R I t J a b a A R R A A K L I F M T C A h 0 Colloids and Surfaces B: Biointerfaces 159 (2017) 454–467 Contents lists available at ScienceDirect Colloids and Surfaces B: Biointerfaces jo ur nal ho me p ag e: www.elsev ier .com/ locate /co lsur fb eview Article mmunoliposomes: A review on functionalization strategies and argets for drug delivery osimar O. Eloya,∗, Raquel Petrilli b, Lucas Noboru Fatori Trevizana, Marlus Chorilli a School of Pharmaceutical Sciences of Araraquara, São Paulo State University, UNESP, Department of Drugs and Medicines, Araraquara, SP, Brazil School of Pharmaceutical Sciences of Ribeirão Preto, São Paulo State University, USP, Department of Pharmaceutical Sciences, Ribeirão Preto, SP, Brazil r t i c l e i n f o rticle history: eceived 28 April 2017 eceived in revised form 26 July 2017 ccepted 29 July 2017 vailable online 5 August 2017 eywords: iposomes mmunoliposomes unctionalization onoclonal antibodies argeted delivery a b s t r a c t Nanoparticles, especially liposomes, have gained prominence in the field of drug delivery for the treat- ment of human diseases, particularly cancer; they provide several advantages, including controlled drug release, protection of the drug against degradation, improved pharmacokinetics, long circulation, and passive targeting to tumors and inflammatory sites due to the enhanced permeability and retention effect. The functionalization of liposomes with monoclonal antibodies or antibody fragments to generate immunoliposomes has emerged as a promising strategy for targeted delivery to and uptake by cells over- expressing the antigens to these antibodies, with a consequent reduction in side effects. In this review, we address functionalization strategies for the non-covalent and covalent attachment of monoclonal anti- bodies and their fragments to liposomal surfaces. The main reaction occurs between the sulfhydryl groups of thiolated antibodies and maleimide-containing liposomes. Furthermore, we explore the main targeting possibilities with these ligands for the treatment of a variety of pathologies, including HER2- and EGFR- positive cancers, inflammatory and cardiovascular diseases, infectious diseases, and autoimmune and neurodegenerative diseases, which have not previously been reviewed together. Overall, many studies have shown selective delivery of immunoliposomes to target cells, with promising in vivo results, partic- ularly for cancer treatment. Although clinical trials have been conducted, immunoliposomes have not yet received clinical approval. However, immunoliposomes are promising formulations that are expected to become available for therapeutic use after clinical trials prove their safety and efficacy, and after scaling issues are resolved. © 2017 Elsevier B.V. All rights reserved. ontents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 2. Monoclonal antibodies and their fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 3. Strategies to conjugate monoclonal antibodies or their fragments to liposomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456 4. Monoclonal antibody- and antibody fragment-mediated targeted delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458 4.1. Vascular targeting: cardiovascular diseases and cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458 4.2. Cancer targeting: HER2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 4.3. Cancer targeting: EGFR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460 4.4. Cancer targeting: other receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461 4.5. Infectious disease targeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Autoimmune and degenerative disease targeting. . . . . . . . . . . . . . . . . . 5. Immunoliposomes: stimulation of endogenous immune response . . . . . . . ∗ Corresponding author at: School of Pharmaceutical Sciences, Department of Drugs a raraquara, São Paulo 14801-902, Brazil. E-mail address: josimar.eloy@gmail.com (J.O. Eloy). ttp://dx.doi.org/10.1016/j.colsurfb.2017.07.085 927-7765/© 2017 Elsevier B.V. All rights reserved. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .462 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 nd Medicines, São Paulo State University (UNESP), Rodovia Araraquara-Jau, km. 1, dx.doi.org/10.1016/j.colsurfb.2017.07.085 http://www.sciencedirect.com/science/journal/09277765 http://www.elsevier.com/locate/colsurfb http://crossmark.crossref.org/dialog/?doi=10.1016/j.colsurfb.2017.07.085&domain=pdf mailto:josimar.eloy@gmail.com dx.doi.org/10.1016/j.colsurfb.2017.07.085 J.O. Eloy et al. / Colloids and Surfaces B: Biointerfaces 159 (2017) 454–467 455 6. Immunoliposomes: clinical development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 7. Conclusion and future perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464 . . . . . . 1 m r t o d t w b h s c a s f w d t a s c e i t w s r p i r s i p s a p t m c [ n a i l d b a f a l c g References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction Nanoparticle drug delivery systems have potential for the treat- ent of a variety of diseases, and their use in medicine is increasing apidly [1,2]. Approximately 40% of small molecule drugs for cancer reatment, for instance, have low aqueous solubility, so the devel- pment of drug delivery systems capable of encapsulating these rugs, enhancing their aqueous solubility, and delivering them to arget sites is highly desirable [3]. Within this context, liposomes, hich are lipid vesicles containing an aqueous core enclosed by a iocompatible lipid membrane, can be useful to encapsulate both ydrophilic and lipophilic drugs [4,5]. These systems are able to electively deliver the drug to a target site, improve its pharma- okinetics and pharmacological effects, and avoid local irritation nd drug toxicity [5]. Due to their success, many examples of lipo- omes are currently available in the clinic. The first liposomal drug ormulation approved by the Food and Drug Administration (FDA) as Doxil ® in 1995 [6,7]. Since then a variety of liposomal drug elivery products have been developed and are now available on he market, such as DaunoXome ® , DepoCyt ® , Myocet ® , Lipodox ® , nd Marqibo ® [8]. Different targeting strategies can be employed with these ver- atile nanocarriers to optimize selective delivery. For instance, ancer tissues and inflammatory sites have a leaky vasculature that nhances the accumulation of liposomes compared to free drug, n a phenomenon named the enhanced permeability and reten- ion (EPR) effect, which results in passive targeting of liposomes ith a mean diameter of 100–200 nm [1,4,9,10]. Additionally, lipo- ome surfaces can be coated with polyethylene glycol (PEG), which educes interactions with blood components and binding to plasma roteins, thereby preventing interactions with opsonins and reduc- ng capture by the reticular endothelial system. This mechanism is elated to shielding the liposome surface, hydrophilicity, and repul- ion between the coated liposome and blood components, resulting n longer circulation time [11]. Active targeting is another possible approach, which is accom- lished by coupling targeting ligands to the surface of liposomes, uch as monoclonal antibodies or their fragments (fragment ntigen-binding (Fab) and single-chain variable fragment (scFv)), roteins, peptides, or carbohydrates. Thus, liposomes can be selec- ively taken up by cells that overexpress the receptor for the oiety, which has the potential to improve therapeutic out- omes by increasing efficacy and reducing off-target toxicity 1,12]. Among the different moieties that can be covalently or on-covalently attached to the liposome surface, antibodies and ntibody fragments are the most widely employed, producing mmunoliposomes [13]. A monoclonal antibody is formed by 2 heavy (H) chains and 2 ight (L) chains, which are further divided into constant and variable omains (CH/CL and VH/VL, respectively), stabilized by disulfide ridges. They are able to selectively bind to a specific part of an ntigen; thus, they represent a growing class of medicines available or targeted therapies for a variety of diseases [14]. Commonly, the ntibody is attached to the distal end of a PEG chain in PEGylated iposomes, using a variety of methods addressed herein [15]. The orrect orientation of the PEG-terminal antibody facilitates anti- en recognition more effectively than the random orientation of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464 antibodies attached to the liposomal bilayer of PEGylated immuno- liposomes [16]. Immunoliposomes have been studied and reviewed, with a focus on functionalization strategies, comparisons with antibody–drug conjugates and immunotoxins, and applications for cancer therapy [17–19]. However, this field has undergone rapid advancement in recent years, with many studies reporting new functionalization methods and targets in not only cancer therapy but other diseases, which have not been previously reviewed. In this paper, we will address the main aspects related to the development of immunoliposomes using a variety of antibodies or antibody fragments for the treatment of cancer, inflammatory and cardiovascular diseases, infectious pathologies, and autoimmune and degenerative diseases. In doing so, we will review and discuss the different functionalization strategies and targeting approaches used both in vitro and in vivo, with a special focus on preclinical findings. In addition, we will highlight the immunoliposomal formulations that are currently in clinical development. 2. Monoclonal antibodies and their fragments To put the functionalization of liposomes with antibodies into perspective, it is important to provide a brief overview of antibody origin, structure, and function. Antibodies are freely circulating molecules that perform two main functions in the immune sys- tem. First, they recognize and bind antigens, which requires great antibody diversity, due to the huge variety of antigens. Second, antibodies play an important role in eliminating antigens [20]. Anti- bodies belong to the class called globulins because of their globular structure and are, thus, immunoglobulins. Monoclonal antibodies bind to a specific epitope and are gen- erated via hybridoma technology, in which a B cell clone is immortalized by fusion with myeloma cells [20]. The B cells are fused with an immortalized myeloma cell line, then cultured in vitro to eliminate non-hybridized cells. The first cell culture contains a mixture of antibodies; for this reason, a sequence of cell isolations is performed to acquire a final isolate culture that produces a unique type of monoclonal antibody [21,22]. The origin of each clinical monoclonal antibody can be iden- tified by the suffix in its name; for instance, murine antibodies end with –omab, chimeric mouse–human antibodies (the entire antigen-specific variable domain of a mouse antibody is added onto the constant domain of a human antibody) end with –ximab, humanized antibodies (the murine hypervariable region is added onto a human antibody framework) end with –zumab, and human antibodies end with –umab [23]. The structure of an antibody determines several features of its participation in the immune response. These features are important for their specificity and biological activity. Antibodies are divided into 3 main parts: 2 Fab portions that are responsible for anti- gen recognition and specificity, and the fragment crystalline (Fc) portion that binds to receptors on target cells and enables comple- ment fixation, which triggers effector functions that eliminate the antigen (Fig. 1a). The Fabs are formed by the connection of the L and H chains. Furthermore, both the L and H chains are connected by disulfide bonds, and each chain is comprised of domains. The first domain is highly variable (the VL or VH domain) in terms of amino acid sequence from one antibody to another. The second and subsequent domains have little variation of amino acids, being 456 J.O. Eloy et al. / Colloids and Surfaces B: Biointerfaces 159 (2017) 454–467 F ns, 2 i f a con ( tion. ( d [ I u � l ( i a b s e a b r d a i b o o a F o v t ig. 1. (a) Layout of the antibody molecule. The structure consists of 4 peptide chai ragment antigen-binding regions constituted by a highly variable domain (VL) and CH2 and CH3). (b) Antibody reduction and acidification. (c) Enzymatic papain diges esignated CL or CH1, CH2, and CH3 domains, as shown in Fig. 1 20,23–28]. There are 5 classes of immunoglobulins: IgA, IgD, IgE, IgG, and gM. The major differences between these classes of immunoglob- lins are the components of their heavy chains, termed � (alpha), (delta), � (epsilon), � (gamma), and � (micro), respectively. The ight chains are designated by the Greek letters � (kappa) and � lambda) [25]. Regarding the functionalization of liposomes with mmunoglobulins, it is important to know that most studies use IgG s the main model. IgG is the predominant immunoglobulin in the lood, lymph fluid, cerebrospinal fluid, and peritoneal fluid. IgG is ubdivided into 4 subclasses: IgG1, IgG2, IgG3, and IgG4. The differ- nces between the subclasses influence their chemical properties nd the resulting biological responses [27]. Liposomes are functionalized not only with whole antibodies, ut also with isolated Fabs. The first fragmentations of antibodies esulted from enzymatic digestion, reduction, or changes in pH. As epicted in Fig. 1b, reduction and acidification can cause the sep- ration of the L and H chains. Fabs and portions the of Fc can be solated by papain enzymatic digestion (Fig. 1c). Moreover, anti- odies can be further separated using pepsin enzymatic digestion f the C-terminal half of the Fc region. This results in the formation f 2 Fab regions attached to half of an Fc fragment, termed F(ab’)2, nd the fragment that corresponds to the C-terminal region of the c portion, termed pFc’(c) (Fig. 1d) [20,25]. As shown in Fig. 1e, ther fragments include the variable domain, a single chain of the ariable domain (scFv), a variable domain disulfide-stabilized, and he single variable heavy domain [21,26]. dentical light chains (L), and 2 identical heavy chains (H). There are also 2 identical stant domain (CL), and the fragment crystalline constructed of 2 constant domains d) Enzymatic pepsin digestion. (e) Monovalent antibody fragments. 3. Strategies to conjugate monoclonal antibodies or their fragments to liposomes A variety of strategies have been reported for the preparation of long-circulating liposomes functionalized with antibodies or their fragments. Most strategies involve conjugation of the antibodies to the distal ends of PEG chains on the PEG-lipid components of the liposomes, on which the inert methoxy group at the end of the PEG chain has been replaced with chemical groups useful for antibody conjugation [15]. Different methods have been developed for this purpose, including non-covalent and covalent, which are more commonly reported, approaches. Antibody derivatization and conjugation have been thoroughly reviewed [17]. Herein, we will briefly focus on the most frequently described methods and high- light some of the characterization results. For covalent attachment, it must be taken into consideration that the antibody molecule contains functional chemical groups, such as amines and carboxylates, which are prone to modification for targeting purposes. Furthermore, the sulfhydryl group plays a key role as a targeting group, and has been extensively reported. Nonetheless, this group occurs infrequently in antibody molecules and must be generated by either the reduction of disulfide groups or through appropriate thiolation agents [17]. Thiolated antibod- ies containing sulfhydryl groups will then react with antibodies containing a chemically reactive molecule, like a lipid-containing maleimide, forming a thioether linkage (Fig. 2) [29–32]. On the other hand, sulfhydryl groups may undergo oxidation with con- sequent disulfide crosslink formation, which can be prevented by oxygen removal from buffers and the addition of ethylenediamine tetraacetic acid (EDTA) to chelate the metal ions that are able to catalyze these reactions [17]. J.O. Eloy et al. / Colloids and Surfaces B: Biointerfaces 159 (2017) 454–467 457 ydryl i t m a o e c t p h o i [ o c a e o q a r a w a c w Fig. 2. Reaction between a thiolated antibody and sulfh The thiolation of antibodies employing Traut’s reagent (2- minothiolane) is one of the most popular strategies employed o date. The cyclic group 2-imidothioester can react with pri- ary amines, thereby opening the ring structure and generating free sulfhydryl [33]. A variety of papers have reported the use f Traut’s reagent for thiolation for the further conjugation of sev- ral monoclonal antibodies, including trastuzumab [29,32,34,35], etuximab [36,37], and others [38–41], onto liposomal surfaces hrough the distal reactive maleimide terminus of the PEGylated- hospholipid maleimide group. Overall, authors have reported igh functionalization efficiencies, as shown by protein analysis f immunoliposome fractions, and the maintenance of the primary ntegrity of antibodies, as shown by polyacrylamide electrophoresis 32,35,37]. Fragments of monoclonal antibodies have been loaded nto liposomes using the same strategies. For instance, Gao and o-workers reported anti-epidermal growth factor receptor (EGFR) nd anti-HER2 tyrosine kinase receptor (also known as ErbB2, c- rbB2 or HER2/neu) antibody Fab’ conjugation to liposomes based n a thioether bond. They characterized the conjugates by the uantification of Fab’ on the surface of the liposomes and evalu- ted the fragment integrity through electrophoresis [42–44]. Other eports of Fab’ liposome functionalization involved targeting the myloid-beta (A ) complex, for which the intact binding affinity as demonstrated. Additionally, the presence of 1 thiol group per ntibody was confirmed, and electrophoresis under non-reducing onditions analysis demonstrated that intact antibody molecules ere bound to the liposomes [45]. -containing liposome for immunoliposome formation. Alternative molecules have been reported for antibody thiolation prior to covalent linkage with liposomes through maleimide-containing lipids. Many studies employed 3-(2- pyridyldithiolpropionic acid)-N-hydroxysuccinimide ester (SPDP) to achieve this purpose. Some studies employed the lipid N-4-(p-Maleimidylphenylbutyryl) dipalmitoyl- L-cY-phosphatidylethanolamine (MPB-PE) for antibody linkage [31,46,47]. Alternatively, other lipids have been used, such as N-(6-maleimidocaproyloxy)-dipalmitoyl phosphatidylethanolethanolamine and poly(ethylene glycol)- �-distearoyl phosphatidylethanolamine-maleimide [48,49]. Interestingly, Zalba et al. prepared oxaliplatin-loaded liposomes targeted against EGFR for the treatment of colorectal cancer. For this purpose, the authors compared the functionalization of liposomes with the whole cetuximab antibody or the Fab’ portion, which were both confirmed by electrophoresis. Moreover, the coupling efficiency was similar for both the monoclonal antibody and the fragment, with a linear increase between 5 and 30 �g of protein per �mol of lipid followed by a plateau in the range of 30–40 �g of protein per �mol of lipid [50]. Another popular thiolation reagent is N-succinimidyl S-acetylthioacetate (SATA), which enables antibody reaction with MPB-PE or succinimidyl 4-(N-malemidomethylcyclohexane)-1-carboxylate [51–54]. In another study, assessment of SATA modification using 10×–100× molar excess relative to the antibody concentration indicated that sulfhydryl group generation, coupling efficiency, the amount of antibody per �mol lipid, and the antibody number per liposome were higher when a higher concentration of SATA was used; 4 es B: B h o r t m f i b f ( p w c v 4 w m t h o l t r t t e l a o i v t [ g p b r c w p [ t a a I c t t m t h w s a a l A o o r e s 58 J.O. Eloy et al. / Colloids and Surfac owever, this strategy inhibited cell targeting, probably because f massive crosslinking of the NH2 groups near the antigen ecognition sites of the antibodies [30]. Some other chemical groups have shown utility for thiola- ion for further reaction with PEGylated lipids conjugated with aleimide [55–57]. Gagné et al. tested 2-mercaptoethylamine-HCl or reducing anti-HLA-DR Fab’ portions, in order to minimize the mmunogenicity associated with the Fc portion of the whole anti- ody, for the delivery of indinavir for HIV therapy. The anchor lipid or conjugation was distearoylphosphatidylethanolamine[poly- ethyleneglycol)2000] maleimide. The liposomes bearing the Fab’ ortions were 2.3-fold less immunogenic than the liposomes ith the whole IgG [58]. Using a different strategy, Shin et al. onjugated an anti-CD133 antibody to liposomes; unlike the pre- ious reports discussed herein, they reacted the antibody with -(maleimidophenyl) butyrate and conjugated it to the lipid, which as first thiolated with dithiothreitol [59]. Thiolated antibodies can also react with groups other than aleimide. For example, vinylsulfone groups at the distal PEG ermini on the surface of liposomes (such as PEG-vinylsulfone-N- ydroxy-succinimidyl ester) can react with the free thiol groups n monoclonal antibodies [60]. Rather than reacting the thio- ated antibody with maleimide-containing liposomes through a hioether bond, other studies have reported different chemical eactions. For instance, Bendas et al. functionalized liposomes hrough N-glutaryl PE as a membrane anchor for preparing he liposomes through an amide bond. For protein linkage, 1- thyl-3(3-dimethylaminopropyl)carbodiimide was incubated with iposomes, followed by gel permeation chromatography. Then, ntibody was added, followed by incubation, then separation f free antibody by gel permeation chromatography. Interest- ngly, the authors compared this protocol to functionalization ia thiolated antibody reaction with maleimide, and found that hey obtained better results with the amide reaction approach 61]. Another study reported functionalization using a carboxyl roup. Monoclonal antibodies were conjugated to N-glutaryl- hosphatidylethanolamine in the presence of octylglucoside y using N-hydroxysulfosuccinimide as a carboxyl-activating eagent [62]. Another approach involves the formation of a ovalent bond between periodate-oxidized antibody molecules ith liposomes containing hydrazide-derivatized distearoyl phos- hatidylethanolamine, through hydrazone linkage formation 63,64]. Finally, another covalent approach involves lipids con- aining the nitrile group, which does not require prior antibody ctivation or derivatization [65–67]. Additionally, covalent chemical bonds between liposomes and ntibodies can be formed with a post-insertion approach (Fig. 3). n this case, liposomes are first prepared in parallel to micelles ovalently bound to antibodies, which are then incubated together o facilitate antibody transfer to liposomes. The authors argued hat the post-insertion technique is a simple, flexible, and effective ethod for preparing targeted liposomal drugs for clinical applica- ions [68]. Using this approach, different antibodies or fragments ave been attached to liposomes, particularly through reactions ith maleimide. Examples include immunoliposomes targeted to oluble Leishmania antigens, EFGR for glioma, endoglin (CD105) nd fibroblast activation protein (FAP), and HER2 for breast cancer, mong others [34,36,69,70]. The orientation of the covalently attached antibody on the iposome surface influences its association with the tumor cells. ntibodies can be attached on conventional liposomes in a random rientation, on pegylated liposomes also in a random orientation r at the terminal end of the PEG-chain in a site-specific manner, endering the antibody molecules with their antigen-binding site xposed and protruding from the liposome. It has been previously hown that the attachment of the CC52 antibody at the terminal iointerfaces 159 (2017) 454–467 end of the PEG-chain of immunoliposomes strongly enhanced the association with CC531 cells even at a relatively low antibody den- sity [16]. Furthermore, the spacer length for antibody attachment can potentially influence the coupling efficiency. Fleiner et al., 2001 compared ethylene glycol, tetraethylene glycol, PEG 400, PEG 1000, dodecyl as thiol-reactive coupling lipids. They found that polar spacers (tetraethylene glycol) achieved a higher coupling efficiency than a nonpolar spacer with approximately the same length (dode- cyl) and the best results were obtained using coupling lipids with a long polar spacer (PEG 1000). The authors argued that a long and polar spacer (e.g., PEG) might be more suitable to present the thiol- reactive maleimide group at the liposome surface [71]. Finally, a long PEG chain has the potential to enhance the targeting effect, because the nanoparticle can better associate with the cells [72]. Finally, antibodies can be non-covalently linked to liposomes, although this strategy is less common. The reaction between biotin and neutravidin or streptavidin is particularly useful. A recent paper demonstrated that cellular uptake of immunolipo- somes is more efficient if the antibody is conjugated through the streptavidin–biotin complex instead of the maleimide group, through PEG-biotin and PEG-maleimide, respectively, for brain- targeted delivery. The authors argued that the cause is likely related to the higher availability of antibody on liposomal surfaces pro- vided by the biotin–streptavidin complex, which could be related to the length and conformation of PEG on the liposome surface [73]. 4. Monoclonal antibody- and antibody fragment-mediated targeted delivery 4.1. Vascular targeting: cardiovascular diseases and cancer Nanoparticles, such as liposomes, are naturally prone to vas- cular targeting due to the EPR effect in the leaky vasculature of cancers and inflamed areas [74]. In cancer blood vessels, there are tight cell–cell junctions with openings of up to 400–600 nm between endothelial cells, which enable passive targeting to take place [75]. Besides the leakiness of the tumor blood vessels, there are tumor-specific endothelial cell changes that can be exploited for the development of actively targeted nanoparticles [76]. In order to be useful for endothelial-targeted delivery, an endothelial determi- nant must be present on the lumen of the endothelium, making it accessible to carriers circulating in the blood stream [77]. Several endothelium-expressed molecules have been proposed as target- ing options to the vascular wall, including vascular endothelial growth factor receptors (VEGF-R) in cancer-associated angiogen- esis and leukocyte adhesion molecules in diseases associated with inflammation [78]. Other targets include platelet endothe- lial cell adhesion molecule 1 (PECAM-1), intercellular adhesion molecule 1 (ICAM-1), transmembrane glycoproteins constitutively expressed on endothelial cells, integrins (specifically �v 3 and �v 5), selectins, and the family of tumor endothelial markers (especially TM) [77]. In response to inflammation, endothelial cells express cell adhe- sion molecules for the recruitment of leukocytes to inflammatory sites. Among these receptors, the carbohydrate-binding selectins initiate and regulate the adhesion cascade [79]. Selectin-directed antibodies have been evaluated as a targeting moiety for lipo- somes, and target sensitivity was evaluated [61,80]. Promising results have been reported by Spragg and coworkers, who demon- strated that E-selectin-targeted immunoliposomes for doxorubicin (dox) delivery mediated marked cytotoxicity when incubated with activated human umbilical vein endothelial cells (HUVECs) that express E-selectin, but not when incubated with non-activated HUVECs [52]. Additionally, cellular uptake of E-selectin-targeted immunoliposomes by activated endothelial cells was attributed to J.O. Eloy et al. / Colloids and Surfaces B: Biointerfaces 159 (2017) 454–467 459 on thr t t e i t i m a s m r H L h t ( b t e t l a a c t t t i l v F d t t c t a R p r [ i c i a w t Fig. 3. Immunoliposome formati he endocytic pathway [65]. Using a different approach, administra- ion of P-selectin-targeted immunoliposomes containing vascular ndothelial growth factor (VEGF) for delivery to the affected areas n myocardial infarction resulted in increased anatomical and func- ional vessels, with a substantial improvement in cardiac function n animals [81]. Adhesion molecules, including endothelial leukocyte adhesion olecule-1 (ELAM), vascular cell adhesion molecule 1 (VCAM-1), nd ICAM-1, play an important role in the inflammatory process, uch as in atherosclerosis through the recruitment of circulatory onocytes [82,83]. For instance, anti-ICAM-1 liposomes have been eported to bind human bronchial epithelial cells (BEAS-2B) and UVECs in a specific, dose-, and time-dependent manner [51]. iposomes targeted with anti-ICAM-1 and anti-ELAM antibodies ave been developed for vascular targeting with applications for he treatment of cardiovascular diseases [84]. Echogenic liposomes responsive to ultrasound stimuli) targeted with anti-VCAM-1 anti- ody were evaluated as a tool for molecular imaging of atheroma; he authors found that pre-treatment with nitric oxide-loaded chogenic liposomes plus ultrasound improved the contrast effec- iveness [31]. Interestingly, anti-VCAM-1 antibody conjugated to iposomes was able to catalyze blood coagulation reactions, with pplications for controlling thrombogenesis [48]. Finally, VCAM-1 nd E-selectin antibodies have been used in combination for vas- ular targeting in inflammation caused by IL-1� and TNF-� [85]. Considering the important role of angiogenesis in tumor growth, argeting the endothelium could serve as an approach for cancer reatment. Thus, VCAM-1 has been reported for tumor vasculature argeting by Gosk et al., who observed that anti-VCAM-1 targeted mmunoliposome exhibited specific binding to activated endothe- ial cells and that immunoliposomes accumulated in vivo in tumor essels [67]. Additionally, other targets are relevant in this process. or example, kinase insert domain-containing receptor VEGF is pre- ominantly expressed on tumor vessels, representing a promising arget for tumor angiogenesis inhibition [38]. Endoglin, a protein of he transforming growth factor receptor complex, has potential as a ancer vasculature target; dox-loaded immunoliposomes directed o endothelial cells, using a scFV fragment against endoglin, medi- ted enhanced cytotoxicity toward endothelial cells [86]. Finally, abenhold et al. developed immunoliposomes targeted with bis- ecific single-chain antibodies against endoglin and FAP, which esulted in strong interactions with a human fibrosarcoma cell line 70]. Finally, bevacizumab is an FDA-approved, recombinant, human- zed monoclonal antibody for the treatment of metastatic colorectal ancer in combination with 5-fluorouracil (5-FU). It targets and nactivates all isoforms of VEGF, thereby inhibiting angiogenesis nd tumor growth [87]. Within this context, cationic liposomes, hich preferentially target the tumor vasculature, were func- ionalized with bevacizumab and had high cellular uptake into ough the post-insertion method. endothelial cells [88]. However, other bevacizumab immunolipo- somes have not been further developed. 4.2. Cancer targeting: HER2 HER2 contributes to the regulation of epithelial cell prolifera- tion and survival, and is a member of the HER family, which also includes the HER1 (EGFR), HER3, and HER4 receptors. HER2 is a type 1 transmembrane glycoprotein that contains 3 distinct regions: an N-terminal extracellular domain, a single �-helix transmembrane domain, and an intracellular tyrosine kinase domain. Although the HER2 receptor has no known ligand, it can heterodimerize with the other receptors, resulting in autophosphorylation of the tyro- sine residues within the cytoplasmic domain, which initiates signal transduction via the PI3K/AKT and RAS/MAPK pathways [89,90]. Overexpression of HER2, found in approximately 20%–25% of cases of breast cancer, is an adverse prognostic indicator of survival, due to the key role of HER2 in this process. Besides the more aggressive phenotype, it is associated with a greater likelihood of lymph node involvement and increased resistance to endocrine therapy [91]. The recombinant, humanized monoclonal antibody trastuzumab (Herceptin ® ) was the first anti-HER2 agent that improved survival in metastatic patients [92]. Blocking HER2 signaling is a promising strategy to inhibit tumor growth. Trastuzumab is able to induce apoptosis in breast cancer cells through antibody-dependent cellular cytotoxicity. After binding to HER2 and reducing signaling, trastuzumab increases p27Kip1 levels, promoting cell cycle arrest and apoptosis. However, not all HER2-positive patients respond to trastuzumab therapy, and HER2-overexpressing tumors can develop resistance to trastuzumab monotherapy; thus, combination therapy with other anti-cancer drugs with distinct mechanisms of action, such as paclitaxel (PTX), dox, or lapatinib, improves the therapeutic out- come. Furthermore, as an overexpressed cyto-membrane protein, HER2 is considered an attractive marker for targeted drug delivery to tumor cells [90,93–95]. Anthracyclines, particularly dox (Adriamycin ® ), have been employed for the treatment of breast cancer, both in the adju- vant and metastatic settings. However, the low therapeutic index of dox is a major drawback. Due to dox cardiotoxicity, it is not clinically combined with trastuzumab. However, PEGylated lipo- somal dox is effective for breast cancer treatment and reduces the toxicity associated with conventional dox, including myelo- suppression, alopecia, nausea, vomiting, and, most importantly, cardiac toxicity. Also, Doxil ® , the first liposome approved by the FDA, is a commercially available dox liposome for cancer treat- ment, including breast tumors [96]. Combining trastuzumab with liposomal dox has been proving clinical benefit [97,98]. Within this context, anti-HER2 immunoliposomes have been developed for dox delivery to breast cancer. Fab’ portions against the HER2 4 es B: B r i l t t c u p b [ i t t t p P m H A p [ e H n l k p f a e r R s a a t e c p i w f v l [ s s r ( g d t m a w o l [ m m [ 60 J.O. Eloy et al. / Colloids and Surfac eceptor were conjugated to dox-loaded liposomes and evaluated n HER2-overexpressing human breast cancer xenografts. Immuno- iposomes behaved better: penetrated and accumulated in the umors and increased antitumor cytotoxicity with less systemic oxicity than free dox [99]. Trastuzumab fragments, have been suc- essfully conjugated to dox-loaded liposomes, resulting in better ptake in HER2-expressing cells, long circulation with favorable harmacokinetics, and potent in vivo anticancer activity, which was etter in the immunoliposome group compared to the other groups 100]. PTX, a taxane drug that inhibits microtubule stabilization dur- ng mitosis, has been widely employed in combination with rastuzumab in the clinic, with significant improvements in the ime of disease progression, duration of response, and time to reatment failure, resulting in increased patient survival com- ared to monotherapy [93]. Yang and collaborators developed TX-loaded immunoliposomes using trastuzumab as the targeting oiety. PEGylated immunoliposomes had higher cellular uptake in ER2-positive cancer cell lines than non-functionalized liposomes. lso, the circulation time of the immunoliposome formulation was rolonged compared to the commercial PTX formulation (Taxol ® ) 34]. Subsequently, the authors observed that immunoliposomes xhibited superior efficacy compared to Taxol ® or liposomes in a ER2-positive BT-474 xenograft model. Conversely, in the HER2- egative MDA-MB-231 xenograft model, immunoliposomes and iposomes had the same tumor distribution and antitumor activity. Mammalian target of rapamycin (mTOR) is a serine/threonine inase member of the cellular phosphatidylinositol 3-kinase (PI3K) athway, which plays a critical role in cell survival. There- ore, inhibitors of mTOR, such as rapamycin (RAP), also known s sirolimus, have been shown to reduce the growth of sev- ral tumors. Additionally, mTOR inhibition may overcome the esistance of some HER2-positive tumors to trastuzumab [101]. ecently, RAP was loaded with high efficiency into immunolipo- omes functionalized with trastuzumab. The formulations were ble to cause cytotoxicity to HER2-positive SKBR3 cells [35]. In nother study, RAP was combined with PTX in trastuzumab- argeted immunoliposomes to utilize RAP and PTX synergistic ffects. The immunoliposomes showed higher uptake in SKBR3 ells and more effectively controlled tumor growth in vivo com- ared to liposomes [32]. Moreover, other drugs have been loaded into HER2-targeted mmunoliposomes. For instance, melittin, an antimicrobial peptide ith anticancer activity, was encapsulated into trastuzumab- unctionalized liposomes. Immunoliposomes decreased cancer cell iability in a dose–response manner and in correlation with their evel of HER2 expression, showing specific binding to SKBR3 cells 29]. PE38KDEL, an immunotoxin, was loaded into immunolipo- omes targeted to the HER2 receptor through Fab’ portions. Cell tudies indicated receptor-specific binding and internalization, esulting in enhanced cytotoxicity [42]. Small interference RNA siRNA)-based gene therapy was able to silence a breast cancer tar- et gene, with consequent SKBR3 cell invasion inhibition, when elivered by anti-HER2 immunoliposomes [43]. Finally, hyperthermia is a promising strategy for adjuvant cancer reatment, particularly using magnetite nanoparticles. Hyperther- ia is able to kill cancer cells directly, and also cause activation of nticancer immunity. Within this context, magnetite nanoparticles ere developed for specific tumor accumulation through targeting f the HER2 receptor, and results showed a HER2-mediated antipro- iferative effect on SKBR3 cells under an alternating magnetic field 49]. In another paper, anti-HER2 immunoliposomes containing agnetite nanoparticles caused tumor regression in the hyperther- ic group, which was sustained for 10 weeks after hyperthermia 102]. iointerfaces 159 (2017) 454–467 4.3. Cancer targeting: EGFR EGFR belongs to the ErbB family of receptor tyrosine kinases and is involved in the pathogenesis and progression of different types of cancers, including lung, head and neck, brain, colon, and breast car- cinomas [103]. The intracellular domains of ErbB receptors contain a highly-conserved tyrosine kinase domain, while the extracellu- lar domains are less conserved; the latter are bound and activated by growth factors, generating complex signal transduction path- ways that involve the Ras/ERK signaling cascade [104]. As a strategy for target inhibition, anti-EGFR monoclonal antibodies bind to the extracellular domain of the receptor, blocking ligand-induced EGFR tyrosine kinase activation. Cetuximab (Erbitux ® ), a human–murine chimeric IgG, binds to EGFR with higher affinity than natural lig- ands, such as TGF-� and EGF, causing receptor inactivation and cell apoptosis. Clinically, cetuximab has been used in combination with other chemotherapeutics, such as platinum drugs or 5-FU, for the treatment of squamous cell carcinoma of the head and neck, and colorectal cancers [105]. Another therapeutic option is the fully human IgG antibody panitumumab, which is also employed for the treatment of metastatic colorectal cancer, and associated with overcoming cetuximab-induced resistance [106]. A variety of drugs have been loaded into EGFR-targeted immunoliposomes. For cancer treatment, some small molecule drugs have been reported, including dox, sodium borocaptate, gem- citabine, 5-FU, oxaliplatin, and boron compound [18,36,37,40,50]. Immunoliposomes functionalized with cetuximab were tested in vitro in EGFR-overexpressing cell lines. Receptor-specific intracel- lular drug delivery by targeted liposomes correlated with receptor density, reaching up to 3-fold higher levels than with non-targeted liposomes [50]. Cetuximab acted synergistically with 5-FU on EGFR-positive skin squamous cell carcinoma A431 cells [37]. Also, cetuximab mediated potent uptake of boron in EGFR-positive F98 glioma cells [36]. Using an anti-EGFR Fab’, immunoliposomes enabled enhanced cytotoxicity and uptake in EGFR-overexpressing hepatocellular carcinoma cells [18]. In murine xenograft models of cancer in vivo, specific targeting to glioma tumors has been demon- strated in a system that combined imaging with drug delivery [40]. Interestingly, targeting with an anti-EGFR Fab’ or cetuximab was compared in a colorectal xenograft model, and the Fab’-targeted liposomes outperformed the cetuximab-targeted liposomes in both tumor accumulation and tumor growth control [50]. Stimuli-responsive nanoparticles are an interesting approach for triggered release upon intrinsic or extrinsic stimuli [107]. Ther- mosensitive liposomes, with the ability to trigger drug release upon temperature increase, are being researched as an alter- native to improve therapeutic efficacy. Thermodox ® , for dox delivery, is a formulation that has reached clinical trials [3]. In the field of targeted delivery, Fab’ portions of cetuximab were covalently attached to thermosensitive liposomes for dox delivery. The liposomes facilitated EGFR-mediated cellular association and temperature-dependent release, with consequent enhanced tumor cell cytotoxicity, demonstrating the potential of the nanocarrier for further evaluation [108]. RNA interference is an endogenous regulatory pathway that causes sequence-specific gene silencing, which can be used as a potent, targeted therapeutic tool applicable to a variety of diseases, including cancer [109]. Within this context, EGFR- targeted immunoliposomes have been used for siRNA delivery. Gao et al. developed an siRNA-loaded immunoliposome function- alized with anti-EGFR Fab’ for hepatocellular carcinoma therapy, and observed enhanced cellular uptake and gene silencing activ- ity compared to the untargeted liposome both in vitro and in vivo. Furthermore, higher tumor accumulation was observed for the immunoliposomes [44]. In a subsequent work, the authors further improved their therapeutic strategy by co-encapsulating ribonu- es B: B c i s o s a e s i t E o t 4 2 s s d t t c t i f I l s w d e f p r a d T c T M J.O. Eloy et al. / Colloids and Surfac leotide reductase M2 siRNA and dox in an anti-EGFR Fab’-targeted mmunoliposome. Results showed potent cytotoxicity, apopto- is, and senescence-inducing activity, with consequent reduced rthotopic tumor weight [18]. Finally, an anti-EGFR Fab’ mediated iRNA-loaded liposome delivery to the EGFR-overexpressing hep- tocellular carcinoma cells SMMC-7721 [110]. Drugs other than conventional chemotherapeutics can be xplored through EGFR targeting. For instance, celecoxib, a elective cyclooxygenase-2 (COX-2) inhibitor, was loaded into mmunoliposomes targeted with cetuximab. The rationale for his approach is that there is crosstalk between COX-2 and GFR. The outcome of this approach was higher uptake in EGFR- verexpressing cells, resulting in higher cytotoxicity using the argeted liposomes [111]. .4. Cancer targeting: other receptors Antibody-based therapy for cancer has been used over the past 0 years and is now one of the most successful and important trategies to treat patients with hematological malignancies and olid tumors [112]. The outcome of this therapy, targeted drug elivery to cancer cells, is an interesting and promising field due to he many targeted receptors overexpressed on cancer cells, in addi- ion to HER2, EGFR, and the vascular receptors, compared to normal ells. The level of expression of these receptors must be sufficient o allow delivery of anticancer drugs in sufficient amounts, and an deal target receptor will generally be expressed on the apical sur- ace of a cancer cell and not within its cytoplasm or nucleus [113]. n Table 1, we summarized a series of studies reporting targeted iposomes using antibodies for a variety of cancer receptors. Most tudies reported enhanced specific cell binding and internalization ith immunoliposomes compared to untargeted liposomes. Among others, potential targets for cancer treatment and iagnosis are the folate and transferrin receptors, which are over- xpressed in a variety of cancer types. Folate receptor is necessary or folate uptake and DNA synthesis and is commonly overex- ressed in ovarian cancer [114]. Farletuzumab is an anti-folate eceptor monoclonal antibody, which has been clinically evalu- ted in combination with carboplatin and pegylated liposomal oxorubicin, and a safe profile has been demonstrated [115]. herefore, a logical approach for ovarian targeted delivery of anti- ancer drugs would the synthesis of liposomes functionalized with able 1 onoclonal antibody and antibody fragments-mediated targeted delivery to cancer. Antibody Cancer type Drug Anti- T-24 Bladder carcinoma Pheopho Mab CC52 Colon cancer 5-Fluoro Mab 2C5 Lung carcinoma and mammary adenocarcinoma Doxorub Anti-B-cell lymphoma Mab LL2 Lymphoma Doxorub Anti-CD32 and anti-CD2 antibodies Leukemia Oligonu Anti-CD19 Myeloma Doxorub Mab AF-20 Hepatocarcinoma Fluoresc Anti-ovarian carcinoma Fab’ Ovarian carcinoma Not emp Anti-CD133 Mab Glioblastoma Gemcita anti-FAP (fibroblast activator protein)’ scFv antibody Metastatic fibrosarcoma Fluoresc Anti-CD19, anti-CD20 and anti-CD37 Mab Leukemia Fluoresc Anti-transferrin scFv antibody fragment Several cancer cell lines and prostate cancer (in vivo) Plasmid iointerfaces 159 (2017) 454–467 461 farletuzumab. Transferrin receptor is involved with iron cellular uptake and is overexpressed in many cancer types, usually cor- related with higher level of malignancy [116]. For example, an anti-transferrin receptor monoclonal antibody, OX26, was con- jugated onto plasmid DNA-loaded liposomes as an approach to enhance the transport to blood brain barrier (BBB), where trans- ferrin receptor is overexpressed. A further functionalization was done with chlorotoxin for specific binding to glioma cells and for facilitated binding to matrix metalloproteinases (MMP-2). Results evidenced higher in vitro transfection and, in vivo, immunolipo- somes reduced tumor volume and increased survival rate [117]. An interesting cancer target is represented by MMPs, a family of zinc-dependent endopeptidases, which are a driving factor for can- cer progression and patient prognosis. MMPs are present in nearly all human cancers and are involved in remodeling the extracellular matrix in tumor microenvironments [118]. For this reason, MMP2- responsive liposomes have been constructed for tumor targeting. Liposomes were conjugated with TAT, a cell penetrating peptide, to enhance the intracellular uptake, and the formulation was further functionalized with a tumor cell-specific anti-nucleosome mono- clonal antibody (mAb 2C5). The liposomal composition included an MMP2-sensitive bond between PEG and lipids, which undergoes cleavage in the tumor by the highly expressed extracellular MMP2, causing the removal of PEG chains for improved uptake. When PEG chain is cleaved, TAT peptide is then exposed, resulting in TAT peptide-mediated endocytosis, causing increased cell uptake. This strategy allows the prevention of the nonspecific intracellular uptake on the way to the tumor by steric shielding of TAT peptide with the long-chain of PEG [119]. In another work, Fab’ portions of antibody against MT1-MMP were conjugated to dox-encapsulating liposomes, resulting in significant tumor growth suppression [56]. 4.5. Infectious disease targeting Immunoliposomes may be employed for antiviral therapy, with advantages over traditional antiviral therapies. Viruses display a varied repertoire of proteins for that can be targeted with antibod- ies. Furthermore, antibodies against viral proteins are more specific than those against cancer cell receptors because cancerous and nor- mal cells share most receptor proteins and only differ in expression levels. Importantly, immunoliposomes may be able to target not Main findings Reference rbide a Enhanced uptake and cytotoxicity [46] deoxyuridine Cell uptake by endocytic process [16,53] icin Recognition, binding and cytotoxicity to cancer cells [120] icin Increased cellular association and internalization and better cytotoxicity than untargeted liposome [60] cleotides Improved cellular uptake [121] icin Receptor- mediated internalization and selective cytotoxicity [64] ent dye Improved specific cell interaction [122] loyed Binding to cancer cells [123] bine and bevacizumab Increased cytotoxicity and antitumor efficacy in xenograft model [59] ent dye Image-guided detection of the spontaneous metastases and accumulation in mice models [124] ent dye Dual-ligand immunoliposome enable a better strategy of personalized treatment of B-cell malignancies [125] DNA Enhanced in vitro and in vivo transfection. In vivo gene delivery and expression [126] 4 es B: B o t v C n A n a p i i l i a c ( c F i a v s f e t v c t t g b S i t [ l i g F i t i l E c t N t t t w f n e p a c 4 p cytoplasmic inclusions known as Lewy bodies, rich in aggregated �- synuclein. Cellular uptake of the targeted immunoliposomes in the cultured brain endothelial cell line hCMEC/D3 was twice as efficient as that of untargeted liposomes [143]. 62 J.O. Eloy et al. / Colloids and Surfac nly the virus but also infected cells presenting viral proteins on heir surfaces [127]. Some papers have reported immunoliposomes targeted to iruses. For HIV treatment, HLA could serve as a target because D4+ T cells express substantial levels of the HLA-DR determi- ant of the major histocompatibility complex class II molecules. nti-HLA immunoliposomes loaded with the antiviral drug indi- avir were successfully delivered to lymphoid tissues and were s effective as the free drug in vitro. Also, liposomes bearing Fab’ ortions were 2.3-fold less immunogenic than liposomes bear- ng the entire IgG [58]. In another approach, phosphatidylserine mmunoliposomes targeted with antibodies that bind HIV-1 virus- ike particles were initially protected from macrophage uptake, n order to provide enough time to circulate through the body nd achieve maximum virus binding [128]. Moreover, anti-CD4 onjugated immunoliposomes containing 2 antiretroviral drugs nevirapine and saquinavir) inhibited viral proliferation at a lower oncentration than free drugs [41]. Other viruses have been targeted with immunoliposomes. or instance, for targeting to influenza A (H5N1) viral infection, mmunoliposomes were functionalized with a humanized scFv ntibody against the hemagglutinin (HA) of H5N1, for influenza irus nucleoprotein siRNA delivery. The authors demonstrated pecific binding to HA-expressing Sf9 cells, enhanced siRNA trans- ection efficiency, and a pronounced silencing effect [129]. Falco t al. encapsulated melittin, a pore-forming lytic amphiphilic pep- ide with antiviral activity, in immunoliposomes targeted against iral hemorrhagic septicemia rhabdovirus (VHSV). The authors laimed that immunoliposomes were able to inhibit VHSV infec- ivity by 95.2% via direct inactivation of the virus [127]. Active targeting with pathogen-binding ligands conjugated to he surfaces of nanoparticles may be an effective strategy to tar- et bacteria [130]. Within this context, immunoliposomes have een developed as an approach for targeting the oral bacterium treptococcus oralis; however, the results were not very promis- ng, considering that the positively charged liposomes adsorbed o the bacteria with greater affinities than the immunoliposomes 131]. Later, the authors developed another immunoliposome oaded with the bactericides chlorhexidine and triclosan for S. oralis mmobilized in polystyrene. The immunoliposomes enhanced rowth inhibition of S. oralis, compared to free bactericide [132]. or detection purposes, Staphylococcus enterotoxin B fluorescent mmunoliposomes were developed for immunochromatographic esting [54]. For protozoa targeting, some approaches have been described, ncluding the development of chloroquine- and primaquine-loaded iposomes functionalized with antibodies against Plasmodium. xciting in vivo results in mice showed that immunoliposomes leared the pathogen below detectable levels [30]. More recently, he same research group used polyclonal antibodies against the TS-DBL1� N-terminal domain of a Plasmodium falciparum pro- ein for targeting lumefantrine-containing liposomes for malaria reatment, with marked inhibition of parasite growth [133]. Addi- ionally, targeting against P. falciparum-infected red blood cells as achieved with immunoliposomes carrying chloroquine and osmidomycin for improved efficacy. Importantly, increasing the umber of antibodies on the liposome surface correspondingly nhanced antiparasitary performance [134]. Finally, immunoli- osomes have been reported as vaccines targeted to Leishmania ntigens, which induced stronger cell-mediated immunity in mice ompared to untargeted liposomes [69]. .6. Autoimmune and degenerative disease targeting Although targeting autoimmune and degenerative diseases is a romising field, it has not been thoroughly explored with immuno- iointerfaces 159 (2017) 454–467 liposomes. An important application for them is the treatment of rheumatoid arthritis, an autoimmune disease, which is the leading cause of disability. In this disease, degradation of the car- tilage matrix is followed by the loss of proteoglycans and other proteins from the surface, which exposes type II collagen (CII) fibrils and makes them accessible to CII antibodies. The most common treatments for rheumatoid arthritis involve nonsteroidal anti-inflammatory drugs, corticosteroids, and antirheumatic drugs and biological agents, including antibodies [135]. For the purpose of arthritis treatment and diagnosis, near infrared immunolipo- somes conjugated with CII antibody were developed. Researchers demonstrated selective binding and quantified cartilage degrada- tion in vivo in guinea pigs [136]. Lupus nephritis is a serious consequence of systemic lupus ery- thematous, an autoimmune disease that also affects other organs, including the skin, pericardium, lungs, and nervous system. Lupus nephritis may lead to potentially fatal renal and cardiovascular damage [137]. Within this context, anti-alpha integrin immuno- liposomes (anti-glomerular mesangial cells) were loaded with a fluorescent dye and specifically delivered in vivo, predominantly to glomeruli, with little nonspecific uptake by CD11b+ cells [138]. Targeting the blood–brain barrier is an interesting approach for the treatment of Alzheimer’s disease, a neurodegenerative brain pathology that causes a decline in cognitive abilities. Loureiro and coworkers developed dual ligand immunoliposomes for drug delivery to the brain and demonstrated cellular uptake in brain capillary cells, as well as the ability to cross the blood barrier in vivo [73]. The accumulation of extracellular A is a determi- nant for the development of Alzheimer’s disease. Thus, a logical approach for its treatment would be the use of anti-A , which has been attempted, through the development of anti-A immunoli- posomes able to capture A . The formulation was able to reduce circulating and brain levels of A in aged animals. Interestingly, the therapeutic efficacy of the immunoliposome treatment was superior to free monoclonal antibody administration [45]. More- over, immunoliposomes have already been employed to detect A using time-of-flight secondary ion mass spectrometry and were demonstrated to specifically bind A , providing information on lipid–protein interactions [139]. It is noteworthy that anti-A immunoliposomes have been shown to deposit in post-mortem A brain samples, confirming the potential of the immunoliposomal strategy for Alzheimer’s disease therapy and diagnosis [140]. Finally, another approach for delivery to the brain is the use of liposomes modified with transferrin antibodies; given that brain capillary endothelial cells express transferrin receptors, this strat- egy is potentially applicable to the treatment of brain degenerative diseases [141]. More recently, targeting with the anti-transferrin OX26 monoclonal antibody was combined with lactoferrin func- tionalization for the delivery of the selective NK3 receptor agonist senktide, which is typically unable to cross the blood–brain bar- rier, resulting in higher brain levels of senktide [142]. Furthermore, liposomes loaded with epigallocatechin-3-gallate, a natural antiox- idant, have been functionalized with the anti-transferrin receptor antibody OX26 and anti-�-synuclein, as a strategy to cross the blood–brain barrier. The purpose was to treat Parkinson’s disease, a motor and cognitive neurodegenerative disorder characterized by impaired dopamine production and the presence of neuronal es B: B 5 r a g m e p t e d r s t d a b l a B i a s i i a l c m T r a t e t ( g t a A v a a m l n 6 o t c a t o f p l c u J.O. Eloy et al. / Colloids and Surfac . Immunoliposomes: stimulation of endogenous immune esponse Cytotoxic T lymphocytes CD8+ play a major role in protection gainst intracellular infection. In this process, presentation of anti- ens requires antigen presenting cells (APCs) and association with ajor histocompatibility complex (MHC) class I. There are sev- ral strategies to induce dendritic cells (DCs) to present antigen eptides. Vaccine delivery systems, such as liposomes, promote he uptake of loaded antigens into APCs, enhancing the gen- ration of immune response. Moreover, immunoliposomes have emonstrated increased ability for immune system stimulation, epresenting a promising strategy to target DCs, leading to efficient timulation of CD8+ as well as CD4+ T cells [144,145]. Therefore, argeting DCs is considered a very promising approach for the evelopment of vaccines [146]. Noteworthy, the direct delivery of ntigens to DCs via antibodies has been recently reviewed [147]. For instance, immunoliposomes have been loaded with Solu- le Leishmania Antigens (SLA) for delivery to a murine model of eishmaniasis. Results showed that liposomes might be effective djuvant systems to induce protection against L. major challenge in ALB/c mice, but stronger cell mediated immune responses were nduced when immunoliposomes were utilized, owing to the inter- ction between IgG and Fc-gamma receptors in DCs [69]. Another trategy for DC targeting involves the C-type lectin receptors, ncluding the DEC-205 target. To this aim, anti-human DEC-205 mmunoliposomes showed increased uptake by DCs and were vailable for antigen processing [148]. Another very interesting approach for immune system stimu- ation involves the use of synthetic single strand oligodeoxynu- leotides (ODNs) containing unmethylated cytosine-guanine (CpG) otifs which mimic the conserved microbial products and bind oll-like receptor 9 (TLR9), potentiating both humoral and cellular esponses [149]. This strategy can be used for effective induction of nti-tumor immunity. Nanoparticles for enhancing the immunos- imulatory effect of CpG-ODN were recently reviewed [150]. For xample, it has been shown that p5 HER-2/neu derived pep- ides encapsulated in 1,2-dioleoyl-3-trimethylammonium propane DOTAP) cationic liposomes co-administered with CpG-ODN reatly enhanced the cytotoxic T lymphocytes response, causing umor progression inhibition. The antitumor vaccine property was ttributed to induction of both CD8+ and CD4+ responses [151]. dditionally, systemic targeting of CpG-ODN to the tumor microen- ironment has been achieved through chemical conjugation with n anti-HER-2/neu monoclonal antibody. The conjugate retained its bility to bind HER-2/neu+ tumors, activate DCs and induce antitu- or responses [152]. Finally, anti-DCs immunoliposomes would ikely be promising CpG-OND vaccines, however this approach eeds to be investigated. . Immunoliposomes: clinical development After the introduction of liposomal dox, Doxil ® , on the market ver 20 years ago, a series of liposomes are now available for the reatment of a variety of diseases, including breast and ovarian can- er, Kaposiı́s sarcoma, acute lymphoblastic leukemia, meningitis, nd fungal infections, such as aspergillosis, as well as for anes- hesia, Furthermore, some clinical trials of liposomes are ongoing r recently concluded [96]. A thorough review on nanomedicines or cancer treatment was recently published [153]. It included the romising formulation Thermodox ® , a dox-loaded thermosensitive iposome, whose phase 3 trial was recently completed for hepato- ellular carcinoma [154]. Furthermore, the variety of reports of successful preclinical se of immunoliposomes, addressed in this paper, led to some iointerfaces 159 (2017) 454–467 463 clinical trials. Currently, a phase 2 clinical trial for an anti-EGFR immunoliposome loaded with dox is recruiting patients with EGFR-positive, triple-negative breast cancer [155]. In addition, a phase 1 trial for an anti-HER2 liposome functionalized with scFv antibody fragments for the delivery of dox has concluded, which revealed that, as a monotherapy or in combination with trastuzumab, the immunoliposome could be effective for man- aging previously treated, HER2-positive breast cancer. A phase 2 trial (HERMIONE) aimed to evaluate the efficacy and safety of the formulation plus trastuzumab in patients with refractory HER2- positive, advanced/metastatic breast cancer [156]. However, the clinical trial was stopped due to the absence of clinical benefit evidence. Additionally, a phase 1 trial is recruiting volunteers to evaluate a liposome for targeted delivery of plasmid DNA for gene therapy to solid tumors via the transferrin receptor, using scFv anti- body fragments [157]. Finally, a melanoma vaccine was evaluated in a phase 1 clinical trial. The formulation Lipovaxin-MM consists of a liposome loaded with melanoma antigens and IFN�, targeted with a single domain antibody fragment (VH) against a dendritic cell-specific intracellular adhesion molecule [158]. Despite the success of liposomes for drug delivery and the recent exciting development of monoclonal antibodies for func- tionalization, no immunoliposome is yet commercially available. However, antibody–drug conjugates (ADCs), which consist of a cytotoxic agent chemically linked to an antibody, therefore com- bining the target selectivity of antibodies with the potency of cytotoxic agents, are available. Brentuximab vedotin was approved and is commercially available for the treatment of Hodgkin’s lymphoma and T-DM1, a trastuzumab emtansine conjugate, was approved for breast cancer treatment. Besides the approval of these ADCs, encouraging clinical responses with safety profile have been observed for other ADCs, which have been previously reviewed [159]. On the other hand, in comparison to ADCs, immunoli- posomes have progressed slower to the clinic, which could be attributed to several factors, including the high development cost associated with their development and the lack of successful incor- poration of a variety of anticancer drugs [19]. Challenges in bringing the bench to bedside in immunolipo- some development involve successful industrial production, which requires a multi-step preparation procedure involving antibody production, coupling, liposome formulation and drug loading, posing challenges for both manufacturing and analytics [160]. For this purpose, Wicki et al., 2015 have developed a large- scale, GMP-compliant production process of anti-EGFR targeted immunoliposomes. Several criteria were considered, such as sta- bility, sterility, pH, drug concentration, endotoxin concentration, leakage, particle size and uptake. The authors claimed that their process was robust, reliable, reproducible and thus suitable for the production of anti-EGFR-targeted nanocarriers in a quantity and quality necessary for clinical trials, which revealed the pharma- cokinetics profile and stability of the nanocarrier, in vivo [161]. 7. Conclusion and future perspective The clinical success of liposomes combined with the develop- ment of monoclonal antibody-based therapies led to the design of immunoliposomes to target diseases while reducing side effects and increasing efficacy. The rationale behind this approach is that targeting ligands may offer the advantage of improved cel- lular uptake once the nanocarrier arrives at the target, although ligand-targeted nanomedicines are subjected to the same physio- logical localization as ligand-lacking nanomedicines and, thus, have comparable biodistribution. As described herein, the development of chemical conjugation strategies, particularly through reactions of the sulfhydryl groups of thiolated antibodies and maleimide- 4 es B: B c i b f f h f t a c s H o l r r t t t a a A d # R 64 J.O. Eloy et al. / Colloids and Surfac ontaining liposomes, has enabled a rapid advancement in the mmunoliposome field. Furthermore, progress in monoclonal anti- ody engineering allowed the development of less immunogenic ragments, such as the scFv portion, commonly used for liposome unctionalization. Overall, the major focus of immunoliposomes as been on the treatment of many types of cancer, although these ormulations have shown potential for the treatment of inflamma- ory, infectious (e.g. HIV and malaria), autoimmune (e.g. arthritis), nd degenerative diseases (e.g. Alzheimer’s disease). Regarding ancer treatment, several targets could be explored, but a con- iderable number of studies have directed their efforts toward ER2 for breast cancer treatment and EGFR for different types f solid tumors. Overall, studies have reported enhanced cellu- ar uptake, resulting in increased cytotoxicity. Moreover, many esearch groups have conducted in vivo studies, with exciting esults obtained using cancer xenograft models. Consequently, hese promising results have led to clinical trials for the evalua- ion of safety and efficacy, yet no immunoliposome has reached he market. However, immunoliposomes are expected to become reality in medicine as soon as more studies reach clinical trials, nd after some issues of industrial scale-up are resolved. cknowledgement The authors would like to thank Fundaç ão do Amparo à Pesquisa o Estado de São Paulo for financial support (grants # 16/02723-8, 16/22042-5 and # 17/05930-7). eferences [1] G.T. Noble, J.F. Stefanick, J.D. Ashley, T. Kiziltepe, B. Bilgicer, Ligand-targeted liposome design: challenges and fundamental considerations, Trends Biotechnol. 32 (2014) 32–45, http://dx.doi.org/10.1016/j.tibtech.2013.09. 007. [2] M. Cheraghi, B. Negahdari, H. Daraee, A. Eatemadi, Heart targeted nanoliposomal/nanoparticles drug delivery: an updated review, Biomed. Pharmacother. 86 (2017) 316–323, http://dx.doi.org/10.1016/j.biopha.2016. 12.009. [3] J.O. Eloy, M.C. Souza, R. Petrilli, J.P.A. Barcellos, R.J. Lee, J.M. Marchetti, Liposomes as carriers of hydrophilic small molecule drugs: strategies to enhance encapsulation and delivery, Colloids Surf. B: Biointerfaces 123 (2014) 345–363, http://dx.doi.org/10.1016/j.colsurfb.2014.09.029. [4] L. Sercombe, T. Veerati, F. Moheimani, S.Y. Wu, A.K. Sood, S. Hua, Advances and challenges of liposome assisted drug delivery, Front. Pharmacol. 6 (2015) 1–13, http://dx.doi.org/10.3389/fphar.2015.00286. [5] R. Gharib, H. Greige-Gerges, S. Fourmentin, C. Charcosset, L. Auezova, Liposomes incorporating cyclodextrin–drug inclusion complexes: current state of knowledge, Carbohydr. Polym. 129 (2015) 175–186, http://dx.doi. org/10.1016/j.carbpol.2015.04.048. [6] S. Clerc, Y. Barenholz, Loading of amphipathic weak acids into liposomes in response to transmembrane calcium acetate gradients, BBA – Biomembranes 1240 (1995) 257–265, http://dx.doi.org/10.1016/0005- 2736(95)00214-6. [7] Y. Barenholz, (Chezy) Doxil® – the first FDA approved nano-drug: lessons learned, J. Control. Release 160 (2012) 117–134. [8] Y. Dou, K. Hynynen, C. Allen, To heat or not to heat: challenges with clinical translation of thermosen- sitive liposomes, J. Control. Release 249 (2017) 63–73, http://dx.doi.org/10.1016/j.jconrel.2017.01.025. [9] Y. Matsumura, H. Maeda, A new concept for macromolecular therapeutics in cnacer chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor agents Smancs, Cancer Res. (1986) 6387–6392. [10] H. Maeda, J. Wu, T. Sawa, Y. Matsumura, K. Hori, Tumor vascular permeability and the EPR effect in macromolecular therapeutics: a review, J. Control. Release 65 (2000) 271–284. [11] V.P. Torchilin, Multifunctional nanocarriers, Adv. Drug Deliv. Rev. 64 (2012) 302–315. [12] R. Petrilli, J.O. Eloy, J.M. Marchetti, R.F.V. Lopez, R.J. Lee, Targeted lipid nanoparticles for antisense oligonucleotide delivery, Curr. Pharm. Biotechnol. 15 (2014) 847–855. [13] E. Paszko, M.O. Senge, Immunoliposomes, Curr. Med. Chem. 19 (2012) 5239–5277. [14] P.J. Kennedy, C. Oliveira, P.L. Granja, B. Sarmento, Antibodies and associates: partners in targeted drug delivery, Pharmacol. Ther. (2017) http://doi.org/ 10.1016/j.pharmthera.2017.03.004. [15] S. Zalipsky, C.B. Hansen, D.E.L. Menezes, T.M. Allen, Polyethylene glycol-grafted immunoliposomes, J. Control. Release 39 (1996) 153–161, http://dx.doi.org/10.1016/0168-3659(95)00149-2. iointerfaces 159 (2017) 454–467 [16] G.A. Koning, J.A.A.M. Kamps, G.L. Scherphof, Efficient intracellular delivery of 5-fluorodeoxyuridine into colon cancer cells by targeted immunoliposomes, Cancer Detect. Prev. 26 (2002) 299–307, http://dx.doi. org/10.1016/S0361-090X(02)00087-9. [17] A.S. Manjappa, K.R. Chaudhari, M.P. Venkataraju, P. Dantuluri, B. Nanda, C. Sidda, K.K. Sawant, R.S.R. Murthy, Antibody derivatization and conjugation strategies. application in preparation of stealth immunoliposome to target chemotherapeutics to tumor, J. Control. Release 150 (2011) 2–22, http://dx. doi.org/10.1016/j.jconrel.2010.11.002. [18] J. Gao, H. Chen, Y. Yu, J. Song, H. Song, X. Su, W. Li, X. Tong, W. Qian, H. Wang, J. Dai, Y. Guo, Biomaterials inhibition of hepatocellular carcinoma growth using immunoliposomes for co-delivery of adriamycin and ribonucleotide reductase M2 siRNA, Biomaterials 34 (2013) 10084–10098, http://dx.doi. org/10.1016/j.biomaterials.2013.08.088. [19] P. Sapra, B. Shor, Monoclonal antibody-based therapies in cancer: advances and challenges, Pharmacol. Ther. 138 (2013) 452–469, http://dx.doi.org/10. 1016/j.pharmthera.2013.03.004. [20] P.J. Delves, S.J. Martin, D.R. Burton, I.M. Roitt, Roitt’s Essential Immunology, 11th ed., Blackwell, Singapore, 2006, pp. 37–52, Chapter 3. [21] M. Little, S. Kipriyanov, F. Le Gall, G. Moldenhauer, Of mice and men: hybridoma and recombinant antibodies, Immunol. Today 21 (2000) 364–370, http://dx.doi.org/10.1016/S0167-5699(00)01668-6. [22] J.K.H. Liu, The history of monoclonal antibody development – progress, remaining challenges and future innovations, Ann. Med. Surg. 3 (2014) 113–116, http://dx.doi.org/10.1016/j.amsu.2014.09.001. [23] N.A.P.S. Buss, S.J. Henderson, M. McFarlane, J.M. Shenton, L. De Haan, Monoclonal antibody therapeutics: history and future, Curr. Opin. Pharmacol. 12 (5) (2012) 615–622, http://dx.doi.org/10.1016/j.coph.2012. 08.001. [24] B. Lu, M.R. Smyth, R. O’Kennedy, Oriented immobilization of antibodies and its applications in immunoassays and immunosensors, Analyst 121 (1996) 29R–32R, http://dx.doi.org/10.1039/an996210029r. [25] V. Calich, C. Vaz, Imunologia, Revinter Ltda, Rio de Janeiro, 2001, pp. 57–64, Chapter 4. [26] A.P. Chapman, PEGylated antibodies and antibody fragments for improved therapy: a review, Adv. Drug Deliv. Rev. 54 (4) (2002) 531–545 https://doi. org/10.1016/S0169-409X(02)00026-1. [27] R. Coico, G. Sunshine, E. Benjamini, Immunology: A Short Course, 5th ed., John Wiley and Sons, Hoboken, 2003, pp. 39–57, Chapter 4. [28] V. Joosten, C. Lokman, C.A.M.J.J. van den Hondel, P.J. Punt, The production of antibody fragments and antibody fusion proteins by yeasts and filamentous fungi, Microb. Cell Fact. 2 (2003), http://dx.doi.org/10.1186/1475-2859-2-1. [29] E. Barrajón-Catalán, M.P. Menéndez-Gutiérrez, A. Falco, A. Carrato, M. Saceda, V. Micol, Selective death of human breast cancer cells by lytic immunoliposomes: correlation with their HER2 expression level, Cancer Lett. 290 (2010) 192–203, http://dx.doi.org/10.1016/j.canlet.2009.09.010. [30] E. Moles, P. Urbán, M.B. Jiménez-Díaz, S.V. Morilla, I.A. Barturen, M.A. Busquets, X.F. Busquets, Immunoliposome-mediated drug delivery to plasmodium-infected and non-infected red blood cells as a dual therapeutic/prophylactic antimalarial strategy, J. Controll. Release 210 (2015) 217–229, http://dx.doi.org/10.1016/j.jconrel.2015.05.284. [31] H. Kim, P.H. Kee, M.R. Moody, M.E. Klegerman, D. Vela, S.L. Huang, D.D. McPherson, S.T. Laing, Nitric-oxide enhanced molecular imaging of atheroma using vascular cellular adhesion molecule 1-targeted echogenic immunoliposomes, Ultrassound, Med. Biol. 41 (2015) 1701–1710, http://dx. doi.org/10.1016/j.ultrasmedbio.2015.02.002. [32] J.O. Eloy, R. Petrilli, D.L. Chesca, F.P. Saggioro, R.J. Lee, J.M. Marchetti, Anti-HER2 immunoliposomes for co-delivery of paclitaxel and rapamycin for breast cancer therapy, Eur. J. Pharm. Biopharm. 115 (2017) 159–167, http://linkinghub.elsevier.com/retrieve/pii/S093964111630786X%0A http:// www.ncbi.nlm.nih.gov/pubmed/28257810. [33] R.R. Traut, A. Bollen, T.T. Sun, J.W.B. Hershey, J. Sundberg, L.R. Pierce, Methyl 4-mercaptobutyrimidate as a cleavable cross-linking reagent and its application to the escherichia coli 30S ribosome, Biochemistry 12 (1973) 3266–3273. [34] T. Yang, M.K. Choi, F.D. Cui, J. Sun, Preparation and evaluation of paclitaxel-loaded PEGylated immunoliposome, J. Control. Release 120 (2007) 169–177, http://dx.doi.org/10.1016/j.jconrel.2007.05.011. [35] J.O. Eloy, R. Petrilli, R.W. Brueggemeier, J.M. Maldonado, R.J. Lee, Rapamycin-loaded immunoliposomes functionalized with trastuzumab: a strategy to enhance cytotoxicity to HER2-positive breast cancer cells, Anticancer Agents Med. Chem. 1 (2016) 1–9. [36] X. Pan, G. Wu, W. Yang, R.F. Barth, W. Tjarks, R.J. Lee, Synthesis of cetuximab-immunoliposomes via a cholesterol- based membrane anchor for targeted delivery of a neutron capture therapy (NCT) agent to glioma cells, Bioconjug. Chem. 18 (2007) 101–108, http://dx.doi.org/10.1021/bc060174r. [37] R. Petrilli, J.O. Eloy, R. Lopez, R. Lee, Cetuximab immunoliposomes enhances delivery of 5-FU to skin squamous carcinoma cells, Anticancer Agents Med. Chem. 16 (2016), http://dx.doi.org/10.2174/1871520616666160526110913, 1–1 http://www.eurekaselect.com/openurl/content.php?genre=article. [38] P. Benzinger, G.M. Baron, P. Reusch, G. Siemeister, J.T. Kley, D. Marmé, C. Unger, U. Massing, Targeting of endothelial KDR receptors with 3G2 immunoliposomes in vitro, Biochimica et Biophysica Acta – Biomembranes 1466 (2000) 71–78, http://dx.doi.org/10.1016/S0005-2736(00)00172-3. dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.tibtech.2013.09.007 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.biopha.2016.12.009 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.1016/j.colsurfb.2014.09.029 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.3389/fphar.2015.00286 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/j.carbpol.2015.04.048 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 dx.doi.org/10.1016/0005-2736(95)00214-6 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0035 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 dx.doi.org/10.1016/j.jconrel.2017.01.025 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0045 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0050 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0055 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0060 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0065 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 http://doi.org/10.1016/j.pharmthera.2017.03.004 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/0168-3659(95)00149-2 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/S0361-090X(02)00087-9 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.jconrel.2010.11.002 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.biomaterials.2013.08.088 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 dx.doi.org/10.1016/j.pharmthera.2013.03.004 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0100 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/S0167-5699(00)01668-6 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.amsu.2014.09.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1016/j.coph.2012.08.001 dx.doi.org/10.1039/an996210029r dx.doi.org/10.1039/an996210029r dx.doi.org/10.1039/an996210029r dx.doi.org/10.1039/an996210029r dx.doi.org/10.1039/an996210029r dx.doi.org/10.1039/an996210029r dx.doi.org/10.1039/an996210029r http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0125 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 https://doi.org/10.1016/S0169-409X(02)00026-1 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 http://refhub.elsevier.com/S0927-7765(17)30508-8/sbref0135 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1186/1475-2859-2-1 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.canlet.2009.09.010 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.jconrel.2015.05.284 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 dx.doi.org/10.1016/j.ultrasmedbio.2015.02.002 http://linkinghub.elsevier.com/retrieve/pii/S093964111630786X%0A http://linkinghub.elsevier.com/retrieve/pii/S093964111630786X%0A http://linkinghub.elsevier.com/retrieve/pii/S093964111630786X%0A http://linkinghub.elsevier.com/retrieve/pii/S093964111630786X%0A http://linkinghub.elsevier.com/retrieve/pii/S09396