Ligand-Modified Human Serum Albumin Nanoparticles for Enhanced

Jul 28, 2015 - Nadja Noske,. ∥. Christine Günther,. ∥. Klaus Langer,. † and Erwin Gorjup*,§. †. Institute of Pharmaceutical Technology and B...
0 downloads 0 Views 2MB Size
Page 1 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

1

Ligand-modified human serum albumin

2

nanoparticles for enhanced gene delivery

3

Jennifer Look†#, Nadine Wilhelm‡#, Hagen von Briesen‡, Nadja Noske§, Christine Günther§,

4

Klaus Langer†, Erwin Gorjup‡*

5 6 7 8 9 10 11 12 13 14 15 16



Institute of Pharmaceutical Technology and Biopharmacy, University of Muenster, Corrensstr. 48, Muenster 48149, Germany ‡

Fraunhofer Institute for Biomedical Engineering, Ensheimer Straße 48, 66386 St. Ingbert, Germany §

apceth GmbH & Co. KG, Max-Lebsche-Platz 30, 81377 Munich, Germany

#

Both authors contributed equally to this work.

*

To whom correspondence should be addressed: Erwin Gorjup, Fraunhofer Institute for Biomedical Engineering, Ensheimer Straße, 48, 66386 St. Ingbert, Germany, Tel: +49 (0) 6894/980-274, Fax: +49 (0) 6894/980-185, [email protected]

ACS Paragon Plus Environment

1

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 40

17

ABSTRACT

18

The development of non-viral gene delivery systems is a great challenge to enable safe gene

19

therapy. In this study, ligand-modified nanoparticles based on human serum albumin (HSA)

20

were developed and optimized for an efficient gene therapy. Different glutaraldehyde

21

crosslinking degrees were investigated to optimize the HSA nanoparticles for gene delivery. The

22

peptide sequence arginine-glycine-aspartate (RGD) as well as the HIV-1 transactivator of

23

transduction sequence (Tat) are well known as promising targeting ligands. Plasmid DNA loaded

24

HSA nanoparticles were covalently modified on their surface with these different ligands. The

25

transfection potential of the obtained plasmid DNA loaded RGD- and Tat-modified nanoparticles

26

was investigated in vitro and optimal incubation conditions for these preparations were studied.

27

It turned out, that Tat-modified HSA nanoparticles with the lowest crosslinking degree of 20%

28

showed the highest transfection potential. Taken together, ligand-functionalized HSA

29

nanoparticles represent promising tools for efficient and safe gene therapy.

30 31

KEYWORDS

32

albumin, nanoparticle, modification, gene delivery, human mesenchymal stem cell

33

ACS Paragon Plus Environment

2

Page 3 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

34

INTRODUCTION

35

Gene therapy is a high-potential therapeutic strategy for the treatment of diseases which are

36

based on a genetic defect. Currently, most clinical trials in gene therapy target cancer. These

37

multifactorial diseases can be cured by gene therapy through delivery of e.g. suicide genes which

38

destroy malignant cells by enzymatic function 1. Genetic modified mesenchymal stem cells

39

(MSC) are suitable to function as a delivery agent for suicide genes in vitro and in vivo in cancer

40

treatment 2-4. In most studies the MSC were genetically modified ex vivo which reduces side

41

effects for the patients. Unfortunately, efficient genetic modification of MSC is current

42

exclusively achieved by viral vectors 5.

43

Virus-based gene delivery is extremely potent and ensures long-term expression of genes 6, 7.

44

However, application in gene therapy is limited due to serious drawbacks like carcinogenicity,

45

immunogenicity, inflammation or high-cost production 8-13. Consequently, non-viral options for

46

gene transfer were developed such as lipoplexes, polyplexes based on poly-L-lysine (PLL) or

47

poly amidoamine (PAMAM), and new nanomaterials such as quantum dots or silica

48

nanoparticles 14-19. Beside the advantages of a low immune response or low cost production in

49

large quantities, most of the non-viral vectors were not applied for gene therapy due to low

50

transfection efficiency or high cytotoxicity in vitro 20, 21. However, human serum albumin based

51

nanoparticles (HSA-NP) are known to be nontoxic, non-immunogenic and biodegradable 22-24. In

52

addition, clinical studies proved the promising use of Abraxane™, a HSA-based nanoparticulate

53

drug, for breast cancer therapy 25. Nevertheless, unmodified HSA nanoparticles are inefficient in

54

gene delivery; mostly due to their negative surface charge, which on the one hand impedes

55

binding of the negatively charged plasmid DNA and on the other hand impedes the cellular

56

uptake of the vectors 26. To overcome this hurdle, Fischer et al. developed cationized HSA

ACS Paragon Plus Environment

3

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 40

57

nanoparticles 27. However, these NP were still not able to sufficiently transfect cells without an

58

additional endosomolytic agent. Combination of HSA with polyethylenimine (PEI), known for

59

its positive effect on endosomal release, led to an increased gene expression in HEK293 cells 28.

60

In contrast, PEI-HSA combined nanoparticles interact with cells nonspecifically, which limits

61

their application in vivo. Therefore, modification of HSA nanoparticles with specific ligands for

62

an efficient uptake seems to be a suitable strategy for non-viral gene delivery. Studies have

63

reported that modification of nanoparticles with the arginine-glycine-aspartate (RGD)-containing

64

peptide led to an efficient transfer across the cell membrane of integrin-positive cells such as

65

B16F10 or HUVEC 29. Moreover, Gojgini et al. showed a RGD-mediated gene delivery in

66

mouse MSC with hyaluronic acid hydrogels in a scaffold for local gene therapy 30. Beside

67

specific ligands, cell penetrating peptides (CPP) like the HIV-1 transactivator of transduction

68

sequence (Tat) facilitate cellular uptake of a large variety of cargos 31-33. Suk et al. demonstrated

69

that modification of PEI/DNA complexes with Tat peptides enhanced gene transfection

70

efficiency up to 14-fold in neuronal cells 34. Nevertheless, most of the studies included cytotoxic

71

nanoparticle formulations such as dendrimers or PEI or showed poor transfection efficiency.

72

The aim of this study was to modify HSA nanoparticles with different ligands like RGD or Tat in

73

order to achieve a non-viral and biocompatible ex vivo gene delivery system for the gene therapy

74

with human mesenchymal stem cells (hMSC). Biocompatibility of the gene delivery system is

75

especially important for a future genetic modification of stem cells due to the fact that

76

mesenchymal stem cells are known to differentiate spontaneously in vitro due to stress situations

77

35, 36

78

In the present study, the efficiency of modified HSA-NP for non-viral ex vivo gene delivery was

79

investigated. HSA-nanoparticles were prepared using an ethanol desolvation method and

.

ACS Paragon Plus Environment

4

Page 5 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

80

characterized the RGD- and Tat-modified delivery systems with regard to particle size, surface

81

charge (zeta potential), ligand binding, and plasmid release. The transfection potential of the

82

different modifications was assessed in varying incubation media and was successfully even in

83

absence of an endosomolytic agent in HEK293T cells. Results were promising and indicate that

84

the study needs to follow up with an optimization of the formulation for non-viral ex vivo gene

85

delivery in the hard-to-transfect hMSC.

86

ACS Paragon Plus Environment

5

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 40

87

EXPERIMENTAL SECTION

88

Materials

89

Human serum albumin (fraction V) and glutaraldehyde 25% solution were obtained from Sigma

90

(Steinheim, Germany). The succinimidyl ester of methoxy poly (ethylene glycol) hexanoic acid

91

(mPEG5000-SHA) and the crosslinker NHS-PEG5000-Maleinimide (NHS-PEG5000-Mal) were

92

purchased from JenKem Technology (Plano, USA). The peptides RGD and RAD were obtained

93

from Peptides International (Louisville, USA) and HIV Tat 48-60 Cys peptide from Innovagen

94

AB (Lund, Sweden). All chemicals were of analytical grade and used as received.

95

Preparation of surface-modified plasmid-loaded HSA nanoparticles

96

HSA nanoparticles were prepared by a desolvation technique as described previously

97

(Steinhauser et al., 2009). In principle, 1 ml human serum albumin solution (20 mg/ml, pH 6.0)

98

was incubated with 100 µg of the respective plasmid (pEGFP-N1, pcMV-Luc) for 15 min under

99

constant stirring (550 rpm) at room temperature (RT). Nanoparticle preparation was performed

100

by dropwise addition of 2.7 ml ethanol 96% (v/v) at a rate of 1 ml/min under stirring (550 rpm).

101

After the desolvation process the nanoparticles were stabilized by crosslinking with

102

glutaraldehyde. Therefore volumes of 2.36 µl and 11.80 µl of a glutaraldehyde solution 8%

103

(m/v) were added, which corresponds to a theoretical calculated crosslinking degrees of 20% and

104

100%, respectively. It was assumed that a glutaraldehyde concentration of 100% enables a

105

theoretic crosslinking of 60 primary amino groups present in one HSA molecule but does not

106

necessarily lead to a quantitative HSA crosslinking. The crosslinking process was performed for

107

at least 12 h under constant stirring (220 rpm) of the NP suspension at RT. Particles were

108

purified in two cycles by centrifugation (14,000 g, 8 min) and redispersion of the pellet in

ACS Paragon Plus Environment

6

Page 7 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

109

phosphate buffer (pH 8.0). For redispersion a Thermomixer Comfort (Eppendorf AG, Hamburg,

110

Germany) and sonication were used.

111 112

Surface modification with RGD and RAD peptides

113

First step of surface modification with RGD and RAD peptide was the activation of

114

nanoparticles with the heterobifunctional crosslinker NHS-PEG5000-Mal. Therefore, NHS-

115

PEG5000-Mal was dissolved in phosphate buffer (pH 8.0) and added in a 11fold molar excess

116

(0.88 mg per mg NP) to the nanoparticle suspension. After incubation for 1 h in a Thermomixer

117

(600 rpm) at RT the nanoparticles were purified in two cycles by centrifugation at 14,000 g for 8

118

min and redispersion in phosphate buffer (pH 8.0). In the second coupling reaction step an

119

equimolar amount of the deacetylated RGD and RAD peptide was added to the nanoparticles

120

(11.08 µg RGD, 11.29 µg RAD per mg NP), respectively. For deacetylation of the peptides an

121

aliquot of 100 µl of deacetylation solution (0.5 M hydroxylamine, 25 mM EDTA in PBS, pH

122

7.2-7.5) was added to the peptide solution in PBS. After an incubation of at least 12 hours in the

123

Thermomixer (600 rpm, RT) the nanoparticles were purified by centrifugation (14,000 g, 8 min)

124

and redispersion in water. The supernatant after centrifugation was collected for quantification of

125

free peptide.

126

Surface modification with TAT peptide

127

For NP surface modification with Tat peptide the first step of coupling reaction was performed

128

as described above. In brief, the crosslinker NHS-PEG5000-Mal was added in a 11fold molar

129

excess to the purified plasmid-loaded nanoparticles. After incubation and purification the Tat

130

peptide was given to the activated NP in a 1:8 molar ratio (3.5 µg per mg NP). After incubation

131

in the Thermomixer (600 rpm, RT) for at least 12 hours the nanoparticle suspension was purified

ACS Paragon Plus Environment

7

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

132

once and the supernatant collected for quantification of the uncoupled peptide amount. In

133

addition, particles were PEG-modified as a negative control. Therefore, plasmid-loaded

134

nanoparticles in phosphate buffer (pH 8.0) were mixed with a 11fold molar excess of

135

mPEG5000-SHA and purified two times after an incubation of 1 h in the Thermomixer (600

136

rpm, RT).

137

Particle characterization

138

The resulting particle yield after purification was determined gravimetrically. Therefore an

139

aliquot (20.0 µl) of the respective nanoparticle sample was put in micro weighing dishes (VWR

140

International GmbH, Darmstadt, Germany) and dried for 2 h at 80°C. The content of the

141

nanoparticles was calculated from the difference of the empty and the nanoparticle-filled dish.

142

Determination of particle diameter and polydispersity index was performed by photon

143

correlation spectroscopy (PCS) with a Zetasizer Nano ZS (Malvern Instruments GmbH,

144

Herrenberg, Germany). The measurements were carried out at 22°C and a scattering angle of

145

173°. Particle diameter was calculated from the intensity of the scattered light (Z-average).

146

Investigations on the nanoparticle surface charge were measured by laser Doppler

147

microelectrophoresis using the zetapotential mode of the same instrument. Before measurement,

148

the samples were diluted with purified water to a concentration of 0.05 mg/ml.

149

Digestion of plasmid-loaded nanoparticles and agarose gel electrophoresis

150

An aliquot of 2 mg NP with maximum 10 µg plasmid was digested with proteinase K solution

151

followed by plasmid extraction via a silica-based spin column. For these digestion and extraction

152

steps DNeasy® Blood & Tissue Kit (QIAGEN GmbH, Hilden, Germany) was used. Agarose gel

153

(0.6%) was prepared with broad range agarose (Carl Roth GmbH & Co. KG, Karlsruhe,

154

Deutschland) in TAE buffer. The marker, control plasmid and the extracted plasmids from the

ACS Paragon Plus Environment

8

Page 9 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

155

different nanoparticle batches were loaded onto the gel together with 2 µl of the nucleic acid

156

stain SafeWhite (NBS Biologicals Ltd., Cambridgeshire, UK), respectively. Electrophoresis was

157

performed at a constant voltage of 80 V for 1 h in TAE buffer. Bands corresponding to the

158

plasmid were detected under UV light and photographed.

159

Entrapment efficiency of plasmid-loaded HSA nanoparticles

160

The amount of plasmid incorporated into the HSA nanoparticles was measured indirectly in the

161

supernatant after nanoparticle purification. The content of free plasmid DNA was determined

162

using Quant-iTTM PicoGreen® dsDNA Assay Kit (Life Technologies, Eugene, USA) according

163

to the operating instruction. The fluorescence was measured by a microplate reader SynergyTM

164

Mx (BioTek Instruments GmbH, Bad Friedrichshall, Germany) at excitation and emission

165

wavelengths of 480 nm and 520 nm, respectively. The incorporated amount of plasmid in the

166

nanoparticles was calculated by the difference between the used amount per mg NP and the

167

detected free amount of plasmid per mg NP in the supernatants of the purification process.

168

Storage of HSA nanoparticles

169

To analyze the storage stability of the HSA nanoparticles, the presence of free plasmid DNA was

170

analyzed with unmodified 20% and 100% crosslinked nanoparticles under four different

171

conditions over 28 days. The following storage conditions were investigated: purified water at

172

8°C, cell medium DMEM with 10% fetal bovine serum (FBS), cell medium DMEM with 10%

173

heat inactivated FBS and cell medium OptiMEM®, all at 37°C, respectively. The heat

174

inactivation of FBS was performed in a water bath at 56°C for 30 min. Typically, lyophilized

175

nanoparticles were resuspended in the respective medium at a concentration of 2.5 mg/ml and

176

aliquoted in separate microcentrifuge tubes for each time point and then stored at 8°C and 37°C,

177

respectively. At predetermined time intervals the tubes were withdrawn and centrifuged at

ACS Paragon Plus Environment

9

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 40

178

20,238 g for 30 min. The supernatants were collected and the amount of released DNA was

179

measured fluorimetrically with Quant-iTTM PicoGreen® dsDNA Assay Kit in a microplate

180

reader as described above.

181

Peptide quantification by HPLC analysis

182

The amount of RGD, RAD, and Tat peptide bound to nanoparticle surface was calculated as the

183

difference between the total amount of the initial peptide added and the amount of peptide

184

measured in the supernatant obtained during the purification steps. The peptide amount was

185

determined by a C18-RP-HPLC method on a Gemini® 5 µm NX C18 110 Å column (250 x 4.6

186

mm) with a gradient elution program. As mobile phase 0.1% trifluoro acetic acid in purified

187

water (A) and acetonitrile (B) were used. For the RGD and RAD peptide the column was

188

equilibrated with 90% A and separation was performed at a flow rate of 1.0 ml/min with a linear

189

gradient (eluent A : eluent B) with the following steps: 0 min (90:10), 8 min (50:50), 10 min

190

(90:10), and 13 min (90:10). For the Tat peptide the column was also equilibrated with 90% A

191

and the separation was performed with the steps: 0 min (90:10), 10 min (70:30), 12 min (90:10),

192

and 15 min (90:10). In both cases the injection volume was 20.0 µl. The detection was performed

193

using a diode array detector by measuring the absorbance at 220 nm.

194

Cell culture

195

All cells were cultured in a humidified atmosphere at 5% CO2 and 37°C. Medium was changed

196

twice a week and cells were subcultured at a maximum confluence of 80-90%. Human epithelial

197

kidney (HEK) 293T cells were cultured with culture medium (DMEM, 10% fetal bovine serum

198

(FBS) and 100 U/ml penicillin and 100 µg/ml streptomycin), unless stated otherwise. Human

199

mesenchymal stem cells (hMSC) were isolated from bone marrow of the caput femoris and

200

cultured in α-MEM supplemented with 15% FBS, 100 U/ml penicillin and 100 µg/ml

ACS Paragon Plus Environment

10

Page 11 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

201

streptomycin. Isolated hMSC were characterized by the expression of the surface marker CD29,

202

CD44, CD73, CD90, CD105, CD106 and HLA-ABC and absence of CD34, CD45, CD133 and

203

HLA-DR, as well as by their adipogenic and osteogenic differentiation capacity.

204

Flow cytometry analysis

205

HEK 293T cells and human mesenchymal stem cells were seeded 24 h before nanoparticle

206

incubation at 7.5x104 or 1.5x104 cells/cm2, respectively. Cells were treated with 50 µg

207

nanoparticles per cm2 growth area in fresh culture medium. After incubation for 24 h, cells were

208

washed with PBS and harvested. After fixation by 10 g/l PFA and 8.5 g/l NaCl in PBS, pH 7.4

209

for 30 min cells were analyzed by flow cytometry with 10,000 cells per sample, using

210

FACSCalibur and CellQuest Pro software (Becton Dickinson, Heidelberg, Germany).

211

Nanoparticles could be detected via their autofluorescence at 488/520 nm.

212

CLSM analysis

213

Cells were grown on culture slides (Becton Dickinson, Heidelberg, Germany) and incubated with

214

HSA nanoparticles as described above for binding analysis. After 48 h of incubation, cells were

215

washed with PBS and cytoplasm was stained with CellTracker™ Blue CMAC dye (Life

216

Technologies, Darmstadt, Germany) according to the manufacturer´s instructions. Cells were

217

fixed with ice-cold 70% ethanol for 5 min and covered with VECTASHIELD HardSet Mounting

218

medium (Vector Laboratories, Burlingame, CA, USA). Samples were stored at 4°C until analysis

219

with a TCS SP8 confocal microscope (Leica microsystem, Heidelberg, Germany).

220

Gene expression analysis

221

Cells were seeded 24 h before transfection experiments as described before. Culture medium was

222

replaced with fresh medium, fresh medium supplemented with 0.1 mM chloroquine or

ACS Paragon Plus Environment

11

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 40

223

OptiMEM® (Life Technologies, Darmstadt, Germany) and added with 50 µg/cm2 nanoparticles.

224

Incubation medium was changed after 24 h and cells were cultured in culture medium for further

225

48 h. Expression of the reporter gene eGFP was analyzed with a fluorescence microscope (IX71,

226

Olympus, Hamburg, Germany). Luciferase activity was quantified 72 h post-transfection using

227

luciferase assay system (Promega GmbH, Mannheim, Germany) following manufacturer´s

228

protocol. Luciferase activity was measured in relative light units (RLU) using Tecan Infinite 200

229

microplate reader (Tecan, Mainz, Germany).

230

ACS Paragon Plus Environment

12

Page 13 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

231

RESULTS

232

Preparation of peptide-modified plasmid-loaded HSA nanoparticles

233

Plasmid-loaded nanoparticles based on human serum albumin (HSA) were prepared by a well-

234

established desolvation method (Fig. 1A). After incubation of the plasmid with the HSA

235

solution, the protein precipitates with the plasmid DNA due to ethanol addition. Particles were

236

stabilized by crosslinking with two different amounts of glutaraldehyde which led to crosslinking

237

degrees of 20% and 100%, respectively. After purification RGD and Tat peptides, as well as

238

their negative controls RAD and PEG (Fig. 1C-E), were attached to the particle surface.

239

Therefore, in the first reaction step a bifunctional PEG-based crosslinker was used for NP

240

activation. In the second step the thiol-reactive maleinimide part of the crosslinker was reacted

241

with the thiol group containing peptides (Fig. 1B). RGD- and RAD-modified nanoparticles with

242

20% crosslinking degree were obtained in a diameter range of about 250 nm (Table 1). The sizes

243

of RGD- and RAD-modified 100% crosslinked particles were in the range of 190 nm and were

244

significantly smaller (p