RNAi nanomedicines: challenges and opportunities within the immune system

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2010 Nanotechnology 21 232001

(http://iopscience.iop.org/0957-4484/21/23/232001)

Download details:

IP Address: 132.66.249.35

The article was downloaded on 20/09/2012 at 09:06

Please note that terms and conditions apply.

View the table of contents for this issue, or go to the journal homepage for more

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING NANOTECHNOLOGY

Nanotechnology 21 (2010) 232001 (13pp) doi:10.1088/0957-4484/21/23/232001

TOPICAL REVIEW

RNAi nanomedicines: challenges andopportunities within the immune systemShiri Weinstein and Dan Peer1

Laboratory of Nanomedicine, Department of Cell Research and Immunology, George S WiseFaculty of Life Science, IsraelandCenter for Nanoscience and Nanotechnology, Tel Aviv University, Tel Aviv, 69978, Israel

E-mail: peer@post.tau.ac.il

Received 26 February 2010, in final form 21 April 2010Published 13 May 2010Online at stacks.iop.org/Nano/21/232001

AbstractRNAi, as a novel therapeutic modality, has an enormous potential to bring the era ofpersonalized medicine one step further from notion into reality. However, delivery of RNAieffector molecules into their target tissues and cells remain extremely challenging. Majorattempts have been made in recent years to develop sophisticated nanocarriers that couldovercome these hurdles. This review will present the recent progress with the challenges andopportunities in this emerging field, focusing mostly on the in vivo applications with specialemphasis on the strategies for RNAi delivery into immune cells.

1. Introduction

RNA interference (RNAi) is a natural cellular mechanismfor RNA-guided regulation of gene expression. Thisregulation is carried out by double stranded ribonucleic acid(dsRNA) that suppresses the expression of specific geneswith complementary nucleotide sequences, either by degradingspecific messenger RNA (mRNA) or by blocking mRNAtranslation. RNAi can be activated exogenously by expressingshort hairpin RNA (shRNA) and microRNA (miRNA)duplexes with viral vectors, or by incorporating synthetic smallinterfering RNAs (siRNAs), double stranded miRNA mimeticsmolecules, and anti-miRNA oligonucleotides (antagomirs)directly into the cell cytoplasm [1–3]. The viral vectortools offer advantages such as long sustained gene silencing,and ease expression of multiple copies of RNAi moleculesfrom one transcript. However, due to safety concerns andinterference to the endogenous mechanism of miRNAs, wechoose not to discuss them in this review.

siRNA is a chemically synthesized double stranded RNA(dsRNA) of 19–23 base pairs with 2-nucleotides unpairedin the 5′-phosphorylated ends and unphosphorylated 3′-ends [4, 5]. Inside the cell cytoplasm, siRNAs are incorporated

1 Author to whom any correspondence should be addressed.

into RNA induced silencing complex (RISC), a protein–RNAcomplex that separates the strands of the RNA duplex anddiscards the passenger (sense) strand. The guide (anti-sense)strand then guides RISC to anneal and cleave the target mRNAor block its translation [2] (see figure 1). By recycling the targetmRNA, the RISC complex that incorporates the guide strand,may show a silencing effect for up to seven days in dividingcells and for several weeks in non-dividing cells. Furthermore,repeated administration of siRNA can result in stable silencingof its target [6].

miRNAs are endogenous non-coding RNAs involved inpost-transcriptional gene expression regulation. They areprimarily transcribed as pri-miRNAs, long hairpin structuredtranscripts, and processed by RNAse III Drosha into 70–100nucleotide pre-miRNAs, which are exported by Exportin 5to the cytoplasm. There, RNAse III Dicer generates dsRNAof approximately 22 nucleotides, the mature miRNA. Fromthis step, the mechanism is similar to that described abovefor siRNAs—the anti-sense strand of the mature miRNA isincorporated to microRNA containing RNA induced silencingcomplex (miRISC), whereas the other strand is degraded. ThemiRISC usually hybridizes to partially complementary bindingsites on the 3′ untranslated region of the target mRNAs, andsilences their expression by interfering with translation orby initiating their degradation. The partial complementary

0957-4484/10/232001+13$30.00 © 2010 IOP Publishing Ltd Printed in the UK & the USA1

Nanotechnology 21 (2010) 232001 Topical Review

Figure 1. A schematic representation of the RNAi mechanism. The endogenous pathway of RNAi begins in the nucleus, where thepri-miRNA molecule is processed by RNAse III enzyme named Drosha and becomes pre-miRNA. Exportin 5 exports the pre-miRNA into thecytoplasm where it is processed by RNAse III Dicer and become the mature miRNA. From this point, the endogenous pathway is very similarto the exogenous pathway, in which synthetic RNAi molecules (siRNAs, miRNA mimetics and antagomirs) are inserted into the cellcytoplasm. The 19–23 double stranded RNA molecule is then incorporated into RISC (or miRISC), were the passenger strand is released andthe guide strand mediates its target mRNA degradation or translation inhibition.

enables each individual miRNA to regulate many differentmRNA targets (in fact, a single miRNA is predicted to targetbetween 100 and 200 mRNAs [7], and one specific mRNAto be regulated by different miRNAs [3, 8]). RNAi couldbe exploited for miRNAs therapeutics by offering differenttactics, both for silencing miRNA expression, for examplewith antagomirs, and for miRNA replacement, for examplewith double stranded miRNA mimetics (siRNA structureanalogous molecules that are equivalent to the endogenousmature miRNA Dicer product). These strategies usually result

in an increase or decrease of the miRNA’s regulated proteins’expression levels, respectively [9, 10].

Knocking down every gene of interest with siRNA,clusters of co-regulated genes with miRNA mimetics orendogenous miRNAs with antagomirs make the addressingof otherwise ‘undruggable’ targets (i.e. molecules withoutligand binding domains or those that have a structuralhomology with other important molecules in the cell) possible.The elimination of clinical safety concerns associated withviral vectors and the reduced likelihood for interference to

2

Nanotechnology 21 (2010) 232001 Topical Review

the endogenous microRNA machinery (which could happendue to saturation of enzymes or transport proteins), makeRNAi molecules a promising new platform for therapy inpersonalized medicine [11].

Despite this promise, utilizing RNAi for therapeutics is nota trivial task. For example, due to the large molecular weight(∼13 kDa), the net negative charge and their hydrophilicity,the efficiency with which naked molecules of siRNA crossthe plasma membrane and enter the cell cytoplasm is verylow [2, 12]. When injected intravenously, in addition torapid renal clearance and susceptibility to degradation byRNAses, unmodified naked siRNAs are recognized by Toll-like receptors (TLRs). This often stimulates the immunesystem, hence provoking an interferon response as well ascomplement activation, cytokine induction, and coagulationcascades. Beside the undesired immune activation, thoseeffects can suppress gene expression globally, generating off-target and misinterpreted outcomes [12, 13]. Most of theseobstacles are relevant for other naked RNA molecules such asmiRNA mimetics and antagomirs. Therefore, there is a clearneed for appropriate delivery systems for RNAi, all of whichhave to utilize cellular mechanisms for internalization, cargorelease (from the carriers and escape from the endosomes),cargo accumulation in the cell cytoplasm and RISC activation.This review will present the recent progress in this emergingfield, focusing on the in vivo applications for RNAi delivery,with special emphasize on the strategies for RNAi deliveryto leukocytes, which are intrinsically hard to transfect anddifficult to target. Most of the strategies discussed here havebeen developed for siRNA delivery. However, the recentaccumulated knowledge regarding miRNA mechanisms andfunctions, along with the consolidated understanding of thehuge therapeutic potential of this emerging field, make itreasonable to assume that in the near future those strategiesas well as new ones will be adapted for antagomirs or miRNAmimetics delivery.

2. Cellular delivery strategies for RNAi

Most of the methods commonly used for in vitro or ex vivodelivery of RNAi are conventional transfection methods.Studies with purely physical methods such as microinjectionand electroporation [14–19], as well as studies that usedcalcium co-precipitation [20], commercial cationic polymersand lipids [4, 21–27] and cell penetrating peptides [28–32],have demonstrated effective knockdown of desired genes.Except for the physical methods (in which the cell is subjectedto an injection of small volumes of RNAi molecules directlyinto the cell cytoplasm or to a burst of electricity thatcauses pores in the membrane, hence elevating the abilityof extracellular material to enter into cell), all the methodsshare a main characterization—a positive (cationic) chargethat enables the complexing of the RNAi molecules and theinteraction with the negatively charged plasma membrane.

2.1. Translation of RNAi into clinical practice

Silencing of gene expression in vitro is a great tool forfunctional and validation studies. Nevertheless, understanding

gene expression in a disease model by validating a specificgenes’ role in vivo, along with the potential to inducetherapeutic gene silencing, open new avenues for utilizingRNAi as a novel therapeutic modality and brings the era oftheranostics (personalized medicine) a step further from avision to a potential reality.

Despite the large diversity of available methods forin vitro and ex vivo RNAi delivery mentioned above, there areadditional hurdles to overcome in translating these methodsinto clinical practice. As we detail below, the biggest hurdlefacing the translation of RNAi therapeutics into the clinic istheir delivery.

2.2. In vivo delivery of RNAi molecules

Local delivery of RNAi molecules has been demonstrated invarious animal models [31, 33–37] and is employed in severalongoing clinical trials [38]. Based on local injections ofnaked or cationic lipid/polymer-formulated RNAi molecules,this method of treatment has demonstrated very promisingoutcomes. Nevertheless, regarding human therapeutics, thiskind of delivery is relevant only for mucosal diseases andsubcutaneous or other accessible tissues.

Systemic delivery of RNAi molecules is the mostchallenging task in this field. While cellular and local deliverystrategies have to deal with the need for internalization,release, and accumulation of the RNAi molecules in the cellcytoplasm, delivery strategies for systemic treatment of anentire animal additionally enforces one to deal with the RNAimolecules’ interactions with blood components, uptake by thereticuloendothelial cells and immune stimulation, degradationby RNAses, renal clearance, anatomical barriers (such asthe liver which tends to accumulate foreign bodies), andpenetration into the target tissue [39].

Systemic delivery of naked RNAi molecules may occurby the hydrodynamic method. This method, whose precisemechanism is unsolved yet, involves rapid injection of a largevolume of RNAi molecules in physiologic solutions (about10% of the body weight administered within 5–10 s) [40, 41].Hepatocytes cells in the liver are the main target of thisapproach. Different studies for siRNA delivery were done withthis method, demonstrating functional knockdown of specificgenes in animal liver [40–43]. Nevertheless, due to volumeoverload side effects (such as right-sided heart failure), thehydrodynamic method is not relevant for human therapeuticuse.

Naked siRNAs have been successfully utilized fortargeting the kidney. When systematically administrated, largeamounts of naked siRNA are excreted by the glomerulus(which excretes any molecule with molecular weight lessthan 40 kDa) and reabsorbed in the proximal tubule. Theaccumulation of free siRNA in the kidney is 40 times higherthan in any other organ, an ideal property for selectivegene therapy. Studies in rat models for renal injuryindicated functional silencing of p53, a major proapoptoticgene, and renal protection, both in single and multipleinjection administration [44]. QPI-1002, a drug fromQuark Pharmaceuticals, is based on these studies, and being

3

Nanotechnology 21 (2010) 232001 Topical Review

evaluated in phase I clinical trials for systemic deliveryof p53-siRNA in acute renal injury and delayed graftfunction [39].

Another strategy for using naked siRNAs systematically istheir stabilization with locked nucleic acids (LNAs), syntheticRNA-like high affinity nucleotide analogs. Although acomplete modification of siRNA with LNAs is impossibledue to significant structural changes, siRNAs incorporatedwith a few LNAs did not lose their silencing ability [45].LNA-modified siRNAs targeting a xenograft cancer modelhave shown to be much more stable in the serum andeffective in inducing silencing of GFP expression [46]. Inaddition, those LNA-modified siRNAs reduced off-targetgene regulation compared with unmodified siRNA. Anotherstudy [47] demonstrated that intravenously administratedLNA-modified siRNAs efficaciously reduced EGFP expressionin the mouse lung bronchoepithelium.

Although more stable in serum, and resistant to nucleasesthan unmodified siRNAs, the LNA-modified siRNAs are stillsubject to renal clearance. Therefore, utilizing naked siRNAssystematically is less applicable when the target organ is notthe kidney. Hence, strategies for systemic delivery of siRNA(and other RNAi molecules) must relay on nanocarriers.

Typically, the formal definition of nanotechnologyrequires that the nano-device or its essential componentsbe man-made and between 1 and 100 nm at least in onedimension [48]. However, from a biological standpoint,the dimensions limitations of nanotechnology devices couldbe more flexible according to the operational and uniqueneeds of biological systems and the breakthrough potentialfor patient care [48–51]. Most of the carriers described inthis review are within the formal dimensions of nano-devices,hence considered as nanocarriers. The other particles, whosediameter exceeds the strict 100 nm limitation, are still in closeproximity and possess great clinical potential (see table 1 forcarrier dimensions).

The engineering concepts and design of nanocarriers forRNAi delivery are summarized in box 1. Those general rulesintend to solve or weaken the hurdles that nanocarriers facewhen systemically administrated into an organism (rodent,non-human primate or human). For example, in order to avoidundesired and probably toxic accumulation of the deliverysystem components in the body, the nanocarriers should bemade from fully degradable materials and should act onspecific cells or tissues while avoiding damaging others [51].

To date, most of the developed strategies for RNAidelivery have been oriented and tested with siRNAs andnot with miRNA mimics or antagomirs, probably as aconsequence of the later discovery of the miRNA mechanismand its function (and hence the later understanding of theirexquisite therapeutic potential) in comparison with siRNA.Consequently, most of the knowledge shared in this reviewfocuses on the development of siRNA nanocarriers. However,we believe that most (if not all) of the strategies outlinedhere could easily be modified to serve as miRNA mimics orantagomir delivery vehicles.

The systemic siRNA delivery strategies are dividedinto two major categories: passive and active (targeted)

Box 1. Engineering concepts and design ofnanocarriers for RNAi delivery.

General requirements from nanocarriers for systemic adminis-tration:

• The nanocarrier should be made from well characterized,easily functionalized and biocompatible materials.

• It should have an extended circulating half-life and abilityto evade the RES system.

• It should avoid the induction of immune responses byevading the immune system.

• It should have a low rate of aggregation in order to avoidtoxic accumulation in the body.

• It should be small enough to penetrate its target tissues butnot too small to avoid renal clearance.

• It should accumulate in the target cells over other cells (ortissues).

• For higher effectiveness the nanocarrier should be eithersoluble or colloidal under aqueous conditions.

• It should have a long shelf life.

Additional requirements from carriers for systemic RNAidelivery:

• To internalize into the target cell in a specific and selectivemanner.

• To ensure efficient release of RNAi molecules in target cellcytoplasm while escaping from the endosome.

• To ensure minimal recognition by RNA sensors inside thecell (such as TLR3, RIG-I and MDA5).

delivery. Passive delivery exploits the inherited tendency ofnanoparticles to accumulate in cancerous or highly inflamedtissues due to the enhanced permeability and retention (EPR)effect or in organs of the reticuloendothelial system (RES), alsoknown as the mononuclear phagocytic system. The EPR effectis the increased permeability of the blood vessels in tumorsand highly inflamed tissues caused by rapid and defectiveangiogenesis, and the dysfunctional lymphatic drainage thatretains the accumulated nanoparticles [52]. Different studieshave shown that the efficacy of extravasation into tumorsis higher with particles whose diameters are smaller then200 nm [51, 53–56]. The RES is part of the immune system,consisting of phagocytic cells located in reticular connectivetissue, primarily monocytes and macrophages. These cellsaccumulate in lymph nodes, the spleen and the liver, and takeup foreign particles believed to be intruders in the body, such asviruses, bacteria and parasites of different types, sizes, shapesand charges [57, 58]. Hence, it is not surprising that majorattempts have been made to develop siRNA delivery systemsfor treating different liver diseases. Active (targeted) deliveryis based on specific antibodies, ligands or ligand mimeticsthat direct the nanocarriers to specific target cells and tissuesin order to achieve maximal therapeutic benefit, decrease theamount of drug required, and avoid nonspecific silencing andtoxicity in bystander cells [51].

4

Nanotechnology 21 (2010) 232001 Topical Review

Table 1. Nanocarriers for systemic RNAi delivery.

Nanocarriertype

Hydrodynamicdiameter (nm) Target tissue RNAi molecule Reference

SNALP ∼100 Livera ApoB-siRNA, Ebola’s Zairstrand polymerase-siRNA,transthyretin-siRNA

[58, 60, 63, 64]

DOPC liposomes ∼100 Tumorsb PAR-1-siRNA, EphA2-siRNA [65, 66]Cationic lipidoidscontaining liposomes

∼50 Livera ApoB-siRNA, Factor VII-siRNA [67]

65–250 Factor VII-siRNA/factor VII,ApoB, PCSK9, Xbp1-siRNAs

[68]

HK peptides 81–402 Tumorsb Raf-1-siRNA RHBDF1-siRNA [70, 71]150–200

Atelocollagen 1.5 Tumorsb FGF-4-siRNA, RPN2-siRNA,human papillomavirus 18 E6 andE7-siRNAs, Bcl-xL-siRNA,miR-16

[30, 73, 76]

Cholesterol/LDL/α-tocopherol/lithocholicacid/lauric acid conjugates

3–15c Liver miR-122 antagomir/apoB-siRNA [77–80]

HDL conjugates 3–15c Liver, gut, kidney andsteroidogenic organs

miR-122 antagomir/ apoB-siRNA [77]

Dynamic polyconjugates 8–12 Hepatocytes (liver) apoB-siRNA [81]PEI nanoplexes ∼100 Certain cancers and tumor

vasculature overexpress av

integrins

VEGF R2-siRNA [82]

CDP particles 50–70 Cancers with Tf receptorsupregulation

EWS-FLI1-siRNA,RRM2-siRNA,HIF-2alpha-siRNA

[83, 84]

Antibody–protaminefusion proteins

3–15 Breast cancer cells expressErbB2, leukocytes/activatedleukocytes

c-myc, MDM2, VEGF-siRNA,cy3 labeled siRNA (proof ofconcept)

[51, 87, 98]

RVG-9R 3–15 Neurons expressacetylcholine receptors

Antiviral FvEJ -siRNA [51, 88]

Aptamers 81 Prostate cancer cells andtumor

Plk1-siRNA [90]

Targeted Cationicliposomes

100–145 Hepatic stellate cells/solidtumors/dendritic cells

gp46-siRNA, HER2-siRNA,EGFR-siRNA,

[91, 93],

CpG oligonucleotide 3–15 Myeloid cells and B cells Stat3-siRNA [51, 94]scFvCD7-Cys 3–15 T cells CCR5/Vif/Tat-siRNA [51, 95]DC3-9dR 3–15 Dendritic cells TNF-α-siRNA [51, 96]Immunoliposomes 50–92 Dendritic cells CD40-siRNA [97]I-tsNPs ∼100 Gut leukocytes/lymphocytes

and monocytesCyclin D1-siRNA/CCR5-siRNA [105, 107]

a Passive targeting is achieved due to RES uptake.b Passive targeting is achieved due to the EPR effect.c Although there is no report showing size measurement we assume it is in the range of other conjugates and fusion proteins.

2.3. Passive systemic RNAi delivery

Stable nucleic acid–lipid particle (SNALP) is a ∼100 nmnon-targeted liposome with low cationic lipid contentthat encapsulates siRNAs and is coated with a diffusiblepolyethylene glycol–lipid (PEG–lipid) conjugate [59, 60]. ThePEG–lipid coat stabilizes the particle during formation andprovides a neutral and hydrophilic exterior that prevents rapidsystemic clearance. The lipid bilayer contains a mixture ofcationic and fusogenic lipids, and enables the internalizationof the SNALP and its endosomal escape while releasing thesiRNAs payload. Biodistribution study indicates that most(28%) of the siRNAs carried by the SNALPs were accumulatedin the liver (and only 0.3% in the lungs). Functional studyof SNALP encapsulated ApoB-siRNA has shown significantreduction in ApoB mRNA levels and as a result a reduction

in the levels of protein expression, serum cholesterol and low-density lipoprotein (the therapeutic benefit) [59]. Despite thepresence of cationic lipids known to trigger toxicities [61, 62],mice and non-human primates revealed no adverse effectsexcept for liver enzyme release. Based on these results,a clinical trial is now being conducted to test the abilityof SNALPs to deliver siRNAs for liver cancer treatment aswell as for reduction of serum cholesterol levels. SNALPencapsulating siRNA against the polymerase gene of the Zairestrain has been shown to protect guinea pigs from the lethalchallenge of the Ebola virus [63]. In a recent study [64],the researchers tried to determine the best lipid formulation,regarding efficacy and safety, for siRNA delivery by SNALPs.They found that the SNALP formulation of the lipid DLin-KC2-DMA was well tolerated in both rodent and non-human

5

Nanotechnology 21 (2010) 232001 Topical Review

primates and exhibited in vivo activity at siRNA doses as lowas 0.01 mg kg−1 in rodents, as well as 0.3 mg kg−1 in non-human primates (in comparison with at least 1 mg kg−1 forthe siRNAs that were used in previous SNALPs systems).Other formulations of cationic liposomes, with larger cationiclipid content than the SNALPs, have induced effective genesilencing, but they also show cytokine induction and toxicityand thus cannot be used for clinical evaluation.

Considering the significant toxicities that have beenassociated with cationic liposomes, neutral charge liposomesare very promising carriers for systemic delivery of siRNAs.1,2-dioleoyl-sn-glycero-3-phosphatidylcholine (DOPC) non-pegylated liposomes encapsulating siRNA against differentmolecules expressed on melanoma and ovarian cancersinhibited tumor growth in human xenograft models [65, 66].The accumulation of these liposomes in the cancerous tissuesis based on the EPR effect.

Bio-inspired materials for RNAi delivery are the cationiclipidoid (synthetic lipid-like molecules) that is formulated withnatural lipids to form liposomes for siRNA delivery. Lipidoidcontaining liposomes have been shown to deliver siRNAsthat induce effective gene silencing (80% reduction in twodifferent genes: ApoB and Factor VII mice’s mRNA levels)in the liver. A single intravenous injection of cationic lipidoidcontaining liposome encapsulated ApoB-siRNA resulted in a50% decrease in the protein level three days and up to twoweeks after the treatment. Although no immune response wasindicated, increases in the serum levels of two liver enzymessuggested liver toxicity [67]. Using combinatorial synthesisand screening of a different class of materials Love et al[68] identified a particular lipidoid formulation that enablesgene silencing with very low doses of siRNAs (less than0.01 mg kg−1 and 0.03 mg kg−1 in the liver of mice or non-human primates, respectively), and without observed adverseside effects. This formulation has demonstrated simultaneoussilencing of five specific genes after a single intravenousinjection [68].

HK peptides are an additional effective delivery strategyfor siRNAs. This system is based on the addition of histidinesinto poly-lysine peptides. While lysine is important for bindingthe siRNAs, histidines stabilize the particles and have animportant role in buffering acidic endosomes, thereby leadingto endosomal disruption and payload release. Specific ratiosand patterns of histidine and lysine have been found to augmentthe siRNA delivery while carriers with a higher ratio ofhistidine to lysine content seemed to be more effective [69].HK peptides carrying Raf-1-siRNA or human rhomboidfamily-1-siRNAs induced significant silencing of target genesand growth inhibition of tumor xenografts [70, 71].

Atelocollagen is a biomaterial consisting of a low-immunogenic fraction of pepsin-digested type I collagenfrom calf dermis. This biomaterial exhibits a stick-likestructure, 1.5 nm in diameter, 300 nm in length, and300 kDa in molecular weight [72], similar in dimensions tocarbon nanotubes but made from naturally occurring protein(collagen). Rich in positively charged residues (lysine andhydroxylysine), it complexes the negatively charged siRNAsand interacts with the plasma membrane, thus helping to

incorporate the siRNAs into the cells. Although these particleshave not been modified to target tumors, passive targetingdue to the EPR effect causes the selective accumulationwithin the cancerous tissues, as shown in several studies withdifferent tumor xenografts (i.e. [30, 73–75]). Initial studiesindicated that atelocollagen particles can be administeredsafely without induction of cytokines or observed toxicity tothe tissues. Another study demonstrated that atelocollagen-miR-16 complexes injected into mouse tail veins efficientlydelivered miR-16 to tumor cells on bone tissues in a xenograftmodel of prostate cancer. Functionally, those complexessignificantly inhibited the growth of prostate tumors [76].

2.4. Active (targeted) systemic RNAi delivery

siRNAs conjugated to other molecules is a common strategyfor active delivery. Cholesterol–siRNA conjugate is oneexample. The specificity of this delivery system is determinedby the lipoprotein to which the cholesterol–siRNAs conjugatesare attached in the circulation. When the conjugates bindLDL, the particles are mainly taken up by the liver dueto its LDL receptor expression, whereas when they bindHDL, they accumulate in the liver, the gut, the kidneyand steroidogenic organs, all of which express scavengerreceptor class B, type I (SR-BI) receptors, which bindHDL [77]. Cholesterol-ApoB-siRNA conjugate, as well as α-tocopherol [78] and lithocholic acid or lauric acid conjugatedto ApoB-siRNA [79], reduced serum cholesterol and ApoBmRNA levels in the liver. The same strategy was alsoemployed for antagomir delivery to the mouse liver. SpecificmiR-122 silencing in mouse liver was demonstrated with2′ O-methylated antagomirs against miR-122 containing acholesterol moiety in their 5′ end. The silencing was longlasting and continued even 23 days post-antagomir injection.This miRNA silencing results in upregulation of several genesinvolved in cholesterol biosynthesis and downregulation ofserum cholesterol levels [80]. Another example for the strategyof conjugating siRNAs to other molecules is the dynamicpolyconjugates [81]. This system includes membrane-activepolymers whose activity is masked until reaching the acidicenvironment of the endosomes. Using N-acetylgalactosamine,which binds to the asialoglycoprotein receptor, this systemtarget hepatocytes. Like the SNALPs, these particles, whencarrying ApoB-siRNAs, decreased ApoB mRNA levels in theliver.

PEI nanoplexes carrying siRNAs have also inducedfunctional silencing in subcutaneously transplanted tumorsin nude mice. Those particles are composed of RGD(Arg-Gly-Asp) peptide coupled via PEG (required for longerhalf-life, and reduced immunogenicity) to polyethylenimine(PEI, a cationic polymer that in addition to its ability tocondense nucleic acids, has a pH-buffering property thatdisrupts endosomes, thus enabling it to reach the cytoplasm).When complexed with siRNAs, due to electrostatic interactionsbetween the cationic polymers and anionic siRNAs, someRGD-PEG-PEI molecules form a polyplex, with the positivelycharged RGD-PEG components exposed on its surface. Thetargeting ability of this particle is based on the overexpression

6

Nanotechnology 21 (2010) 232001 Topical Review

of av integrins, to which RGD peptides bind, in certain cancersand in tumor vasculature [82].

Cyclodextrin containing polycation (CDP) particles havebeen also successfully used for siRNA delivery into mousesubcutaneous tumors [83]. CDP is a polymer with a cyclicoligomeric glucose backbone that when complexed withsiRNA assembles into a colloidal 50–70 nm particle. Toachieve targeting, transferin-coupled PEG is attached to thesurface of the particles to exploit the upregulation of Tfreceptors in cancers. However, despite being considerablyless toxic than conventional cationic polymers (such as PEI),safety experiments on non-human primates revealed that in thehigh concentration tested, injection of these particles inducedelevation in blood urea (which might indicate kidney toxicity).In addition, a mild increase in serum liver enzyme levels anda mild increase in IL-6 levels (IL-6 is considered as a globalmarker for inflammation) were observed. Multiple injectionsof the particles induced antibodies to human-Tf. Despite thosedisadvantages, the efficacy of Tf-coupled CDP containingsiRNAs for ovarian cancer and melanoma treatment is beingevaluated now in a clinical trial with early positive results forgene silencing [84, 85].

Antibody–protamine fusion protein carriers (3–15 nm indiameter) are unique systems for systemic siRNA delivery.These fusion proteins are made by molecular biologytechniques (cloning of fragment of an antibody together withprotamine (a basic protein) in a single transcript) and expressedin bacteria or mammalian cells. Protamines are relativelysmall (molecular weight of about 5–8 kDa) and highly basicproteins composed of 55–79% arginine residues [86]. Thesefusion proteins are build of a fragment of an antibody,which acts as a targeting agent, directing these carriers intoa specific cell surface receptor, and the protamine is ansiRNA binding protein, which neutralizes and condenses thesiRNAs. ErbB2-protamine fusion protein in complex withsiRNA significantly inhibited the growth of breast cancercells [87]. A similar strategy was developed for targeting thebrain, overcoming the obstacle of the blood–brain barrier. Thisstrategy is based on a short peptide, derived from rabies virusglycoprotein (instead of antibody), with nonamer arginineresidues at the carboxy terminus (RVG-9R), important for theglycoprotein–siRNA binding. The ability to target brain tissueby these particles is achieved due to RVG binding to specificreceptors (acetylcholine receptors) on neurons. Intravenousinjections of RVG-9R-anti-GFP-siRNAs into transgenic GFPmice resulted in a specific decrease in GFP expression inthe brain. Furthermore, mice treated with intravenous RVG-9R—antiviral—siRNAs were protected against fatal viralencephalitis. Repeated treatments with those particles did notinduce inflammatory cytokines or anti-peptide antibodies [88].

Aptamer–siRNA chimeras are completely RNA basedparticles for specific delivery of siRNAs. This approachrelies only on the fact that structured RNAs are capableof binding a variety of proteins with high affinity andspecificity. The chimera includes both a targeting moiety,the aptamer, and an RNA-silencing moiety, the siRNA [89].Modified aptamer–siRNA chimeras, in which the aptamerportion mediates binding to PSMA (a cell surface receptor

overexpressed in prostate cancer cells and tumor vascularendothelium) with siRNA against Plk1 caused a pronouncedregression of PSMA-expressing tumors in athymic mice aftersystemic administration [90]. This approach eliminates variousside effects and is considered to have low immunogenicity.Additional advantages are the possibility to synthesize largequantities at a relatively low cost, and the smaller size ofaptamers compared with that of antibodies (<15 kDa versus150 kDa), which promotes better tissue penetration.

Different formulations of targeted cationic liposomesserved for selective targeting of hepatic stellate cells (whichare the major cell population involved in the formation ofscar tissue in response to liver damage, named fibrosis) orsolid tumors. Stellate cells express receptors for retinol,the binding protein which efficiently uptakes vitamin A.Based on these, injection of cationic liposomes coupled tovitamin A and complexed with siRNA to a murine keyfibrogenesis factor (gp46) into cirrhotic mice, silenced thespecific gene in mouse liver and resolved fibrosis [91]. 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) liposomesencapsulating HER2-siRNAs and containing histidine–lysinepeptides (to enhance the escape from the endosomes)and a single-chain antibody fragment targeting transferrinreceptors (elevated on the membranes of tumor cells) ontheir surface, have been targeted to a tumor xenograftand inhibited its growth [92]. Anisamide–PEG–liposomes–polycation–DNA (anisamide–PEG–LPDs) are unilamellarcationic liposomes coated with PEG-linked anisamide (asmall-molecule compound binding sigma receptors) on theirsurface, and a protamine-condensed mixture of siRNAs anda carrier calf thymus DNA in their core. EncapsulatingEGFR-siRNA, anisamide–PEG–LPDs injected intravenouslyinto tumor-bearing mice, have been shown to increase mousesensitivity to chemotherapy [93]. Unfortunately, these particlesinduced a significant increase in serum cytokine levels, thusweakening the potential for clinical therapeutic use. However,it is important to note that cytokine response is not alwaysdeleterious to therapy and there are cases when immuneactivation could enhance the therapeutic effects.

2.5. Targeted delivery systems for leukocytes

Utilizing RNAi to manipulate gene expression in leukocytesholds great promise for the drug discovery field, as well asfor facilitating the development of new therapeutic platformsfor leukocyte-implicated diseases such as inflammation, bloodcancers, and leukocyte-tropic viral infections. However, dueto their resistance to conventional transfection methods and totheir dispersing in the body, systemic delivery to leukocytesis even more challenging than the systemic delivery to otherorgans and tissues.

Kortylewski et al [94] used siRNAs synthetically linkedto a CpG oligonucleotide agonist of TLR9 (figure 2(A)) fortargeting myeloid cells and B cells (both are key componentsof the tumor microenvironment) that express this receptor.These particles simultaneously silenced stat3 by siRNA andactivated TLR responses by their agonists. Consequently,they effectively shifted the tumor microenvironment from

7

Nanotechnology 21 (2010) 232001 Topical Review

Figure 2. Schematic representation of targeted delivery systems forleukocytes. (A) siRNA synthetically linked to a CpG (B) schematicillustration of the 9 arginine (9R) strategy either with a scFv (B) orwith a ligand such as DC3 (not shown here) (C) immunoliposomescontaining siRNAs (D) scFv-protamine fusion protein loaded withsiRNAs (E) integrin-targeted stabilized nanoparticle (I-tsNP)entrapping siRNAs.

pro-oncogenic to anti-oncogenic (by causing activation oftumor associated immune cells and potent antitumor immuneresponses). This strategy is a good example of the case inwhich immune activation enhances the therapeutic effect.

Two studies from the same group presented newlydeveloped siRNA delivery systems for treating viral infections(figure 2(B)). scFvCD7-Cys is a single-chain antibody againstCD7 (a surface antigen present on the majority of humanT lymphocytes) that was modified to include a cysteine(Cys) residue for conjugation to a 9 arginine (9R) peptide.This conjugate was used for targeted delivery of CCR5 (achemokine receptor that function as co-receptor for HIV entryinto T lymphocytes) and Vif/Tat (HIV replication proteins)—siRNAs payloads into T cells, and has been demonstrated tosuppress HIV infection in humanized mice without inducingtoxicity in their target cells [95]. A similar approach fortreating dengue virus infected cells employed DC3 (12-merpeptide that targeting dendritic cells (cells that sample theirenvironment))-9R for targeting, with TNF-α (which playsa major role in dengue pathogenesis) or specific highlyconserved sequence in the viral envelope-siRNAs. These

complexes significantly reduced virus induced production ofTNF-α and succeeded in suppressing the viral replication inmonocyte derived dendritic cells and macrophages in vitro.In vivo, treatment of mice with intravenous injection of DC3-9R-complexes carrying TNF-α–siRNAs effectively suppressedthis cytokine production by dendritic cells [96].

Zheng et al [97] have also developed a strategy for specificdelivery of siRNAs into dendritic cells. Their strategy is basedon immunoliposomes decorated with monoclonal antibodyagainst DEC-205 (figure 2(C)), a dendritic cell specific protein.Those particles containing anti-CD40 siRNAs, when injectedintravenously to mice, demonstrated selective siRNA uptake inimmune organs and functional silencing of CD40 that resultedin immune modulation.

Antibody–protamine fusion carriers have also been shownto be an efficient siRNA delivery system to leukocytes.In a fundamental study, Song et al [87] have designed aprotamine–antibody fusion protein by fusing protamine to theC terminus of the heavy chain Fab fragment of an HIV-1 envelope antibody. These particles complexed siRNAstargeted against the HIV-1 capsid gene gag, inhibited HIVreplication in hard to transfect, HIV infected primary Tcells ex vivo. Based on this system, we have developedour approach [98] for targeting leukocytes, which is basedon integrins, cell adhesion molecules that mediate cell–cell and cell–matrix interactions [99]. We have developedsingle-chain variable fragment antibody (scFv)—protaminefusion proteins (figure 2(D)) utilizing the lymphocyte functionassociated antigen-1 (LFA-1) integrin, which is expressed inall leukocytes’ subtypes, for selective targeting. The use ofLFA-1 for targeting leukocytes is supported by its exclusiveexpression on leukocytes, its constitutive internalization andrecycling activity and its ability to undergo activation-dependent conformational changes. Using these antibody–protamine fusion proteins we have demonstrated selectivedelivery of siRNAs into leukocytes both in vitro and in vivo.Importantly, neither lymphocyte activation nor interferonresponse was indicated [98]. Furthermore, by targetingthese fusion proteins to the high affinity conformationof LFA-1 that characterizes activated lymphocytes, wedemonstrated even more selective gene silencing, whichunlike most immunosuppressive therapies, could provide away to overcome the unwanted immune stimulation withoutglobal immunosuppressive effects on bystander immune cells.Additionally, due to the prevalenve of aberrant affinitymodulation of integrins in a variety of leukocyte-implicationdiseases [100, 101], targeting the high affinity conformation ofLFA-1 seems to be very promising therapeutic strategy [98].

Next, in order to increase payload and achieve morerobust targeted gene silencing, we have generated integrin-targeted and stabilized nanoparticles (I-tsNP, figures 2(E)and 3). I-tsNPs were engineered by the guidelines outlinedin box 1. Namely, we have developed neutral lipidbased nanoparticles, made from phospholipids and cholesterolthat was extruded under polycarbonate filters to form auniform ∼80 nm (after extrusion and before adding targetingmoieties) in diameter nanoparticles. The particles have beensurface functionalized with a polysugar named hyaluronan

8

Nanotechnology 21 (2010) 232001 Topical Review

Figure 3. Schematic illustration of the process involved in generating I-tsNPs. A multilamellar vesicle (MLV) is made by the lipid filmmethod and then extrusion is carried out under filters with defined sizes. The extrusion forms an ∼80 nm diameter unilamellar vesicle (ULV)from MLV. The cross-linker EDAC is then used to covalently bind the glycosaminoglycan hyaluronan and attached it to the ULV. Amonoclonal antibody (mAb) to the integrin is covalently attached to hyaluronan-coated nanoparticles by EDAC/NHS reaction and generatesthe I-tsNPs that are purified with a size exclusion column. Rehydrating lyophilized I-tsNPs with water containing protamine-condensedsiRNAs entrap the condensed siRNAs within the nanoparticle.

(HA), a naturally occurring glycosaminoglycan. The HAcoating endow the carriers with long circulation (similar topolyethylene glycol—PEG), a scaffold for antibody bindingand a built-in cryoprotection in a cycle of lyophilization andrehydration [102–105]. The targeting ability of the particleshas been achieved by attaching a monoclonal antibody againstβ7 integrin (which is highly expressed in gut leukocytes)to the HA-coated nanoparticles [105]. Made from naturalbiomaterials, these nanoparticles (final diameter of ∼100 nm)offer a safe platform for siRNA delivery, avoiding cytokineinduction and liver damage. Enabling the usage of low dosesof siRNAs, this system, in addition to advantages such ashigh payload capacity (∼4000 siRNA molecules per particle)and low off-target effects and toxicities, is economicallypromising. To examine the targeting ability specifically tothe gut, we labeled the nanoparticles with non-exchangeable3H–cholesterol and examine the biodistribution to differentorgans in the present of colitis. We found that the particleshave preferential accumulation in the gut (both the colon andthe small intestine) with reduce accumulation in the liver andspleen (figure 4(A)). We have recently shown, using positronemission tomography (PET) imaging, that the targeting agenton the particles’ surface, the antibody against β7 integrin(labeled with 64Cu), selectively targets leukocytes that hometo the gut and therefore could direct the antibody-decoratedparticles into sites of gut inflammation such as colitis [106].Utilizing the I-tsNP system, we identified cyclin D1 (CD1),a regulator protein of the entry into, and the progressionthroughout the cell cycle, as a potential novel target forinflammatory bowel diseases (such as Ulcerative Colitis andCrohn’s disease). Mice that were intravenously injected withsiRNAs against CD1 every two days could suppress intestinalinflammation whilst maintaining their weight compare tocontrol groups that dramatically decreased in bodyweight(figure 4(B)). Interestingly, β7 I-tsNPs with CD1-siRNA shiftsthe inflammatory response from a pro-inflammatory to more ofan anti-inflammatory response, by decreasing the levels of thepro-inflammatory cytokines TNF-α and IL-12p40 compared toall controls (figure 4(C)), hence revealing that CD1 operates asa master regulator of the pro-inflammatory cytokines.

Since I-tsNP is a platform, by changing the surfacedirecting agent and the payload inside the particles, it ispossible to target other cell types and deliver different siRNAsthat will silence specific genes. I-tsNPs surface modified with

an LFA-1 integrin-targeted antibody, were used for deliveryof CCR5-siRNAs to human lymphocytes and monocytes.This system has been shown to protect mice from HIVinfection [107]. LFA-1 I-tsNPs with CCR5-siRNAs did notinduce interferon response or TNF-α (inflammatory cytokine)secretion, hence strengthening their potential for clinicalrelevance.

3. Summary

In this topical review, we described a variety of deliverysystems made using different types of material and provideda rationale and some engineering concepts for designing newnanocarriers for RNAi delivery. Although there is no clinicalapproved RNAi delivery system yet, we are convinced thatin the coming years this situation will change drastically.We base this assumption on one of the major advantages ofRNAi delivery systems—the relative ease of altering themfor purposes other than the original purpose. Changing thetargeting agent in active delivery systems (by replacing theantibody or the ligand decorating the nanoparticle surface)or the payloads inside the nanoparticle (by using differentsequences for targeting different genes, or by changingthe RNAi molecules, for example from siRNA to miRNAmimetics), both in active and passive delivery systems, opensnew avenues for treating a diversity of diseases as wellas adjusting the treatment to the unique molecular geneexpression abnormalities of a specific patient in a personalizedapproach [11]. Still, since no ‘magic bullet’ is available, thereis the need to look for novel, effective methods to deliverRNAi effector molecules to specific cell types. This will bean interdisciplinary effort bringing together material scientists,engineers, chemists, physicists and biologists to solve thisfundamental problem.

Much data has been accumulated in recent yearsconcerning the specific roles of miRNA in health anddisease [10, 108, 109]. Nevertheless, this data still has notbeen incorporated into therapeutic tools. It is likely thatin the near future the existing strategies for siRNA deliverywill also be exploited for therapeutic delivery of miRNAmimetics and antagomirs. This emerging field, consistentwith the miRNA natural function, holds enormous therapeuticpromise, for treating immune system pathologies and others,affecting simultaneously many targets mRNAs, or multiple

9

Nanotechnology 21 (2010) 232001 Topical Review

Figure 4. Representing experimental results with β7 I-tsNP. (A) β7 I-tsNP is a specific RNAi delivery system into gut leukocytes as indicatedby the biodistribution of radiolabeled β7 I-tsNPs in mice with or without colitis. Biodistribution of the nanoparticles was measures 12 h afterinjection in a total of six mice per group in three independent experiments. SP—spleen, PLN—peripheral lymph node, PP—Peyer’s patches(a lymphoid organ in the gut), MLN–mesenteric lymph node, SI—(small intestine), ND—not detectable. (B + C) delivery of CD1-siRNA byβ7 I-tsNP alleviated intestinal inflammation in induced colitis. Mice were treated with intravenous injections of CD1- or luciferase (Luci, acontrol sequence of gene that does not exist in mice)-siRNAs entrapped in either β7 I-tsNPs or IgG-sNPs (nanoparticles covalently attach tonon-binding mAb), or naked CD1-siRNA, at days 0, 2, 4, and 6 (three independent experiments with a total of six mice per group).(B) Changes in body weight (one of the common parameters for mouse health evaluation) due to the different treatments. (C) mRNAexpression of CD1 and cytokines in the gut due to the different treatments. mRNA expression was measured by qRT-PCR with homogenizedcolon samples harvested at day 9 after the first injection. ∗∗ indicate p < 0.01, ∗∗∗ indicate P < 0.001.

gene networks, and hence enabling a more comprehensiveregulation specifying the desired cell phenotype.

Acknowledgments

This work was supported by grants from the Alon Foundation,the Marie Curie IRG-FP7 of the European Union, and theLewis trust USA to Dan Peer.

References

[1] Sledz C A and Williams B R 2005 RNA interference inbiology and disease Blood 106 787–94

[2] de Fougerolles A, Vornlocher H P, Maraganore J andLieberman J 2007 Interfering with disease: a progressreport on siRNA-based therapeutics Nat. Rev. Drug Discov.6 443–53

[3] Liu Z, Sall A and Yang D 2008 MicroRNA: an emergingtherapeutic target and intervention tool Int. J. Mol. Sci.9 978–99

[4] Elbashir S M, Harborth J, Lendeckel W, Yalcin A,Weber K and Tuschl T 2001 Duplexes of 21-nucleotideRNAs mediate RNA interference in cultured mammaliancells Nature 411 494–8

[5] Amarzguioui M, Rossi J J and Kim D 2005 Approaches forchemically synthesized siRNA and vector-mediated RNAiFEBS Lett. 579 5974–81

[6] Bartlett D W and Davis M E 2006 Insights into the kinetics ofsiRNA-mediated gene silencing from live-cell andlive-animal bioluminescent imaging Nucleic Acids Res.34 322–33

[7] Krek A et al 2005 Combinatorial microRNA targetpredictions Nat. Genet. 37 495–500

[8] Singh S K, Pal Bhadra M, Girschick H J and Bhadra U 2008MicroRNAs–micro in size but macro in function FEBS J.275 4929–44

[9] Esau C C and Monia B P 2007 Therapeutic potential formicroRNAs Adv. Drug Deliv. Rev. 59 101–14

10

Nanotechnology 21 (2010) 232001 Topical Review

[10] Love T M, Moffett H F and Novina C D 2008 Not miR-lysmall RNAs: big potential for microRNAs in therapy J. All.Clin. Immunol. 121 309–19

[11] Kedmi R and Peer D 2009 RNAi nanoparticles in the serviceof personalized medicine Nanomedicine 4 853–5

[12] Robbins M, Judge A, Ambegia E, Choi C, Yaworski E,Palmer L, McClintock K and MacLachlan I 2008Misinterpreting the therapeutic effects of small interferingRNA caused by immune stimulation Human Gene. Therapy19 991–9

[13] Kleinman M E et al 2008 Sequence- and target-independentangiogenesis suppression by siRNA via TLR3 Nature452 591–7

[14] Ma M et al 2009 Decreased cofilin1 expression is importantfor compaction during early mouse embryo developmentBiochim. Biophys. Acta 1793 1804–10

[15] Bose S, Leclerc G M, Vasquez-Martinez R and Boockfor F R2010 Administration of connexin43 siRNA abolishessecretory pulse synchronization in GnRH clonal cellpopulations Mol. Cell Endocrinol. 314 75–83

[16] Wiese M, Castiglione K, Hensel M, Schleicher U,Bogdan C and Jantsch J 2009 Small interfering RNA(siRNA) delivery into murine bone marrow-derivedmacrophages cells by electroporation J. Immunol. Methods353 102–10

[17] Honjo K et al 2009 MDR1a/1b gene silencing enhances drugsensitivity in rat fibroblast-like synoviocytes J. Gene Med.12 219–27

[18] Grandjean V, Gounon P, Wagner N, Martin L, Wagner K D,Bernex F, Cuzin F and Rassoulzadegan M 2009 ThemiR-124-Sox9 paramutation: RNA-mediated epigeneticcontrol of embryonic and adult growth Development136 3647–55

[19] Vidic S, Markelc B, Sersa G, Coer A, Kamensek U, Tevz G,Kranjc S and Cemazar M 2010 MicroRNAs targetingmutant K-ras by electrotransfer inhibit human colorectaladenocarcinoma cell growth in vitro and in vivo CancerGene Therapy at press

[20] Donze O and Picard D 2002 RNA interference in mammaliancells using siRNAs synthesized with T7 RNA polymeraseNucleic Acids Res. 30 e46

[21] Zhang M, Bai C X, Zhang X, Chen J, Mao L and Gao L 2004Downregulation enhanced green fluorescence protein geneexpression by RNA interference in mammalian cells RNABiol. 1 74–7

[22] Gosain A K, Machol J A t, Gliniak C and Halligan N L 2009TGF-beta1 RNA interference in mouse primary dura cellculture: downstream effects on TGF receptors, FGF-2, andFGF-R1 mRNA levels Plast. Reconstr. Surg. 124 1466–73

[23] Cheng S Q, Wang W L, Yan W, Li Q L, Wang L andWang W Y 2005 Knockdown of survivin gene expressionby RNAi induces apoptosis in human hepatocellularcarcinoma cell line SMMC-7721 World J. Gastroenterol.11 756–9

[24] Baker B E, Kestler D P and Ichiki A T 2006 Effects of siRNAsin combination with Gleevec on K-562 cell proliferationand Bcr-Abl expression J. Biomed. Sci. 13 499–507

[25] Karimi M H, Ebadi P, Pourfathollah A A, Soheili Z S,Samiee S, Ataee Z, Tabei S Z and Moazzeni S M 2009Immune modulation through RNA interference-mediatedsilencing of CD40 in dendritic cells Cell Immunol.259 74–81

[26] Zhou J Y, Ma W L, Fei J, Ding D P, Shi R, Jiang L andZheng W L 2006 Effects of microRNA miR-181a on geneexpression profiles of K562 cells Nan Fang Yi Ke Da XueXue Bao 26 606–9

[27] Migliore C, Petrelli A, Ghiso E, Corso S, Capparuccia L,Eramo A, Comoglio P M and Giordano S 2008MicroRNAs impair MET-mediated invasive growth CancerRes. 68 10128–36

[28] Crombez L, Charnet A, Morris M C, Aldrian-Herrada G,Heitz F and Divita G 2007 A non-covalent peptide-basedstrategy for siRNA delivery Biochem. Soc. Trans. 35 44–6

[29] Muratovska A and Eccles M R 2004 Conjugate for efficientdelivery of short interfering RNA (siRNA) into mammaliancells FEBS Lett. 558 63–8

[30] Minakuchi Y et al 2004 Atelocollagen-mediated syntheticsmall interfering RNA delivery for effective gene silencingin vitro and in vivo Nucleic Acids Res. 32 e109

[31] Takei Y, Kadomatsu K, Yuzawa Y, Matsuo S andMuramatsu T 2004 A small interfering RNA targetingvascular endothelial growth factor as cancer therapeuticsCancer Res. 64 3365–70

[32] Puebla I, Esseghir S, Mortlock A, Brown A, Crisanti A andLow W 2003 A recombinant H1 histone-based system forefficient delivery of nucleic acids J. Biotechnol.105 215–26

[33] Bitko V, Musiyenko A, Shulyayeva O and Barik S 2005Inhibition of respiratory viruses by nasally administeredsiRNA Nat. Med. 11 50–5

[34] Palliser D, Chowdhury D, Wang Q Y, Lee S J, Bronson R T,Knipe D M and Lieberman J 2006 An siRNA-basedmicrobicide protects mice from lethal herpes simplex virus2 infection Nature 439 89–94

[35] Kim B et al 2004 Inhibition of ocular angiogenesis by siRNAtargeting vascular endothelial growth factor pathway genes:therapeutic strategy for herpetic stromal keratitis Am. J.Pathol. 165 2177–85

[36] Trang P et al 2009 Regression of murine lung tumors by thelet-7 microRNA Oncogene 29 1580–7

[37] Nagata Y, Nakasa T, Mochizuki Y, Ishikawa M, Miyaki S,Shibuya H, Yamasaki K, Adachi N, Asahara H andOchi M 2009 Induction of apoptosis in the synovium ofmice with autoantibody-mediated arthritis by theintraarticular injection of double-stranded MicroRNA-15aArthritis Rheum 60 2677–83

[38] Novobrantseva T I, Akinc A, Borodovsky A andde Fougerolles A 2008 Delivering silence: advancements indeveloping siRNA therapeutics Curr. Opin. Drug Discov.Devel. 11 217–24

[39] Peer D and Shimaoka M 2009 Systemic siRNA delivery toleukocyte-implicated diseases Cell Cycle 8 853–9

[40] Zender L and Kubicka S 2007 Suppression of apoptosis in theliver by systemic and local delivery of small-interferingRNAs Methods Mol. Biol. 361 217–26

[41] De Souza A T et al 2006 Transcriptional and phenotypiccomparisons of Ppara knockout and siRNA knockdownmice Nucleic Acids Res. 34 4486–94

[42] Saito Y et al 2007 Osteopontin small interfering RNA protectsmice from fulminant hepatitis Human Gene. Therapy18 1205–14

[43] Lewis D L and Wolff J A 2007 Systemic siRNA delivery viahydrodynamic intravascular injection Adv. Drug Deliv. Rev.59 115–23

[44] Molitoris B A et al 2009 siRNA targeted to p53 attenuatesischemic and cisplatin-induced acute kidney injury J. Am.Soc. Nephrol. 20 1754–64

[45] Braasch D A, Jensen S, Liu Y, Kaur K, Arar K, White M Aand Corey D R 2003 RNA interference in mammalian cellsby chemically-modified RNA Biochemistry 42 7967–75

[46] Mook O R, Baas F, de Wissel M B and Fluiter K 2007Evaluation of locked nucleic acid-modified smallinterfering RNA in vitro and in vivo Mol. Cancer Ther.6 833–43

[47] Glud S Z, Bramsen J B, Dagnaes-Hansen F, Wengel J,Howard K A, Nyengaard J R and Kjems J 2009 NakedsiLNA-mediated gene silencing of lung bronchoepitheliumEGFP expression after intravenous administrationOligonucleotides 19 163–8

11

Nanotechnology 21 (2010) 232001 Topical Review

[48] Ferrari M 2005 Cancer nanotechnology: opportunities andchallenges Nat. Rev. Cancer 5 161–71

[49] LaVan D A, McGuire T and Langer R 2003 Small-scalesystems for in vivo drug delivery Nat. Biotechnol.21 1184–91

[50] Whitesides G M 2003 The ‘right’ size in nanobiotechnologyNat. Biotechnol. 21 1161–5

[51] Peer D, Karp J M, Hong S, Farokhzad O C, Margalit R andLanger R 2007 Nanocarriers as an emerging platform forcancer therapy Nat. Nanotechnol. 2 751–60

[52] Iyer A K, Khaled G, Fang J and Maeda H 2006 Exploiting theenhanced permeability and retention effect for tumortargeting Drug Discov. Today 11 812–8

[53] Couvreur P and Vauthier C 2006 Nanotechnology: intelligentdesign to treat complex disease Pharm. Res. 23 1417–50

[54] Yuan F, Dellian M, Fukumura D, Leunig M, Berk D A,Torchilin V P and Jain R K 1995 Vascular permeability in ahuman tumor xenograft: molecular size dependence andcutoff size Cancer Res. 55 3752–6

[55] Torchilin V P 2005 Recent advances with liposomes aspharmaceutical carriers Nat. Rev. Drug Discov. 4 145–60

[56] Hobbs S K, Monsky W L, Yuan F, Roberts W G, Griffith L,Torchilin V P and Jain R K 1998 Regulation of transportpathways in tumor vessels: role of tumor type andmicroenvironment Proc. Natl Acad. Sci. USA 95 4607–12

[57] Maeda H, Sawa T and Konno T 2001 Mechanism oftumor-targeted delivery of macromolecular drugs,including the EPR effect in solid tumor and clinicaloverview of the prototype polymeric drug SMANCSJ. Control Release 74 47–61

[58] Matsumura Y and Maeda H 1986 A new concept formacromolecular therapeutics in cancer chemotherapy:mechanism of tumoritropic accumulation of proteins andthe antitumor agent smancs Cancer Res. 46 6387–92

[59] Zimmermann T S et al 2006 RNAi-mediated gene silencing innon-human primates Nature 441 111–4

[60] Morrissey D V et al 2005 Potent and persistent in vivoanti-HBV activity of chemically modified siRNAs Nat.Biotechnol. 23 1002–7

[61] Lv H, Zhang S, Wang B, Cui S and Yan J 2006 Toxicity ofcationic lipids and cationic polymers in gene deliveryJ. Control Release 114 100–9

[62] Ma Z, Li J, He F, Wilson A, Pitt B and Li S 2005 Cationiclipids enhance siRNA-mediated interferon response in miceBiochem. Biophys. Res. Commun. 330 755–9

[63] Geisbert T W et al 2006 Postexposure protection of guineapigs against a lethal ebola virus challenge is conferred byRNA interference J. Infect Dis. 193 1650–7

[64] Semple S C et al 2010 Rational design of cationic lipids forsiRNA delivery Nat. Biotechnol. 28 172–6

[65] Villares G J et al 2008 Targeting melanoma growth andmetastasis with systemic delivery of liposome-incorporatedprotease-activated receptor-1 small interfering RNACancer Res. 68 9078–86

[66] Landen C N Jr, Chavez-Reyes A, Bucana C, Schmandt R,Deavers M T, Lopez-Berestein G and Sood A K 2005Therapeutic EphA2 gene targeting in vivo using neutralliposomal small interfering RNA delivery Cancer Res.65 6910–8

[67] Akinc A et al 2008 A combinatorial library of lipid-likematerials for delivery of RNAi therapeutics Nat.Biotechnol. 26 561–9

[68] Love K T et al 2010 Lipid-like materials for low-dose, in vivogene silencing Proc. Natl Acad. Sci. USA 107 1864–9

[69] Leng Q, Goldgeier L, Zhu J, Cambell P, Ambulos N andMixson A J 2007 Histidine-lysine peptides as carriers ofnucleic acids Drug News Perspect. 20 77–86

[70] Leng Q, Scaria P, Lu P, Woodle M C and Mixson A J 2008Systemic delivery of HK Raf-1 siRNA polyplexes inhibitsMDA-MB-435 xenografts Cancer Gene Therapy15 485–95

[71] Yan Z, Zou H, Tian F, Grandis J R, Mixson A J, Lu P Y andLi L Y 2008 Human rhomboid family-1 gene silencingcauses apoptosis or autophagy to epithelial cancer cells andinhibits xenograft tumor growth Mol. Cancer Ther.7 1355–64

[72] Ochiya T, Takahama Y, Nagahara S, Sumita Y, Hisada A,Itoh H, Nagai Y and Terada M 1999 New delivery systemfor plasmid DNA in vivo using atelocollagen as a carriermaterial: the Minipellet Nat. Med. 5 707–10

[73] Honma K, Iwao-Koizumi K, Takeshita F, Yamamoto Y,Yoshida T, Nishio K, Nagahara S, Kato K andOchiya T 2008 RPN2 gene confers docetaxel resistance inbreast cancer Nat. Med. 14 939–48

[74] Fujii T et al 2006 Intratumor injection of small interferingRNA-targeting human papillomavirus 18 E6 and E7successfully inhibits the growth of cervical cancer Int. J.Oncol. 29 541–8

[75] Mu P, Nagahara S, Makita N, Tarumi Y, Kadomatsu K andTakei Y 2009 Systemic delivery of siRNA specific to tumormediated by atelocollagen: combined therapy using siRNAtargeting Bcl-xL and cisplatin against prostate cancer Int. J.Cancer 125 2978–90

[76] Takeshita F et al 2010 Systemic delivery of syntheticmicroRNA-16 inhibits the growth of metastatic prostatetumors via downregulation of multiple cell-cycle genesMol. Ther. 18 181–7

[77] Wolfrum C et al 2007 Mechanisms and optimization of invivo delivery of lipophilic siRNAs Nat. Biotechnol.25 1149–57

[78] Nishina K, Unno T, Uno Y, Kubodera T, Kanouchi T,Mizusawa H and Yokota T 2008 Efficient in vivo deliveryof siRNA to the liver by conjugation of alpha-tocopherolMol. Ther. 16 734–40

[79] Lorenz C, Hadwiger P, John M, Vornlocher H P andUnverzagt C 2004 Steroid and lipid conjugates of siRNAsto enhance cellular uptake and gene silencing in liver cellsBioorg. Med. Chem. Lett. 14 4975–7

[80] Krutzfeldt J, Rajewsky N, Braich R, Rajeev K G, Tuschl T,Manoharan M and Stoffel M 2005 Silencing of microRNAsin vivo with ‘antagomirs’ Nature 438 685–9

[81] Rozema D B et al 2007 Dynamic PolyConjugates for targetedin vivo delivery of siRNA to hepatocytes Proc. Natl Acad.Sci. USA 104 12982–7

[82] Schiffelers R M, Ansari A, Xu J, Zhou Q, Tang Q, Storm G,Molema G, Lu P Y, Scaria P V and Woodle M C 2004Cancer siRNA therapy by tumor selective delivery withligand-targeted sterically stabilized nanoparticle NucleicAcids Res. 32 e149

[83] Hu-Lieskovan S, Heidel J D, Bartlett D W, Davis M E andTriche T J 2005 Sequence-specific knockdown ofEWS-FLI1 by targeted, nonviral delivery of smallinterfering RNA inhibits tumor growth in a murine modelof metastatic Ewing’s sarcoma Cancer Res. 65 8984–92

[84] Davis M E 2009 The first targeted delivery of siRNA inhumans via a self-assembling, cyclodextrin polymer-basednanoparticle: from concept to clinic Mol. Pharm. 6 659–68

[85] Davis M E, Zuckerman J E, Choi C H, Seligson D, Tolcher A,Alabi C A, Yen Y, Heidel J D and Ribas A 2010 Evidenceof RNAi in humans from systemically administered siRNAvia targeted nanoparticles Nature 464 1067–70

[86] Andrabi S M 2007 Mammalian sperm chromatin structure andassessment of DNA fragmentation J. Assist. Reprod. Genet.24 561–9

12

Nanotechnology 21 (2010) 232001 Topical Review

[87] Song E et al 2005 Antibody mediated in vivo delivery ofsmall interfering RNAs via cell-surface receptors Nat.Biotechnol. 23 709–17

[88] Kumar P, Wu H, McBride J L, Jung K E, Kim M H,Davidson B L, Lee S K, Shankar P and Manjunath N 2007Transvascular delivery of small interfering RNA to thecentral nervous system Nature 448 39–43

[89] McNamara J O 2nd, Andrechek E R, Wang Y, Viles K D,Rempel R E, Gilboa E, Sullenger B A and Giangrande P H2006 Cell type-specific delivery of siRNAs withaptamer–siRNA chimeras Nat. Biotechnol. 24 1005–15

[90] Dassie J P, Liu X Y, Thomas G S, Whitaker R M, Thiel K W,Stockdale K R, Meyerholz D K, McCaffrey A P,McNamara J O 2nd and Giangrande P H 2009 Systemicadministration of optimized aptamer–siRNA chimeraspromotes regression of PSMA-expressing tumors Nat.Biotechnol. 27 839–49

[91] Sato Y et al 2008 Resolution of liver cirrhosis using vitaminA-coupled liposomes to deliver siRNA against acollagen-specific chaperone Nat. Biotechnol. 26 431–42

[92] Pirollo K F, Rait A, Zhou Q, Hwang S H, Dagata J A, Zon G,Hogrefe R I, Palchik G and Chang E H 2007 Materializingthe potential of small interfering RNA via a tumor-targetingnanodelivery system Cancer Res. 67 2938–43

[93] Li S D, Chen Y C, Hackett M J and Huang L 2008Tumor-targeted delivery of siRNA by self-assemblednanoparticles Mol. Ther. 16 163–9

[94] Kortylewski M et al 2009 In vivo delivery of siRNA toimmune cells by conjugation to a TLR9 agonist enhancesantitumor immune responses Nat. Biotechnol. 27 925–32

[95] Kumar P et al 2008 T cell-specific siRNA delivery suppressesHIV-1 infection in humanized mice Cell 134 577–86

[96] Subramanya S et al 2009 Targeted delivery of siRNA tohuman dendritic cells to suppress Dengue viral infectionand associated proinflammatory cytokine productionJ. Virol. 84 2490–501

[97] Zheng X et al 2009 A novel in vivo siRNA delivery systemspecifically targeting dendritic cells and silencing CD40genes for immunomodulation Blood 113 2646–54

[98] Peer D, Zhu P, Carman C V, Lieberman J andShimaoka M 2007 Selective gene silencing in activatedleukocytes by targeting siRNAs to the integrin lymphocyte

function-associated antigen-1 Proc. Natl Acad. Sci. USA104 4095–100

[99] Shimaoka M, Takagi J and Springer T A 2002 Conformationalregulation of integrin structure and function Annu. Rev.Biophys. Biomol. Struct. 31 485–516

[100] Cauli A, Yanni G, Pitzalis C, Challacombe S and Panayi G S1995 Cytokine and adhesion molecule expression in theminor salivary glands of patients with Sjogren’s syndromeand chronic sialoadenitis Ann. Rheum. Dis. 54 209–15

[101] Tanaka Y et al 1998 Constitutive chemokine productionresults in activation of leukocyte function-associatedantigen-1 on adult T-cell leukemia cells Blood 91 3909–19

[102] Peer D, Florentin A and Margalit R 2003 Hyaluronan is a keycomponent in cryoprotection and formulation of targetedunilamellar liposomes Biochim. Biophys. Acta 1612 76–82

[103] Peer D and Margalit R 2004 Loading mitomycin C inside longcirculating hyaluronan targeted nano-liposomes increasesits antitumor activity in three mice tumor models Int. J.Cancer 108 780–9

[104] Peer D and Margalit R 2004 Tumor-targeted hyaluronannanoliposomes increase the antitumor activity of liposomalDoxorubicin in syngeneic and human xenograft mousetumor models Neoplasia 6 343–53

[105] Peer D, Park E J, Morishita Y, Carman C V andShimaoka M 2008 Systemic leukocyte-directed siRNAdelivery revealing cyclin D1 as an anti-inflammatory targetScience 319 627–30

[106] Dearling J L, Park E J, Dunning P, Baker A, Fahey F,Treves S T, Soriano S G, Shimaoka M, Packard A B andPeer D 2010 Detection of intestinal inflammation byMicroPET imaging using a (64)Cu-labeled anti-beta(7)integrin antibody Inflamm. Bowel Dis. at press

[107] Kim S S et al 2009 RNAi-mediated CCR5 Silencing byLFA-1-targeted Nanoparticles Prevents HIV Infection inBLT Mice Mol. Ther. 18 370–6

[108] Tong A W and Nemunaitis J 2008 Modulation of miRNAactivity in human cancer: a new paradigm for cancer genetherapy? Cancer Gene Therapy 15 341–55

[109] O’Connell R M, Rao D S, Chaudhuri A A andBaltimore D 2010 Physiological and pathological roles formicroRNAs in the immune system Nat. Rev. Immunol.10 111–22

13