Lipid-based nanoparticles in the systemic delivery of ... - Future Medicine

Review For reprint orders, please contact: [email protected]

Lipid-based nanoparticles in the systemic delivery of siRNA RNAi therapeutics are believed to be the future of personalized medicine and have shown promise in early clinical trials. However, many physiological barriers exist in the systemic delivery of siRNAs to the cytoplasm of targeted cells to perform their function. To overcome these barriers, many siRNA delivery systems have been developed. Among these, lipid-based nanoparticles have great potential owing to their biocompatibility and low toxicity in comparison with inorganic nanoparticles and viral systems. This review discusses the hurdles of systemic siRNA delivery and highlights the recent progress made in lipid-based nanoparticles, which are categorized based on their key lipid components, including cationic lipid, lipoprotein, lipidoid, neutral lipid and anionic lipid-based nanoparticles. It is expected that these lipid nanoparticle-based siRNA delivery systems will have an enabling role for personalized cancer medicine, where siRNA delivery will join forces with genetic profiling of individual patients to achieve the best treatment outcome. KEYWORDS: lipid n lipoprotein n liposome n nanoparticle n RNAi n siRNA delivery n systemic delivery barrier

Since the discovery of the RNAi in 1998 by Fire et al. [1], the first validation of specific gene knockdown in mammalian cells [2] and the first clinical trial of siRNA for age-related macular degeneration in 2004 [3], RNAi therapeutics gained the world’s attention and became an attractive and promising technique in personalized treatment of a broad range of diseases, including cancer, liver and immune-related diseases. RNAi is found in the cytoplasm [4]. As illustrated in Figure 1, siRNA, a double strand of RNA, incorporates into the RNA-induced silencing complex (RISC), causing unwinding of its double strand. The sense strand of siRNA is then removed from RISC, and the activated RISC with the antisense strand serves as a template for the binding of complementary mRNA, inducing mRNA degradation. As siRNA only functions when it reaches the cytoplasm of cells that produce the targeted gene, systemic siRNA delivery encounters many barriers from its administration all the way to reaching the target gene to be fully functional.

to be ‘cohesive’ enough to minimize degradation or disintegration. The delivery system must also be effective at minimizing nonspecific opsonization, phagocytosis and immune activation, while also presenting surface properties that promote interaction with the desired cellular targets [5]. In addition, prolonging the blood circulation of siRNA is also important for its effective delivery as naked siRNA is eliminated from blood just 5 min after intravenous injection [4–8].

Barriers in the systemic delivery of siRNA Stability in the blood stream After intravenous injection, the first aim is to keep the siRNA stable in the bloodstream. Naked siRNA is easily degraded by many endogenous enzymes and aggregated by serum proteins in the blood, which requires the siRNA delivery system

Transport across the vascular endothelium The endothelium acts as a semiselective barrier between the vessel lumen and surrounding tissue, controlling the passage of materials into tissues. As listed in Table 1, the normal capillary endothelium can be divided into three types: continuous endothelium, fenestrated capillaries and discontinuous capillaries [9]. The size limitation of normal endothelium structure suggests that siRNA delivery systems should have small sizes (≤150 nm) to readily cross the vascular endothelial barrier. It is also known that the regional structure of blood vessels changes in the inflammation sites and solid tumors, where the enhanced permeation and retention effect on drug delivery has been widely observed. Small nanoparticles (NPs) with diameter of less than 500 nm, usually less than 150 nm, showed significant enhanced permeation and retention effects in ‘leaky’ tumor blood vessels [10].

10.2217/NNM.13.192 © 2014 Future Medicine Ltd

Nanomedicine (2014) 9(1), 105–120

Qiaoya Lin1,2,3, Juan Chen1, Zhihong Zhang3 & Gang Zheng*1,2 Ontario Cancer Institute & Techna Institute, University Health Network, Toronto, ON, Canada 2 Department of Medical Biophysics, University of Toronto, Toronto Medical Discovery Tower 5-363, 101 College Street, Toronto, ON, M5G 1L7, Canada 3 Britton Chance Center for Biomedical Photonics, Wuhan National Laboratory for Optoelectronics, Huazhong University of Science & Technology, Wuhan, China *Author for correspondence: [email protected] 1

part of

ISSN 1743-5889

105

Review

Lin, Chen, Zhang & Zheng

Cytoplasm i

ii RNAi siRNA

a

Endosome

b

RISC

Nucleus

Target mRNA cleavage

Nanomedicine © Future Medicine Ltd (2014)

Figure 1. In vivo siRNA systemic delivery barriers and the mechanism of RNAi. (A) Stability in the blood stream; (B) transport across the vascular endothelial barrier; (C) diffusion through the extracellular matrix; (D) delivery into the cytoplasm by (Di) endosomal escape and (Dii) direct cytosolic delivery. (Dia) The siRNAs or siRNA nanoparticles were trapped in the endosome and (Dib) the siRNAs were released from the endosome into the cytoplasm.

Diffusion through the extracellular matrix After crossing the vascular endothelium, the siRNA delivery vehicle has to diffuse through the extracellular matrix (ECM), which is composed of gels of polysaccharides and fibrous 106

Nanomedicine (2014) 9(1)

proteins, and acts as the structural support for the animal cells. The tight structure of the ECM hinders the diffusion of larger NPs, especially in poorly permeable tumor tissue. For example, Cabral et al. demonstrated that 30-nm micelles could effectively penetrate deep tumor tissue of future science group

Lipid-based nanoparticles in the systemic delivery of siRNA

pancreatic adenocarcinoma while the 70-nm micelles were retained around the vasculature, indicating that only NPs smaller than 50 nm can penetrate poorly permeable tumors [11]. In human glioblastoma (U87) and melanoma (Mu89), it was demonstrated that an increase of the particles’ molecular size decreased their interstitial diffusion [12]. In addition, the transmission electron microscopy results showed the interfibrillar spacing between bundles of aligned and compact fibrils in U87 mouse tumors was 20–42 nm [12]. Altogether, the small size of nanocarriers is superior for effective delivery of siRNA to poorly permeable tissues. In addition, the charge of the NPs could also influence particle diffusion. Stylianopoulos et al. demonstrated that neutral particles diffuse faster than charged particles [13]. Delivery into the cytoplasm After diffusion through the ECM, the siRNA delivery system needs to transport through the cell membrane, reach the cytoplasm and release the siRNA payload. The naked siRNA with a negative charge cannot readily cross the negatively charged cell membrane. The siRNAs ferried by nanocarriers usually enter into cells via endocytosis pathways, such as macropinocytosis, clathrin-mediated endocytosis (CME) and caveolae-/lipid raft-mediated endocytosis [14,15]. Macropinocytosis and CME usually drive the siRNA carrier into the endosome (Figure 1Dia), where mature endosomes fuse easily with lysosomal vesicles, resulting in enzymatic destruction of siRNA. Therefore, escaping from the endosome is further required for effective siRNA delivery (Figure 1D ib ). The caveolae-/lipid raft-mediated endocytosis is a lesser characterized pathway compared with macropincytosis and CME [16,17]. However, it is has been demonstrated that some lipid raft-related delivery could bypass the endosomal route to achieve direct cytosolic delivery of drugs (Figure 1Dii) [18–20], opening a new avenue for enhanced cytosolic siRNA delivery. To overcome the barriers in the systemic delivery of siRNA, variable nonviral siRNA delivery vehicles have been developed, including lipid-based NPs [21–24], polymers [25] or lipid polymer hybrid NPs [26], hydrogels [27], microbubbles [28], inorganic NPs, such as silica [29], gold [30], quantum dot [31], iron oxide [32,33] and carbon nanotubes [34,35], and other materials, such as oligonucleotide NPs [36] and exosomes [37]. Among them, the lipid-based NPs are in the most advanced stage of development and have shown favorable biocompatible and future science group

Review

biodegradable properties in comparison with inorganic carriers and viral vectors. This review will focus on the development of lipid-based NPs for the systemic delivery of siRNAs. The lipid-based NPs reviewed here are divided into the following types, based on their key component lipids: cationic lipid-based NPs; lipoprotein-related NPs; and other lipid NPs, including lipid-like material-, neutral lipid- and anionic lipid-based NPs.

Cationic lipid-based siRNA delivery Cationic liposomes are the most common lipidbased nanocarrier used for siRNA delivery. siRNA loading onto the cationic liposome is mainly attributed to the electrostatic interaction between the anionic siRNA and cationic lipids. The cationic lipids facilitate the particles’ intracellular uptake and further endosomal escape. Some cationic liposomes have been widely used for siRNA delivery, such as 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP)–siRNA, N-(1-[2,3-dioleyloxy]propyl)-N,N,N-trimethylammonium chloride– siRNA and Lipofectamine® 2000 (Life Technologies, MD, USA)–siRNA complexes [38–41], which are simply formulated by cationic lipids and siRNAs. However, these complexes generally have a high electric charge density on their surface that readily induce nonspecific interactions with serum proteins and immunogenic response, resulting in their rapid removal from the blood stream [4,42,43]. Development of advanced cationic lipid-based siRNA delivery system Stable loading of siRNA

To stabilize the cationic lipid-based siRNA delivery system, many neutral lipids, such as cholesterol (Chol) or 1,2-dioleoyl-sn-glycerol3-phosphoethanolamine (DOPE) [44], have been added to the complex formulation. These neutral lipids not only stabilize the formation but also enhance the cellular uptake of particles, which will be discussed later (see the ‘Improving the cellular uptake & enhancing siRNA release into the cytoplasm’ section). Table 1. Normal capillary endothelium. Type

Organs

Size (nm)

Continuous

Brain, skeletal, cardiac and smooth muscles, lung, skin, subcutaneous and mucous membranes

≤1.8–2.0

Fenestrated

Kidney, small intestine and salivary glands

≤11

Discontinuous Liver, spleen and bone marrow

www.futuremedicine.com

Up to 150

107

Review


In comparison with surface loading, siRNA loading in the NP core should provide better delivery stability. To improve the siRNA core loading efficiency, many helper cationic polymers, such as protamine, were introduced to precondense siRNA into the core of a liposome. A good example is the liposome–polycation–DNA (LPD) NPs developed by Li et al. [45]. In this NP, protamine interacts with the nucleic acid to form a negatively charged compact core, cationic liposomes composed of DOTAP/Chol collapse onto the core via charge–charge interaction and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine–PEG is then coated onto the outer surface of the particles. Thus, the siRNA delivery is supported and shielded by both the cationic lipid bilayer and PEG coating [45]. To improve siRNA loading efficacy, many efforts have been undertaken to optimize the key component of liposome cationic lipids. Santel et al. synthesized a series of b-l-arginyl-2, 3-l-diaminopropionic acid-N-palmityl-N-oleylamide trihydrochloride (AtuFECT) cationic lipids and validated that AtuFECT01 enables enhancement of the siRNA-binding ability when compared with commercial cationic lipids, such as DOTAP or N-(1-[2,3-dioleyloxy]propyl)-N,N,N-trimethylammonium chloride [7]. The ability of the AtuFECT-formulated liposome to deliver siRNA (the resultant AtuFECT–liposome–siRNA complex is termed as AtuPLEX) that inhibits CD31 and Tie2 in the vasculature of mice has also been demonstrated [7]. Later, the same group extended the application of the optimized AtuPLEX to treat advanced solid tumor by targeting PKN3 (termed Atu027) and achieved significant inhibition of tumor growth [46], lymph node metastasis [46] and lung metastasis formation [47]. Notably, a Phase I clinical trial of Atu027 for the treatment of advanced solid cancers has been completed, and a further clinical test using a combination of Atu027 and gemcitabine is currently ongoing [201,202]. The stable loading of siRNA depends not only on the formulation components but also on siRNA-integrating methods. Buyens et al. published a comprehensive review on this aspect, which covers variable methods for siRNA loading, such as simple mixing, direct hydration of a lipid film by a siRNA-contained solution and the ethanol dilution method [48]. Among them, the simple mixing method usually gave the poorest siRNA encapsulation efficacy [48]. Furthermore, the stable systemic delivery of siRNA can be improved by chemical 108


modifications of the siRNAs backbone, such as 2´-O-methyl and 2´-fluoro RNA. Chol conjugation to the siRNA sequence also enhanced siRNA loading efficacy on lipid-based NPs, and exhibited higher biologic activity compared with unmodified siRNA [49]. Improving siRNA delivery pharmacokinetics

To improve the blood stability and the in vivo pharmacokinetics properties, many other helper lipids were introduced into the siRNA delivery system, such as PEGylated lipid (PEG lipid) [48]. The PEGylation effect of prolonging the circulation time of many carriers in the blood has been reviewed by many groups [48,50,51], including, but not limited to, the influence of acyl chain length in PEG lipids and their molar percentage in the liposome composition [52]. Sonoke et al. found that the PEG lipids with long acyl chains showed better siRNA delivery pharmacokinetics and efficiency when compared with those with the short or unsaturated chains [53]. The percentage of PEG lipids in the siRNA delivery system should consider both blood circulation time and cell uptake efficiency. A higher percentage of PEG lipids usually gives a longer blood circulation time but weakens the cellular uptake and subsequent endosomal escape. Therefore, an optimal density of PEG lipids is required [48,51]. Yagi et al. developed a ‘wrapsome’ NP, comprised of a core of siRNA/cationic DOTAP that was fully enveloped by a neutral lipid bilayer containing egg phosphatidylcholine and PEG lipid in a weight ratio of 24:14.8. They found that the wrapsome could improve the stability and systemic circulation of siRNA, thus resulting in enhanced specific gene knockdown and significant anti-tumor activity in vivo [54]. To overcome the PEGylation-induced problem of cell uptake and endosomal escaping, Carmona et al. developed a pH-sensitive PEGylated liposome by postcoupling a PEG-2000 dialdehyde on the surface of particles composed of cationic cholesteryl polyamine–N1-cholesteryloxycarbonyl-3, 7-diazanonane-1, 9-diamine, neutral lipids (DOPE) and Chol–PEG350 aminoxy lipid via oxime linkage. The oxime linkage of the NPs is stable at pH 7, but is prone to decomposition in an acidic environment (pH 5.5 and below), which induces the release of PEG coating from particles, thus granting acidic pHtriggered cell uptake and endosome release of nucleic acids. This formulation demonstrated effective RNAi delivery for on the control of hepatitis B virus (HBV) virus replication [55]. future science group


Hatakeyama et al. developed another PEGcleavable siRNA delivery system, termed a ‘multifunctional envelope-type nanodevice’ (MEND), composed of DOTAP, DOPE, Chol, 1,2-distearoyl-sn-glycero-3-phosphoethanolamine–PEG and a PEG–peptide–DOPE (PPD). PPD has a peptide sequence-linked PEG and DOPE lipid moiety together, so that the PEG is removed when the peptide linker is cleaved by the target molecule, such as MMP. It has been demonstrated that PPD-formulated NPs accelerated both cellular uptake and endosomal escape in MMP-rich tumor environments, compared with a conventional PEG fomulation, resulting in potent silencing (~70%) activity in vivo with no observed hepatotoxicity and innate immune stimulation [56]. Notably, the PEGylation methods also influence the stability of siRNA delivery. For example, simply mixing siRNAs with PEGylated liposomes results in poor loading of siRNAs onto the outer surface of particles, causing the premature release of siRNA, whereas post-PEG-coating on the siRNA liposome could enable more stable siRNA delivery. This aspect has been discussed and reviewed in detail by Buyens et al. [48]. Although the beneficial PEGylation effect on siRNA delivery is well acknowledged, it has been found that repeated injection of PEGylated NPs could induce the loss of their long circulation characteristics owing to the production of the anti-PEG IgM [57]. The phenomenon of accelerated blood clearance serves as a reminder of the limitation of PEG-coated NPs. In addition, other challenges in the context of the pharmaceutical development of lipid-based siRNA therapeutics have been reviewed extensively by Gindy et al., including the development of a robust manufacturing process, the setting of appropriate product specifications and controls, development of strategies to assess and ensure product stability, and the evaluation of product comparability throughout development [58]. Improving biocompatibility & reducing immunotoxicity

To reduce the immunotoxicity of siRNA delivery, some biogenic materials were included in the cationic lipid-based siRNA delivery system. For example, hyaluronic acid, a biogenic component distributed widely in the ECM, was introduced into LPD-NPs by Chono et al. to develop a liposome–protamine–hyaluronic acid (LPH)-NP [59]. Hyaluronic acid provides LPH-NP multivalent charges to enhance the particle condensation while containing no immunostimulatory future science group

Review

CpG motifs. This LPH-NP delivery of siRNA induced a 80% silence of luciferase activity in the metastatic B16F10 tumor in the lungs at a low, single, intravenous injection dose of 0.15 mg siRNA/kg [59], while not causing obvious immunotoxicity at a dose range of 0.15–1.2 mg siRNA/kg. Recently, they employed LPH-NP in the delivery of siRNA for silencing of CD47, a ‘self-marker’ that is usually overexpressed on the surface of cancer cells to enable them to escape from immunosurveillance, and achieved effective inhibition of melanoma tumor growth and lung metastasis [60]. To improve the biocompatibility of cationic lipid delivery systems, Yang et al. developed a lipid–polymer hybrid NP prepared by a singlestep nanoprecipitation of a formulation of cationic lipids, BHEM–Chol and amphiphilic polymers. Such hybrid NPs exhibited excellent stability in serum and showed significant improvement on biocompatibility compared with the pure BHEM–Chol particles [26]. A similar toxicity reduction effect has also been observed for other hybrid NPs with involvement of polyglutamate [61–63]. Improving the cellular uptake & enhancing siRNA release into the cytoplasm

To enhance the cellular uptake and further endosomal escaping of siRNA, some neutral helper lipids, such as DOPE and 1,2-distearoyl-sn-glycero-3-phosphocholine were added to the cationic lipid-based siRNA delivery system. The role of such fusogenic lipids has been reviewed by Wasungu et al. [44]. DOPE profoundly affects the polymorphic features of the liposome–siRNA complex (termed lipoplexes) in that it promotes the transition from a lamellar to a hexagonal phase, thus inducing fusion and disruption of the membrane. The optimized 3-b-(N-[N´,N´-dimethylaminoethane] carbamoyl) Chol/DOPE-based lipoplexes have been found to improve the transfection efficiency of the lipoplex [64]. Khoury et al. prepared a NP using the cationic lipid RPR209120 in combination with DOPE. This NP enables effective systemic delivery of DNA, resulting in successful silencing of TNF-a in collageninduced arthritis of mice, and complete cure of collagen-induced arthritis via weekly intravenous administration [65]. Since then, this formulation for siRNA delivery to myeloid cells has been successfully applied [66,67]. By replacing the protamine/DNA core of the LPD NPs with a pH-sensitive calcium www.futuremedicine.com

109

Review


phosphate (CaP) core, Yang et al. recently developed a biodegradable lipid/calcium/phosphate NP (LCP-NP). CaP is biocompatible, bio degradable material and native to the body, as it is the principle mineral component of teeth and bones. It is hypothesized that the CaP core could be rapidly dissolved at acidic endosome pH, which will increase the osmotic pressure and cause NP disassembly, endosome swelling and, finally, siRNA release into the cytoplasm. The study showed that the LCP-NP releases more cargo to the cytoplasm compared with the LPD formulation, leading to a significant (~40-fold in vitro and ~fourfold in vivo) improvement in siRNA delivery. More recently, they coformulated three siRNAs against MDM2, c-myc and VEGF in a LCP-NP. Such a formulation caused simultaneous silencing of the three oncogenes in metastatic nodules, and resulted in significant inhibition of lung metastases (~70–80%) at a relatively low dose (0.36 mg/kg) without any observed toxicity [68]. Recently, to enhance endosomal escape, Harashima et al. synthesized a pH-sensitive cationic lipid, YSK05, which contains a tertiary amine group for pH sensitivity. They incorporated this lipid into the above mentioned MEND siRNA delivery system, and found that the YSK05–MEND combination had a better ability for endosomal escape than other MEND-containing conventional cationic lipids [69,70]. Some other new cationic lipids have also been reported for improving siRNA delivery, such as N´,N´-dioctadecyl-N-4, 8-diaza10-aminodecanoylglycine amide [71] and 1,2-dilinoleyloxy-3-dimethylaminopropane, which will be the subject of later discussion [23].

Active targeting delivery

Incorporating an active target motif, such as transferrin, RGD, EGF and various cell-penetration peptides, to the cationic lipid-based siRNA delivery system could significantly improve the intracellular uptake and endosomal release of siRNA. Daka et al. [72] and Ogris et al. [73] have reviewed this aspect extensively. Example of an advanced cationic lipid-based siRNA delivery system: stable nucleic acid lipid particles & their application Among the cationic lipid-based siRNA vehicles, several advanced delivery systems have gained worldwide attention owing to their high efficiency in systemic siRNA delivery methods, such as LPD, AtuPLEX, stable nucleic acid lipid particle (SNALP) systems, LPD and AtuPLEX, and have been well reviewed previously [74,75]. Here, we take SNALP as an example to highlight its approach in overcoming barriers of siRNA systemic delivery. SNALP & its application

SNALP is constructed by three basic lipids: an ionizable cationic lipid (1,2-dilinoleyloxy3-dimethylaminopropane), a neutral helper lipid, including Chol and fusogenic lipids, and a PEG lipid (Table 2). In SNALPs, the backbonemodified siRNA is encapsulated within a closed shell of a cationic-zwitterionic lipid bilayer, furnished with an outer PEG shield (Figure 2A). The lipid bilayer contains a mixture of cationic and fusogenic lipids to enable cellular uptake and further endosomal escape owing to electrostatic interactions between the negatively charge cell

Table 2. Stable nucleic acid lipid particle formulations. Stable nucleic acid lipid particle formulations in systemic delivery

Target siRNA and modification (yes/no)

Animal model

Ref.

DSPC:Chol:PEG–C–DMA:DLin–DMA (20:48:2:30) –siRNA

Hepatitis B virus (yes)

Mouse model of hepatitis B virus replication

[23]


ApoB (yes)

Cynomolgus monkeys

[76]


Cell cycle proteins PLK1 (yes) Intrahepatic mouse tumor models and s.c. tumor models

DSPC:Chol:PEG–C–DMA:cationic lipid (10:40:10:40) –siRNA

Factor VII (yes)

[78–80]

C57BL/6 mice

[24]

DPPC:PEG–C–DMA:Chol:DLin–KC2–DMA (7.1:1.4:34.3:57.1) –siRNA ApoB, transthyretin (yes)

Cynomolgus monkeys

[24]

PEG–DMG:Chol:Octyl–CLinDMA (2:48:50) –siRNA

Rosa26-LSL-Luc mouse strain, Adeno-liver-Luc mice

[77]

Luciferase (yes)

Chol: Cholesterol; DLin–DMA: 1,2-dilinoleyloxy-3-dimethylaminopropane; DLin–KC2–DMA: 2,2-dilinoleyl-4-dimethylaminoethyl-(1,3)dioxolane; DMG: Dimyristoylglycerol; DPPC: Dipalmitoylphosphatidylcholine; DSPC: 1,2-distearoyl-sn-glycero-3-phosphocholine; Octyl–CLinDMA: 2-([8-([3b]cholest-5-en-3-yloxy)octyloxy)-N,N-dimethyl-3-([9Z,12Z]-octadeca-9,12-dien-1-yloxy)propan-1-amine; PEG–C–DMA: 3-N-([-methoxy PEG 2000]carbamoyl)-1,2dimyristyloxypropylamine; s.c.: Subcutaneous.

110


future science group


Transport through the PM

Review

Cytosolic release of siRNAs PM

In vitro HeLa cells

PM 50 nm

200 nm Trapped in the EE compartment pH 5.5–6.5

M LE

EE Golgi

EE 50 nm

200 nm Cationic lipids PEG lipids

Trapped in the Lys pH >4.5

500 nm LE

200 nm

50 nm

LE Golgi Lys 200 nm

50 nm

Cytosolic siRNA–gold (%)

Neutral lipids siRNA

Trapped in the LE compartment pH >5.5

50 nm

1.5 1.0 0.5

50 nm

0.5 1.0 1.5 2.0 3.0 4.0 5.0 6.0 Time (h) Hepatocytes

HeLa

Figure 2. Stable nucleic acid lipid particle siRNA delivery system. (A) Stable nucleic acid lipid particle (SNALP) siRNA delivery system. (B) Visualization of SNALP–siRNA–gold by electron microscopy. (C) Ultrastructural analysis of SNALP in vitro trafficking. SNALP–siRNA–gold detected in HeLa cells in vitro, by electron microscopy. SNALP–siRNA–gold was found in the extracellular matrix close to the PM and inside the EE compartment, LE compartment and Lys within cells. Magnified images (right panels) permit appreciation of the subcellular localization of siRNA–gold. (D & E) Cytosolic release of siRNA. (D) siRNA–gold concentrates within the LE compartment of HeLa cells in vitro and (E) the quantification of cytosolic siRNA–gold release kinetics in a liver section in vivo (solid line) and in HeLa cells in vitro (dashed line). The error bars represent the standard error. EE: Early endocytic; LE: Late endocytic; Lys: Lysosome; PM: Plasma membrane. Adapted with permission from [82] .

membrane and cationic lipids. The coated PEG provides a neutral, hydrophilic exterior to shield the cationic bilayer, protect the nucleic acid core against degradation by nucleases, sterically stabilize the particles against disassembly in collagen networks and prevent nonspecific binding to cells, thus making NPs ‘cohesive’ and ‘stealthy’ to prevent siRNA degradation and rapid clearance in the bloodstream. The SNALPs are approximately 70–150 nm in diameter (Figure 2B), and their major accumulation (±standard deviation) was found in mouse liver (28 ± 1.7% of injected dose administered) and spleen (8.2 ± 2.8% of injected dose administratered) [23]. Their biodistribution is related to the property of NPs, such as size, lipid and PEGylation features. In general, the PEGylated NPs are readily navigated to the mononuclear phagocyte systems, such as the liver and spleen. The therapeutic efficacy of SNALP–siRNA against HBV was validated in a HBV replication mouse model by Morrissey et al. [23]. They found future science group

that SNALP delivery prolonged the circulation time of siRNA and achieved efficient inhibition of serum HBV by daily intravenous injections of 3 mg/kg/day of siRNA [23]. Zimmermann et al. further demonstrated the efficient systemic delivery of SNALP–siRNA in nonrodent species. They administered intravenous injections of SNALP–siRNA against the ApoB gene at single doses of 1 or 2.5 mg/kg on cynomolgus monkeys and achieved potent gene knockdown of ApoB mRNA (>90%). Significant downregulation of ApoB protein, serum Chol and low-density lipoprotein (LDL) levels was also observed as early as 24 h after treatment and lasted for 11 days with a 2.5-mg/kg siRNA dose [76]. More recently, Semple et al. found a highperformance lipid, 2,2-dilinoleyl-4-dimethyl aminoethyl-(1,3)-dioxolane by screening various ionizable cationic lipids, and demonstrated the excellent activity of the 2,2-dilinoleyl4-dimethylaminoethyl-(1,3)-dioxolane-based SNALP in effectively silencing the endogenous www.futuremedicine.com

111

Review


hepatic gene at doses of siRNA as low as 0.01 mg/kg in rodents and 0.1 mg/kg in nonhuman primates in vivo [24]. Meanwhile, Tao et al. developed another optimized SNALP system by screening cationic lipids and adjusting PEG lipid density, and achieved potent gene knockdown in the liver (90%) of mice [77]. Besides the application of SNALP in liver diseases, Judge et al. demonstrated the successful delivery of siRNA to solid tumors in mice targeting the PLK1 gene, resulting in significant inhibition of subcutaneous Hep3B tumor growth [78]. Notably, the SNALP system encapsulated with self-amplifying RNA has been reported as a new vaccine platform by Geall et al., which substantially increases immunogenicity by eliciting broad, potent and protective immune responses [79]. Altogether, it is not surprising that the SNALP system has entered or completed multiple clinical trials for systemic siRNA delivery, such as VEGF and KSP in liver cancer, ApoB in liver disease, TTR in transthyretin amyloidosis, PLK1 in cancer and Elola in Ebola virus disease [75]. The impact of physiological constrains, such as tumor vasculature, on the efficiency of siRNA delivery has also been investigated. Li et al. found that SNALP predominantly delivered siRNA to areas adjacent to functional tumor blood vessels by analyzing the spatial distribution of localized target knockdown within tumor sections relative to tumor hypoxia [80]. This pheno menon suggests that it is probably not easy for the SNALP–siRNA complex to cross the ECM with a size of 70–150 nm; therefore, most of them are trapped in the areas adjacent to blood vessels [80]. To improve the transport of SNALP across the ECM barrier, Rudorf et al. developed a small SNALP, called a ‘mono-NALPs’, which was self-assembled by solvent exchange from solution containing siRNA mixed with the four lipid components DOTAP, DOPE, 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-PEG-2000. The mono-NALP has a similar core–shell structure as SNALP, but is only 30 nm in diameter, which should facilitate its transport in vivo across the ECM to achieve deeper tissue penetration [81]. Recently, the mechanism of cellular uptake, intracellular transport and the endosomal escape of siRNA by SNALP delivery had been investigated. It has been shown that SNALPs entered cells by both constitutive and inducible pathways in a cell type-specific manner using CME and 112


macropinocytosis (Figure 2C) [82]. However, the escape of siRNAs from the endosome into the cytosol occurred at low efficiency (