Layer-by-Layer Nanoparticles as an Efficient siRNA Delivery Vehicle for SPARC Silencing

1790 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheimwileyonlinelibrary.com

full papers

Layer-by-Layer Nanoparticles as an Effi cient siRNA Delivery Vehicle for SPARC Silencing

Yang Fei Tan , Raghavendra C. Mundargi , Min Hui Averil Chen , Jacqueline Lessig , Björn Neu , Subbu S. Venkatraman , * and Tina T. Wong*

1. Introduction

‘RNA interference’ (RNAi) – a gene therapy fi rst discov-

ered by Fire, Mello and co-workers in the late 1990s – refers

to the specifi c down-regulation of proteins in target cells

or organs. [ 1–3 ] Gene expression of a multitude of undesired

proteins can be altered through post-transcriptional gene

silencing performed by the introduction of small interfering

RNA (siRNA) molecules into the cytoplasm. [ 1,4 ] siRNA, a

double stranded 21–23 nucleotides RNA duplex, is incorpo-

rated into the RNA-induced silencing complex (RISC) to

cleave in to single strand by Argonaute 2 (Ago2) and pre-

vents protein translation of its complementary mRNA by

means of RNAse activation. [ 5 ] Due to its simplicity and the

low-dose effect, RNAi can be regarded as a promising tool

for an elegant, curative treatment of a wide range of dis-

eases.. Various approaches have been reported for delivering DOI: 10.1002/smll.201303201

Effi cient and safe delivery systems for siRNA therapeutics remain a challenge. Elevated secreted protein, acidic, and rich in cysteine (SPARC) protein expression is associated with tissue scarring and fi brosis. Here we investigate the feasibility of encapsulating SPARC- siRNA in the bilayers of layer-by-layer (LbL) nanoparticles (NPs) with poly( L -arginine) (ARG) and dextran (DXS) as polyelectrolytes. Cellular binding and uptake of LbL NPs as well as siRNA delivery were studied in FibroGRO cells. siGLO-siRNA and SPARC -siRNA were effi ciently coated onto hydroxyapatite nanoparticles. The multilayered NPs were characterized with regard to particle size, zeta potential and surface morphology using dynamic light scattering and transmission electron microscopy. The SPARC-gene silencing and mRNA levels were analyzed using ChemiDOC western blot technique and RT-PCR . The multilayer SPARC-siRNA incorporated nanoparticles are about 200 nm in diameter and are effi ciently internalized into FibroGRO cells. Their intracellular fate was also followed by tagging with suitable reporter siRNA as well as with lysotracker dye; confocal microscopy clearly indicates endosomal escape of the particles. Signifi cant (60%) SPARC-gene knock down was achieved by using 0.4 pmole siRNA /µg of LbL NPs in FibroGRO cells and the relative expression of SPARC mRNA reduced signifi cantly (60%) against untreated cells. The cytotoxicity as evaluated by xCelligence real-time cell proliferation and MTT cell assay, indicated that the SPARC- siRNA -loaded LbL NPs are non-toxic. In conclusion, the LbL NP system described provides a promising, safe and effi cient delivery platform as a non-viral vector for siRNA delivery that uses biopolymers to enhance the gene knock down effi ciency for the development of siRNA therapeutics.

Nanoparticles

Y. F. Tan,[†] T. T. Wong Singapore Eye Research Institute 11 Third Hospital Avenue, 168751 , Singapore E-mail: [email protected]

R. C. Mundargi,[†] M. H. A. Chen, J. Lessig, S. S. Venkatraman, T. T. Wong School of Materials Science and Engineering Nanyang Technological University 50 Nanyang Avenue, 639798 , Singapore E-mail: [email protected]

B. Neu Faculty of Life Sciences Rhine-Waal University of Applied Sciences Landwehr 4 D-47533 , Kleve , Germany

[†]Yang Fei Tan and Raghavendra C. Mundargi contributed equally to this work.

small 2014, 10, No. 9, 1790–1798

http://doi.wiley.com/10.1002/smll.201303201

Layer-by-Layer Nanoparticles as an Efficient siRNA Delivery Vehicle for SPARC Silencing

1791www.small-journal.com© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

siRNA into the cytoplasm, such as polymeric nanoparticles

(NPs), liposomes and surface modifi cations by folate, choles-

terol, biotin or fl uorescent molecules. [ 6–8 ]

To improve gene silencing effi ciency, viral vectors

have been utilized for siRNA delivery as well. Neverthe-

less, overcoming viral vector oncogenicity and immuno-

genicity remains a signifi cant barrier for viral-based siRNA

delivery. [ 9 ] The poor cellular uptake of naked siRNA , its

rapid degradation by RNAses and the diffi culty of targeting

of siRNA to systemic disease sites are currently limiting

the widespread use of siRNA therapeutics. [ 6 ] To overcome

this, lipid or polymer-based siRNA delivery systems [ 9,10 ]

have been successfully used for local siRNA delivery, par-

ticularly to ocular, intradermal, liver, neural, pulmonary tar-

gets. [ 11 ] In addition to effective cellular uptake, non-toxicity/

non-immunogenicity of the carriers and effective intracel-

lular delivery of siRNA are essential for RNAi to function

as therapeutics. [ 7 ] Therefore, current research focuses on

non-viral vectors, such as polymer-based nanoparticles to

overcome these challenges. However, most non-viral car-

riers lack acceptable effi cacy and possess a high level of

cytotoxicity. [ 8 ]

The layer-by-layer (LbL) self assembly of polycations

and polyanions on colloids was fi rst described by Donath

et al. [ 9 ] The gentle assembly based on electrostatic interac-

tions between positively and negatively charged polymers is

a simple and versatile method with high applicability. Former

studies focused on micro/nanoparticles such as polysty-

rene latex, [ 11 ] silica [ 12–14 ] and melamine formaldehyde [ 15–17 ]

and the more biocompatible templates with calcium car-

bonate, [ 18–20 ] poly(D,L-lactide- co -glycolide) [ 21–23 ] (PLGA)

fl at templates, [ 24 ] as well as biological cells. [ 25,26 ] Although

the LbL technique has already been applied for sub-micron

sized particles in a few reported approaches [ 27,28 ] such as

gold nanoparticles, quantum dots, [ 29–32 ] PLGA NPs, [ 22,23 ]

poly- L -lactic acid NPs, [ 33,34 ] the use of biocompatible and

biodegradable nanoparticles, as hydroxyapatite, is not only

novel but also potentially much less cytotoxic than the

aforementioned candidates. Layer-by-layer nanoparticles

possess signifi cant advantages as siRNA carriers. As highly

customizable carrier systems, they have the potential to pro-

vide high uptake effi ciency, as well as for localized siRNA

release as shown in Scheme 1 . Polyelectrolytes of the par-

ticle surface can interact with several macromolecules of

the cytoplasm, thus facilitating multilayer decomposition. [ 12 ]

The gradual degradation of the biopolymer layers occurring

within cells allows release of the multilayer incorporated

cargo.

Secreted protein, acidic and rich in cysteine (SPARC) is

a calcium-binding matricellular protein that modulates cell-

extracellular matrix. There is a strong association between

elevated expression of SPARC and tissue scarring and

fi brosis. Increased expression of SPARC has been observed in

fi brotic disorders and targeting of SPARC expression to mod-

ulate fi brosis has been evaluated as a potential therapeutic

approach. Here we hypothesize that a LbL system based on

hydroxyapatite (HA) will provide effi cient gene knockdown

in fi broblasts. We investigate knockdown effi ciency with a

layer by layer construction using poly- L -arginine (ARG),

dextran sulfate (DXS), optimized for cellular penetration

and defoliation.

This carrier incorporates the SPARC- siRNA for the intra-

cellular transport and for targeting the SPARC-mRNA in

fi broblasts (FibroGRO). In order to study cellular binding,

uptake, intracellular processing of SPARC- siRNA coated

onto LbL NPs, siGLO-Green transfection indicator (siGLO)

was used as a multilayer constituent in LbL NPs. Nano-

particle uptake and cellular transfection were investigated

in FibroGRO cells by means of confocal microscopy and

fl ow cytometry. The targeted protein expression levels in

FibroGRO cells are evaluated by ChemiDOC® western blot

technique, while the mRNA levels (post LbL treatment) was

quantifi ed by using RT-PCR. The cytotoxicity of LbL NPs

was evaluated by xCelligence real-time cell proliferation and

MTT cell assay. (Supplementary data).

2. Results and Discussion

2.1. Fabrication and Functionalization of LbL NPs

Layer-by-Layer (LbL) NPs were fabricated using commer-

cially available HA NPs as templates coated with the oppo-

sitely charged biopolymer layers ARG, DXS with SPARC

siRNA in the bilayers. The SPARC- siRNA coatings in the

bilayers and concentration of the polyelectrolytes were opti-

mised to build the multilayered NPs, appreciable siRNA

loading and to prevent agglomeration. Since siRNA coating

into the NP multilayer is only meaningful if layer delami-

nation with a subsequent siRNA release into the cytoplasm

takes place, the defoliation was studied by means of siRNA

reporter molecules incorporated into LbL multilayer. To

evaluate cellular uptake of LbL NPs, siGLO - siRNA green

transfection indicator was coated in the bilayers of LbL NPs

as a negatively charged layer; the particle size of siGLO-

LbL NPs is 485 nm with zeta potential of +43 mV. The actual

amount of siGLO coated on bilayers is 0.4 pmole/μg of

LbL nanoparticles with coating effi ciency of 98% ± 0.2. The

Scheme 1. Schematic representation of layer-by-layer nanoparticles defoliation in the cells and siRNA release.

small 2014, 10, No. 9, 1790–1798

Y. F. Tan et al.

1792 www.small-journal.com © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

full papers

amount reported here is deemed suffi cient for gene silencing

by other authors. [ 32 ]

The infl uence of LbL on the size and surface charge

with SPARC-siRNA loaded LbL NPs was also investigated

and results are depicted in Table 1 . The SPARC- siRNA

coated LbL NPs yields a size of 350 nm with zeta potential

of +43 mV. The coating effi ciency of SPARC- siRNA is 97%

with 0.4 pmole SPARC/μg of LbL nanoparticles. For gene

silencing studies, this concentration (0.4 pmole/µg of LbL

nanoparticles) was used to treat the FibroGRO cells.

2.2. LbL NP-Cell Interaction and Cellular Uptake

Cellular binding is facilitated by electrostatic interaction

between positively charged LbL NPs and the cell mem-

brane. Consequently, the charge of the outermost LbL NPs

layer plays a key role in the fi rst step of LbL NPs/cell inter-

action. Therefore, cellular uptake by FibroGRO was studied

with LbL NPs containing ARG as outermost multilayer

component. Apart from charge, size is an important param-

eter infl uencing the cellular uptake rate, the nanoparticles are

easily taken up by cells compared to microparticles. LbL NPs

binding and uptake were analyzed both qualitatively using

CLSM and quantitatively by using fl ow cytometry (FCM)

(2.5 ng LbL NPs/cell). Figure 1 shows typical CLSM images

of a considerable number of LbL NPs inside and attached

to FibroGRO (a-c) cells after one day of incubation. Con-

trols containing LbL NPs with unlabeled siRNA are shown

in traces d-f. According to the transmission (b,e) and overlay

(c,f) pictures, the fl uorescent particles were only present

at cell membranes and in the cytoplasm, but not inside the

nuclei. Nuclei stain by means of DAPI supports this fi nding

that no particles can be found in the nuclei. This indicates that

FibroGRO cells were able to internalise the siGLO-loaded

LbL nanoparticles, however the particles were too large for

nuclear entry. This result corroborates with the fi ndings of

Zhou et al. [ 27 ] that 400 nm PLGA nanoparticles with chitosan/

alginate multilayer are incorporated into hepatocytes, but no

nuclei localization could be found in the confocal images. [ 22 ]

The LbL nanoparticles uptake in

FibroGRO cells was studied by fl ow

cytometer measurements involving trypan

blue quenching to avoid surface bond

nanoparticles on the cells. ( Figure 2 ). The

nanoparticles internalized in the cells are

protected from trypan blue quenching as

living cells do not allow trypan blue pene-

tration. However, particles bound towards

the outer surface of the cell membrane

will be exposed to trypan blue and subse-

quently be quenched.

The histogram of siGLO-loaded LbL

NPs treated FibroGRO cells before (black

graph) and after (light grey graph) trypan

blue quenching is shown. FibroGRO cells

treated with LbL NPs containing unla-

beled siRNA (grey graph) serve as control.

Representative data of three independent

measurements are shown.

The comparison of the fl uorescence

intensities of FibroGRO cells treated with

control LbL NPs containing unlabeled

siRNA with Gmean-8 ± 0.1 and siGLO

loaded-LbL NPs after quenching with

Gmean-100 ± 0.7 clearly indicates cel-

lular uptake of LbL nanoparticles. The

fl uorescence intensity shift after treating

Table 1. Particle size and zeta potential measurements of LbL Nanoparticles.

Layer-by-Layer nanoparticles Mean particle size (nm) Dispersity ± SD Zeta potential (mV ± SD)

HA 121 ± 5 0.23 ± 0.01 −22 ± 1

HA/ARG/ 230 ± 54 0.2 ± 0.04 57.5 ± 8.3

HA/ARG/DXS/ 245 ± 20 0.2 ± 0.08 −63.3 ± 10.7

HA/ARG/DXS/ARG/ 347 ± 44 0.3 ± 0.09 43.1 ± 13.5

HA/ARG/DXS/ARG/SPARC/ 375 ± 107 0.3 ± 0.07 −35.6 ± 8.1

HA/ARG/DXS/ARG/SPARC/ARG/ 350 ± 94 0.3 ± 0.23 43.9 ± 9.8

Figure 1. The cellular interaction of siGLO containing LbL NPs. Shown are confocal laser scanning microscopy images of (a–c) FibroGRO cells one day after incubation with siGLO-loaded LbL nanoparticles. (a) The dot-like green fl uorescence inside the cell bodies demonstrates the uptake of siGLO-loaded LbL nanoparticles by FibroGRO which is especially illustrated by (b) the transmission light and (c) the overlay of both. Unlabeled controls (FibroGRO cells incubated with LbL NPs containing unlabeled siRNA) are shown in traces (d–f). The transmission is demonstrated in trace (e) whereas the overlay can be seen in trace (f). Nuclei are stained by means of DAPI in order to clearly distinguish from the cytoplasm. Shown are representative confocal images of three independent measurements.

small 2014, 10, No. 9, 1790–1798



LbL NPs to FibroGRO cells before and after quenching indi-

cates a strong LbL NPs interaction with FibroGRO cells as

a cellular uptake process. Trypan blue addition to FibroGRO

cells treated with siGLO containing LbL NPs resulted in a

rather marginal fl uorescence intensity shift. Since the cellular

fl uorescence intensity after trypan blue quenching is higher

than the fl uorescence intensity of the control cell group, this

indicates successful internalization of the LbL NPs by the

FibroGRO cells. [ 35,36 ]

2.3. Intracellular Processing of siGLO-loaded LbL Nanoparticles

In order to obtain information on internalization and intra-

cellular localization of the SPARC -siRNA coated LbL NPs,

staining with the Lysotracker was performed during LbL NPs

uptake. As gene silencing takes place at the mRNA stage in

the cytosol, [ 37 ] the fate of the NPs inside the cells and the

intracellular localization must ensure the availability of siRNA

in this compartment. Escape from the endosomes is important

for the effective delivery of siRNA into the cytosol. [ 38 ]

Figure 3 depicts live cell confocal images captured at var-

ious particle incubation times: at four hours post LbL incuba-

tion (Figure 3 a) we see cells surrounded by the LbL NPs but

also some spotty green fl uorescence within the cells, indicating

the presence of siGLO within the NPs. The spotty nature of

Figure 3. The localization of siGLO-loaded LbL NPs within FibroGRO cells. Shown are CLSM investigations of the nanoparticle uptake by FibroGRO cells and a concomitant lysosomal stain. The staining with lysotracker red was performed at different time points; (a) 4 h, (b) 24 h and (c) 72 h. The arrows in Figure 3 b and 3 c point to co-localized regions of NPs and endosomes (orange colour). At 72 hours, the diffuse green regions in Figure 3 c are due to the coalescence of siRNA molecules that have been ‘released’ from the NPs. Scale bar represents 50 µm. Shown are representative confocal images of three independent measurements.

Figure 2. Flow cytometric detection of cellular nanoparticle uptake. HA nanoparticles, coated with ARG, DXS and siGLO were coincubated with FibroGRO cells for one day.

small 2014, 10, No. 9, 1790–1798

Y. F. Tan et al.


full papersthe fl uoroscence is attributed to siGLO still

incorporated within NPs (possibly aggre-

gated). As the incubation time increased to

24 h a colocalization of the green- siGLO

fl uorescence and the red lysotracker was

visible (orange or yellow colour, overlaid in

Figure 3 b) indicating LbL particle localiza-

tion inside endosomes. At 72 hours, Figure 3 c

shows spotty fl uorescent (green) regions as

well as diffuse regions within the cells. This

is attributed to successful escape of NPs

incorporating siGLO from the endosomes

with some concurrent or sequential release

of free siGLO molecules that then coalesce

to form diffuse green regions.

The release of siGLO associated with

LbL NPs is inferred from fl uorescence

extension and coalescence of the spots and

subsequent diffuse distribution of fl uorescence throughout

the cells. [ 39 ] No siRNA-associated fl uorescence was observed

for untreated cells (blanks), and the co-localization of green-

siGLO and endosomes followed by release of siGLO indicates

LbL NPs were successfully taken up into the FibroGRO cells

and that the siGLO release is gradual. [ 40 ] In addition to the

average LbL average particle size of 200 nm as determined

by transmission electron microscopy ( Figure 4 a and 4 b), the

spherical shape and smooth morphology of SPARC -siRNA

coated LbL NPs also probably facilitate the effective cel-

lular uptake. [ 40 ] The FibroGRO cells uptake the poly( L -

arginine) coated LbL NPs by endocytosis, [ 3 ] as observed by

endosomal colocalization [ 39 ] and further the siGLO-LbL NPs

escape from endosomes by pore formation in the endosomal

lipid bilayers. [ 41–43 ] In the case of non-viral vectors such as

positively charged NPs and surface modifi ed PLGA NPs the

major mechanism for cellular uptake is by endocytosis. [ 38,39 ]

Endosomal acidifi cation begins almost immediately upon

the nanoparticle’s scission from the plasma membrane, as its

lumen no longer communicates with the surrounding intra-

cellular media. [ 3 ] The ARG peptide (with a pKa of ∼11.5) is

expected to be protonated at pH’s much lower than 11, and

so when they enter endosomes, are already fully positively

charged (no proton sponge effect). Subsequently, the positively

charged LbL NPs interact with phospholipid membranes to

form pores large enogh for the NPs to squeeze through. This

escape is confi rmed by the spotty green regions (which are

not single particles, but could be aggregates). Subsequently,

the layers defoliate to release siGLO molecules in the cytosol,

and these molecules will coalesce to form diffuse green regions

that grow over time of incubation. Herce et al., 41 and Huang 42

have reported that the binding of cationic peptides to the lipid

bilayers leads to internal stress that can be suffi ciently strong

to create pores in the endosomal lipid membranes.

2.4. Gene Knock Down Studies

We report the targeted protein levels evaluation by a stain-

free technology. [ 44 ] Figure 5 depicts the protein separation

from all the lysed protein samples loaded in to the stain free

Figure 4. TEM images of SPARC-siRNA coated fi ve layer LbL nanoparticles; scale bar (a) 200 nm; (b) 50 nm.

Figure 5. Protein separation and antibody treatment using Biorad Chemidoc System: (A) Protein separation in stain free gels. (B) Protein transfer to PVDF membrane using Transblot system. (C) The bands corresponding to SPARC (43 kDa) after substrate treatment viewed in Chemidoc imager with following sample sequence; (1) LbL NPs with SPARC- siRNA layer, (2) LbL NPs with SPARC- siRNA layer, (3) Blank cells, (4) Lipofectamine with SPARC- siRNA , (5) Lipofectamine with scrambled- siRNA .

small 2014, 10, No. 9, 1790–1798



gels, the bands corresponding to SPARC protein (43 kDa)

in gel as well as in membranes clearly suggest the protein

separation and transfer to the membrane by stain free (SF)

technique. The bands corresponding to SPARC (43 kDa)

after substrate treatment indicate the reduction in protein

expression levels with LbL NPs containing SPARC- siRNA

and lipofectamine with SPARC- siRNA but in case of blank

cells and lipofectamine-scramble siRNA there was no decline

in the band intensity indicating no reduction in the protein

expression levels.

The normalization of band intensity

against total protein was performed using

Chemidoc Image Lab™ software and the

data is reported as % knock down with

reference to control cells in Figure 6 . Per-

centage knock down for SPARC loaded-

LbL NPs is 60% and for SPARC with

lipofectamine as positive control is 89%.

The statistical analysis on knock down

effi ciency of LbL NPs compared to blank

cells, as well as scrambled-siRNA indi-

cates signifi cant difference (p < 0.001).

Hence the data indicates the reduction

in the targeted protein expression by

SPARC- siRNA delivered through the LbL

NPs. There was no knock down with non-

specifi c siRNA indicating effectiveness of

LbL NPs as non-viral delivery vector.

In the present study, we have demon-

strated that SPARC- siRNA loaded-LbL

NPs composed of ARG as polycation and

DXS as polyanion could effectively deliver

siRNA into fi broblast cells with much

lower cytotoxicity. Moreover, SPARC- siRNA delivered by

LbL NPs could specifi cally reduce target protein expression

levels in fi broblast cells. These LbL NPs based on HA as core

have several advantages over chemical conjugates of siRNA

based delivery vectors. First, the formulation steps are very

simple, reducing the time for preparation of siRNA delivery

systems. Toxicity of the LbL based siRNA delivery vehicles

appeared to be low, as indicated by cytotoxicity studies. This

fi nding represents an important advantage for future in vivo

applications of LbL NPs systems, because a limiting factor

for gene delivery using other non-viral particles, has been

toxicity.

2.5. Quantitative RT-PCR Studies

To confi rm that the SPARC-protein knockdown is medi-

ated by decreased mRNA expression in FibroGRO cells,

we measured the relative SPARC and RPL13 mRNA levels

post 96 h LbL-SPARC and Lipofectamine-SPARC treatment

( Figure 7 ). The mRNA levels were reduced to 0.4 ± 0.04 fold

(relative to control or untreated cells) with LbL-SPARC

NPs treated cells and is comparable to mRNA levels with

Lipofectamine-SPARC treated cells. There was no reduc-

tion in mRNA levels associated with scrambled- siRNA and

untreated cells, similar results were reported by Wong et al. 45

on knock down of SPARC protein in Human Tenon’s cap-

sule fi broblasts (HTF). The signifi cant (p < 0.01) reduction in

mRNA levels on treatment with LbL NPs clearly indicates

effi cient delivery of SPARC- siRNA into the cells for silencing

the SPARC gene. Furthermore, our study demonstrates that

LbL NPs are effective in decreasing the mRNA levels as well

as SPARC protein expression in the fi broblast cells.

Fibrosis, which is the secretion and deposition of the cell

extracellular matrix (ECM), is a frequent result of various

Figure 6. Percentage Knock down effi ciency of SPARC-siRNA loaded LbL nanoparticles in fi broGRO cells, 96 h post nanoparticles treatment. LbL nanoparticles were composed of one layer of SPARC-siRNA, Blank cells and Lipofectamine (LF) with SPARC-siRNA, LF with scrambled-siRNA. *p < 0.001. Results were obtained from three independent experiments.

Figure 7. Relative mRNA levels of SPARC and RPL13 showing effi cient gene silencing using SPARC- siRNA loaded LbL NPs in FibroGRO cells. Values are shown as relative to control on 96 h post LbL treatment with LbL-SPARC siRNA, Lipofectamine-SPARC siRNA. Untreated cells and Lipofectamine-scrambled siRNA are used as controls. *p < 0.01. Results were obtained from three independent experiments (n = 3). Mean ± SD.

small 2014, 10, No. 9, 1790–1798

Y. F. Tan et al.


full papersdiseases such as hypertension, diabetes, liver cirrhosis and

infl ammatory processes. [ 46–48 ] In fi brosis, there is elevated

SPARC expression indicating the involvement of the SPARC

protein in modulating ECM interactions. [ 48 ] SPARC expres-

sion and up-regulation has been reported in multiple types

of fi brosis, both in human tissues and animal models. [ 46 ]

Researchers have shown that inhibition of SPARC expression

decreases fi brosis involving dermal, hepatic, renal, pulmonary,

intestinal fi brosis and glaucoma. [ 47 ] Seet et al., reported that

the reduction of SPARC improved surgical success in a sur-

gical mouse model of ocular scarring. [ 49 ] Hence, the targeting

of SPARC expression has been identifi ed as a potential ther-

apeutic strategy for wound modulation and reducing scarring

since SPARC down-regulation also resulted in delayed cell

migration, reduced collagen contractility and lower expres-

sions of profi brotic and pro-infl ammatory genes. [ 45 ] Our

study with LbL nanoparticles targeting SPARC protein and

reducing the mRNA levels in FibroGRO cells clearly confi rm

a promising approach for the use of RNAi as a therapeutic

strategy for minimizing fi brosis.

3. Conclusion

In this study, we have developed biocompatible hydroxyapa-

tite based layer-by-layer nanoparticles incorporating SPARC -siRNA with arginine and dextran as bilayers. The SPARC

siRNA -loaded LbL nanoparticles were successfully taken up

by the fi broblasts to deliver the siRNA . To our knowledge,

this is the fi rst time that the SPARC -siRNA encapsulated LbL

NPs have been reported to demonstrate successful knock

down of SPARC-protein in FibroGRO cells. The initial signif-

icant cellular binding is facilitated by the positively charged

surface layer (ARG), leading to uptake by endocytosis, which

is indicted by co-localization of NPs with endosomes. This is

followed by endsosomal escape of (mostly) intact NPs with

the charged outer layer facilitating pore formation within

the endosomes. Up to 60% SPARC-protein knock down

was achieved with a single LbL NPs treatment. Relative

SPARC and RPL13 mRNA levels signifi cantly were reduced

to 0.4 ± 0.04 folds, demostrating that effi cient SPARC gene

silencing can be achieved by the LbL approach. In addition

there was no observed cytotoxicity associated with LbL NPs

treatment in fi broblasts supporting the safety of LbL NPs as

a potential non-viral vector for siRNA delivery. In conclusion,

these fi ndings support the promising development of layer-

by-layer nanoparticles as a non viral therapeutic approach to

knock down mRNA .

4. Experimental Section

Materials : Hydroxyapatite (HA) nanoparticles (<200 nm), poly- L -arginine hydrochloride (ARG) (M W >70,000) and Dextran sul-fate sodium salt (DXS) (M W >36,000) were procured from Sigma (Singapore) and MP Biomedicals LLC (Singapore). siGLO green transfection indicator was obtained from Dharmacon, Thermo Sci-entifi c (Singapore). Lysotracker Red DND-99 and Opti-MeM reduced serum medium were from Invitrogen (Singapore). FibroGRO human foreskin fi broblasts were from Millipore (Singapore). Dulbecco’s

modifi ed Eagle’s medium (DMEM with high glucose and with L-glutamine), Dulbecco’s PBS without calcium and magnesium, trypsin-EDTA, penicillin-streptomycin and fetal bovine serum (FBS) were from PAA laboratories (Singapore). Trypan blue stock solution (0.4%) was purchased from Sigma-Aldrich (Singapore). SPARC-siRNA (sense-AACAAGACCUUCGACUCUUUC: antisense- GGAAGA-GUCGAAGGUCUUGUU)] Mw-13369 g/mole and scrambled siRNA (sense- GCUCACAGCUCAAUCCUAAUC: antisense-GAUUAGGAUU-GAGCUGUGAGC) were procured from Bio-Rev, South Korea.

Preparation of Layer-by-Layer Nanoparticles : Hydroxyapatite nanoparticles (NPs) were suspended in 0.2 µm fi ltered purifi ed water (Millipore), vortexed for fi ve minutes and collected by centrif-ugation (ST16R, Sorvall) at 12000 rpm for one min, the procedure is repeated to wash the HA nanoparticles. The nanoparticles sus-pension is added to equal amount of 0.5 mg/mL poly( L -arginine, ARG) followed by vortex and sonication for ten minutes and the ARG coated NPs were collected and washed using sodium chlo-ride.. The ARG coated NPs were resuspended in sodium chloride and added to 0.5 mg/mL dextrin sulphate (DXS) as anionic layer and in case of SPARC-siRNA, 4 μL of 20 nM solution is used in the siRNA coating with incubation time of thirty min. Final layer of LbL coating was achieved by adding siRNA coated LbL NPs to ARG fol-lowed by incubation for 10 minutes. Samples for size and zeta potential measurements were collected for each layer, in case of siRNA layer the measurements were carried out in nanodrop (V3.7, Thermoscientifi c) at 230 nm to estimate the siRNA coating on the LbL NPs. We have incorporated siGLO, reporter-siRNA, SPARC-siRNA in LbL NPs separately and the LbL NPs were designated as follows: HA|ARG|DXS|ARG|s iRNA |ARG.

Electrophoretic Mobility : The electrophoretic mobility of the LbL NPs after coating of each layer was measured in 0.2 μm fi l-tered 0.1 M Nacl at room temperature using a Malvern Zetasizer 2000 (Malvern Instruments). The zeta-potential (ζ-potential) was calculated from the electrophoretic mobility (μ) using the Smolu-chowski function ζ = μη/ε where η and ε are the viscosity and per-mittivity of the solvent, respectively.

Cell Culture and Cellular Uptake of LbL NPs : FibroGRO cells derived from human foreskin were cultured in DMEM (10% FBS, penicillin/streptomycin) and incubated at 37 °C, 5% CO 2 . Cells were detached using trypsin-EDTA and passaged at ratios of 1:2 to 1:4.

4 × 10 3 cells were seeded in eight well glass-bottom chamber slide (Nalgene Nunc International, Napperville, IL, USA; size of one well: 0.7 cm × 0.7 cm, in DMEM/10% FBS) overnight. The medium was changed to Opti-Mem before the addition of 10 μg of siGLO-loaded LbL NPs. Further post-incubation for 4 hrs, CLSM meas-urements were done by replacing medium with PBS. DAPI stain for cell nucleus counter stain was performed by adding 300 μL of 300 nM DAPI dilactate to the cells for 5 min and rinsed with PBS. For tracking of acidic organelles within the cells, lysotracker red was added according to manufacturer’s instructions. Confocal micrographs were captured using a confocal laser scanning micro-scope (CLSM), Zeiss LSM 510.

For fl ow cytometric analysis of cell-particle interaction, 2 × 10 5 FibroGRO cells per well were seeded in 6 well plates and cultured overnight. The medium was changed to opti-mem before addition of siGLO-loaded LbL NPs. Following 4 hours incubation, the cells were detached by incubation in trypsin-EDTA for 5 mins at 37 °C and mixed gently to achieve single cells before addition of twice

small 2014, 10, No. 9, 1790–1798



the volume of complete medium. Complete removal of the medium was achieved by centrifugation at 300 rcf and subsequent washes with PBS. The cells were re-suspended in PBS and constantly stored on ice and in the dark until fl ow cytometry measurement. For quenching 10 μL of 0.4% trypan blue was added to the cells to remove membrane-bound LbL NPs. [ 35 ] Flow cytometry of the cells was executed immediately and each experiment was repeated three times and one representative data is presented.

Flow Cytometry (FCM) : The fl uorescence intensities of the labeled particles as well as cells incubation with the particles were investigated by FCM (FACS Calibur, Becton Dickinson, USA). siGLO Green was detected in the FL-1 channel (band pass: 530 ± 15 nm) after a laser excitation at 488 nm. The 10 4 events were detected in the relevant regions for carrier/cell interaction and analyzed using the WinMDI2.9 software.

Confocal Laser Scanning Microscopy (CLSM) : The visualiza-tion of NP/cell interaction and uptake was carried out by means of CLSM (Zeiss, LSM 510 Meta, Jena, Germany). To detect siGLO Green fl uorescence intensities an Ar/Kr laser (excitation wave-length: 488 nm) was used, while red fl uorescence (Lysotracker Red) was detected with a He/Ne laser (excitation wavelength: 543 nm). siGLO and Lysotracker Red emissions were detected with band pass fi lters (505–530 nm and 560–615 nm).

Gene Knock Down Evaluation by Western Blot : Fibroblasts (1.2 × 10 5 , 2 mL) were seeded into a 6-well chamber in complete DMEM medium and cultured for 24 h, SPARC siRNA functionalized LbL NPs, lipofectamine with SPARC and scrambled-siRNA samples containing 400 pmol siRNA were treated to the cells. Cells without any treatment were used as blank cells and optimem was changed to complete DMEM medium after 24 h NPs incubation and cells were harvested, lysed after four days of incubation.

Protein extraction was carried out after preset incubation time of 96 h using cell lyses buffer, the extracted protein in each well was analysed by using coomassi blue at 595 nm. Equivalent amount (30 μg) of protein from different samples and ladder (pre-cision plus TM ) are loaded in to the gel . After protein separation the imaging was performed using a ChemiDoc to validate the protein separation step, further the protein transfer to the PVDF membrane was performed by using Trans-blot® (Bio-Rad) system at 200 V for about 40 min. The membranes after protein transfer were visu-alised in Chemidoc imager to validate the protein transfer to the membrane, further the membranes were blocked in 5% skim milk for 1 h at 20 °C. [ 39 ]

Antibody treatment was accomplished by membranes probing with primary antibody (mouse monoclonal IgG1, Santa Cruz) 1:1000 dilution in 2.5% skim milk and incubated for 1 h at 20 °C, further the membrane was washed with TBST for three time inter-mittently at 10 min. Horseradish peroxidise linked anti-mouse secondary antibody (Jackson Immuno Labs) diluted to 1:10000 in 2.5% skim milk. The membrane with secondary antibody was incubated at 20 °C for 1 h and washed using TBST for six times intermittently at 5 min. Bands were digitally visualized using chemiluminescent substrate in ChemiDoc imager and images were captured using Quantity One software (Bio-Rad). Total protein images were obtained after protein transfer to the PVDF membrane and all the blots were normalized against control cells. For rela-tive quantifi cation of SPARC protein levels, band corresponding to SPARC at 43 kD was selected with triplicate runs and mean value with standard deviation is reported.

RNA Preparation, cDNA Synthesis and Quantitative Real-Time PCR : To study the mRNA levels post LbL-SPARC NPs treatment, cells were plated in six-well tissue culture plates at a density of 120,000 cells/well in DMEM medium for 24 h. SPARC-siRNA loaded LbL NPs, lipofectamine-SPARC and Lipofectamine-scrambled siRNA each corresponding to 400 pmol of siRNA were treated to the cells and mRNA expression was evaluated four days post-treatment.

Total RNA was extracted using the RNeasy Kit (Qiagen, Sin-gapore) according to the manufacturer’s instructions. Firststrand cDNA was synthesized with 250 ng total RNA extract and 1 uL of 50 ng/μL random hexamer primer (Invitrogen Co. Singapore) with Superscript III reverse transcriptase (Invitrogen Co. Singa-pore) according to the manufacturer’s instructions. Quantitative real-time PCR (qPCR) was performed in a total volume of 10 uL in 384-well microtiter plates. Each reaction consisted of 0.5 μL of fi rst-strand reaction product, 0.25 μL each of upstream and downstream primers (10 μM each), 5 μL of Power SYBR Green PCR Master Mix (Applied BioSystems, CA, USA) and 4 μL of DNase-RNasefree distilled water (Sigma-Aldrich Corp., MO, USA). Ampli-fi cation and analysis of cDNA fragments were carried out by use of the Roche LightCycler 480 System (Roche Diagnostics Corp, Indianapolis, USA). All PCR reactions were performed in triplicate. All mRNA levels were measured as CT threshold levels and were normalized with the corresponding 60S ribosomal protein L13’ (RPL13) CT values. Values are expressed as fold increase/decrease over the corresponding values for untreated FibroGRO control cells by the 2 −ΔΔCT method. The primers used for RT-PCR were mentioned in Table 2 .

Statistics : Each experiment was repeated at least three times. All data are expressed as mean ± standard deviation (SD). All values are presented as means of triplicate measurements; Statistical sig-nifi cance ( P < 0.001) was determined using two-tailed t -tests.

Acknowledgments

This work was supported by the Singapore National Medical Research Council (TTW). Authors acknowledge the fi nancial support from AcRF, MOE, Singapore and thank Dr. Scott A. Irvine for fruitful discussions.

[1] A. Fire , S. Xu , M. K. Montgomery , S. A. Kostas , S. E. Driver , C. C. Mello , Nature 1998 , 391 , 806 .

[2] H. L. Jiang , Y. K. Kim , R. Arote , J. W. Nah , M. H. Cho , Y. J. Choi , T. Akaike , C. S. Cho , J Controlled Release 2007 , 117 , 273 .

[3] J. S. Appelbaum , J. R. LaRochelle , B. A. Smith , D. M. Balkin , J. M. Holub , Chem. Biol. 2012 , 19 , 819 .

[4] R. K. M. Leung , P. A. Whittaker , Pharmacol. Ther 2005 , 107 , 222 .

Table 2. Primers used in real-time PCR.

RPL13 gene Forward primer CATCGTGGCTAAACAGGTACTG

Reverse primer GCACGACCTTGAGGGCAGCC

SPARC gene Forward primer GTG CAG AGG AAA CCG AAG AG

Reverse primer TGT TTG CAG TGG TGG TTC TG

small 2014, 10, No. 9, 1790–1798

Y. F. Tan et al.


full papers

Received: October 8, 2013Published online: February 8, 2014

[5] M. T. McManus , P. A. Sharp , Nat. Rev. Genet. 2002 , 3 , 737 . [6] P. Guo , O. Coban , N. M. Snead , J. Trebley , S. Hoeprich , S. Guo ,

Y. ShuP , Adv. Drug Delivery Rev. 2010 , 62 , 650 . [7] M. Kapoor , D. J. Burgess , S. D. Patil , Int. J. Pharm. 2012 , 427 , 35 . [8] S. J. Tan , P. Kiatwuthinon , Y. H. Roh , J. S. Kahn , D. Luo , Small

2011 , 7 , 841 . [9] K. A. Whitehead , R. Langer , D. G. Anderson , Nat. Rev. Drug Discov.

2009 , 8 , 129 . [10] J. Xu , C. Jin , S. Hao , G. Luo , D. Fu , Expert. Opin. Biol. Ther. 2010 ,

10 , 73 . [11] A. de Fougerolles , H.-P. Vornlocher , J. Maraganore , J. Lieberman ,

Nat. Rev. Drug Discov. 2007 , 6 , 443 . [12] E. Donath , G. B. Sukhorukov , F. Caruso , S. A. Davis , H. Moehwald ,

Angew. Chem. Int. Ed. 1998 , 37 , 2201 . [13] G. Decher , Science 1997 , 277 , 1232 . [14] H. K. Na , M. H. Kim , K. Park , S. R. Ryoo , K. E. Lee , H. Jeon , R. Ryoo ,

C. B. Hyeon , D. H. Min , Small 2012 , 8 , 1752 . [15] U. C. Reibetanz , E. Typlt , J. Hofmann , E. Donath , Macromol. Biosci.

2006 , 6 , 153 . [16] X. Yang , X. Han , Y. Zhu , Colloids Surf. A 2005 , 264 , 49 . [17] A. Yu , I. R. Gentle , G. Q. Lu , J. Colloid Interface Sci 2009 , 333 , 341 . [18] C. Gao , S. Moya , H. Lichtenfeld , A. Casoli , H. Fiedler , E. Donath ,

H. Möhwald , Macromol. Mater. Eng. 2001 , 286 , 355 . [19] N. Kato , P. Schuetz , A. Fery , F. Caruso , Macromolecules 2002 , 35 ,

9780 . [20] I. L. Radtchenko , G. B. Sukhorukov , S. Leporatti , G. B. Khomutov ,

E. Donath , H. Möhwald , J. Colloid Interface Sci. 2000 , 230 , 272 . [21] S. De Koker , S. T. Naessens , B. G. De Geest , P. Bogaert ,

J. Demeester , S. C. De Smedt , J. Grooten , Adv. Funct. Mater. 2007 , 17 , 3754 .

[22] D. V. Volodkin , N. I. Larionova , G. B. Sukhorukov , Biomacromol-ecules 2004 , 5 , 1962 .

[23] A. A. Antipov , D. Shchukin , Y. Fedutik , A. I. Petrov , G. B. Sukhorukov , H. Moehwald , Colloids Surf. A 2003 , 224 , 175 .

[24] Y. W. Yang , P. Y. J. Hsu , Biomaterials 2008 , 29 , 2516 . [25] S. Kakade , D. S. Manickam , H. Handa , G. Mao , D. Oupický , Int. J.

Pharm. 2009 , 365 , 44 . [26] J. Zhou , G. Romero , E. Rojas , L. Ma , S. Moya , C. Gao , J. Colloid

Interface Sci. 2010 , 345 , 241 . [27] J. Zhou , S. Moya , L. Ma , C. Gao , J. Shen , Macromol. Biosci. 2009 ,

9 , 326 . [28] G. Decher , J. D. Hong , Macromol. Chem. Macromol. Symp. 1991 ,

46 , 321 .

[29] B. Neu , B. A. Voigt , R. Mitlohner , S. Leporatti , C. Y. Gao , E. Donath , H. Kiesewetter , H. Moehwald , J. Microencapsul. 2001 , 18 , 385 .

[30] S. Moya , L. Dähne , A. Voigt , S. Leporatti , E. Donath , H. Möhwald , Colloids Surf. A 2001 , 183 , 27 .

[31] Y. Wang , A. S. Angelatos , F. Caruso , Chem. Mater. 2007 , 20 , 848 . [32] S. K. Lee , M. S. Han , S. Asokan , C. H. Tung , Small 2011 , 7 , 364 . [33] G. Schneider , G. Decher , Nano Lett. 2004 , 4 , 1833 . [34] K. S. Mayya , B. Schoeler , F. Caruso , Adv. Funct. Mater. 2003 , 13 ,

183 . [35] U. Reibetanz , M. H. A. Chen , S. Mutukumaraswamy , Z. Y. Liaw ,

B. H. L. Oh , Venkatraman , E. S. Donath , B. Neu , Biomacromol-ecules 2010 , 11 , 1779 .

[36] U. Reibetanz , D. Halozan , M. Brumen , E. Donath , Biomacromol-ecules 2007 , 8 , 1927 .

[37] C. Huang , M. Li , C. Chen , Q. Yao , Expert Opin. Ther. Targets 2008 , 12 , 637 .

[38] I. A. Khalil , K. Kogure , H. Akita , H. Harashima , Pharmacol. Rev. 2006 , 58 , 32 .

[39] M. Benfer , T. Kissel , Eur. J. Pharm. Biopharm. 2012 , 80 , 247 . [40] M. Caldorera-Moore , N. Guimard , L. Shi , K. Roy , Expert Opin. Drug

Delivery 2010 , 7 , 479 . [41] H. W. Huang , Phys. Rev. Lett. 2004 , 92 , 19 . [42] H. D. Herce , A. E. Garcia , J. Litt , R. S. Kane , P. Martin , N. Enrique ,

A. Rebolledo , V. Milesi , Biophys. J. 2009 , 97 , 1917 . [43] A. K. Varkouhi , M. Scholte , G. Storm , H. J. Haisma , J. Controlled

Release 2011 , 151 , 220 . [44] A. Gürtler , N. Kunz , M. Gomolka , S. Hornhardt , A. A. Friedl ,

K. McDonald , J. E. Kohn , A. Posch , Anal. Biochem. 2012 , 433 , 105 .

[45] L.-F. Seet , R. Su , L. Z. Toh , T. T. Wong , J. Cell. Mol. Med. 2012 , 16 , 1245 .

[46] J. Trombetta-Esilva , A. D. Bradshaw , Open Rheumatol. J. 2012 , 6 , 146 .

[47] H. Jarvelainen , A. Sainio , M. Koulu , T. N. Wight , R. Penttinen , Pharmacol. Rev. 2009 , 61 , 198 .

[48] P. Bornstein , E. H. Sage , Curr. Opin. Cell Biol. 2002 , 14 , 608 . [49] Li-F. Seet , R. Su , V. A. Barathi , W. S. Lee , R. Poh , Y. M. Heng ,

E. Manser , E. N. Vithana , T. Aung , M. Weaver , E. H. Sage , T. T. Wong , PLoS One 2010 , 5 , 9415 .

small 2014, 10, No. 9, 1790–1798

Documents

Layer-by-Layer Nanoparticles as an Efficient siRNA Delivery Vehicle for SPARC Silencing