Delivery of siRNA to the Mouse Lung via a Functionalized Lipopolyamine

original article

© The American Society of Gene & Cell Therapy

Delivery of siRNA to the Mouse Lung via a Functionalized Lipopolyamine Kevin J Polach1, Majed Matar1, Jennifer Rice1, Gregory Slobodkin1, Jeff Sparks1, Richard Congo1, Angela Rea-Ramsey1, Diane McClure1, Elaine Brunhoeber1, Monika Krampert2, Andrea Schuster2, Kerstin Jahn-Hofmann2, Matthias John2, Hans-Peter Vornlocher2, Jason G Fewell1, Khursheed Anwer1 and Anke Geick2 EGEN Inc., Huntsville, Alabama, USA; 2Roche Kulmbach GmbH, Kulmbach, Germany

1

We have designed a series of versatile lipopolyamines which are amenable to chemical modification for in vivo delivery of small interfering RNA (siRNA). This report focuses on one such lipopolyamine (Staramine), its functionalized derivatives and the lipid nanocomplexes it forms with siRNA. Intravenous (i.v.) administration of Staramine/siRNA nanocomplexes modified with methoxypolyethylene glycol (mPEG) provides safe and effective delivery of siRNA and significant target gene knockdown in the lungs of normal mice, with much lower knockdown in liver, spleen, and kidney. Although siRNA delivered via Staramine is initially distributed across all these organs, the observed clearance rate from the lung tissue is considerably slower than in other tissues resulting in prolonged siRNA accumulation on the timescale of RNA interference (RNAi)-mediated transcript depletion. Complete blood count (CBC) analysis, serum chemistry analysis, and histopathology results are all consistent with minimal toxicity. An in vivo screen of mPEG modified Staramine nanocomplexes-containing siRNAs targeting lung cell-specific marker proteins reveal exclusive transfection of endothelial cells. Safe and effective delivery of siRNA to the lung with chemically versatile lipopolyamine systems provides opportunities for investigation of pulmonary cell function in vivo as well as potential treatments of pulmonary disease with RNAibased therapeutics. Received 15 June 2011; accepted 2 September 2011; published online 11 October 2011. doi:10.1038/mt.2011.210

Introduction The safe and efficient delivery of nucleic acids to target cells in vivo remains a fundamental problem for the development of RNA- and DNA-based therapeutics. The RNA interference (RNAi) pathway offers the potential to advance the treatment of disease through the specific silencing of gene products not “druggable” by conventional therapies.1–3 This specificity is provided through base pairing of small interfering RNAs (siRNAs) with target mRNA transcripts, thus making RNAi-based therapeutics accessible to rational design. In addition, the molecular machinery responsible

for RNAi-mediated gene silencing is ubiquitous across many cell types allowing for intervention with many types of disease provided the siRNA can be delivered into the cytoplasm of target cells within the required tissue. Solving the complexity of siRNA delivery is the focus of ongoing research4–18 where approaches can be grouped into two categories based on the route of administration: local delivery directly to tissues of interest and systemic delivery to a broad range of tissues. Cationic lipid nanocomplexes have received considerable attention as systemic delivery vehicles for siRNA as they offer protection from nuclease degradation in circulation, increase the siRNA residence time in the blood, mediate interactions with negatively charged nucleic acid cargo and target cell membranes and promote cellular uptake by endocytosis.7,19,20 Delivery via lipid nanocomplexes shifts siRNA biodistribution from the kidneys, the site of accumulation and clearance for “naked” siRNA upon intravenous (i.v.) injection, to other tissues including the lung, liver, and spleen.20 In vivo application of cationic lipid delivery systems by i.v. injection faces three major obstacles: (i) inefficient delivery, as the required dose of complex often exceeds the amount required for activity by orders of magnitude, (ii) systemic toxicity and innate immune responses,21,22 as the highly charged lipid nanocomplexes interact with opsonizing proteins, and (iii) siRNA accumulation in and clearance from the liver, which limit applications to other target tissues. Potentially, each of these issues may be addressed through covalent modification of the lipids with chemical and biological moieties that alter the in vivo behavior of the lipid nanocomplexes. This general approach has been used in other systems which show target gene knockdown after i.v. injection.7,23,24 Therapeutic applications of siRNA have appeared in clinical trials and include potential treatments for macular degeneration, respiratory syncytial virus infection, liver cancer, and other solid tumors and hypercholesterolemia.17 We have developed a lipopolyamine (Staramine) for in vivo delivery of siRNA. An essential feature of Staramine is that it is amenable to covalent modification which allows the introduction of functional groups to improve the safety and efficiency of siRNA delivery for in vivo applications. In this article, we describe a functionalized Staramine formulation that provides for safe and effective delivery of siRNA to lung endothelium following intravenous administration. The physicochemical properties, in vivo

Correspondence: Kevin J Polach, EGEN Inc., 601 Genome Way, Suite 3100, Huntsville, Alabama 35806, USA. E-mail: [email protected] Molecular Therapy vol. 20 no. 1, 91–100 jan. 2012

91

© The American Society of Gene & Cell Therapy

Staramine-mediated siRNA Delivery

Results Generation of Staramine nanocomplexes The primary goal in the synthesis strategy was to produce a highly effective siRNA delivery platform based on a lipopolyamine core structure (Staramine) and its functionalized derivatives (Figure  1). We synthesized two modified Staramine molecules by covalent attachment of methoxypolyethylene glycol (mPEG): Star-mPEG550, a polydisperse mPEG with an average molecular weight of 550 Da and Star-mPEG515, a monodisperse mPEG with a precise molecular weight of 515 Da. Nanocomplexes were formed with Staramine and the mPEG modified Staramine (10:1 molar ratio) and siRNA (20:1 molar ratio). Several studies presented here were performed with complexes containing either nonspecific control siRNAs (siNon) or siRNAs targeting the Caveolin-1 transcript (siCav-1), a widely expressed gene essential to caveolae formation. The average particle size and zeta-potential for these lipid nanocomplexes from multiple independent preparations are listed in Table 1. In all cases, the particle size measured for the nanocomplexes was ~100 nm; the addition of either mPEG modifier led to a partial reduction in zeta-potential but all formulations remained cationic (25–45 mV). Nanocomplexes were formed equally well with all siRNAs tested (Supplementary Figure S1) and resulted in particles with similar physicochemical properties (Supplementary Table S1).

following i.v. injections of Star:Star-mPEG550/siCav-1 nanocomplexes into ICR mice. Chemokine and cytokine transcript levels and cytokine protein levels were determined through quantitative reverse transcription PCR (qPCR) and enzyme-linked immunosorbent assay (ELISA), respectively (Figure 2a,b). Upon administration of nanocomplexes (40 µg siRNA per animal), there were modest changes in the levels of inflammatory chemokines and their mRNA transcripts in the lungs compared to a 5% dextrose injection. A volcano plot comparing the changes in gene expression to the P value associated with each data set revealed four genes with highly significant changes in expression of greater than fourfold: Ccl2, Ccl7, Ccl12, and IL1r2 (Figure  2a). Circulating cytokine protein levels were measured by enzyme-linked immunosorbent assay protein analysis of whole blood collected and pooled from four treated animals at 4 hours postinjection (Figure  2b). Among the cytokines assessed in this study, only granulocyte colony-stimulating factor showed an increase upon treatment with mPEG550-modified Staramine nanocomplexes

a

1.E-05 1.E-04

p Value

distribution, safety, gene silencing efficacy, and potential therapeutic applications of this lung siRNA delivery system are described.

IL1r2 Ccl12

1.E-03

Ccl7 Ccl2

1.E-02 1.E-01

Assessment of toxicity of Staramine and mPEG modified Staramine nanocomplexes As cationic delivery systems are often triggers of systemic toxicity, local and systemic inflammatory responses were evaluated

1

3

5

7

9

Table 1  Physical properties of various nanocomplexes Particle size (nm)

Zeta-potential (mV)

Staramine

103.6 ± 14.9

44.5 ± 12.7

Star:Star-mPEG550

94.0 ± 27.7

24.6 ± 7.0

Star:Star-mPEG515

91.6 ± 16.2

33.5 ± 6.3

Abbreviations: mPEG, methoxypolyethylene glycol. Mean ± SD.

g

Fa G CS CM F -C SF

TN

N-

A

Cytokine

IF

12

17

IL

10

IL

IL

Figure 1 Structures for the Staramine and methoxypolyethylene glycol (mPEG) modified Staramine lipopolyamines. For monodisperse mPEG515, n = 11. For polydisperse mPEG550, n is reported as the average number of ethylene glycol units (11.75), based on the manufacturer’s specified molecular weight.

IL

−5

6

0 4

n

5

IL

O

10

IL

O

O

15

2

O

20

IL

H N

25

−10

92

−1

30

1B

N

O

Nanocomplexes

−3

40

1A

H N H N

b

IL

N

NH2

O

O

−5

Log2 (Fold Difference)

Relative absorbance units (StarPEG550/Dextrose)

H N

−7

35

H N O

1.E+00

Figure 2 Toxicity profiles with Star:Star-mPEG550/siCav-1 nanocomplexes. Injections of Star:Star-mPEG550/siCav-1 nanocomplexes result in minimal systemic toxicity as assessed through chemokine and cytokine expression levels. (a) Chemokine and cytokine transcript levels were determined by quantitative reverse transcription PCR (qPCR) arrays with total RNA purified from homogenized lung tissue isolated 4 hours postinjection from five independent experiments. Results are plotted as a volcano plot depicting fold expression change (x-axis, vertical lines represent fourfold changes) versus t-test P values (y-axis, horizontal line represents P value = 0.01). (b) Cytokine expression levels were also assessed by protein assay (ELISA) for blood samples collected 4 hours postinjection; blood samples collected from four animals were pooled for single point analysis. Results are expressed as fold changes in absorbance between Star:Star-mPEG550 nanocomplex injections and 5% dextrose injections. mPEG, methoxypolyethylene glycol.

www.moleculartherapy.org vol. 20 no. 1 jan. 2012

© The American Society of Gene & Cell Therapy

Staramine-mediated siRNA Delivery

a

Star:Star-mPEG550 particles

5% dextrose

Relative Value

100 90 80 70 60 50 40 30 20 10 0

p = 0.052

L)

./u

BC

u ho

(t

BC

)

uL

il/

(m

B

L)

/d

(g

b

T

C

H

G

H

R

W

(%

)

eu

N

%

l(

hi

p tro

2000 1800 1600 1400 1200 1000 800 600 400 200 0

p = 0.078

)

(%

te

y oc

ph

m Ly

)

o

on

)

)

%

l(

hi

p no

si

Eo

M

5% dextrose

(%

te

y oc

Platelet count (thous./uL)

) %

l(

hi

p so

p = 0.11

Ba

Star: Star-mPEG550 particles

300 250 Relative Value

compared to the dextrose injection. Cytokines typically induced via the innate immune response, interferon-α, interleukin-6, and tumor necrosis factor-α showed no significant change in expression at 6 hours postinjection (data not shown), consistent with PCR array analysis. Complete blood count (CBC) analysis of mouse serum suggested a twofold increase in the average number of circulating neutrophils (Figure  3a), though this change was within the experimental error of the collected data sets (P = 0.052). No significant change in lymphocyte, monocyte or platelet populations were observed (P = 0.078, 0.068, and 0.11, respectively). Serum chemistry analysis revealed no significant increase in aspartate aminotransferase or alanine transaminase levels that typically accompany liver toxicity and no change in creatinine or urea (BUN) levels indicative of compromised renal function (Figure 3b). A comparison of acute toxicities resulting from injections of Staramine, Star:Star-mPEG550, and Star:Star-mPEG515 nanocomplexes (measured through a toxicity index derived from cage-side observations) show mitigation of the dose-dependent toxicity obtained with Staramine nanocomplexes upon addition of mPEG (Supplementary Figure S2) consistent with improved safety. Immunohistochemical analysis of lung tissue after treatment with Star:Star-mPEG550/siGFP nanocomplexes was performed to evaluate infiltration of peripheral neutrophils (Figure 4). In mice treated with Star:Star-mPEG550/siRNA nanocomplexes, Gr-1 (a marker for granulocytes) positive cells were detected almost exclusively in the blood vessels (Figure 4a), similar to treatment with phosphate-buffered saline (Figure  4b). In contrast, in animals treated with lipopolysaccharide inhalation as positive control for lung inflammation a significant amount of Gr-1 positive cells was found in the alveolar area (Figure 4c). Omission of the Gr-1 antibody has demonstrated the specificity of the observed signals (Figure  4d). Hematoxylin and eosin (H&E) staining for tissues collected at 72 hours postinjection revealed no significant morphological changes for tissues treated with Star:Star-

200 150 100 50 0 )

a at

(U

T AL

ph

K

AL

s ho

) /L

)

/L

/L

se

(U

ST

A

l ta

To

P

L)

n

ei

ot

pr

(g

)

g/

n

Bu

g/

(m

C

e

in

in

at

re

)

g/

(m

se

(m

co

lu

G

)

dL

dL

dL

/d

(U

Figure 3 Complete blood count (CBC) and serum chemistry analysis with Star:Star-mPEG550/siCav-1 nanoparticles. Whole blood and serum chemistry analyses were assessed for mice treated with either 5% dextrose or Star:Star-mPEG550/siCav-1 nanocomplexes [single intravenous (i.v.) injection of 40 µg]. Samples were collected 24 hours postinjection for (a) CBC analysis and (b) serum chemistry analysis as described in the Materials and Methods section; results plot averages and SD for samples from five animals. mPEG, methoxypolyethylene glycol.

a

c

e

b

d

f

Figure 4 Lung histopathology with intravenous (i.v.) injection of Star:Star-mPEG550/siGFP nanocomplexes. Neutrophil infiltration was examined by staining lung tissue sections from mice treated with either (a) Star:Star-mPEG550/siGFP nanocomplexes or (b) phosphate-buffered saline (PBS) with an anti Gr-1 antibody. (c) Sections from lipopolysaccharide (LPS)-treated animals (20 minutes inhalation of 0.5 mg/ml LPS) were used as a positive control for lung inflammation. (d) Negative control tissues generated by omission of GR-1 antibody in staining procedure. Hematoxylin and eosin (H&E) staining of lung tissues collected 72 hours postinjection with either (e) Star:Star-mPEG550/siGFP nanocomplexes or (f) PBS. mPEG, methoxypolyethylene glycol.

Molecular Therapy vol. 20 no. 1 jan. 2012

93

© The American Society of Gene & Cell Therapy

Staramine-mediated siRNA Delivery

nanocomplexes (Figure 5b) or Star:Star-mPEG550/siCav-1 nanocomplexes (Figure 5c) revealed a persistence of the siRNA in lung tissue that exceeded that observed in the liver, spleen, and kidney. For the Staramine/siCav-1 nanocomplexes, nearly half of the material present in the lung at the 1-hour time point remained at the 24-hour time point, whereas most of the accumulated siRNA was cleared from the liver, spleen, and kidney over that same time frame. The clearance rate of siRNA with the PEGylated particles was again slower in the lung than in other tissues. Target mRNA knockdown was examined in lung and liver tissue at 24, 48, and 96 hours after a single i.v. administration of Staramine/siCav-1 nanocomplexes or Star:Star-mPEG550/ siCav-1 nanocomplexes (40 µg siRNA total dose). Identical injections were performed with complexes containing either the nonspecific control siRNA or the Cav-1 targeting siRNA. The Cav-1 expression levels were determined by qPCR from whole-organ homogenates of lung and liver tissue. For Staramine/siCav-1 nanocomplexes, an ~60% reduction of the Cav-1 transcript is evident at 48 hours postinjection compared to Staramine/ siNon-treated animals (Figure 5d); this knockdown persists for 96 hours. For the Star:Star-mPEG550/siCav-1 nanocomplexes, a similar level of Cav-1 knockdown is evident at 48 hours, but appears to subside at 96 hours. Cav-1 transcript levels in the liver are not significantly different than that observed with the nonspecific control siRNA.

mPEG550 particles (Figure 4e) compared to phosphate-buffered saline injections (Figure 4f). Similar analysis of liver tissue (H&E staining, Gr-1 staining, and F4/80 staining for macrophages) and spleen tissue (H&E staining and F4/80 staining) at 72 hours postinjection also showed no significant infiltration of macrophages (Supplementary Figures S3 and S4). Some splenomegaly was observed with Star:Star-mPEG550/siGFP nanocomplex treatment compared to phosphate-buffered saline injections.

Target gene knockdown and biodistribution with PEGylated Staramine systems In order to compare the in vivo behavior of the core Staramine system and the mPEG-modified Staramine system, in vivo activity assays were assessed following i.v. injection of Staramine/ siCav-1 nanocomplexes and Star:Star-mPEG550/siCav-1 nanocomplexes. The initial clearance of i.v. injected Staramine/siCav-1 nanocomplexes (40 µg siRNA) from the blood and corresponding accumulation in the lung and liver was assessed for samples collected 5 minutes after injection (Figure 5a) using a stem-loop qPCR assay. At this early time, the majority of the siRNA appeared in the liver tissue, with lower levels observed in the lungs and blood. In order to develop a more complete picture of the in vivo biodistribution, changes in siRNA accumulation were assessed for multiple organs as a function of time. Kinetic analyses of the biodistributions for siRNA delivered by both Staramine/siCav-1

a

Lung

b

Blood

1

Lung

Lung

0.01

20

40

60

60

80

0.01 0.001

0

20

d

Spleen

Kidney

0.1

0

Spleen

40

100

Time (hours post-injection)

1

0.001

Kidney

0.1

Percent of dextrose treated sample

10

Liver

1

Staramine biodistribution

c

Lung

10

Micrograms of siRNA / gram tissue

Micrograms of siRNA

10

0.1

Micrograms of siRNA / gram tissue

Liver

100

80

Time (hours post-injection)

100

siNon

160 140 120 100 80 60 40 20 0

* ** ***

siCav−1

*** *** *

24 48 96 24 48 96 24 48 96 24 48 96 Lung

Liver

Staramine

Lung

Liver

Star:Star−mPEG550

Figure 5 Small interfering RNA (siRNA) biodistribution and target gene knockdown with Staramine and Star:Star-mPEG550/siCav-1 nanocomplexes. Systemic administration of Staramine/siCav-1 nanocomplexes results in siRNA retention and target gene knockdown in the lung tissue. The distribution of siRNA and subsequent target gene knockdown was determined across mouse tissues following intravenous (i.v.) injection of Staramine/siCav-1 and Star:Star-mPEG550/siCav-1 nanocomplexes into the tail vein of ICR mice. Results depict averages and SD for three experiments. (a) Initial distributions of siRNA across lung, liver, and blood were determined via the stem-loop quantitative reverse transcription PCR (qPCR) assay with tissues collected 5 minutes postinjection. Injections of (b) Staramine/siCav-1 nanocomplexes or (c) Star:Star-mPEG550/siCav-1 nanocomplexes were followed by measurements of siRNA accumulation and clearance from lung, liver, spleen, and kidney over 96 hours and plotted as micrograms of siRNA detected per gram of tissue. (d) Cav-1 transcript depletion was determined by qPCR analysis of homogenized lung or liver tissue collected 24, 48, or 96 hours after injection of Staramine or Star:Star-mPEG550/siCav-1 nanocomplexes. P values were calculated from two-tailed, equal variance student t-tests (*P > 0.05, **P < 0.05, ***P < 0.01). mPEG, methoxypolyethylene glycol.

94

www.moleculartherapy.org vol. 20 no. 1 jan. 2012

© The American Society of Gene & Cell Therapy

Staramine-mediated siRNA Delivery

Identification of the transfected lung cell type and verification of RNAi-mediated knockdown In order to better characterize the lung specific delivery resulting from systemic administration of Star:Star-mPEG550/ siCav-1 nanocomplexes, a series of siRNAs targeting established marker mRNAs for various cell types within the lung were used to identify those cell types transfected most effectively. These studies were performed in a green fluorescent protein (GFP) expressing transgenic mouse line with either GFP targeting siRNAs or cell type marker siRNAs injected on day 1 and day 3 at a dose of 2 mg/kg. Lung and liver tissues were collected 24 hours after the second injection and homogenized for Quantigene analysis of target transcripts. While GFP targeting siRNAs produce only modest GFP transcript knockdown in whole lung homogenates, the screen of marker proteins to several lung epithelial cell types and fibroblasts (data not shown) as well as leukocytes (CD45), endothelial GFP

CD45

b

CD31

GFP

140%

140%

120%

120%

Relative mRNA expression

Relative mRNA expression

a

cells (CD31), and GFP reveals significant knockdown of the CD31 transcript, resulting in 70% knockdown relative to nonspecific siRNA control groups (Figure 6a). Administration of Staramine/siRNA nanocomplexes harboring siRNAs targeting marker mRNAs for other cell types did not result in significant depletion of the endothelial cell marker mRNAs. Further studies used an additional siRNA targeting CD31 (siCD31-2) and an siRNA targeting a second endothelial cell marker transcript, Tie-2, also resulted in significant target gene knockdown relative to nonspecific siRNA controls: 75% knockdown with CD31-1, 70% knockdown with CD31-2, and 67% knockdown with Tie-2 (Figure  6b). Western blot analysis of whole lung homogenates showed a significant reduction of CD31 protein in the CD31 siRNA-treated animals when compared to animals receiving the Staramine/siGFP nanocomplex (Figure 6c). RNAi-mediated transcript cleavage was confirmed by 5′-RACE analysis (Figure 6d).

100% 80% 60% 40%

80% 60% 40% 20% 0%

siGFP

c kDA

CD31

100%

20% 0%

Tie−2

siCD31

siCD45

siGFP

1

2

3

siGFP

PBS

d

siCD31

4

165

39

1

2

3

Specific siRNA M

4

siCD31−1 siCD31−2 siTie−2

1

2

3

PBS

non−Specific siRNA 4

1

2

3

4

H 2O M

CD31 600

600

100

100

β−actin

Figure 6 RNA interference (RNAi)-mediated target gene knockdown in the pulmonary endothelium. Systemic administration of Star:StarmPEG550 /siRNA nanocomplexes results in siRNA delivery specifically to the pulmonary endothelium (n = 4). (a) A series of Star:Star-mPEG550/siRNA nanocomplexes were prepared carrying siRNAs targeting established cell type-specific marker proteins and administered by intravenous (i.v.) injection to enhanced GFP (eGFP)-transgenic mice. Knockdown of the target gene transcript was determined by Quantigene 2.0 analysis of lung tissue collected 24 hours after the second of two injections by comparison to untreated samples [phosphate-buffered saline (PBS) injections] and samples from mice treated with nanocomplexes-containing nonspecific siRNAs, the mean of the latter being set to 100%. (b) A second siRNA sequence targeting CD31 (siCD31-2) and a third siRNA sequence targeting Tie-2 were complexed with Star:Star-mPEG550 and administered by i.v. injection. Target gene knockdown was assessed at 24 hours after the second of two injections as before. (c) Protein samples collected from whole lung homogenates of animals treated with Star:Star-mPEG550/siRNA nanocomplexes carrying siRNAs targeting either the GFP transcript or the CD31 transcript were subjected to western blot analysis. β-Actin was used as an internal blotting control to compare protein loading to CD31 target protein levels. (d) Lung tissue was collected from mice treated with Star:Star-mPEG550/siRNA nanocomplexes-containing either siRNA targeting CD31 or nonspecific siRNA (four mice per group) at 24 hours after the second of two injections. Total RNA was collected from whole lung homogenates for each mouse and subjected to 5′ RACE analysis. RNAi-mediated cleavage of the CD31 transcript was confirmed by the presence of the 420 base pairs (bp) nucleotide PCR product. M—100 bp markers; H2O—water template control. mPEG, methoxypolyethylene glycol.

Molecular Therapy vol. 20 no. 1 jan. 2012

95

© The American Society of Gene & Cell Therapy

Staramine-mediated siRNA Delivery

Modification of Staramine nanocomplexes with monodisperse mPEG515 Use of the polydisperse mPEG550 modifiers could result in suboptimal performance of the nanocomplexes due to the variable surface properties associated with the heterogeneous mPEG lengths. Lung siRNA accumulation and clearance for nanocomplexes modified either with the polydisperse mPEG550 or with the monodisperse mPEG515 were compared. In contrast to the Star:Star-mPEG550/ siCav-1 nanocomplexes, which had highly variable siRNA clearance rates, the clearance rate of siRNA delivered with Star:StarmPEG515/siCav-1 nanocomplexes was much more uniform and relatively slow where ~70% of the material present at 1 hour remained after 24 hours and ~25% remained after 168 hours (Figure 7a). Results are plotted as individual measurements from multiple experiments. Cav-1 transcript knockdown was observed (~50%) in the lung tissue with Star:Star-mPEG515/siCav-1 nanocomplexes at 2 days postinjection with significant knockdown persisting during the 10-day study period (Figure 7b).

Discussion The success of RNAi-based therapeutics is dependent on modular siRNA delivery systems that allow for chemical modifications to fine tune biodistribution as well as biological activity. We have evaluated the in vivo properties of a versatile lipopolyamine (Staramine) and its chemically functionalized derivatives in an attempt to improve the utility of the lipopolyamine for in vivo delivery of siRNA. The intravenous administration of siRNA with the nonfunctionalized Staramine core resulted in siRNA retention and corresponding RNAi-mediated target gene knockdown predominantly in the mouse lung. Chemical conjugation of mPEG modifiers to Staramine, either a monodisperse mPEG515 or a polydisperse mPEG550, significantly improved formulation stability and safety without affecting the gene silencing activity in the lung. This covalent attachment of mPEG to the Staramine lipopolyamine, rather than simple adsorption to the liposomal surface, is expected to provide greater resistance to displacement by serum proteins and Star:Star-mPEG550

10

b

Star:Star-mPEG515

1

0.1

0.01

0.001

siNon 140

Percent of untreated cells

Micrograms siRNA / gram tissue

a

improve serum stability. The lung distribution and gene silencing observed with the Staramine systems is in contrast to the typical systemic behavior of cationic lipid nanocomplexes, which generally show the largest gene knockdown in liver ­tissue17,25–28 with only transient accumulation of siRNA in the lung. The accumulation of cationic lipid delivery systems in the liver may be attributed to passive mechanisms involving the particular physiology of the liver, in which the fenestrated endothelium provides the potential for particle uptake, or through endogenous targeting mechanisms involving apolipoproteins. Intravenous injection of Staramine/siCav-1 nanocomplexes results in rapid transfer of siRNA from the blood to the liver with both the nonfunctionalized and mPEG modified Staramine systems, a result that mirrors recent observations with the lipidoid-siRNA nanocomplex LPN01.19 Staramine nanocomplexes are not delivered preferentially to the lung, the first capillary bed encountered in circulation, as the majority of the material detected in the stemloop assays is recovered from the liver tissue shortly after injection. Rather, the fraction of material that does accumulate in the lung is retained over a much longer period than material in the liver, spleen, or kidney. Accumulation of the Staramine systems in the lung tissue is independent of moderate changes in particle size (80–150 nm) and zeta-potential (20–45 mV) and is observed with both PEGylated and non-PEGylated delivery systems alike. The precise mechanism for siRNA retention in the lung remains unknown, though aggregate trapping seems unlikely as the greatest siRNA retention and longest target gene knockdown duration is observed with Star:Star-mPEG515/siCav-1 nanocomplexes which display reduced aggregation in serum. Delivery of siRNA exclusively to the endothelial cells of the lung may indicate direct transfer of the nanocomplexes from circulation to the cells lining the capillary bed. Although a greater mass amount of siRNA is recovered from the liver tissue, the siRNA content per unit tissue mass is nearly equivalent across these two organs shortly after administration. The correlation observed between siRNA retention and target gene knockdown in lung endothelial cells is consistent

0

50

100

150

Time (hours post−injection)

200

250

P=0.04

siCav−1

P<0.01

P<0.01

96

168

P<0.02

120 100 80 60 40 20 0

48

240

Time (hours post−injection)

Figure 7 Target gene knockdown and small interfering RNA (siRNA) delivery with Star:Star-mPEG515/siCav-1 nanocomplexes. siRNA accumulation and target gene knockdown in the mouse lung were determined following injection of Star:Star-mPEG515/siCav-1 nanocomplexes. (a) Clearance of the siRNA from the lung tissue was monitored over 10 days and plotted as independent points from three different experiments in terms of micrograms of siRNA detected per gram of tissue. (b) Knockdown of the Cav-1 target gene transcript was determined by quantitative reverse transcription PCR (qPCR) analysis of homogenized lung tissue by comparison to untreated samples (5% dextrose injections) and samples from mice treated with nanocomplexes-containing nonspecific siRNAs. Results are averages and SD for three experiments, P values are calculated from twotailed, equal variance student t-tests. mPEG, methoxypolyethylene glycol.

96

www.moleculartherapy.org vol. 20 no. 1 jan. 2012

© The American Society of Gene & Cell Therapy

with successful delivery and incorporation of siRNAs into RNAinduced silencing complex, though the reasons for faster siRNA clearance rates from and minimal target gene knockdown in the liver, spleen, and kidney are unknown. One possible explanation may lie with the type or types of cells transfected in each organ. The rapid clearance of siRNA from the liver tissue could correspond to uptake by Kupffer cells, monocytes, and macrophages. Though the use of PEGylated nanocarriers generally decreases association with the reticuloendothelial system, the mPEGs used in the Staramine system are of relatively small size (~500 Da) and may still be subject to clearance. The rate of clearance observed with the Staramine systems is quite rapid compared to rates observed with other delivery systems where target gene knockdown in hepatocytes is observed19 consistent with delivery to other cell types. Intravenous administration of cationic delivery systems can result in systemic toxicity corresponding to electrostatically driven aggregation of particles in the presence of serum proteins, including opsonizing proteins, such as immunoglobulins and fibronectin, as well as components of the complement system.29 This aggregation has been observed for the Staramine core system through changes in particle size upon addition of serum (Supplementary Figure S1) and is reduced by the addition of mPEG modified Staramine lipopolyamines to the lipid nanocomplexes consistent with the mPEG mediated improvements in safety. These same modifications result in significant improvements to the delivery safety measured through the toxicity index (Supplementary Figure S1) consistent with the accepted view that PEG improves the stability of nanocomplexes in vivo.29–31 Cationic lipid nanocomplexes can also be proinflammatory through interaction of both the lipid and the nucleic acid components with the Toll-like receptors of monocytes/macrophages, culminating in the induction of chemokine and cytokine gene expression.14,32 The nonspecific siRNA and the Cav-1 siRNA are modified by the manufacturer for optimal serum stability and minimal cell toxicity. The modest cytokine induction observed in vivo upon systemic administration of Star:Star-mPEG550/siCav-1 nanocomplexes involves Ccl2, Ccl7, and Ccl12, monocyte chemotactic proteins induced in lung injury33–35 and responsive to type I interferon signaling.36,37 These cytokines act as homing ligands for the Ccr2 receptor present on monocytes/macrophages and activated T-cells. The significance of this cytokine increase is not clear because the changes in expression observed for these homing ligands do not result in a corresponding increase in Ccr2 gene expression expected upon monocyte/macrophage recruitment and do not correspond to an increase in circulating monocyte populations in CBC analysis. Similar increases in granulocyte colony-stimulating factor expression may correspond to the apparent doubling in the average number of circulating neutrophils, though the collected data sets for the neutrophil increase are not statistically different at the 5% significance level (P = 0.052). siRNAs synthesized for CD31, CD45, GFP, and Tie-2 included selective incorporation of 2′-O-methyl modifications which abrogate an inflammatory response. These modified siRNAs were tested in-house with human peripheral blood mononuclear cell and showed no sign of cytokine induction.38 Immunohistochemistry of lung tissue collected after treatment with Star:Star-mPEG550/siGFP showed no infiltration of Gr-1 positive cells into the lung tissue related to administration of the delivery system and H&E staining revealed no significant morphological Molecular Therapy vol. 20 no. 1 jan. 2012

Staramine-mediated siRNA Delivery

changes compared to phosphate-buffered saline controls. The lack of induction of tumor necrosis factor-α and interferon-α as well as the minor induction of interleukin-6, are also consistent with minimal systemic toxicity. Collectively, this data suggests minimal toxicity associated with the PEGylated Staramine delivery system. Localized delivery of siRNA to the lungs has been of particular therapeutic interest and several different types of carriers have been implemented as delivery vehicles for intratracheal (i.t.) and intranasal delivery, including cationic liposomes, polysaccharide nanoparticles, cationic polymers and dendrimers.39–43 Recent work on siRNA delivery to the lung via liposomal complexes has compared i.t. or intranasal delivery routes to systemic i.v. delivery.23,44–46 i.t. delivery of naked siRNA and siRNA lipoplexes has resulted in delivery to the pulmonary epithelium, whereas i.v. delivery of siRNA-lipoplexes resulted in significant siRNA accumulation in endothelial cells. Though i.t. delivery resulted in ~20% target gene knockdown in epithelial cells, this activity was accompanied by significant inflammation.45 Repeated i.v. injections (once every 24 hours over 4 days) of lipoplexes carrying siRNA targeting either the VE-cadherin transcript or the protein kinase N3 transcript (both expressed predominantly in the endothelium) resulted in ~50% target gene knockdown 24 hours after the final injection. A second series of studies comparing i.t. instillation and i.v. injections of naked siRNA and siRNA lipoplexes revealed significant changes in the biodistribution of fluorescently labeled siRNA with epifluorescent and confocal microscopy.44 i.v. injections of naked siRNA resulted in accumulation in the kidneys and rapid renal excretion, whereas injection of siRNA lipoplexes led to accumulation of fluorescent signal in the heart, lung, spleen, and liver. After a single administration, fluorescent images show a gradual decrease in the signals present in the heart and lung 2 hours after injection, whereas signals in the liver and spleen persist over 20 hours. siRNA delivery with Star:Star-mPEG550 and Star:StarmPEG515 nanocomplexes has a substantially different kinetic behavior compared to the i.v. injection of the lipoplexes described in these earlier studies, where Staramine promotes siRNA persistence in the lungs and rapid clearance from the liver. Delivery of siRNA with Staramine/siCav-1 nanocomplexes modified with monodisperse mPEG515 resulted in a consistently slow clearance rate for the siRNA from the lung, compared to those composed of Star:Star-mPEG550, which showed more variable rates of clearance over multiple experiments. The prolonged residence of the siRNA in lung tissue when delivered with monodisperse mPEG515 modified Staramine/siCav-1 nanocomplexes results in a similar amount of target gene knockdown compared to the work described above (~50%), but requires only a single administration of nanocomplexes and persists for up to 10 days. As expression of the Cav-1 gene is not restricted to the pulmonary endothelial cells, this knockdown percentage may only reflect activity in the subpopulation of cells that are reached by the delivery system. This hypothesis is supported by studies comparing knockdown of GFP transcripts expressed throughout the lung and CD31 transcripts expressed predominantly in endothelial cells. i.v. administrations of polydisperse mPEG550 modified Staramine/siCD31 (2 mg/kg, two injections) led to ~75% depletion of the CD31 transcript, whereas similar administrations of siGFP nanocomplexes led to ~20% knockdown of GFP transcript. The improved siRNA 97

Staramine-mediated siRNA Delivery

© The American Society of Gene & Cell Therapy

retention and duration of knockdown of siRNA delivered with Star:Star-mPEG515/siCav-1 nanocomplexes compared to the Star:Star-mPEG550/siCav-1 nanocomplexes and nonfunctionalized Staramine does not correlate with any measured change in the physical properties of the nanocomplex or with the efficiency of siRNA binding onto the nanocomplexes. With this duration of target gene knockdown, dosing regimens including weekly treatments with siRNA nanocomplexes become feasible. The inherent accumulation of Staramine/siCav-1 nanocomplexes to the lung may provide a promising platform for drug development focused on pulmonary tissue. Significant potential may lie in the treatment of lung cancer, where lung accumulation may act synergistically with carrier uptake in lung tumors through the enhanced permeability and retention of the tumor vasculature. This application may also benefit from the addition of specific targeting ligands to the chemically versatile Staramine nanocomplexes to facilitate uptake by tumor cells. Such applications will require further development of these delivery systems and a greater understanding of the underlying mechanisms dictating biodistribution and retention of nanocomplexes.

siRNA was selected from a set of four individual siRNA molecules, all targeting the Cav-1 transcript, through an in vitro assay of target gene knockdown. These siRNAs carry proprietary modifications reported to improve nuclease stability while resulting in minimal cellular toxicity. The sequences for sense and antisense strands of nonspecific control and Cav-1 targeting siRNAs were as follows: siNon sense: 5′ UGGUUUACAUGUCGACUAAUU 3′ siNon antisense: 5′ UUAGUCGACAUGUAAACCAUU 3′ siCav-1 sense: 5′ GUCCAUACCUUCUGCGAUCUU 3′ siCav-1 antisense: 5′ GAUCGCAGAAGGUAUGGACUU 3′ siRNAs were received from the manufacturer annealed and lyophilized. Stocks were resuspended at 5 mg/ml in water for injection and stored at −20 °C. Sequences for primers and probes used in stem-loop qPCR assays were as follows: Stem-loop primer: 5′ G TCGTATCCAGTGCAGGGTCCGAGGTA TTCGCACTGGATACGACAAGTCCATA 3′ siCav-1 forward: 5′ CGCGCGGATCGCAGAAGG 3′ siCav-1 reverse: 5′ GTGCAGGGTCCGAGGT 3′ siCav-1 probe: 5′ TCGCACTGGATACGACAAGTCCA 3′ Primers and probes were received from the manufacturer lyophilized; stocks were resuspended at 500 nmol/l in 10 mmol/l Tris (pH 8.0) and stored at −20 °C. Taqman Gene Expression Assays for detection of the Cav-1 transcript were obtained from Applied Biosystems (Carlsbad, CA).

Materials and Methods

Formulation of Staramine liposomes and siRNA cargo. Staramine synthesis and purification has been verified through analytical high-performance liquid chromatography, mass spectrometry, and NMR (data not shown). Chloroform solutions of Staramine alone or 10:1 mixtures of Staramine and Star-mPEG550 or Star-mPEG515 were rotary-evaporated to a film. The flask of liposome film was held under high vacuum overnight. Water for injection (B. Braun Medical, Bethlehem, PA) was added to the film to give the desired Staramine concentration and bath sonicated for ~30 minutes using Branson Water Bath Sonicator model 2510 followed by a probe sonication for 5 minutes (using a continuous pulse sonication with an output wattage of 5–10 watts (rms) (Model 100; Fisher Scientific Sonic Desmembrator, Pittsburg, PA). The liposome solution was filtered through a 0.2-µm filter, diluted with 5% dextrose (generating an isotonic solution) and mixed with the desired amount of siRNA. Particle size and zeta-potential of the complexes was measured with 90Plus/BI-MAS Particle sizer with BI-Zeta option (Brookhaven Instruments, Holtsville, NY). Particle size and zeta-potential measurements were conducted using Brookhaven 90 Plus Particle Size and Zeta-Potential Analyzer. The complexation efficiency was determined by gel retardation assay and Quant-iT RiboGreen RNA assays (Life Technologies, Carlsbad, CA). The gel retardation assay was performed by loading the Staramine/ siRNA solution on 1% agarose gel and electrophoresed at 100 V for 1 hour. To release the siRNA from the Staramine nanoparticles, 10% TritonX-100 was added to the complex and the solution was loaded on the agarose gel. The RiboGreen assay was used to determine the percentage of siRNA complexed with Staramine; the fluorescently stained siRNA was measured before and after the disruption of complexes with 10% of Triton X-100.

Synthesis of siRNAs. siRNAs directed against GFP, CD31, CD45, and Tie-

2, used in this study were all synthesized at Roche Kulmbach, Germany and consisted of a 21-nucleotide sense strand and a 21-nucleotide antisense strand resulting in a 19 base pairs double strand and two-nucleotide overhang at the 3′-end of both strands. GFP-siRNA: sense 5′-AAcGAGAAGcGcGAucAcAdT*dT-3′, antisense 5′-UGUGAUCGCGCUUCUCGUUd T*dTC-3′; CD31-siRNA-1: sense 5′ –GacGAuGcGAuGGuGuAuAdT*dT-3′, antisense 5′-puAuAcACcAUCGcAUCGUCdT* dT-3′; CD31-siRNA-2: sense 5′-AccAcGAuuGAGuAcGAGGdT*dT-3′, antisense 5′-pCCUCGuACUcAAUCGUGGUdT *dT-3′, CD45-siRNA:7,23,24 sense 5′-cuGGcuGAAuuucAGAGcAdT*dT-3′, antisense 5′-UGCUCUGAAAUUcAGCcAGdT* dT-3′; Tie-2-siRNA: sense 5′-GAAGAuGcAGuGAuuuAcAdT*dT-3′, antisense 5′-UGuAAAUcACUGcAUCUUCdT* dT-3′. The lower-case letters represent 2′-O-methyl-modified nucleotides; asterisks represent phosphorothioate linkages. RNA oligonucleotides were synthesized using commercially available 5′-O-(4,4, 0 -dimethoxytrityl)3′-O-(2-cyanoethyl-N,N-diisopropyl) phosphoramidite monomers of uridine (U), 4-N-benzoylcytidine (CBz), 6-N-benzoyladenosine (ABz) and 2-Nisobutyrylguanosine (GiBu) with 2′-O-t-butyldimethylsilyl protected phosphoramidites and the corresponding 2′-O-methyl phosphoramidites according to standard solid phase oligonucleotide synthesis protocols.47 After cleavage and deprotection, RNA oligonucleotides were purified by anion-exchange high-performance liquid chromatography and characterized by electrospray mass spectrometry. RNA with phosphorothioate backbone at a given position was achieved by oxidation of phosphite with Beaucage reagent48 during oligonucleotide synthesis. To generate siRNAs from RNA single strands, equimolar amounts of complementary sense and antisense strands were mixed and annealed, and siRNAs were further characterized by capillary gel electrophoresis. Nucleic acids. Two siRNAs (siNon and siCav-1) were purchased from Dharmacon (Lafayette, CO) (siSTABLE, in vivo purified). The Cav-1

98

In vivo experiments. Staramine/siRNA nanocomplexes were diluted into

sterile water and injected into the tail vein of female ICR mice or transgenic mice expressing enhanced GFP. Doses were adjusted by altering the concentration of particles in each injection; injection volumes were held constant at 10 µl/g. At various times postinjection, organs were collected and immediately frozen in liquid nitrogen for storage at −80 °C. All procedures used in animal studies at EGEN (Huntsville, AL) were performed in accordance with local, state, and federal regulations and approved by the Institutional Animal Care and Use Committee. Female ICR mice were obtained from Harlan Laboratories (Houston, TX) and ranged from 8 to 10 weeks of age (17–22 g) at the time of each study. All procedures used in animal studies at Roche Kulmbach were performed in accordance with the German Animal Protection Law of 2006, and permission was obtained from

www.moleculartherapy.org vol. 20 no. 1 jan. 2012

© The American Society of Gene & Cell Therapy

the local Veterinary Office (Regierungspräsidium Mittelfranken, Ansbach, Germany; permission no. 54-2532.1-5/09). Enhanced GFP-transgenic mice49 were bred in-house. qPCR. Total RNA was collected from cells grown in culture using RNEasy kits (Qiagen, Valencia, CA) and tissues using Tri-Reagent (Molecular Research Center, Cincinnati, OH) according to the manufacturer’s protocols. These RNA stocks were used as templates for synthesis of complementary DNAs (cDNAs) using random hexamer primers and reverse transcriptase (Multiscribe; Applied Biosystems). The reverse transcription reaction contained: 10 mmol/l Tris (pH 8.3), 5.5 mmol/l MgCl2, 50 ng total RNA, 2.5 µmol/l random hexamer primer, 500 µmol/l each dNTP, 0.4 unit/ µl RNAse inhibitor, 1.25 unit/µl reverse transcriptase. Taqman primers and probe sets (Applied Biosystems) were used in qPCR to amplify the transcript of interest from the corresponding cDNA template. The qPCR contained: 10 mmol/l Tris (pH 8.3), 50 mmol/l KCl, 5 mmol/l MgCl2, 50 ng total RNA, 900 nmol/l forward and reverse primers, 250 nmol/l probe, 200 µmol/l each dNTP, 0.1 unit/ml Taq polymerase. All changes were measured against an internal control using primers specific for housekeeping genes (GAPDH or mPpib). Results are presented as the percent of target gene expression observed relative to the dextrose treated control group. Averages and standard deviations are calculated from at least three experiments, P values are calculated from equal variance, two-tailed t-tests. Quantification of mRNA with quantigene assay. The QuantiGene 2.0 assay (Panomics/Affymetrix, Santa Clara, CA) was used to quantify the reduction of mouse CD31, GFP, CD45, and Tie-2 mRNA in lung tissue after siRNA treatment. Tissue samples were collected 24 hours after the second injection. Three aliquots from each pulverized lung were lysed and directly used for CD31, GFP, CD45, and Tie-2 and GAPDH mRNA quantification, and the ratio of CD31 and GAPDH mRNA was calculated and expressed as a group average relative to the mean of all groups treated with unspecific siRNA. Specific QG2.0 probes for detection of CD31 and GAPDH mRNA levels were designed by Panomics/Affymetrix. In vivo siCav-1 biodistribution studies. Staramine nanocomplexes were

formulated with siRNA targeting Cav-1 or the control siNon using the methods described previously. Nanocomplexes were diluted into sterile water and injected into the tail vein of female ICR mice. Injections contained 40 µg (siRNA content in nanocomplexes) in a volume of 200 µl. Organs were collected at multiple times postinjection and immediately frozen in liquid nitrogen for storage at −80 °C. Homogenized tissue samples were prepared in Tri-Reagent (Molecular Research Center) from tissues collected at multiple times post-transfection. Homogenized samples were diluted as needed (minimum dilution of 1:100) in water for stem-loop annealing and extension to generate cDNA templates using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems). The stem-loop annealing reaction contained: 1× RT buffer, 300 nmol/l stem-loop primer, and 10 µl of the diluted homogenized sample (20 µl total reaction volume) and cycled through the following temperature profile in a thermocycler: 94 °C for 10 minutes, 75 °C for 2 minutes, 60 °C for 3 minutes, 50 °C for 3 minutes, 40 °C for 3 minutes, 30 °C for 3 minutes, 4 °C hold. The siCav-1 stem-loop primer and qPCR primers and probe were obtained from Operon with the sequences described above. The annealing reaction was then treated with an elongation premix to produce the final cDNA synthesis reaction: 1 mmol/l dNTPs, 1× RT buffer, 8 units RNAse inhibitor, 50 units reverse transcriptase (30 µl total reaction volume) which was cycled through the following temperature profile: 16 °C for 30 minutes, 42 °C for 30 minutes, 85 °C for 5 minutes. qPCR amplification of the cDNA template was performed with the following temperature profile: 50 °C for 2 minutes, 95 °C for 10 minutes, followed by 40 cycles alternating between 95 °C for 15 seconds and 60 °C for 1 minute. Sample concentrations were determined by comparison to standard curves generated with serial dilutions of the siRNA stock. Uniformity of the RNA preparations was verified with

Molecular Therapy vol. 20 no. 1 jan. 2012

Staramine-mediated siRNA Delivery

primers and probes specific for the endogenous U6 small RNA. Individual samples were normalized to total RNA concentrations for stocks purified from homogenized tissues. Results are presented as micrograms of siRNA (as calculated from the standard curves) or as micrograms of siRNA per gram of tissue, normalized to the average weight of the particular organ. Averages and standard deviations are calculated from at least three experiments. Chemokine/cytokine PCR arrays. Lung tissue samples were collected

from animals treated with Star:Star-mPEG550/siCav-1 nanocomplexes (60 µg dose, 3 mg/kg) at 4 hours postinjection. Total RNA was isolated from homogenized tissues first with Tri-Reagent then followed by Qiagen’s RNEasy kit, both according to manufacturers’ protocols and including the RNeasy on column DNAse treatment, to generate a high-quality RNA preparation. Inflammatory responses were measured with the RT2 Profiler PCR Array System (SABiosciences, Frederick, MD) according to the manufacturer’s protocol. Results depict the fold change in expression of each gene upon treatment with Star:Star-mPEG550/siCav-1 nanocomplexes compared to 5% dextrose injections and are average values from five independent studies. The volcano plot was constructed from the same data set and shows only four genes surpassing the fourfold increase with t-test P values <0.01. Total genes assessed: Abcf1, Bcl6, Cxcr5, C3, Casp1, Ccl1, Ccl11, Ccl12, Ccl17, Ccl19, Ccl2, Ccl20, Ccl22, Ccl24, Ccl25, Ccl3, Ccl4, Ccl5, Ccl6, Ccl7, Ccl8, Ccl9, Ccr1, Ccr2, Ccr3, Ccr4, Ccr5, Ccr6, Ccr7, Ccr8, Ccr9, Crp, Cx3cl1, Cxcl1, Cxcl10, Cxcl11, Cxcl12, Cxcl13, Cxcl15, Pf4, Cxcl5, Cxcl9, Cxcr3, Ccr10, Ifng, Il10, Il10ra, Il10rb, Il11, Il13, Il13ra1, Il15, Il16, Il17b, Il18, Il1a, Il1b, Il1f6, Il1f8, Il1r1, Il1r2, Il20, Il2rb, Il2rg, Il3, Il4, Il5ra, Il6ra, Il6st, Il8rb, Itgam, Itgb2, Lta, Ltb, Mif, Scye1, Spp1, Tgfb1, Tnf, Tnfrsf1a, Tnfrsf1b, Cd40lg, Tollip, Xcr1. Cytokine enzyme-linked immunosorbent assay. Whole blood samples

were collected from animals treated with Star:Star-mPEG550/siCav-1 nanocomplexes (40 µg dose, 2 mg/kg) at 4 hours postinjection. Total protein was prepared from pooled blood samples obtained from four individual animals by adding TENT buffer (10 mmol/l Tris–HCl, pH 8.0, 1 mmol/l EDTA, 0.15 mol/l NaCl, 0.5% Triton X-100) in the presence of protease inhibitors (Calbiochem Protease Inhibitor Cocktail Set I; Fisher, Pittsburg, PA). Cytokine expression levels were then measured with the cytokine enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN). Results are presented as the fold change in absorbance signal for each gene upon treatment with Star:Star-mPEG550/siCav-1 nanocomplexes compared to 5% dextrose injections. Whole blood and serum chemistry analysis. Whole blood samples and

serum samples were collected from animals treated with 5% dextrose or Star:Star-mPEG550/siCav-1 nanocomplexes (40 µg dose, 2 mg/kg) at 24 hours postinjection. CBC and serum chemistry analysis was performed by IDEXX Laboratories (West Sacramento, CA) on whole blood and serum samples shipped at 4 °C via overnight courier.

Immunohistochemistry of lung tissue sections. Lung tissue was fixed by

intratrachial infusion of 1 ml 4% paraformaldehyde and overnight incubation in the fixative. Tissue was dehydrated in a series of graded ethanol and embedded in paraffin. Deparaffinized sections (10 µm) were stained with H&E for general histologic examination. For immunohistochemistry, deparaffinized sections were incubated overnight with a rat antimouse Gr-1 monoclonal antibody (clone RB6-8C5; BD Biosciences, San Jose, CA) diluted 1:50 in 2% bovine serum albumin. The primary antibody was detected using an anti-rat biotinylated antibody, the Vectastain ABC-horseradish peroxidase detection kit and 3,3′-diaminobenzidine as peroxidase substrate (all from Vector Laboratories, Burlingame, CA), according to the manufacturers recommendations.

Protein extraction. Proteins were extracted using TEDT buffer-contain-

ing 20 mmol/l Tris (Sigma, Munich, Germany) pH 7.5, 2 mmol/l EDTA

99

Staramine-mediated siRNA Delivery

(Roth, Karlsruhe, Germany), 500 µmol/l EGTA (Sigma), 1 mmol/l DTT (Roth), 0.5% Triton-X-100 (Sigma). Aliquots of pulverized lung tissue were dissolved in a tenfold amount of buffer and incubated on ice for 30 minutes. The lysates were then vortexed and centrifuged at 16,000g for 5 minutes at 4°C. The supernatant was transferred into a new tube for further processing. Western blot analysis. SDS-PAGE was performed with a 10% resolving

and 4% stacking gel in a Hoefer gel chamber. After electrophoresis, proteins were transferred to a nitrocellulose membrane (Hybond-C extra; Amersham Biosciences, Piscataway, NJ) and probed with antibodies raised against CD31 (Santa Cruz sc-1506) and β-actin (AbD Serotec, Duesseldorf, Germany).

5′-RACE analysis. Total RNA (5 µg) from lung samples of individual animals treated with different siRNAs was ligated to a GeneRacer adaptor (Invitrogen, Carlsbad, CA) without prior treatment. Ligated RNA was reverse transcribed using a gene-specific primer (GSP: 5′-GAAGGACTCCTGCACGGTGACGTATT-3′). To detect cleavage products, PCR was performed using primers complementary to the RNA adaptor (GR5′: 5′-CTCTAGAGCGACTGGAGCACGAGGACACTA-3′) and CD31 mRNA (RP1: 5′-AGCTTGGCAGCGAAACACTAACACGT-3′). Amplification products were resolved by agarose gel electrophoresis and visualized by ethidium bromide staining. The identity of specific PCR products was confirmed by sequencing of the excised 420 base pairs bands of the nested PCR.

SUPPLEMENTARY MATERIAL Figure  S1  Nanocomplex formation with various siRNAs. Figure  S2  Toxicity assessments and serum induced aggregation with Staramine nanocomplexes. Figure  S3  Liver histopathology with i.v. injection of Star:Star-mPEG550 nanocomplexes. Figure  S4  Spleen histopathology with i.v. injection of Star:Star-mPEG550 nanocomplexes. Table  S1  Physical properties of siRNA nanocomplexes.

ACKNOWLEDGMENTS The authors thank Casey Pence, Leslie Wilkinson, David Ulkoski, Amy Pettigrew, Errin Christian, Kirby Wallace, Petra Deuerling, Sabrina Krause, and André Wetzel for skillful technical assistance. K.J.P., M.M., J.R., G.S., J.S., A.R., D.M., E.B., J.G.F., and K.A. are employees of EGEN Inc; M.K., A.S., K.J.-H., M.J., H.-P.V. and A.G. are employees of Roche Kulmbach.

REFERENCES

1. Tiemann, K and Rossi, JJ (2009). RNAi-based therapeutics-current status, challenges and prospects. EMBO Mol Med 1: 142–151. 2. Grimm, D (2009). Small silencing RNAs: state-of-the-art. Adv Drug Deliv Rev 61: 672–703. 3. Vaishnaw, AK, Gollob, J, Gamba-Vitalo, C, Hutabarat, R, Sah, D, Meyers, R et al. (2010). A status report on RNAi therapeutics. Silence 1: 14. 4. Soutschek, J, Akinc, A, Bramlage, B, Charisse, K, Constien, R, Donoghue, M et al. (2004). Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature 432: 173–178. 5. Akhtar, S and Benter, IF (2007). Nonviral delivery of synthetic siRNAs in vivo. J Clin Invest 117: 3623–3632. 6. Samad, A, Sultana, Y and Aqil, M (2007). Liposomal drug delivery systems: an update review. Curr Drug Deliv 4: 297–305. 7. Akinc, A, Zumbuehl, A, Goldberg, M, Leshchiner, ES, Busini, V, Hossain, N et al. (2008). A combinatorial library of lipid-like materials for delivery of RNAi therapeutics. Nat Biotechnol 26: 561–569. 8. Farokhzad, OC (2008). Nanotechnology for drug delivery: the perfect partnership. Expert Opin Drug Deliv 5: 927–929. 9. Li, SD, Chen, YC, Hackett, MJ and Huang, L (2008). Tumor-targeted delivery of siRNA by self-assembled nanoparticles. Mol Ther 16: 163–169. 10. Bennewitz, MF and Saltzman, WM (2009). Nanotechnology for delivery of drugs to the brain for epilepsy. Neurotherapeutics 6: 323–336. 11. Farokhzad, OC and Langer, R (2009). Impact of nanotechnology on drug delivery. ACS Nano 3: 16–20. 12. Gao, K and Huang, L (2009). Nonviral methods for siRNA delivery. Mol Pharm 6: 651–658. 13. Reischl, D and Zimmer, A (2009). Drug delivery of siRNA therapeutics: potentials and limits of nanosystems. Nanomedicine 5: 8–20. 14. Tseng, YC, Mozumdar, S and Huang, L (2009). Lipid-based systemic delivery of siRNA. Adv Drug Deliv Rev 61: 721–731. 15. Whitehead, KA, Langer, R and Anderson, DG (2009). Knocking down barriers: advances in siRNA delivery. Nat Rev Drug Discov 8: 129–138.

100

© The American Society of Gene & Cell Therapy

16. Higuchi, Y, Kawakami, S and Hashida, M (2010). Strategies for in vivo delivery of siRNAs: recent progress. BioDrugs 24: 195–205. 17. Schroeder, A, Levins, CG, Cortez, C, Langer, R and Anderson, DG (2010). Lipid-based nanotherapeutics for siRNA delivery. J Intern Med 267: 9–21. 18. Semple, SC, Akinc, A, Chen, J, Sandhu, AP, Mui, BL, Cho, CK et al. (2010). Rational design of cationic lipids for siRNA delivery. Nat Biotechnol 28: 172–176. 19. Akinc, A, Goldberg, M, Qin, J, Dorkin, JR, Gamba-Vitalo, C, Maier, M et al. (2009). Development of lipidoid-siRNA formulations for systemic delivery to the liver. Mol Ther 17: 872–879. 20. Santel, A, Aleku, M, Keil, O, Endruschat, J, Esche, V, Fisch, G et al. (2006). A novel siRNA-lipoplex technology for RNA interference in the mouse vascular endothelium. Gene Ther 13: 1222–1234. 21. Juliano, R, Bauman, J, Kang, H and Ming, X (2009). Biological barriers to therapy with antisense and siRNA oligonucleotides. Mol Pharm 6: 686–695. 22. Behlke, MA (2006). Progress towards in vivo use of siRNAs. Mol Ther 13: 644–670. 23. Santel, A, Aleku, M, Keil, O, Endruschat, J, Esche, V, Durieux, B et al. (2006). RNA interference in the mouse vascular endothelium by systemic administration of siRNAlipoplexes for cancer therapy. Gene Ther 13: 1360–1370. 24. Davis, ME, Zuckerman, JE, Choi, CH, Seligson, D, Tolcher, A, Alabi, CA et al. (2010). Evidence of RNAi in humans from systemically administered siRNA via targeted nanoparticles. Nature 464: 1067–1070. 25. Akinc, A, Querbes, W, De, S, Qin, J, Frank-Kamenetsky, M, Jayaprakash, KN et al. (2010). Targeted delivery of RNAi therapeutics with endogenous and exogenous ligand-based mechanisms. Mol Ther 18: 1357–1364. 26. Bisgaier, CL, Siebenkas, MV and Williams, KJ (1989). Effects of apolipoproteins A-IV and A-I on the uptake of phospholipid liposomes by hepatocytes. J Biol Chem 264: 862–866. 27. Love, KT, Mahon, KP, Levins, CG, Whitehead, KA, Querbes, W, Dorkin, JR et al. (2010). Lipid-like materials for low-dose, in vivo gene silencing. Proc Natl Acad Sci USA 107: 1864–1869. 28. Tao, W, Davide, JP, Cai, M, Zhang, GJ, South, VJ, Matter, A et al. (2010). Noninvasive imaging of lipid nanoparticle-mediated systemic delivery of small-interfering RNA to the liver. Mol Ther 18: 1657–1666. 29. Alexis, F, Pridgen, E, Molnar, LK and Farokhzad, OC (2008). Factors affecting the clearance and biodistribution of polymeric nanoparticles. Mol Pharm 5: 505–515. 30. Immordino, ML, Dosio, F and Cattel, L (2006). Stealth liposomes: review of the basic science, rationale, and clinical applications, existing and potential. Int J Nanomedicine 1: 297–315. 31. Klibanov, AL, Maruyama, K, Torchilin, VP and Huang, L (1990). Amphipathic polyethyleneglycols effectively prolong the circulation time of liposomes. FEBS Lett 268: 235–237. 32. Robbins, M, Judge, A and MacLachlan, I (2009). siRNA and innate immunity. Oligonucleotides 19: 89–102. 33. Maus, UA, Wellmann, S, Hampl, C, Kuziel, WA, Srivastava, M, Mack, M et al. (2005). CCR2-positive monocytes recruited to inflamed lungs downregulate local CCL2 chemokine levels. Am J Physiol Lung Cell Mol Physiol 288: L350–L358. 34. Moore, BB, Murray, L, Das, A, Wilke, CA, Herrygers, AB and Toews, GB (2006). The role of CCL12 in the recruitment of fibrocytes and lung fibrosis. Am J Respir Cell Mol Biol 35: 175–181. 35. Szymczak, WA and Deepe, GS Jr (2009). The CCL7-CCL2-CCR2 axis regulates IL-4 production in lungs and fungal immunity. J Immunol 183: 1964–1974. 36. Bauer, JW, Baechler, EC, Petri, M, Batliwalla, FM, Crawford, D, Ortmann, WA et al. (2006). Elevated serum levels of interferon-regulated chemokines are biomarkers for active human systemic lupus erythematosus. PLoS Med 3: e491. 37. Lee, PY, Li, Y, Kumagai, Y, Xu, Y, Weinstein, JS, Kellner, ES et al. (2009). Type I interferon modulates monocyte recruitment and maturation in chronic inflammation. Am J Pathol 175: 2023–2033. 38. Judge, AD, Bola, G, Lee, AC and MacLachlan, I (2006). Design of noninflammatory synthetic siRNA mediating potent gene silencing in vivo. Mol Ther 13: 494–505. 39. Beyerle, A, Braun, A, Merkel, O, Koch, F, Kissel, T and Stoeger, T (2011). Comparative in vivo study of poly(ethylene imine)/siRNA complexes for pulmonary delivery in mice. J Control Release 151: 51–56. 40. Ghosn, B, Singh, A, Li, M, Vlassov, AV, Burnett, C, Puri, N et al. (2010). Efficient gene silencing in lungs and liver using imidazole-modified chitosan as a nanocarrier for small interfering RNA. Oligonucleotides 20: 163–172. 41. Günther, M, Lipka, J, Malek, A, Gutsch, D, Kreyling, W and Aigner, A (2011). Polyethylenimines for RNAi-mediated gene targeting in vivo and siRNA delivery to the lung. Eur J Pharm Biopharm 77: 438–449. 42. Merkel, OM, Mintzer, MA, Librizzi, D, Samsonova, O, Dicke, T, Sproat, B et al. (2010). Triazine dendrimers as nonviral vectors for in vitro and in vivo RNAi: the effects of peripheral groups and core structure on biological activity. Mol Pharm 7: 969–983. 43. Varkouhi, AK, Lammers, T, Schiffelers, RM, van Steenbergen, MJ, Hennink, WE and Storm, G (2011). Gene silencing activity of siRNA polyplexes based on biodegradable polymers. Eur J Pharm Biopharm 77: 450–457. 44. Garbuzenko, OB, Saad, M, Betigeri, S, Zhang, M, Vetcher, AA, Soldatenkov, VA et al. (2009). Intratracheal versus intravenous liposomal delivery of siRNA, antisense oligonucleotides and anticancer drug. Pharm Res 26: 382–394. 45. Gutbier, B, Kube, SM, Reppe, K, Santel, A, Lange, C, Kaufmann, J et al. (2010). RNAimediated suppression of constitutive pulmonary gene expression by small interfering RNA in mice. Pulm Pharmacol Ther 23: 334–344. 46. Garbuzenko, OB, Saad, M, Pozharov, VP, Reuhl, KR, Mainelis, G and Minko, T (2010). Inhibition of lung tumor growth by complex pulmonary delivery of drugs with oligonucleotides as suppressors of cellular resistance. Proc Natl Acad Sci USA 107: 10737–10742. 47. Damha, MJ and Ogilvie, KK (1993). Oligoribonucleotide synthesis. The silylphosphoramidite method. Methods Mol Biol 20: 81–114. 48. Iyer, RP, Uznanski, B, Boal, J, Storm, C, Egan, W, Matsukura, M et al. (1990). Abasic oligodeoxyribonucleoside phosphorothioates: synthesis and evaluation as anti-HIV-1 agents. Nucleic Acids Res 18: 2855–2859. 49. Constien, R, Forde, A, Liliensiek, B, Gröne, HJ, Nawroth, P, Hämmerling, G et al. (2001). Characterization of a novel EGFP reporter mouse to monitor Cre recombination as demonstrated by a Tie2 Cre mouse line. Genesis 30: 36–44.

www.moleculartherapy.org vol. 20 no. 1 jan. 2012