Complement activation by PEGylated liposomes containing prednisolone

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx 1

Contents lists available at SciVerse ScienceDirect

European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps

Complement activation by pegylated liposomes containing prednisolone

2 3 4 5 6 7 8 9 10 11 1 2 3 5 14 15 16 17 18 19 20 21 22 23 24

Q1

Jolanda M. van den Hoven a,b,⇑, Reka Nemes b, Josbert M. Metselaar b,c, Bastiaan Nuijen a, Jos H. Beijnen a,b, Gert Storm b,c, Janos Szebeni d,e a

Dept. Pharmacy & Pharmacology, Slotervaart Hospital, Louwesweg 6, 1066 EC Amsterdam, The Netherlands Dept. Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, PO Box 80082, 3508 TB Utrecht, The Netherlands c Dept. Targeted Therapeutics, MIRA Institute, University of Twente, PO Box 217, 7500 AE Enschede, The Netherlands d Dept. Nanobiotechnology and Regenerative Medicine, Faculty of Health, Miskolc University, Budapest, Hungary e Nanomedicine Research and Education Center, Semmelweis University, Bay Zoltan Ltd., Budapest, Hungary b

a r t i c l e

i n f o

Article history: Received 22 August 2012 Received in revised form 28 February 2013 Accepted 8 March 2013 Available online xxxx Keywords: Complement activation PEGylated liposomes Infusion reactions Hypersensitivity reactions

a b s t r a c t Infusion of PEGylated liposomes can give rise to hypersensitivity reactions (HSRs) in a relatively large number of patients. Previously it has been shown that these reactions can be caused by activation of the complement (C) system by a negative charge on the anchor molecule of PEG at the liposomal surface. In this study it is tested whether the activation of the C system by PEG-liposomes could be significantly reduced to values comparable to nonreactive liposomal formulations, by changing the PEGylation-profile on the liposomal surface. Therefore, the formation of C activation markers SC5b-9, C3a, C4d and Bb in normal human serum by both prednisolone loaded and empty liposomes with a variation of PEG chain length, PEG surface concentration, PEG anchor molecule and liposomal size was determined using in vitro assays. The tested liposomes caused no or only mild (30%) activation of C except for one formulation wherein the PEG2000 was anchored to cholesterol (CHOL-PEG2000). The latter liposomes caused paralleling rises in SC5b-9 and Bb levels, suggesting excess activation of the alternative pathway. While the relative safety of weak C activator liposomes remains to be confirmed in vivo, the unique, non-charge and non-antibody-mediated direct conversion of C3 by CHOL-PEG2000 liposomes-although argues against the clinical development of these vesicles-opens new opportunities to understand liposomal C activation at the molecular level. Ó 2013 Published by Elsevier B.V.

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

43 44

1. Introduction

45

Liposomes have been investigated extensively as drug delivery vehicles to increase the therapeutic index of the encapsulated drug, and their versatility to accommodate a wide range of therapeutic agents has been demonstrated in numerous preclinical and clinical settings (Metselaar et al., 2002). Surface modification with a hydrophilic polymer layer, such as poly-ethylene glycol (PEG), opposes their uptake by cells of the mononuclear phagocyte system (MPS) and therefore results in prolonged circulation times. Such long-circulating PEGylated liposomes have demonstrated passive targeting to tumor tissues and inflamed sites, as the locally enhanced vascular permeability allows these drug carriers to extravasate by virtue of the enhanced permeability and retention (EPR) effect. This is one of the mechanisms that improve the therapeutic index of the encapsulated drug (Duncan et al., 2005; Maeda et al., 2000; Reddy, 2000).

46 47 48 49 50 51 52 53 54 55 56 57 58 59

⇑ Corresponding author at: Dept. Pharmacy & Pharmacology, Slotervaart Hospital, Louwesweg 6, 1066 EC Amsterdam, The Netherlands. Tel.: +31 (0)20 512 47 31; fax: +31 (0)20 512 47 53. E-mail address: [email protected] (J.M. van den Hoven).

However, infusion of PEGylated liposomes can give rise to hypersensitivity reactions (HSRs) in a relatively large number of patients (numbers of up to 30% have been reported) (Chanan-Khan et al., 2003; Szebeni et al., 2002, 2012). The reaction usually occurs at the start of infusion and includes symptoms like cardiopulmonary distress, hypo- or hypertension, dyspnea, tachypnea, facial edema and pain in the chest and back. Most of these HSRs are transient and mild, but life-threatening reactions also have been documented in hypersensitive patients (Bates et al., 1995; Chanan-Khan et al., 2003; Laing et al., 1994; Moghimi et al., 2010; Schneider et al., 1998; Szebeni et al., 2012; Vaidya et al., 2002). Since these reactions occur at the first exposure to the drug (without prior sensitization), they are often referred to as ‘pseudoallergy’ and (because of the causal or contributing role of complement activation) the phenomenon is called ‘complement activation related pseudoallergy’ (CARPA) (Szebeni, 2005; Szebeni et al., 2011). Previously it has been shown that C activation by liposomes can be caused by a negative charge on the phospholipid anchor molecule of PEG at the vesicle surface (Moghimi et al., 2006). In an attempt to avoid CARPA, alternative structures for PEG have been developed, like poly(amino acid)s, poly(glycerol) and poly(acrylamide), though these alternative surface structures

0928-0987/$ - see front matter Ó 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.ejps.2013.03.007

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 2 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121

J.M. van den Hoven et al. / European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx

appeared also to induce C activation (Knop et al., 2010; Romberg et al., 2007). To our knowledge, the only long-circulating liposomal formulation that has shown not to activate the C system is a very small-sized (<70 nm) DSPC:CHOL (2:1) liposomal formulation (Szebeni et al., 2002). However, a major disadvantage of these small liposomes is their low internal volume, resulting in a low encapsulation efficiency of water-soluble drugs (Knop et al., 2010). In 1998, Bradley et al. reported that increasing both the PEG chain length and the molar concentration of PEG at the membrane effectively reduces C1q binding, and thereby reduces C activation via the classical pathway (Bradley et al., 1998). This to us was an indication that by changing the properties of the liposomal PEGlayer, the chance of C activation, and, hence, infusion reactions can be minimized while preserving the long circulating profile. The properties of the PEG layer can be altered by changing the PEG chain length, the anchor molecule for PEG in the liposomal bilayer and the grafting density of PEG. Since the severity of C activation is also known to be affected by liposomal size, lamellarity, charge, cholesterol content and the encapsulated drug (Bradley et al., 1998; Szebeni et al., 2012), we also investigated the influence of liposomal size, surface charge, and the presence or absence of encapsulated drug (prednisolone disodium phosphate (PLP)). Several in vitro experiments were performed wherein we measured various C split products, each giving different information on the activation process. Thus, the SC5b-9 enzyme-linked immunosorbent assay (ELISA) assay measures the soluble, non-lytic form of the terminal C complex (TCC), is the end product of activation of the whole C cascade. It is generated by the assembly of C factors C5–C9 and subsequent binding to the regulatory S protein (MüllerEberhard, 1984). The C system is activated via three different routes: the classical, lectin and alternative pathways. Activation of all separate pathways results in activation of the terminal pathway, resulting in formation of SC5b-9 (Kirschfink and Mollnes, 2003). Under normal conditions, activation of all three C pathways results in the formation of C3a from C3, which is also quantified by an ELISA. In the classical and alternative pathways, C3 activation results in the formation of C4d and Bb, respectively, which are quantified by a separate ELISA method, to allow quantitative assessment of the extent of activation of the alternative pathway by the tested formulations.

122

2. Materials and methods

123

2.1. Preparation of liposomes

124

The liposomes were prepared using a film extrusion method (Amselem et al., 1993). Briefly, dipalmitoylphosphatidylcholine (DPPC), 1,2-distearoyl-phosphatidylethanolamine-methyl-polyethyleneglycol conjugate-2000 (DSPE-PEG2000) (both from Lipoid GmbH, Ludwigshaven, Germany), 1,2-distearoyl-phosphatidylethanolamine-methyl-polyethyleneglycol conjugate-550 (DSPE -PEG550, Avanti Polar Lipids Inc., Alabaster, AL, USA), 1,2-distearoyl-SN-glycero-3-phosphatidylethanolamine (DSPE), cholesterolpolyethyleneglycol conjugate-2000 (CHOL-PEG2000) (both from NOF, Grobbendonk, Belgium) and cholesterol (BUFA, Uitgeest, The Netherlands) were dispersed in ethanol in molar ratios as described in Table 1. A lipid film was created by rotary evaporation at 65 °C. The lipid film was hydrated with an aqueous solution containing prednisolone disodiumphosphate (BUFA, Uitgeest, The Netherlands) in a concentration of 139 mg/mL, or phosphate buffered saline (PBS) (B. Braun, Melsungen, Germany) in case of empty liposomes. The resulting coarse dispersion was downsized by multiple extrusion steps through polycarbonate filter membranes with pore sizes of 50 or 100 nm to a final pore size as mentioned in Table 1. The size was confirmed by Dynamic Light

125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143

Scattering (DLS). Subsequently, the unencapsulated prednisolone sodiumphosphate (PLP) was removed by dialysis against PBS using Slide-A-Lyzer dialysis cassettes (Thermo Fisher Scientific, EttenLeur, The Netherlands) with a molecular weight cut-off of 10 kD, with repeated changing of the dialysis medium. All compounds used were of pharmaceutical (Ph. Eur) or highly pure (P99%) grade and were used without any further purification. Since it is the lipid bilayer that causes the C activation, the concentration of bilayer components was kept constant in the tested formulations. Therefore, the lipid content of the liposomal solutions was determined by HPLC and the liposomal solutions were subsequently diluted with PBS to a final total lipid concentration of 55 mM, as these concentrations resulted in successful in vitro analysis previously (Romberg, 2007; Szebeni et al., 2002).

144

2.2. Characterization of liposomes

158

2.2.1. Measurement of size and size distribution The size and size-distribution (polydispersity index, PDI) of liposomes were determined by dynamic light scattering (DLS) with a Malvern ALV CGS-3 system (Malvern instruments Ltd., Malvern, Worcestershire, United Kingdom) with a scattering angle of 90° at 25 °C. Samples were diluted approximately 150 times using phosphate buffered saline (PBS) (B. Braun, Melsungen, Germany) before measurement.

159

2.2.2. Drug and phospholipid analysis Prednisolone phosphate concentrations were determined by high performance chromatography with UV detection (HPLC-UV) using an 1100 series HPLC system consisting of a binary pump, Model G1312A, an autosampler Model G1367A and a UV-detector Model G1314A (all from Agilent technologies, Amstelveen, The Netherlands). A Zorbax Eclipse-XDB-C8 analytical column (750  4.6 mm ID, particle size 5 lm, Agilent Technologies, Palo Alto, California, USA) preceded by a guard column (reversed phase 10  3 mm, Varian, Palo Alto, California, USA) were used. Absorbance was measured at 254 nm. Injection of 10 lL of sample was followed by a linear gradient of 5–90% acetonitrile (Biosolve B.V., Amsterdam, The Netherlands) with 10 mM ammonium formate (Fluka via Sigma–Aldrich, St. Louis, MO, USA). The pH was set at 3.6 using perchloric acid (Merck, Darmstadt, Germany). The flow rate was 1.0 mL/min. Chromatograms were processed using Chromeleon software (Dionex Corporation, Sunnyvale, CA,USA). To determine the amount of (un)encapsulated prednisolone disodium phosphate, an additional dialysis step was performed against PBS using Slide-A-Lyzer dialysis cassette (Thermo Fisher Scientific, Etten-Leur, The Netherlands) with a molecular weight cut-off of 10 kD. A 2 mL sample of the formulation solution was dialyzed against 600 mL of PBS for at least 8 h. Both the permeate and the retentate were analyzed on the above mentioned HPLC-UV system. Lipid concentrations of the separate lipid components were determined by HPLC with evaporative light scattering detection (ELSD) using an 1100 series binary HPLC pump, Model G1312A (Agilent technologies, Amstelveen, The Netherlands), AS 3000 autosampler (Thermo Separation Products, Breda, The Netherlands) and an Alltech Varex MKIII Evaporative Light Scattering Detector (ELSD) (Grace (Alltech), Deerfield, IL, USA). An X-Bridge C18 analytical column (750  4.6 mm ID, particle size 2.5 lm, Waters corporation, Milford, MA, USA) was used. Injection of 30 lL of sample was followed by a linear gradient of 80–100% methanol (Biosolve B.V., Amsterdam, The Netherlands) with 1% triethylamine (Merck, Darmstadt, Germany). The flow rate was 0.4 mL/min. Chromatograms were processed using Chromeleon software.

167

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

145 146 147 148 149 150 151 152 153 154 155 156 157

160 161 162 163 164 165 166

168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 3

J.M. van den Hoven et al. / European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx Table 1 Composition of the tested formulations (molar lipid ratio). Formulation

Liposomal size (nm)

PLP

DPPC

DSPE-PEG2000

DSPE-PEG550

DSPE

CHOL

CHOL-PEG2000

A B C D E F G H I J

80 100 120 80 100 100 100 80 100 100

Yes Yes Yes No No Yes No Yes No No

1.88 1.88 1.88 1.88 1.88 1.88 1.88 1.88 1.88 1.88

0.15 0.15 0.15 0.15 0.15 – – – 0.07 –

– – – – – 0.15 0.15 0.15 – –

– – – – – – – – 0.07 0.15

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.85

– – – – – – – – – 0.15

Abbreviations: PLP, prednisolone sodium phosphate; DPPC, dipalmitoylphosphatidylcholine; DSPE, 1,2-distearoyl-phosphatidylethanolamine; PEG, polyethyleneglycol; CHOL, cholesterol.

206 207 208 209 210 211 212

Prior to HPLC analysis the samples were diluted to a concentration of approximately 1 lg/mL prednisolone phosphate or 3 mg/ mL total lipid. Subsequently, an extraction using dichloromethane (Merck, Darmstadt, Germany), sterile water for injections (B. Braun) and methanol (Biosolve) was performed on the lipid-containing samples, to separate the prednisolone phosphate and the lipid compounds.

218

2.2.3. Measurement of surface charge (zeta-potential) To determine the zeta-potential, the liposomal formulations were diluted 1:100 in 10 mM HEPES buffer. These diluted samples were measured by Laser Doppler Micro-Electrophoresis, using a Zetasizer Nano-Z (Malvern Instruments Ltd., Worcestershire, United Kingdom).

219

2.3. In vitro complement assays in human serum samples

213 214 215 216 217

220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238

2.3.1. SC5b-9. ELISA Whole blood samples of 10 healthy volunteers were collected in 10.0 mL BD VacutainerÒ silicon coated glass serum tubes with no additives (BD, Franklin Lakes, NJ USA). Blood samples were allowed to clot at room temperature and subsequently centrifuged at 3000 rpm for 5 min to collect serum. Serum samples were aliquoted and stored at 20 °C. Frozen samples were rapidly thawed at 37 °C and kept on ice until use. Complement activation was assessed by Microvue SC5b-9 Plus ELISA kits (Quidel Co., SanDiego, CA, USA). Serum from healthy volunteers was incubated with the diluted (55 mM total lipid) liposomal formulations (4:1) in duplicate for 30 min at 37 °C, as this concentration resulted in successful in vitro analysis previously (Romberg, 2007; Szebeni et al., 2002). After incubation the samples were diluted 20-fold in the ‘‘sample diluent’’ of the kit and 100 ll aliquots from this mixture were applied into the wells of the ELISA plate. The assays were performed according to the manual supplied with the kit. The absorbance was measured using a Wallac 1420 Victor 96-wellplate reader (Perkin–Elmer, Waltham, MA,

USA) at 450 nm. SC5b-9 concentrations were calculated using a linear curve fit. All formulations were tested in serum of 10 different individuals. In this article, the percentage of increase in complement activation marker formation compared to PBS (0% activation level) are used to quantify the activation of the complement system. Chanan-Khan et al. previously showed high specificity and sensitivity of SC5b-9 as a biomarker of DoxilÒ-induced hypersensitivity reactions in cancer patients, but only when a more than 2-fold elevation of SC5b-9 levels compared to the baseline SC5b-9 level (0.25 ± 0.12 ug/mL (mean ± SD, n = 50) (Buyon et al., 1992)) was seen. Based on sensitivity and specificity, these data suggest that only more than 2–4-fold (100–300%) increases of SC5b-9 levels can be taken as an elevated risk for the occurrence of hypersensitivity reactions, while a more than 4-fold (or 300%) increase is considered clinically relevant for the occurrence of hypersensitivity reactions (Chanan-Khan et al., 2003). Therefore, we considered a 100–300% (2–4-fold) increase of the SC5b-9 concentration compared to the PBS baseline to be an increased risk on hypersensitivity reactions. In this line of thought, a more than 300% increase in SC5b-9 concentration is considered clinically relevant complement activation, with a high risk for hypersensitivity reactions in the patient. Below 100% and above 30% increase is considered to represent a mild activation of the complement system. A schematic representation of the ELISA assay principle is given in Fig. 1.

239

2.3.2. Other complement activation markers C3a, C4d and Bb concentrations were determined by Microvue C3a Plus ELISA kits, Microvue C4d Plus ELISA kits and Microvue Bb Plus ELISA kits respectively (Quidel Co., SanDiego, CA, USA). Serum samples of the 5 most sensitive individuals (as selected from the SC5b-9 concentrations formed) were incubated with the diluted liposomal formulations (4:1) for 30 min at 37 °C in duplicate. Incubated serum samples were diluted 5000-fold, 70fold and 20-fold for the C3a, C4d and Bb analysis respectively, in

265

Fig. 1. Schematic representation of the ELISA principle.

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264

266 267 268 269 270 271 272 273

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 4

J.M. van den Hoven et al. / European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx

Table 2 Liposome characteristics (concentrations before dilution to 55 mM total lipid concentration). Formulation

A B C D E F G H I J

Mean liposomal size (nm)

PDI

81 97 115 82 97 88 94 81 95 93

0.06 0.06 0.07 0.03 0.17 <0.01 0.05 0.01 0.07 0.03

Encapsulated PLP (mg/mL)

Total lipid concentration (mM)

1.2 3.0 7.4 – – 4.7 – 3.0 – –

55.4 61.5 119 145.3 218.2 98.3 185.8 111.7 155.3 161.2

Abbreviations: PLP, prednisolone sodium phosphate; DPPC, dipalmitoylphosphatidylcholine; DSPE, 1,2-distearoyl-phosphatidylethanolamine; PEG, polyethyleneglycol; CHOL, cholesterol.

274 275 276 277 278 279

the supplied ‘‘sample diluents’’ of the kit and 100 ll aliquots from this mixture were applied into the wells of the ELISA plates. The assays were performed according to the kit manuals. The absorbance was measured using a Wallac 1420 Victor 96-wellplate reader (PerkinElmer, Waltham, MA, USA) at 450 nm. C3a, C4d and Bb concentrations were calculated based on the calibration curves.

280

3. Results

281

3.1. Characteristics

282

288

The liposomal characteristics of the prepared liposomes are given in Table 2. HPLC analysis of the lipid components confirmed the molar lipid ratio of the liposomes as shown in Table 1. Increased PLP concentrations were found upon an increase of the liposomal size, which is a general observation with water-soluble drugs that are (passively) encapsulated in the aqueous interior of the liposome.

289

3.2. In vitro complement activation

283 284 285 286 287

290 291 292 293 294 295 296

First, the formation of the SC5b-9 complex was determined in sera of 10 separate individuals in duplicate. All liposomal formulations induced the formation of SC5b-9 to some extent in a number of sera. Most of these are considered to be mild biologically irrelevant activations, since the SC5b-9 concentrations have increased only 30–100% as compared to incubation with PBS, and are not considered statistically significant different (p > 0.05). However,

the formulation in which PEG2000 was coupled to cholesterol instead of DSPE showed a significant (p < 0.05) 300–1600% increase in SC5b-9 concentration in sera of all tested individuals, indicating that this formulation induced relatively strong C activation (Fig. 2, formulation J). This was confirmed by C3a and Bb analysis, which showed increases of 1100% and 100%, respectively (Fig. 2, formulation J). Formulations containing different amounts of DSPE-PEG were compared regarding their ability to activate the C system. As can be seen from Fig. 2, no significant changes in C activation were seen when the PEG density at the liposomal surface was reduced. With 2.5 mol% of PEG2000 at the liposomal surface (Fig. 2, formulation I) instead of 5 mol% (Fig. 2, formulation E), a comparable mean level of SC5b-9 formation, as well as a comparable number of sera showing (mild) increase of C3a formation were found. Nevertheless, serum of one individual showed >300% increase of C3a formation after incubation with the formulation with lowered surface concentration of PEG2000 (Fig. 2, formulation I), which might be suggestive of a higher risk of hypersensitivity reactions at lower grafting density of PEG, at least in some human sera. It was also hypothesized that liposomes with a shorter PEGchain could cause more activation of the C system. Therefore, formulations were prepared containing PEG550 instead of PEG2000. As can be seen from Fig. 2 (formulation E and G) reducing the PEG chain lengths to PEG550 did not result in a significant increase in the formation of SC5b-9 and C3a. It was expected that an increase in liposomal size could increase C activation marker formation. However, the mean levels of both C3a and SC5b-9, as well as the number of sera of individuals showing (mild) activation, did not significantly change (Fig. 3). An approximately 300% increases of C3a formation as compared to PBS was seen in serum of the same individual for both the smallest (80 nm) and the largest size (120 nm) tested (Fig. 3, formulation A and C). The same was found when comparing different sizes of PEG550 formulations (Formulation G and H, data not shown). Only the largest liposomal size tested (120 nm) showed a significantly increased activation of Bb (Fig. 3), indicating that size might be important for C activation by triggering the alternative pathway. PEGylated liposomes containing PLP tend to show increased C activation as compared to their empty liposomal counterparts (Fig. 4). However, this increase of mean levels of C activation markers is only significant for PEG550 liposomes in the Bb assay. For two of the tested formulations, namely 100 nm PEG2000-liposomes with PLP and 100 nm PEG550-liposomes with PLP (formulations A and F respectively), serum of only one individual revealed a > 300% increase of C3a levels versus PBS. Overall, these observations suggest a limited, pathway-specific impact of intraliposomal PLP on C activation only in some sensitive indiviuals’s sera.

Fig. 2. Formation of complement activation markers after incubation of human serum with 100 nm empty liposomes with 5 mol% DSPE-PEG2000 (E), 5 mol% DSPE-PEG550 (G), 2.5 mol% DSPE-PEG2000 (I) and 5 mol% CHOL-PEG2000 (J). Exact composition of the liposomes is given in Table 1. Formation of the different complement activation markers is given as a% as compared to incubation with PBS (PBS = 0%). Each black dot represents the mean of a duplicate measurement of the complement activation marker in serum of 1 individual. Empty dots represent the mean of all sera, error bars represent the standard deviation.  = significant complement activation marker formation as compared to the other formulations mentioned (p < 0.05).

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 J.M. van den Hoven et al. / European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx

5

Fig. 3. Formation of complement activation markers after incubation with liposomes increasing in size. Human serum was incubated with PLP loaded liposomal formulations containing 5 mol% DSPE-PEG with sizes of approximately 80 nm (A), 100 nm (B) and 120 nm (C). Exact composition of the liposomes is given in Table 1. Formation of the different complement activation markers is given as a% as compared to incubation with PBS (PBS = 0%). Each black dot represents the mean of a duplicate measurement of the complement activation marker in serum of 1 individual. Empty dots represent the mean of all sera, error bars represent the standard deviation.  = significant complement activation marker formation as compared to the other formulations mentioned (p < 0.05).

Fig. 4. PEGylated liposomes with PLP tends to increase activation of the complement system compared their empty counterparts, as was shown for liposomes containing 5 mol% DSPE-PEG2000 of 80 nm (A with encapsulated PLP and D without PLP) and 100 nm (B with encapsulated PLP and E without PLP) respectively, as well as for 100 nm liposomes containing 5 mol% DSPE-PEG550 (F with encapsulated PLP and G without PLP). Each black dot represents the mean of a duplicate measurement of the complement activation marker in 1 individual. Empty dots represent the mean of all individuals, error bars represent standard deviations.  = significant complement activation marker formation as compared to the equivalent empty formulation (p < 0.05).

345 346 347

Considering the uniquely increased C activation by CHOLPEG2000 liposomes (formulation J), we wished to explore the inter-experimental variation of increased C activation, as well as

the role of classical pathway activation in it. To this goal, this formulation was re-tested in 5 reactive sera, measuring by ELISA the formation of SC5b-9, Bb and C4d, taking the non-activating DPPC/

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

348 349 350

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 6

J.M. van den Hoven et al. / European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx

Fig. 5. Complement activation by CHOL-PEG2000 (J) and DSPE-PEG2000 (B) liposomes in normal human sera that showed reactivity for CHOL-PEG2000 liposomes (see Fig. 2, formulation J) relative to PBS (PBS = 100%). Experimental procedures were the same as in Figs. 2–4, except that zymosan (5 mg/mL) was also used as positive control. Error bars represent standard deviations (n = 5).  = significant rise of activation marker (SC5b-9, Bb and C4d) compared to the non-activating formulation B (p < 0.05).

351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367

PEG2000-DSPE liposomes (formulation B) and Zymosan as negative and positive controls. For these liposome preparations we also measured the zeta-potential, which resulted 18.0 ± 1.2 and 1.2 ± 0.3 mV (mean ± SD, n = 4) for formulations B and J, respectively. As shown in Fig. 5, the major rise in the formation of SC5b-9 and Bb caused by CHOL-PEG2000 (J) liposomes vs. DSPE-PEG2000 (B) was reproduced, while C4d showed no difference among these liposome preparations. These findings confirm the major reduction of surface negativity by replacing DSPE with CHOL as PEG anchor, as well as the major increase of C activating power of these nearneutral CHOL-PEG2000 liposomes relative to the negatively charged, but otherwise near equivalent (DSPE-PEG2000) liposomes. Furthermore, these data provide strong evidence for that the excessive C activation of CHOL-PEG2000 liposomes proceeds mainly on the alternative pathway, without involving the classical pathway, at least to a major extent.

368

4. Discussion

369

The objective of this study was to investigate if changing the properties of the PEGlayer on the liposomal surface reduces C activation that has been associated with the hypersensitivity reactions upon infusion of PEGylated liposomes (Chanan-Khan et al., 2003; Szebeni et al., 2002, 2012). Therefore, we measured the concentration of several C activation markers that were formed in vitro after incubation with liposomal formulations with different PEG layer properties. Four C activation markers were selected that are known markers of C activation in different routes and stages. Specifically, C3a is a measure of C3 activation, Bb reflects activation of the alternative pathway, C4d the classical pathway and SC5b-9 provides a measure of the whole cascade until TCC. The sera of the individuals in this study were first screened for SC5b-9 formation since this factor is known to be a sensitive and consistent measure of C activation in human blood (Szebeni, 2012; Szebeni et al., 2003). Previous studies have shown that more than 2–4-fold (100– 300%) increases of SC5b-9 levels in in vitro C assays may be considered as an elevated risk for the occurrence of HSRs in vivo, while a more than 4-fold (or 300%) increase is considered a clinically relevant risk for the occurrence of these reactions (Chanan-Khan et al., 2003). Since all formulations, except the CHOL-PEG2000 formulation showed only mild mean levels of C activation (less than

370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391

100% increased mean concentrations of SC5b-9 as compared to PBS), our data suggest that the PEGylated liposomal formulations studied here have low risk for causing HSRs in man. However, occasional reactions cannot be excluded, as shown by the higher values of C activation (exceeding 100% increase in SC5b-9 concentrations) in sera of specific individuals in our study. Extensive research has been published concerning the occurrence of HSRs after administration of (PEGylated) liposomes. Various groups have shown that factors such as multilamellarity, a large size (>200 nm), a non-circular shape, a high cholesterol content (P71%) a high infusion rate (or bolus injection) all significantly increase the chance of C activation (Moghimi et al., 2011; Szebeni et al., 2000). Previously it was thought that addition of PEG to the liposomal surface would reduce the interaction with plasma proteins and thus with C factors (as was shown in vitro for C1Q) in the systemic circulation (Bradley et al., 1998; Szebeni et al., 2000). However, PEGylated liposomal products, like DoxilÒ, have been shown to induce C activation-related hypersensitivity reactions in patients (Chanan-Khan et al., 2003). Complement activation by liposomes in general has been related to the presence of an anionic phosphate moiety in the PEG-lipid conjugate, though the mechanism of the phenomenon remains elusive (Moghimi et al., 2011). An alternative for PEG without this effect has not yet been found, and therefore PEG still remains the golden standard in the formulation of long circulating liposomes (Knop et al., 2010; Romberg, 2007; Romberg et al., 2007). In this study, it was found that a reduction of PEG chain length caused no significant changes in the activation of the C system in vitro, nor did a reduction of the PEG density at the surface, at least until 2.5 mol%. Previously it has been shown for DoxilÒ, that the presence of doxorubicin influences the activation of the C system (Szebeni et al., 2002, 2012). It was hypothesized that the activation could be caused by a change in physicochemical properties of the liposomal surface due to the formation of elongated doxorubicin crystals in the liposomal aqueous core. In our study however, there also seems to be a tendency to an increased formation of C activation markers after loading the PEGylated liposomes with PLP, which is encapsulated as a solution in the liposomal aqueous interior. We cannot exclude that PLP and other drugs that are encapsulated in dissolved form rather than as crystals, can also cause slight changes to the liposomal formulation that result in an increased chance of C activation. This warrants a more in depth investigation of the physicochemical characteristics of PLP-PEGylated liposomal formulations. The formulation in which cholesterol was used as the anchor molecule for PEG, instead of normally used DSPE, showed unexpectedly strong C activation. Although unsuitable as a potential new drug delivery vehicle in view of the activation results, this formulation might provide a new tool for better understanding the mechanism of C activation by PEGylated liposomes. CHOLPEG2000 differs from DSPE-PEG2000 in that the former has no negatively charged phosphate group in its lipophillic anchor that stabilizes the position of the PEG-lipid-conjugate relative to the membrane surface. Because this charged phosphate group appeared to be the main cause of C activation by PEG-DSPE (Moghimi et al., 2006), it was assumed that replacing it with another, not charged, or less charged PEG carrier might reduce the risk of C activation. However, the opposite was found as CHOL-PEG2000 lead to much more efficient C activation than any other liposome in this study, an activation that was comparable to that of zymosan. This phenomenon of triggering strong C activation by nanoparticles s not limited to liposomes, as for example it has also been encountered in the field of carbon nanotubes (Hamad et al., 2010; Moghimi and Hunter, 2010; Moghimi et al., 2010). Based on the absence of C4d changes, i.e., involvement of the classical pathway, one can rule out a role for antibodies

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457

PHASCI 2719

No. of Pages 7, Model 5G

29 March 2013 J.M. van den Hoven et al. / European Journal of Pharmaceutical Sciences xxx (2013) xxx–xxx

484

(e.g., anti-cholesterol, anti-PEG and anti-phospholipid antibodies (Alving et al., 1977; Swartz et al., 1988; Wassef et al., 1989)) in the excessive C activation by CHOL-PEG2000 vesicles. Increased C activation by near neutral liposomes via the alternative pathway is a phenomenon for which, to our best knowledge, there is no precedence in the C and liposome literature, Although it is likely that cholesterol does not hold the PEG to the liposomes as firmly as DSPE, even if the CHOL-PEG leaves the vesicles during incubation with the serum (in an equilibrium lipid exchange process), it leaves back small, neutral liposomes, which are not supposed to activate C. The phenomenon is clearly puzzling, and may involve rapid structural changes of either the liposomes or CHOL-PEG2000 that leaves the liposomes and may associate with lipoproteins with de novo formation of C activating particles. Future studies will hopefully clarify the puzzle and maybe highlight a novel mechanism of C activation via the alternative pathway. In summary, we found that PLP containing PEGylated liposomes are weak C activators in some, but not all human sera. This suggests a relatively small, but not negligible risk for infusion reactions with these liposomes, i.e., the an occasional rise of a reaction in a hypersensitive individual. This implies a need to either fully eliminate the C activating capability of PLP containing PEGylated liposomes, or develop predictive assays or preventive procedures that enables doctors and nurses to better cope with liposome infusion reactions. Further studies on this phenomenon might also be useful for the elucidation of the mechanisms behind C activation by liposomal formulations in general.

485

Disclosure statement

486 487

JM Metselaar is affiliated to Enceladus Pharmaceuticals which contributed financially to the work reported in this publication.

488

Acknowledgements

489

493

This work was supported by the Dutch Technology Foundation STW, Project 07440, the Hungarian National Office of Technology (Grant NanoMedi) and TAMOP-4.2.1.B-10/2/KONV-2010-0001 (New Hungarian Development Plan, supported by the European Union, and co-financed by the European Social Fund).

494

References

495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517

Alving, C.R., Richards, R.L., Guirguis, A.A., 1977. Cholesterol-dependent human complement activation resulting in damage to liposomal model membranes. J. Immunol. 118, 342–347. Amselem, S., Gabizon, A., Barenholz, Y., 1993. A large-scale method for the preparation of sterile and non-pyrogenic liposomal formulations of defined size distributions for clinical use. In: Gregoriadis, G. (Ed.), Liposome Technology. CRC Press, Boca Raton, Fl, USA, pp. 501–525. Bates, C.M., Carey, P.B., Hind, C.R., 1995. Anaphylaxis due to liposomal amphotericin (AmBisome). Genitourin. Med. 71, 414. Bradley, A.J., Devine, D.V., Ansell, S.M., Janzen, J., Brooks, D.E., 1998. Inhibition of liposome-induced complement activation by incorporated poly(ethylene glycol)-lipids. Arch. Biochem. Biophys. 357, 185–194. Buyon, J.P., Tamerius, J., Belmont, H.M., Abramson, S.B., 1992. Assessment of disease activity and impending flare in patients with systemic lupus erythematosus. Comparison of the use of complement split products and conventional measurements of complement. Arthritis Rheum. 35, 1028–1037. Chanan-Khan, A., Szebeni, J., Savay, S., Liebes, L., Rafique, N.M., Alving, C.R., Muggia, F.M., 2003. Complement activation following first exposure to pegylated liposomal doxorubicin (Doxil): possible role in hypersensitivity reactions. Ann. Oncol. 14, 1430–1437. Duncan, R., Vicent, M., Greco, F., Nicholson, R., 2005. Polymer-drug conjugates: towards a novel approach for the treatment of endrocine-related cancer. Endocr. Relat. Cancer 12, S189–199.

458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483

490 491 492

7

Hamad, I., Al-Hanbali, O., Hunter, A.C., Rutt, K.J., Andresen, T.L., Moghimi, S.M., 2010. Distinct polymer architecture mediates switching of complement activation pathways at the nanosphere-serum interface: implications for stealth nanoparticle engineering. ACS Nano 4, 6629–6638. Kirschfink, M., Mollnes, T.E., 2003. Modern complement analysis. Clin. Vaccine Immunol 10, 982–989. Knop, K., Hoogenboom, R., Fischer, D., Schubert, U.S., 2010. Poly(ethylene glycol) in drug delivery: pros and cons as well as potential alternatives. Angew. Chem. Int. Ed. 49, 6288–6308. Laing, R.B., Milne, L.J., Leen, C.L., Malcolm, G.P., Steers, A.J., 1994. Anaphylactic reactions to liposomal amphotericin. Lancet 344, 682. Maeda, H., Wu, J., Sawa, T., Matsumura, Y., Hori, K., 2000. Tumor vascular permeability and the EPR effect in macromolecular therapeutics: a review. J. Control Release 65, 271–284. Metselaar, J., Mastrobattista, E., Storm, G., 2002. Liposomes for intravenous drug targeting: design and applications. Mini Rev. Med. Chem. 2, 319–329. Moghimi, S.M., Andersen, A.J., Hashemi, S.H., Lettiero, B., Ahmadvand, D., Hunter, A.C., Andresen, T.L., Hamad, I., Szebeni, J., 2010. Complement activation cascade triggered by PEG-PL engineered nanomedicines and carbon nanotubes: the challenges ahead. J. Control Release 146, 175–181. Moghimi, S.M., Hunter, A.C., 2010. Complement monitoring of carbon nanotubes. Nat. Nano 5, 382. Moghimi, S.M., Andersen, A.J., Ahmadvand, D., Wibroe, P.P., Andresen, T.L., Hunter, A.C., 2011. Material properties in complement activation. Adv. Drug Deliv. Rev. 63, 1000–1007. Moghimi, S.M., Hamad, I., Andresen, T.L., Jørgensen, K., Szebeni, J., 2006. Methylation of the phosphate oxygen moiety of phospholipidmethoxy(polyethylene glycol) conjugate prevents PEGylated liposomemediated complement activation and anaphylatoxin production. FASEB J. 20, 2591–2593. Müller-Eberhard, H.J., 1984. The membrane attack complex. Springer Semin. Immunopathol. 7, 93–141. Reddy, K.R., 2000. Controlled-release, pegylation, liposomal formulations: new mechanisms in the delivery of injectable drugs. Ann. Pharmacother. 34, 915– 923. Romberg, B., 2007. Poly(amino acid)s: Next-generation Coatings for LongCirculating Liposomes [PhD thesis], Utrecht University, Utrecht, The Netherlands, pp. 90–99. Romberg, B., Metselaar, J.M., Baranyi, L., Snel, C.J., Bünger, R., Hennink, W.E., Szebeni, J., Storm, G., 2007. Poly(amino acid)s: promising enzymatically degradable stealth coatings for liposomes. Int. J. Pharm. 331, 186–189. Schneider, P., Klein, R.M., Dietze, L., Sohngen, D., Leschke, M., Heyll, A., 1998. Anaphylactic reaction to liposomal amphotericin (AmBisome). Br. J. Haematol. 102, 1108–1109. Swartz Jr, G.M., Gentry, M.K., Amende, L.M., Blanchette-Mackie, E.J., Alving, C.R., 1988. Antibodies to cholesterol. Proc. Natl. Acad. Sci. USA 85, 1902–1906. Szebeni, J., 2012. Hemocompatibility testing for nanomedicines and biologicals: predictive assays for complement mediated infusion reactions. Eur. J. Nanomed. 4, 33–53. Szebeni, J., Baranyi, L., Savay, S., Milosevits, J., Bunger, R., Laverman, P., Metselaar, J., Storm, G., Chanan-Khan, A., Liebes, L., Muggia, F., Cohen, R., Barenholz, Y., Alving, C., 2002. Role of complement activation in hypersensitivity reactions to doxil and hynic PEG liposomes: experimental and clinical studies. J. Liposome Res. 12, 165–172. Szebeni, J., 2005. Complement activation-related pseudoallergy: a new class of drug-induced acute immune toxicity. Toxicology 216, 106–121. Szebeni, J., Baranyi, L., Savay, S., Bodo, M., Morse, D.S., Basta, M., Stahl, G.L., Bünger, R., Alving, C.R., 2000. Liposome-induced pulmonary hypertension: properties and mechanism of a complement-mediated pseudoallergic reaction. Am. J. Physiol. Heart Circ. Physiol. 279, H1319–H1328. Szebeni, J., Baranyi, L., Savay, S., Milosevits, J., Bodo, M., Bunger, R., Alving, C.R., 2003. The interaction of liposomes with the complement system: in vitro and in vivo assays. Methods Enzymol. 373, 136–154. }cs, P., Rozsnyay, Z., Weiszhár, Z., Urbanics, R., Rosivall, L., Cohen, R., Szebeni, J., Bedo Garbuzenko, O., Báthori, G., Tóth, M., Bünger, R., Barenholz, Y., 2012. Liposomeinduced complement activation and related cardiopulmonary distress in pigs: factors promoting reactogenicity of Doxil and Am Bisome. Nanomedicine 8, 176–184. Szebeni, J., Muggia, F., Gabizon, A., Barenholz, Y., 2011. Activation of complement by therapeutic liposomes and other lipid excipient-based therapeutic products: prediction and prevention. Adv. Drug Deliv. Rev. 63, 1020–1030. Vaidya, S.J., Seydel, C., Patel, S.R., Ortin, M., 2002. Anaphylactic reaction to liposomal amphotericin B. Ann. Pharmacother. 36, 1480–1481. Wassef, N.M., Johnson, S.H., Graeber, G.M., Swartz Jr, G.M., Schultz, C.L., Hailey, J.R., Johnson, A.J., Taylor, D.G., Ridgway, R.L., Alving, C.R., 1989. Anaphylactoid reactions mediated by autoantibodies to cholesterol in miniature pigs. J. Immunol. 143, 2990–2995.

Please cite this article in press as: van den Hoven, J.M., et al. Complement activation by pegylated liposomes containing prednisolone. Eur. J. Pharm. Sci. (2013), http://dx.doi.org/10.1016/j.ejps.2013.03.007

518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595