (ultraviolet or visible light), which offers a deeper tissue penetration. Accordingly, Pan et al. designed a NIR light-responsive nanocarrier (UCNPs-Cas9@PEI) for Cas9 RNP delivery based on UCNPs (Figure ). Cas9 RNP was covalently connected on the surface of UCNPs via a photocleavable 4-(hydroxymethyl)-3-nitrobenzoic acid (ONA) molecule. The UCNPs can convert NIR light (980 nm) into local ultraviolet light to cleave ONA bonds, achieving the on-demand release of Cas9 RNP. By loading a sgRNA targeting PLK-1, UCNPs-Cas9@PEI effectively inhibited the tumor progression both in vitro and in vivo through the NIR-controlled genome editing. Similarly, Wu et al. reported a photo-responsive charge-reversal nanovector (UCNP-UVP-P) for the spatiotemporally controlled activation of CRISPR-Cas9 pDNA. The photo-cleavable electropositive polyethylene glycol (UVP) was coated on the surface of UCNPs to condense pDNA via the electrostatic interaction. Under NIR light irradiation (980 nm), the cationic side-chains of UVP can convert into anionic charged chains to release pDNA for effective PLK-1 disruption, and thus effectively inhibiting tumor growth without noticeable organ damage.

Magnetic-responsive CRISPR-Cas9 delivery

Magnetic-responsive refers to a technology that can guide magnetic materials to the targeted region under an external magnetic field, allowing remote-controlled targeted drug delivery. Magnetic-responsive CRISPR-Cas9 delivery can significantly improve the accuracy of genome editing to reduce undesirable off-target effects. Recently, Zhu and colleagues developed a spatially controllable CRISPR-Cas9 system (MNPs-BVs) by complexing MNPs with recombinant baculoviral vectors (BVs). BVs hold large loading capacity and ability to transduce multiple mammalian cells with high efficiency and low cytotoxicity. However, the in vivo function of BVs was inactivated by the complement system, leading to a significant decrease in viral transduction. However, MNP-BVs can be activated locally in vivo in the presence of an external magnetic field, achieving tissue-specific genome editing by facilitating MNP-BV margination and cellular internalization. Kaushik et al. also developed a magnetically guided non-invasive delivery strategy for the on-demand controlled release of CRISPR-Cas9 system based on magneto-electric nanoparticles (MENPs). Upon an external magnetic field, MENPs can cross the BBB to target latent virus and release Cas9 RNP due to polarization changes of MENP surface, thereby effectively eradicating latently HIV-1 infection in the brain.

Ultrasound-responsive CRISPR-Cas9 delivery

Ultrasound shows strong penetration to biological tissues, which can focus energy on the depth of diseased tissues without trauma. Thus, many invasive treatments that are not suitable for surgery can be intervened by ultrasound, and this method has also been reported on the delivery of CRISPR-Cas9 system. For example, Bruhn et al. developed ultrasound-powered nanomotors for intracellular delivery of Cas9 RNP by linking Cas9 with thiofunctional AuNWs via a reversible disulfide bond. Under ultrasound treatment (5 min), nanomotors can directly penetrate the cell plasma membrane, and then release Cas9 RNP to achieve effective genome editing. These nanomotors can disrupt the GFP expression with an efficiency up to 80% within 2 h incubation and display acceptable genome editing efficiency with only 0.6 nM Cas9 RNP, offering a promising approach for the spatiotemporal genome editing. Yin and colleagues developed an ultrasound remote control of RNP-release system (HMME@LipCas9) to knockout NFE2L2, a key protein in maintaining intracellular redox homeostasis, for enhanced sonodynamic therapy (SDT) (Figure ). HMME@Lip-Cas9 can generate ROS to disrupt the endosomal membrane to promote the endosomal escape of Cas9 RNP, and eventually release Cas9 RNP into the cytoplasm to break the oxidative stress defensing system under ultrasound irradiation. This study not only provides a potential strategy for the spatiotemporal genome editing but also represents a therapeutic model combining SDT with CRISPR-Cas9 genome editing. Pu et al. further covalently linked Cas9 RNP onto sonosensitizer-integrated MOFs via ROS-sensitive thioether bonds for enhanced SDT (Figure ). Under external ultrasound stimulation, sonosensitizer (5, 10, 15, 20-tetrakis (4-carboxy-phenyl) porphyrin) loaded in MOFs can generate sufficient ROS to damage DNA of tumor cells. Moreover, ROS can break the thioether bonds for effective Cas9 RNP release, inducing genome disruption of the MTH1 gene to improve the sensitivity of tumor cells to ROS-induced DNA damage. With this sono-controllable genome editing strategy, a significant tumor growth inhibition and prolonged survival rate were achieved.

[IMAGE OMITTED. SEE PDF]

CONCLUSIONS AND PERSPECTIVES

The CRISPR-Cas9 system stands out for its simplicity, accuracy and high genome editing efficiency during the updated iteration of the genome editing system and showed the great promise in scientific research and disease treatments. The CRISPR-Cas9 system can be delivered in the form of either CRISPR-Cas9 pDNA, Cas9 mRNA or sgRNA, or Cas9 RNP. However, each CRISPR-Cas9 form has a strong negative charge and can not enter the targeted cells due to electrostatic repulsion with the negatively charged cell membrane. Therefore, developing a safe and effective delivery system is an essential prerequisite for the application of the CRISPR-Cas9 system. Currently, although various physical methods (e.g., microinjection, electroporation, and hydrodynamic injection) and viral vectors (e.g., AAV, AV, and LV) have been developed to deliver CRISPR-Cas9 system, some challenges remain need to be addressed. Direct delivery of the CRISPR-Cas9 system into targeted cells using physical methods may be effective in cultured cells in vitro, but these physical methods are not suitable for in vivo clinical application. Similarly, virus vectors have demonstrated its high efficiency in CRISPR-Cas9 delivery, while their application was severely limited by the risks of mutagenesis, restriction of insertion size, and immunogenicity.

Compared to viral vectors, non-viral nanocarriers have exhibited multiple advantages in CRISPR-Cas9 delivery, including excellent loading capability, low immunogenicity, and reduced off-target effects. In addition, these non-viral nanocarriers can be adapted to deliver all three forms of CRISPR-Cas9 system. Therefore, non-viral nanocarriers offer new opportunities for the design of CRISPR-Cas9 delivery systems. In this review, we comprehensively reviewed the recent development of non-viral nanocarriers, including polymeric nanoparticles, lipids, DNA nanoclews, inorganic nanomaterials, and CPPs. The advantages and disadvantages of various non-viral nanocarriers are summarized in Table . However, non-viral nanocarriers need to be further improved for more precise and efficient CRISPR-Cas9 genome editing. In the past several years, stimuli-responsive nanocarriers have been designed for the spatiotemporal delivery of CRISPR-Cas9 system in responsive to various exogenous signals (e.g., pH, redox, enzymes, ATP, miRNA, and hypoxia) and external stimuli (e.g., photo, magnetism, and ultrasound). The advantages and disadvantages of various stimuli-responsive nanocarriers are summarized in Table . Moreover, these stimuli-responsive nanocarriers hold a great promise in overcoming multiple physiological and intracellular barriers, thus realizing efficient genome editing and reduced undesirable side effects.

TABLE 1 Advantages and disadvantages of various non-viral nanocarriers for CRISPR-Cas9 genome editing.

Delivery system	CRISPR-Cas9 form	Advantages	Disadvantages	Refs.
Polymeric nanoparticles	PEI	pDNA; Cas9 mRNA and sgRNA; and Cas9 RNP	High charge density and good pH buffering capacity	High biotoxicity of PEI with high molecular weight and relatively short circulation time in vivo	[]
Dendrimer	pDNA; Cas9 mRNA and sgRNA; and Cas9 RNP	Easy of functionalization; low cost; and high efficiency	Relatively short circulation time and unknown long-term toxicity in vivo	[]
Chitosan	pDNA; Cas9 mRNA and sgRNA	Good biocompatibility and biodegradability	Poor in vivo stability and relatively low efficiency	[]
Lipids		pDNA; Cas9 mRNA and sgRNA; and Cas9 RNP	High efficiency and feasible of in vivo application	Low payload capacity	[]
DNA nanostructures		Cas9 RNP	Programmable assembly and precise control of size, shape and function	Poor in vivo stability and potential immune cell recognition	[]
Inorganic nanomaterials	MOFs	pDNA; Cas9 mRNA and sgRNA; and Cas9 RNP	Large surface area; adjustable pore size; and fast endosomal escape	Potential toxicity of metal ions	[]
MSNs	pDNA and Cas9 RNP	Well-defined surface properties; ease of preparation; and high loading capability	Easy of precipitation and unknown long-term toxicity in vivo	[]
BPs	pDNA; Cas9 mRNA and sgRNA; and Cas9 RNP	High payload capacity; good biocompatibility and biodegradability	Low efficiency and poor in vivo stability	[]
AuNPs	pDNA and Cas9 RNP	Ease of preparation and functionalization	Potential toxicity in vivo	[]
Calcium phosphate	pDNA and Cas9 RNP	High payload capacity and good biocompatibility	Low efficiency	[]
CPPs		pDNA; Cas9 mRNA and sgRNA; and Cas9 RNP	High efficiency; safe; and feasible of in vivo application	Need protein engineering and selective sensitive to nuclease and protease	[]

TABLE 2 Advantages and disadvantages of various stimuli-responsive nanocarriers for CRISPR-Cas9 genome editing.

Type	Stimulus	Advantages	Disadvantages	Refs
Internal	pH	Prolonged blood circulation; enhanced endosomes/lysosomes escape; reduced side-effects; and improved tissue-targeting	Narrow pH gradient between targeted lesions and normal tissues; potential off-target effects	[]
Redox	Reduced side-effects and enhanced accumulation in multidrug resistant cell	Unstable in blood circulation; low tissue-targeting; uncontrollable intracellular release; and potential off-target effects	[]
Enzyme	Good stability during blood circulation; reduced side-effects; and improved tissue-targeting	Unstable release due to different level of enzyme in different disease stages	[]
ATP	Reduced side-effects	Unstable in blood circulation and uncontrollable intracellular release	[]
MicroRNA	Cell specific targeting and reduced side-effects	Complex design; unstable in blood circulation	[]
Hypoxia	Prolonged blood circulation; reduced side-effects; and improved tissue-targeting	Low response rate and potential off-target effects	[]
External	Light	Spatiotemporally controllable; easy to focus on smaller areas; and noninvasive and absence of ionizing radiations	Poor tissue penetration ability; potential toxicity of photosensitizers; and danger of over-heating	[]
Magnetic	Spatiotemporally controllable; no danger of overheating; good tissue penetration ability; low toxicity	Relatively costly and difficult to focus on smaller areas	[]
Ultrasound	Spatiotemporally controllable; good tissue penetration ability; low toxicity; and simple equipment	Possible metastatic dissemination due to ultrasound-induced enhancement of vessel permeability; short blood circulation of microbubble	[]

While so many promising results for the CRISPR-Cas9 delivery have been reported in recent years, there are still several critical challenges to be addressed in promoting CRISPR-Cas9 technology from basic research to clinical application. Firstly, the long-term biosafety of nanomaterials needs to be further studied due to accumulation of nanomaterials in normal organs, especially liver. Therefore, the development of biodegradable nanocarriers based on biodegradable chemical bonds (e.g., vinylene bonds, imine bonds, imidazole units, and disulfide bonds) may be a promising solution and should be widely studied. The local administration of nanocarriers into targeted tissues can also achieve higher genome editing efficiency, lower systemic toxicity, and reduced off-target effects. Secondly, targeted groups on the surface of nanocarriers often invalid after the systemic administration, which is mainly contributed to the hindrance of absorbed proteins (protein crown) during blood circulation. Therefore, negatively charged nanocarriers and a chain of suitable length between nanocarriers and targeted groups to avoid the non-specific protein adsorption and expose targeted groups are necessary. Thirdly, the precision of current stimuli-responsive systems in disease treatment is still suboptimal. A combination of multiple stimuli should be considered, including both internal and external stimuli. Additionally, more research should be carried out to explore new and specific stimuli of various diseases, such as miRNA stimuli. Fourthly, limited carriers were developed for gene correction due to the complexity design for co-delivery of CRISPR-Cas9 components and donor DNA templates. With the rapid progress in nanotechnology, more efforts should be made to develop delivery systems for effective HDR-based gene correction. Lastly, current nanocarriers used for CRISPR-Cas9 delivery are still focused on preclinical evaluation, and thus more efforts should be carried out to promote the clinical application of these technologies.

ACKNOWLEDGMENTS

This work was supported by National Key Research and Development Programs of China (Grant Number: 2018YFA0209700), the National Natural Science Foundation of China (NSFC; Grant Numbers. 22077073, 22204001), Natural Science Research Project for Anhui Universities (Grant Number: 2022AH050731), and Open Project of Key Laboratory of Functional Polymer, Ministry of Education (Grant Number: KLFPM202203).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

References

H. Ledford, E. Callaway, Nature 2020, 586, 346.

P. Strzyz, Nat. Rev. Mol. Cell Biol. 2020, 21, 714.

L. Cong, F. A. Ran, D. Cox, S. Lin, R. Barretto, N. Habib, P. D. Hsu, X. Wu, W. Jiang, L. A. Marraffini, F. Zhang, Science 2013, 339, 819.

J. A. Doudna, E. Charpentier, Science 2014, 346, [eLocator: 1077].

A. Katti, B. J. Diaz, C. M. Caragine, N. E. Sanjana, L. E. Dow, Nat. Rev. Cancer 2022, 22, 259.

P. D. Hsu, E. S. Lander, F. Zhang, Cell 2014, 157, [eLocator: 1262].

H. Zhu, C. Li, C. Gao, Nat. Rev. Mol. Cell Biol. 2020, 21, 661.

R. Barrangou, J. A. Doudna, Nat. Biotechnol. 2016, 34, 933.

M. Jinek, K. Chylinski, I. Fonfara, M. Hauer, J. A. Doudna, E. Charpentier, Science 2014, 337, 816.

J. E. Garneau, M.‐E. Dupuis, M. Villion, D. A. Romero, R. Barrangou, P. Boyaval, C. Fremaux, P. Horvath, A. H. Magadan, S. Moineau, Nature 2010, 468, 67.

M. Jinek, F. Jiang, D. W. Taylor, S. H. Sternberg, E. Kaya, E. Ma, C. Anders, M. Hauer, K. Zhou, S. Lin, M. Kaplan, A. T. Iavarone, E. Charpentier, E. Nogales, J. A. Doudna, Science 2014, 343, [eLocator: 1215].

H. Wang, H. Yang, C. S. Shivalila, M. M. Dawlaty, A. W. Cheng, F. Zhang, R. Jaenisch, Cell 2013, 153, 910.

T. Maruyama, S. K. Dougan, M. C. Truttmann, A. M. Bilate, J. R. Ingram, H. L. Ploegh, Nat. Biotechnol. 2015, 33, 538.

M. H. Porteus, N. Engl. J. Med. 2019, 380, 947.

X. Yao, X. Wang, X. Hu, Z. Liu, J. Liu, H. Zhou, X. Shen, Y. Wei, Z. Huang, W. Ying, Y. Wang, Y.‐H. Nie, C.‐C. Zhang, S. Li, L. Cheng, Q. Wang, Y. Wu, P. Huang, Q. Sun, L. Shi, H. Yang, Cell Res. 2017, 27, 801.

L. Xie, J. Sun, L. Mo, T. Xu, Q. Shahzad, D. Chen, W. Yang, Y. Liao, Y. Lu, J. Biol. Eng. 2019, 13, 90.

R. V. Schendel, J. V. Heteren, R. Welten, M. Tijsterman, PLoS Genet. 2016, 12, [eLocator: 1006368].

J.‐H. Seol, E. Y. Shim, S. E. Lee, Mutat. Res., Fundam. Mol. Mech. Mutagen. 2018, 809, 81.

S. Konermann, M. D. Brigham, A. E. Trevino, J. Joung, O. O. Abudayyeh, C. Barcena, P. D. Hsu, N. Habib, J. S. Gootenberg, H. Nishimasu, O. Nureki, F. Zhang, Nature 2015, 517, 583.

D. Bikard, W. Jiang, P. Samai, A. Hochschild, F. Zhang, L. A. Marraffini, Nucleic Acids Res. 2013, 41, [eLocator: 7429].

L. A. Gilbert, M. A. Horlbeck, B. Adamson, J. E. Villalta, Y. Chen, E. H. Whitehead, C. Guimaraes, B. Panning, H. L. Ploegh, M. C. Bassik, L. S. Qi, M. Kampmann, J. S. Weissman, Cell 2014, 159, 647.

M. H. Porteus, D. Carroll, Nat. Biotechnol. 2005, 23, 967.

N. Sun, H. Zhao, Biotechnol. Bioeng. 2013, 110, [eLocator: 1811].

C. A. Lino, J. C. Harper, J. P. Carney, J. A. Timlin, Drug Delivery 2018, 25, [eLocator: 1234].

R. M. Gupta, K. Musunuru, J. Clin. Invest. 2014, 124, [eLocator: 4154].

N. E. Sanjana, L. Cong, Y. Zhou, M. M. Cunniff, G. Feng, F. Zhang, Nat. Protoc. 2012, 7, 171.

H. Li, Y. Yang, W. Hong, M. Huang, M. Wu, X. Zhao, Signal Transduction Targeted Ther. 2020, 5, 6.

Z. WareJoncas, J. M. Campbell, G. Martinez‐Galvez, W. A. C. Gendron, M. A. Barry, P. C. Harris, C. R. Sussman, S. C. Ekker, Nat. Rev. Nephrol. 2018, 14, 663.

C. V. Trung, T. Weber, B. Wefers, W. Wurst, S. Sander, K. Rajewsky, R. Kuehn, Nat. Biotechnol. 2015, 33, 543.

C. Yu, Y. Liu, T. Ma, K. Liu, S. Xu, Y. Zhang, H. Liu, M. L. Russa, M. Xie, S. Ding, L. S. Qi, Cell Stem Cell 2015, 16, 142.

X. He, C. Tan, F. Wang, Y. Wang, R. Zhou, D. Cui, W. You, H. Zhao, J. Ren, B. Feng, Nucleic Acids Res. 2016, 44, e85.

K. Suzuki, Y. Tsunekawa, R. Hernandez‐Benitez, J. Wu, J. Zhu, E. J. Kim, F. Hatanaka, M. Yamamoto, T. Araoka, Z. Li, M. Kurita, T. Hishida, M. Li, E. Aizawa, S. Guo, S. Chen, A. Goebl, R. D. Soligalla, J. Qu, T. Jiang, X. Fu, M. Jafari, C. R. Esteban, W. T. Berggern, J. Lajara, E. Nunez‐Delicado, P. Guillen, J. M. Campistol, F. Matsuzaki, G.‐H. Liu, P. Magistretti, K. Zhang, E. M. Callaway, K. Zhang, J. C. I. Belmonte, Nature 2016, 540, 144.

H.‐X. Wang, M. Li, C. M. Lee, S. Chakraborty, H.‐W. Kim, G. Bao, K. W. Leong, Chem. Rev. 2017, 117, [eLocator: 9874].

Z. Glass, M. Lee, Y. Li, Q. Xu, Trends Biotechnol. 2018, 36, 173.

S. Tong, B. Moyo, C. M. Lee, K. Leong, G. Bao, Nat. Rev. Mater. 2019, 4, 726.

H. Rahimi, M. Salehiabar, J. Charmi, M. Barsbay, M. Ghaffarlou, M. R. Razligh, S. Davaran, R. Khalilov, M. Sugiyama, H. Nosrati, S. Kaboli, H. Danafar, T. J. Webster, Nano Today 2020, 34, [eLocator: 100895].

C. Liu, L. Zhang, H. Liu, K. Cheng, J. Controlled Release 2017, 266, 17.

S. Zhang, J. Shen, D. Li, Y. Cheng, Theranostics 2021, 11, 614.

B. H. Yip, Biomolecules 2020, 10, 839.

T. Fang, X. Cao, M. Ibnat, G. Chen, J. Nanobiotechnol. 2022, 20, 354.

R. Mout, M. Ray, Y.‐W. Lee, F. Scaletti, V. M. Rotello, Bioconjugate Chem. 2017, 28, 880.

C. T. Charlesworth, I. Hsu, A. C. Wilkinson, H. Nakauchi, Nat. Rev. Immunol. 2022, 22, 719.

C. Zhuo, J. Zhang, J.‐H. Lee, J. Jiao, D. Cheng, L. Liu, H.‐W. Kim, Y. Tao, M. Li, Signal Transduction Targeted Ther. 2021, 6, 238.

D. R. Jacobson, O. A. Saleh, Nucleic Acids Res. 2017, 45, [eLocator: 1596].

H. Tang, X. Zhao, X. Jiang, Adv. Drug Delivery Rev. 2021, 168, 55.

A. P. Chandrasekaran, M. Song, K.‐S. Kim, S. Ramakrishna, Prog. Mol. Biol. Transl. Sci. 2018, 159, 157.

D. Wang, P. W. L. Tai, G. Gao, Nat. Rev. Drug Discovery 2019, 18, 358.

T. Wang, J. J. Wei, D. M. Sabatini, E. S. Lander, Science 2014, 343, 80.

F. A. Ran, L. Cong, W. X. Yan, D. A. Scott, J. S. Gootenberg, A. J. Kriz, B. Zetsche, O. Shalem, X. Wu, K. S. Makarova, E. V. Koonin, P. A. Sharp, F. Zhang, Nature 2015, 520, 186.

H. Yin, R. L. Kanasty, A. A. Eltoukhy, A. J. Vegas, J. R. Dorkin, D. G. Anderson, Nat. Rev. Genet. 2014, 15, 541.

M. Naeem, S. Majeed, M. Z. Hoque, I. Ahmad, Cells 2020, 9, [eLocator: 1608].

J. M. Crudele, J. S. Chamberlain, Nat. Commun. 2018, 9, [eLocator: 3497].

C.‐F. Xu, G.‐J. Chen, Y.‐L. Luo, Y. Zhang, G. Zhao, Z.‐D. Lu, A. Czarna, Z. Gu, J. Wang, Adv. Drug Delivery Rev. 2021, 168, 3.

V. Picanco‐Castro, C. G. Pereira, D. T. Covas, G. S. Porto, A. Athanassiadou, M. L. Figueiredo, Nat. Biotechnol. 2020, 38, 151.

M. Foldvari, D. W. Chen, N. Nafissi, D. Calderon, L. Narsineni, A. Rafiee, J. Controlled Release 2016, 240, 165.

C. E. Nelson, C. A. Gersbach, Annu. Rev. Chem. Biomol. Eng. 2016, 7, 637.

L. Li, S. Hu, X. Chen, Biomaterials 2018, 171, 207.

M. J. Mitchell, M. M. Billingsley, R. M. Haley, M. E. Wechsler, N. A. Peppas, R. Langer, Nat. Rev. Drug Discovery 2021, 20, 101.

J. Yan, D. D. Kang, Y. Dong, Biomater. Sci. 2021, 9, [eLocator: 6001].

A. Tasset, A. Bellamkonda, W. Wang, I. Pyatnitskiy, D. Ward, N. Peppas, H. Wang, Nanoscale 2022, 14, [eLocator: 3698].

M. Marangi, G. Pistritto, Front. Pharmacol. 2018, 9, 396.

Q. Hu, P. S. Katti, Z. Gu, Nanoscale 2014, 6, [eLocator: 12273].

W. R. Wilson, M. P. Hay, Nat. Rev. Cancer 2011, 11, 393.

D. Schmaljohann, Adv. Drug Delivery Rev. 2006, 58, [eLocator: 1655].

Y. Lu, A. A. Aimetti, R. Langer, Z. Gu, Nat. Rev. Mater. 2017, 2, [eLocator: 16075].

M. Huo, J. Yuan, L. Tao, Y. Wei, Polym. Chem. 2014, 5, [eLocator: 5183].

W. Cai, T. Luo, L. Mao, M. Wang, Angew. Chem., Int. Ed. 2021, 60, [eLocator: 8596].

Z. Chen, J. Zhu, Z. Chen, M. Du, R. Yao, W. Fu, A. Lei, J. Wang, Plant Biotechnol. J. 2022, 20, [eLocator: 2048].

T. Wan, J. Zhong, Q. Pan, T. Zhou, Y. Ping, X. Liu, Sci. Adv. 2022, 8, [eLocator: 9435].

B. He, W. Peng, J. Huang, H. Zhang, Y. Zhou, X. Yang, J. Liu, Z. Li, C. Xu, M. Xue, H. Yang, P. Huang, Protein Cell 2020, 11, 518.

D. S. D'Astolfo, R. J. Pagliero, A. Pras, W. R. Karthaus, H. Clevers, V. Prasad, R. J. Lebbink, H. Rehmann, N. Geijsen, Cell 2015, 161, 674.

X. Han, Z. Liu, M. C. Jo, K. Zhang, Y. Li, Z. Zeng, N. Li, Y. Zu, L. Qin, Sci. Adv. 2015, 1, [eLocator: 1500454].

Q. Yan, Q. Zhang, H. Yang, Q. Zou, C. Tang, N. Fan, L. Lai, Cell Regener. 2014, 3, 12.

M. Vilarino, S. T. Rashid, F. P. Suchy, B. R. McNabb, T. van der Meulen, E. J. Fine, S. Ahsan, N. Mursaliyev, V. Sebastiano, S. S. Diab, M. O. Huising, H. Nakauchi, P. J. Ross, Sci. Rep. 2017, 7, [eLocator: 17472].

Z. Liu, S. Chen, W. Xie, Y. Song, J. Li, L. Lai, Z. Li, Mol. Ther. 2022, 30, 256.

T. Nomura, M. Yoshikawa, K. Suzuki, K. Mochida, STAR Protoc. 2020, 1, [eLocator: 100023].

P. E. Boukany, A. Morss, W.‐C. Liao, B. Henslee, H. Jung, X. Zhang, B. Yu, X. Wang, Y. Wu, L. Li, K. Gao, X. Hu, X. Zhao, O. Hemminger, W. Lu, G. P. Lafyatis, L. J. Lee, Nat. Nanotechnol. 2011, 6, 747.

M. C. Latella, M. T. Di Salvo, F. Cocchiarella, D. Benati, G. Grisendi, A. Comitato, V. Marigo, A. Recchia, Mol. Ther. Nucleic Acids 2016, 5, [eLocator: 389].

F. Xie, L. Ye, J. C. Chang, A. I. Beyer, J. Wang, M. O. Muench, Y. W. Kan, Genome Res. 2014, 24, [eLocator: 1526].

J. N. Søndergaard, K. Geng, C. Sommerauer, I. Atanasoai, X. Yin, C. Kutter, Commun. Biol. 2020, 3, 319.

T. Suda, D. Liu, Mol. Ther. 2007, 15, [eLocator: 2063].

D. K.‐C. Chiu, V. W.‐H. Yuen, J. W.‐S. Cheu, L. L. Wei, V. Ting, M. Fehlings, H. Sumatoh, A. Nardin, E. W. Newell, I. O.‐L. Ng, T. C.‐C. Yau, C.‐M. Wong, C. C.‐L. Wong, Gastroenterology 2020, 159, 609.

H. Yin, W. Xue, S. Chen, R. L. Bogorad, E. Benedetti, M. Grompe, V. Koteliansky, P. A. Sharp, T. Jacks, D. G. Anderson, Nat. Biotechnol. 2014, 32, 551.

S. Lin, B. Staahl, R. K. Alla, J. A. Doudna, eLife 2014, 3, [eLocator: e04766].

J. L. Gori, P. D. Hsu, M. L. Maeder, S. Shen, G. G. Welstead, D. Bumcrot, Hum. Gene Ther. 2015, 26, 443.

B. Bonamassa, L. Hai, D. Liu, Pharm. Res. 2011, 28, 694.

W. M. Kholosy, M. Visscher, K. Ogink, H. Buttstedt, K. Griffin, A. Beier, J. P. Gerlach, J. J. Molenaar, N. Geijsen, M. de Boer, A. Chatsisvili, J. Biotechnol. 2021, 338, 71.

H. Bloomer, J. Khirallah, Y. Li, Q. Xu, Adv. Drug Deliv Rev. 2022, 181, [eLocator: 114087].

Y. Zhang, J. Yu, H. N. Bomba, Y. Zhu, Z. Gu, Chem. Rev. 2016, 116, [eLocator: 12536].

X. Han, Z. Liu, Y. Ma, K. Zhang, L. Qin, Adv. Biosyst. 2017, 1, [eLocator: 1600007].

Y. Ma, X. Han, O. Quintana Bustamante, R. Bessa de Castro, K. Zhang, P. Zhang, Y. Li, Z. Liu, X. Liu, M. Ferrari, Z. Hu, J. C. Segovia, L. Qin, Integr. Biol. 2017, 9, 548.

L. R. Krilov, Pediatr. Infect. Dis. J. 2005, 24, 555.

G. Wadell, M. L. Hammarskjold, G. Winberg, T. M. Varsanyi, G. Sundell, Ann. N. Y. Acad. Sci. 1980, 354, 16.

E. Ehrke‐Schulz, M. Schiwon, T. Leitner, S. David, T. Bergmann, J. Liu, A. Ehrhardt, Sci. Rep. 2017, 7, [eLocator: 17113].

E. Dormond, M. Perrier, A. Kamen, Biotechnol. Adv. 2009, 27, 133.

M. Garcia, J. Bonafont, J. Martinez‐Palacios, R. Xu, G. Turchiano, S. Svensson, A. J. Thrasher, F. Larcher, M. Del Rio, R. Hernandez‐Alcoceba, M. I. Garin, A. Mencia, R. Murillas, Mol. Ther.‐Methods Clin. Dev. 2022, 27, 96.

C. J. Stephens, E. J. Lauron, E. Kashentseva, Z. H. Lu, W. M. Yokoyama, D. T. Curiel, J. Controlled Release 2019, 298, 128.

Q. Xie, W. Bu, S. Bhatia, J. Hare, T. Somasundaram, A. Azzi, M. S. Chapman, Proc. Natl. Acad. Sci. 2002, 99, [eLocator: 10405].

P. J. Carter, R. J. Samulski, Int. J. Mol. Med. 2000, 6, 17.

Y. Yoon, D. Wang, P. W. L. Tai, J. Riley, G. Gao, J. A. Rivera‐Perez, Nat. Commun. 2018, 9, 412.

Y. Yao, J. Wang, Y. Liu, Y. Qu, K. Wang, Y. Zhang, Y. Chang, Z. Yang, J. Wan, J. Liu, H. Nakashima, S. E. Lawler, E. A. Chiocca, C.‐F. Cho, F. Bei, Nat. Biomed. Eng. 2022, 6, [eLocator: 1257].

A. D. Nahmad, C. R. Lazzarotto, N. Zelikson, T. Kustin, M. Tenuta, D. Huang, I. Reuveni, D. Nataf, Y. Raviv, M. Horovitz‐Fried, I. Dotan, Y. Carmi, R. Rosin‐Arbesfeld, D. Nemazee, J. E. Voss, A. Stern, S. Q. Tsai, A. Barzel, Nat. Biotechnol. 2022, 40, [eLocator: 1241].

V. M. Vogt, M. N. Simon, J. Virol. 1999, 73, [eLocator: 7050].

M. A. Kotterman, T. W. Chalberg, D. V. Schaffer, Annu. Rev. Biomed. Eng. 2015, 17, 63.

J. L. Shirley, Y. P. de Jong, C. Terhorst, R. W. Herzog, Mol. Ther. 2020, 28, 709.

J. Dubrot, S. K. Lane‐Reticker, E. A. Kessler, A. Ayer, G. Mishra, C. H. Wolfe, M. D. Zimmer, P. P. Du, A. Mahapatra, K. M. Ockerman, T. G. R. Davis, I. C. Kohnle, H. W. Pope, P. M. Allen, K. E. Olander, A. Iracheta‐Vellve, J. G. Doench, W. N. Haining, K. B. Yates, R. T. Manguso, Immunity 2021, 54, 571.

D. Heck, M. S. Kowalczyk, D. Yudovich, R. Belizaire, R. V. Puram, M. E. McConkey, A. Thielke, J. C. Aster, A. Regev, B. L. Ebert, Nat. Biotechnol. 2014, 32, 941.

S. Lee, Y.‐Y. Kim, H. J. Ahn, Biomaterials 2021, 272, [eLocator: 120793].

H. Ma, Y. Dang, Y. Wu, G. Jia, E. Anaya, J. Zhang, S. Abraham, J.‐G. Choi, G. Shi, L. Qi, N. Manjunath, H. Wu, Cell Rep. 2015, 12, 673.

L.‐Z. Hu, W. Wang, J.‐L. Xu, Y.‐Y. Jia, M.‐L. Huan, C. Li, S.‐Y. Zhou, B.‐L. Zhang, Nanotechnology 2020, 31, 32.

Z. Zhang, T. Wan, Y. Chen, Y. Chen, H. Sun, T. Cao, S. Zhou, G. Tang, C. Wu, Y. Ping, F.‐J. Xu, J. Huang, Macromol. Rapid Commun. 2018, 40, [eLocator: 1800068].

L. Zhao, Y. Luo, Q. Huang, Z. Cao, X. Yang, Small 2020, 16, [eLocator: 2004879].

C. Liu, T. Wan, H. Wang, S. Zhang, Y. Ping, Y. Cheng, Sci. Adv. 2019, 5, [eLocator: 8922].

T. Taharabaru, R. Yokoyama, T. Higashi, A. F. A. Mohammed, M. Inoue, Y. Maeda, T. Niidome, R. Onodera, K. Motoyama, ACS Appl. Mater. Interfaces 2020, 12, [eLocator: 21386].

S. S. Rohiwal, N. Dvorakova, J. Klima, M. Vaskovicova, F. Senigl, M. Slouf, E. Pavlova, P. Stepanek, D. Babuka, H. Benes, Z. Ellederova, K. Stieger, Sci. Rep. 2020, 10, [eLocator: 4619].

J. Yang, Q. Zhang, H. Chang, Y. Cheng, Chem. Rev. 2015, 115, [eLocator: 5274].

T.‐F. Li, Y.‐Y. Cheng, Y. Wang, H. Wang, D.‐F. Chen, Y.‐T. Liu, L. Zhang, W.‐Z. Han, R.‐D. Liu, Z.‐J. Wang, C.‐M. Yang, C. J. Jafta, D. Clemens, U. Keiderling, Chin. J. Polym. Sci. 2019, 37, 827.

L. Zhao, Q. Wu, Y. Cheng, J. Zhang, J. Wu, T. Xu, J. Am. Chem. Soc. 2010, 132, [eLocator: 13182].

H. Wang, Y. Wang, Y. Wang, J. Hu, T. Li, H. Liu, Q. Zhang, Y. Cheng, Angew. Chem., Int. Ed. 2015, 54, [eLocator: 11647].

L. Ren, J. Lv, H. Wang, Y. Cheng, Angew. Chem., Int. Ed. 2020, 59, [eLocator: 4711].

H. Zhang, T. F. Bahamondez‐Canas, Y. Zhang, J. Leal, H. D. C. Smyth, Mol. Pharm. 2018, 15, [eLocator: 4814].

J. Qiao, W. Sun, S. Lin, R. Jin, L. Ma, Y. Liu, Chem. Commun. 2019, 55, [eLocator: 4707].

P. L. Felgner, M. Holm, H. Chan, Proc. West. Pharmacol. Soc. 1989, 32, 115.

X. Zhou, L. Huang, Biochim. Biophys. Acta, Biomembr. 1994, 1189, 195.

L. Zhang, P. Wang, Q. Feng, N. Wang, Z. Chen, Y. Huang, W. Zheng, X. Jiang, NPG Asia Mater. 2017, 9, e441.

P. Guo, J. Yang, J. Huang, D. T. Auguste, M. A. Moses, Proc. Natl. Acad. Sci. 2019, 116, [eLocator: 18295].

J. B. Miller, S. Zhang, P. Kos, H. Xiong, K. Zhou, S. S. Perelman, H. Zhu, D. J. Siegwart, Angew. Chem., Int. Ed. 2017, 56, [eLocator: 1059].

E. Y. Cho, J.‐Y. Ryu, H. A. R. Lee, S. H. Hong, H. S. Park, K. S. Hong, S.‐G. Park, H. P. Kim, T.‐J. Yoon, J. Nanobiotechnol. 2019, 17, 19.

Q. Cheng, T. Wei, L. Farbiak, L. T. Johnson, S. A. Dilliard, D. J. Siegwart, Nat. Nanotechnol. 2020, 15, 313.

J. A. Zuris, D. B. Thompson, Y. Shu, J. P. Guilinger, J. L. Bessen, J. H. Hu, M. L. Maeder, J. K. Joung, Z.‐Y. Chen, D. R. Liu, Nat. Biotechnol. 2015, 33, 73.

X. Yu, X. Liang, H. Xie, S. Kumar, N. Ravinder, J. Potter, X. d. M. du Jeu, J. D. Chesnut, Biotechnol. Lett. 2016, 38, 919.

E. Haapaniemi, S. Botla, J. Persson, B. Schmierer, J. Taipale, Nat. Med. 2018, 24, 927.

M. Wang, J. A. Zuris, F. Meng, H. Rees, S. Sun, P. Deng, Y. Han, X. Gao, D. Pouli, Q. Wu, I. Georgakoudi, D. R. Liu, Q. Xu, Proc. Natl. Acad. Sci. 2016, 113, [eLocator: 2868].

Y. Li, T. Yang, Y. Yu, N. Shi, L. Yang, Z. Glass, J. Bolinger, I. J. Finkel, W. Li, Q. Xu, Biomaterials 2018, 178, 652.

V. Linko, A. Ora, M. A. Kostiainen, Trends Biotechnol. 2015, 33, 586.

Y. Ke, C. Castro, J. H. Choi, Annu. Rev. Biomed. Eng. 2018, 20, 375.

W. Sun, W. Ji, J. M. Hall, Q. Hu, C. Wang, C. L. Beisel, Z. Gu, Angew. Chem., Int. Ed. 2015, 127, [eLocator: 12197].

W. Sun, J. Wang, Q. Hu, X. Zhou, A. Khademhosseini, Z. Gu, Sci. Adv. 2020, 6, [eLocator: 2983].

E. Lin‐Shiao, W. G. Pfeifer, B. R. Shy, M. Saffari Doost, E. Chen, V. S. Vykunta, J. R. Hamilton, E. C. Stahl, D. M. Lopez, C. R. Sandoval Espinoza, A. E. Deyanov, R. J. Lew, M. G. Poirer, A. Marson, C. E. Castro, J. A. Doudna, Nucleic Acids Res. 2022, 50, [eLocator: 1256].

H.‐C. Zhou, J. R. Long, O. M. Yaghi, Chem. Rev. 2012, 112, 673.

H. Hayashi, A. P. Cote, H. Furukawa, M. O'Keeffe, O. M. Yaghi, Nat. Mater. 2007, 6, 501.

S. K. Alsaiari, S. Patil, M. Alyami, K. O. Alamoudi, F. A. Aleisa, J. S. Merzaban, M. Li, N. M. Khashab, J. Am. Chem. Soc. 2018, 140, 143.

M. Z. Alyami, S. K. Alsaiari, Y. Li, S. S. Qutub, F. A. Aleisa, R. Sougrat, J. S. Merzaban, N. M. Khashab, J. Am. Chem. Soc. 2020, 142, [eLocator: 1715].

X. Xu, O. Koivisto, C. Liu, J. Zhou, M. Miihkinen, G. Jacquemet, D. Wang, J. M. Rosenholm, Y. Shu, H. Zhang, Adv. Ther. 2020, 4, [eLocator: 2000072].

W. Zhou, H. Cui, L. Ying, X.‐F. Yu, Angew. Chem., Int. Ed. 2018, 57, [eLocator: 10268].

M. Hansen‐Bruhn, B. E.‐F. De Avila, M. Beltran‐Gastelum, J. Zhao, D. E. Ramirez‐Herrera, P. Angsantikul, K. V. Gothelf, L. Zhang, J. Wang, Angew. Chem., Int. Ed. 2018, 57, [eLocator: 2657].

Y. Wang, P. K. Shahi, R. Xie, H. Zhang, A. A. Abdeen, N. Yodsanit, Z. Ma, K. Saha, B. R. Pattnaik, S. Gong, J. Controlled Release 2020, 324, 194.

X. Yang, Q. Tang, Y. Jiang, M. Zhang, M. Wang, L. Mao, J. Am. Chem. Soc. 2019, 141, [eLocator: 3782].

Y. Wang, Q. Zhao, N. Han, L. Bai, J. Li, J. Liu, E. Che, L. Hu, Q. Zhang, T. Jiang, S. Wang, Nanomed. Nanotechnol. Biol. Med. 2015, 11, 313.

J. Pang, Y. Luan, X. Yang, Y. Jiang, L. Zhao, Y. Zong, Z. Li, Mini‐Rev. Med. Chem. 2012, 12, 775.

X. She, L. Chen, Z. Yi, C. Li, C. He, C. Feng, T. Wang, S. Shigdar, W. Duan, L. Kong, Mini‐Rev. Med. Chem. 2018, 18, 976.

K. Zivojevic, M. Mladenovic, M. Djisalov, M. Mundzic, E. Ruiz‐Hernandez, I. Gadjanski, N. Z. Knezevic, J. Controlled Release 2021, 337, 193.

A. Noureddine, A. Maestas‐Olguin, E. A. Saada, A. E. LaBauve, J. O. Agola, K. E. Baty, T. Howard, J. K. Sabo, C. R. S. Espinoza, J. A. Doudna, J. S. Schoeniger, K. S. Butler, O. A. Negrete, C. J. Brinker, R. E. Serda, Acta Biomater. 2020, 114, 764.

L. Qin, S. Jiang, H. He, G. Ling, P. Zhang, J. Controlled Release 2020, 318, 50.

W. Liu, A. Dong, B. Wang, H. Zhang, Adv. Sci. 2021, 8, [eLocator: 2003033].

H. Yue, R. Huang, Y. Shan, D. Xing, J. Mater. Chem. B 2020, 8, [eLocator: 11096].

N. Gupta, R. Malviya, Biochim. Biophys. Acta, Rev. Cancer 2021, 1875, [eLocator: 188532].

E. Ju, T. Li, S. R. da Silva, S.‐J. Gao, ACS Appl. Mater. Interfaces 2019, 11, [eLocator: 34717].

R. Mout, M. Ray, G. Y. Tonga, Y.‐W. Lee, T. Tay, K. Sasaki, V. M. Rotello, ACS Nano 2017, 11, [eLocator: 2452].

R. Mout, M. Ray, T. Tay, K. Sasaki, G. Y. Tonga, V. M. Rotello, ACS Nano 2017, 11, [eLocator: 6416].

Y. Jiang, J. Hardie, Y. Liu, M. Ray, X. Luo, R. Das, R. F. Landis, M. E. Farkas, V. M. Rotello, J. Controlled Release 2018, 283, 235.

K. Lee, M. Conboy, H. M. Park, F. Jiang, H. J. Kim, M. A. Dewitt, V. A. Mackley, K. Chang, A. Rao, C. Skinner, T. Shobha, M. Mehdipour, H. Liu, W.‐C. Huang, F. Lan, N. L. Bray, S. Li, J. E. Corn, K. Kataoka, J. A. Doudna, I. Conboy, N. Murthy, Nat. Biomed. Eng. 2017, 1, 889.

B. Lee, K. Lee, S. Panda, R. Gonzales‐Rojas, A. Chong, V. Bugay, H. M. Park, R. Brenner, N. Murthy, H. Y. Lee, Nat. Biomed. Eng. 2018, 2, 497.

R. Shahbazi, G. Sghia‐Hughes, J. L. Reid, S. Kubek, K. G. Haworth, O. Humbert, H.‐P. Kiem, J. E. Adair, Nat. Mater. 2019, 18, [eLocator: 1124].

T. J. Levingstone, S. Herbaj, N. J. Dunne, Nanomaterials 2019, 9, [eLocator: 1570].

B. Y. Liu, X.‐Y. He, C. Xu, L. Xu, S.‐L. Ai, S.‐X. Cheng, R.‐X. Zhuo, Biomacromolecules 2018, 19, [eLocator: 2957].

S. Li, Z. Song, C. Liu, X. L. Chen, H. Han, ACS Appl. Mater. Interfaces 2019, 11, [eLocator: 47762].

M. Li, H. Xie, Y. Liu, C. Xia, X. Cun, Y. Long, X. Chen, M. Deng, R. Guo, Z. Zhang, Q. He, J. Controlled Release 2019, 304, 204.

S. Ramakrishna, A.‐B. K. Dad, J. Beloor, R. Gopalappa, S.‐K. Lee, H. Kim, Genome Res. 2014, 24, [eLocator: 1020].

J. Yin, Q. Wang, S. Hou, L. Bao, W. Yao, X. Gao, J. Am. Chem. Soc. 2018, 140, [eLocator: 17234].

J. Yin, S. Hou, Q. Wang, L. Bao, D. Liu, Y. Yue, W. Yao, X. Gao, Bioconjugate Chem. 2019, 30, 898.

Y. Shen, J. L. Cohen, S. M. Nicoloro, M. Kelly, B. Yenilmez, F. Henriques, E. Tsagkaraki, Y. J. K. Edwards, X. Hu, R. H. Friedline, J. K. Kim, M. P. Czech, J. Biol. Chem. 2018, 293, [eLocator: 17291].

S. Ganta, H. Devalapally, A. Shahiwala, M. Amiji, J. Controlled Release 2008, 126, 187.

N. Deirram, C. Zhang, S. S. Kermaniyan, A. P. R. Johnston, G. K. Such, Macromol. Rapid Commun. 2019, 40, [eLocator: 1800917].

K. Tu, H. Deng, L. Kong, Y. Wang, T. Yang, Q. Hu, M. Hu, C. Yang, Z. Zhang, ACS Appl. Mater. Interfaces 2020, 12, [eLocator: 16018].

Q. Liu, K. Zhao, C. Wang, Z. Zhang, C. Zheng, Y. Zhao, Y. Zheng, C. Liu, Y. An, L. Shi, C. Kang, Y. Liu, Adv. Sci. 2019, 6, [eLocator: 1801423].

Q. Liu, J. Cai, Y. Zheng, Y. Tan, Y. Wang, Z. Zhang, C. Zheng, Y. Zhao, C. Liu, Y. An, C. Jiang, L. Shi, C. Kang, Y. Liu, Nano Lett. 2019, 19, [eLocator: 7662].

J. Liu, J. Chang, Y. Jiang, X. Meng, T. Sun, L. Mao, Q. Xu, M. Wang, Adv. Mater. 2019, 31, [eLocator: 1902575].

Q. Liu, C. Wang, Y. Zheng, Y. Zhao, Y. Wang, J. Hao, X. Zhao, K. Yi, L. Shi, C. Kang, Y. Liu, Biomaterials 2020, 258, [eLocator: 120275].

W. Sun, W. Ji, J. M. Hall, Q. Hu, C. Wang, C. L. Beisel, Z. Gu, Angew. Chem., Int. Ed. 2015, 54, [eLocator: 12029].

H. Yue, X. Zhou, M. Cheng, D. Xing, Nanoscale 2018, 10, [eLocator: 1063].

R. Xie, X. Wang, Y. Wang, M. Ye, Y. Zhao, B. S. Yandell, S. Gong, Adv. Mater. 2022, 34, [eLocator: 2110618].

S. Zhen, Y. Liu, J. Lu, X. Tuo, X. Yang, H. Chen, W. Chen, X. Li, Hum. Gene Ther. 2020, 31, 309.

F. Li, N. Song, Y. Dong, S. Li, L. Li, Y. Liu, Z. Li, D. Yang, Angew. Chem., Int. Ed. 2022, 61, [eLocator: 2116569].

G. Chen, A. A. Abdeen, Y. Wang, P. K. Shahi, S. Robertson, R. Xie, M. Suzuki, B. R. Pattnaik, K. Saha, S. Gong, Nat. Nanotechnol. 2019, 14, 974.

Y. Wang, P. K. Shahi, X. Wang, R. Xie, Y. Zhao, M. Wu, S. Roge, B. R. Pattnaik, S. Gong, J. Controlled Release 2021, 336, 296.

G. Chen, B. Ma, Y. Wang, S. Gong, ACS Appl. Mater. Interfaces 2018, 10, [eLocator: 18515].

Y. Wang, B. Ma, A. A. Abdeen, G. Chen, R. Xie, K. Saha, S. Gong, ACS Appl. Mater. Interfaces 2018, 10, [eLocator: 31915].

Y. Zou, X. Sun, Q. Yang, M. Zheng, O. Shimoni, W. Ruan, Y. Wang, D. Zhang, J. Yin, X. Huang, W. Tao, J. B. Park, X.‐J. Liang, K. W. Leong, B. Shi, Sci. Adv. 2022, 8, [eLocator: abm8011].

H. Sies, Redox Biol. 2017, 11, 613.

H. Chen, J. Tian, W. He, Z. Guo, J. Am. Chem. Soc. 2015, 137, [eLocator: 1539].

X. Yan, Q. Pan, H. Xin, Y. Chen, Y. Ping, Sci. Adv. 2021, 7, abj0624.

L. Zhao, D. Li, Y. Zhang, Q. Huang, Z. Zhang, C. Chen, C.‐F. Xu, X. Chu, Y. Zhang, X. Yang, ACS Nano 2022, 16, [eLocator: 13821].

R. De La Rica, D. Aili, M. M. Stevens, Adv. Drug Delivery Rev. 2012, 64, 967.

C. E. Callmann, C. V. Barback, M. P. Thompson, D. J. Hall, R. F. Mattrey, N. C. Gianneschi, Adv. Mater. 2015, 27, [eLocator: 4611].

Q. Yao, L. Kou, Y. Tu, L. Zhu, Trends Pharmacol. Sci. 2018, 39, 766.

K. Zhang, M. Yang, H. Sun, X. Wang, H. Li, J. Han, Z. Zhang, Sens. Actuators, B 2023, 376, [eLocator: 133023].

J. Yang, Z. Li, M. Shen, Y. Wang, L. Wang, J. Li, W. Yang, J. Li, H. Li, X. Wang, Q. Wu, C. Gong, Adv. Sci. 2021, 8, [eLocator: 2100292].

W. Cai, J. Liu, X. Chen, L. Mao, M. Wang, J. Am. Chem. Soc. 2022, 144, [eLocator: 22272].

L. Li, L. Song, X. Liu, X. Yang, X. Li, T. He, N. Wang, S. Yang, C. Yu, T. Yin, Y. Wen, Z. He, X. Wei, W. Su, Q. Wu, S. Yao, C. Gong, Y. Wei, ACS Nano 2017, 11, 95.

Y. Qian, X. Wang, Y. Liu, Y. Li, R. A. Colvin, L. Tong, S. Wu, X. Chen, Cancer Lett. 2014, 351, 242.

L. E. Low, J. Wu, J. Lee, B. T. Tey, B.‐H. Goh, J. Gao, F. Li, D. Ling, J. Controlled Release 2020, 324, 69.

H. Mollaei, R. Safaralizadeh, Z. Rostami, J. Cell. Physiol. 2019, 234, [eLocator: 12369].

A. Esquela‐Kerscher, F. J. Slack, Nat. Rev. Cancer 2006, 6, 259.

M. Hirosawa, Y. Fujita, C. J. C. Parr, K. Hayashi, S. Kashida, A. Hotta, K. Woltjen, H. Saito, Nucleic Acids Res. 2017, 45, [eLocator: e118].

M. Hirosawa, Y. Fujita, H. Saito, ACS Synth. Biol. 2019, 8, [eLocator: 1575].

X.‐W. Wang, L.‐F. Hu, J. Hao, L.‐Q. Liao, Y.‐T. Chiu, M. Shi, Y. Wang, Nat. Cell Biol. 2019, 21, 522.

V. Petrova, M. Annicchiarico‐Petruzzelli, G. Melino, I. Amelio, Oncogenesis 2018, 7, 10.

X. Li, Y. Pan, C. Chen, Y. Gao, X. Liu, K. Yang, X. Luan, D. Zhou, F. Zeng, X. Han, Y. Song, Angew. Chem., Int. Ed. 2021, 60, [eLocator: 21200].

A. Davis, K. V. Morris, G. Shevchenko, Mol. Ther.‐Methods Clin. Dev. 2022, 25, 158.

Y. Chen, X. Yan, Y. Ping, ACS Mater. Lett. 2020, 2, 644.

E. Ju, K. Dong, Z. Liu, F. Pu, J. Ren, X. Qu, Adv. Funct. Mater. 2015, 25, [eLocator: 1574].

J. F. Suyver, A. Aebischer, D. Biner, P. Gerner, J. Grimm, S. Heer, K. W. Kramer, C. Reinhard, H. U. Gudel, Opt. Mater. 2005, 27, [eLocator: 1111].

P. Wang, L. Zhang, W. Zheng, L. Cong, Z. Guo, Y. Xie, L. Wang, R. Tang, Q. Feng, Y. Hamada, K. Gonda, Z. Hu, X. Wu, X. Jiang, Angew. Chem., Int. Ed. 2018, 57, [eLocator: 1491].

L. Li, Z. Yang, S. Zhu, L. He, W. Fan, W. Tang, J. Zou, Z. Shen, M. Zhang, L. Tang, Y. Dai, G. Niu, S. Hu, X. Chen, Adv. Mater. 2019, 31, [eLocator: 1901187].

S. Deng, X. Li, S. Liu, J. Chen, M. Li, S. Y. Chew, K. W. Leong, D. Cheng, Sci. Adv. 2020, 6, abb4005.

Y. Pan, J. Yang, X. Luan, X. Liu, X. Li, J. Yang, T. Huang, L. Sun, Y. Wang, Y. Lin, Y. Song, Sci. Adv. 2019, 5, aav7199.

H. Yin, B. Zhou, C. Dong, Y. Zhang, J. Yu, Y. Pu, W. Feng, L. Sun, H. Hu, Y. Chen, H. Xu, Adv. Funct. Mater. 2021, 31, [eLocator: 2107093].

L. Shi, W. Wu, Y. Duan, L. Xu, Y. Xu, L. Hou, X. Meng, X. Zhu, B. Liu, Angew. Chem., Int. Ed. 2020, 59, [eLocator: 19168].

E. R. Ruskowitz, C. A. DeForest, Nat. Rev. Mater. 2018, 3, [eLocator: 17087].

Y. Wu, J. Zheng, Q. Zeng, T. Zhang, D. Xing, Nano Res. 2020, 13, [eLocator: 2399].

H. Zhu, L. Zhang, S. Tong, C. M. Lee, H. Deshmukh, G. Bao, Nat. Biomed. Eng. 2019, 3, 126.

A. Kaushik, R. Nikkhah‐Moshaie, R. Sinha, V. Bhardwaj, V. Atluri, R. D. Jayant, A. Yndart, B. Kateb, N. Pala, M. Nair, Sci. Rep. 2017, 7, [eLocator: 45663].

H. Tsuru, H. Shibaguchi, M. Kuroki, Y. Yamashita, M. Kuroki, Free Radical Biol. Med. 2012, 53, 464.

H. Yin, L. Sun, Y. Pu, J. Yu, W. Feng, C. Dong, B. Zhou, D. Du, Y. Zhang, Y. Chen, H. Xu, ACS Cent. Sci 2021, 7, [eLocator: 2049].

Y. Pu, H. Yin, C. Dong, H. Xiang, W. Wu, B. Zhou, D. Du, Y. Chen, H. Xu, Adv. Mater. 2021, 33, [eLocator: 2104641].

T. Sun, Y. S. Zhang, B. Pang, D. C. Hyun, M. Yang, Y. Xia, Angew. Chem., Int. Ed. 2014, 53, [eLocator: 12320].

Y. Lyu, J. Zeng, Y. Jiang, X. Zhen, T. Wang, S. Qiu, X. Lou, M. Gao, K. Pu, ACS Nano 2018, 12, [eLocator: 1801].

T. Lei, M. Guan, J. Liu, H.‐C. Lin, R. Pfattner, L. Shaw, A. F. McGuire, T.‐C. Huang, L. Shao, K.‐T. Cheng, J. B. H. Tok, Z. Bao, Proc. Natl. Acad. Sci. 2017, 114, [eLocator: 5107].

T. Repenko, A. Rix, S. Ludwanowski, D. Go, F. Kiessling, W. Lederle, A. J. C. Kuehne, Nat. Commun. 2017, 8, 470.

E. Polo, M. Collado, B. Pelaz, P. Del Pino, ACS Nano 2017, 11, [eLocator: 2397].